Social Emotions

Don't regret regret - cool new talk by Kathryn Schulz

2011-12-07T16:17:00.001Z

Kathryn Schulz gives an excellent and personal account of what it means to experience regret. In this recently added TED talk, she outlines the main aspects of thought that make up regret and why it can be such a strong feeling using her, admittedly, ultimately minor, regret of getting a tattoo as an example. The important aspect of her talk is that regret in itself is not something that must be avoided - that negative feeling associated is a strong sign that whatever happend (or didn't happen) was genuinely important to us. From this we can use strategies such as reappraising the situation (like Kathryn does in regards to her tattoo) in order to better understand ourselves and how we would like to be in the future.

Moodscope 2: reflections and thoughts

2011-11-29T13:48:00.010Z

One of the more frequently visited posts on this site is the semi-review of Moodscope, written in June last year, and since then, I have received requests to write an update on how I've found it to use. Moodscope is an online tool designed to track the ups and downs of your moods using a simple positive to negative scale based around answering items from the PANAS (positive affect negative affect schedule). These results can be emailed to your friends or other important people in your life so that they are kept in the loop and, if necessary, can respond accordingly. I'll cover how I found using it first and then whether it worked for me.

One of the main comments I had last time is the theory behind the development of the tool. It has been designed based on the understanding of the Hawthorne effect that to simply measure an aspect of psychology is also to improve it. Last time, I considered this to be an incorrect application of the effect and would still stand by this. That isn't to say that increasing self-awareness of our mood cannot also lead to improvements in how we feel - evidence for and against this concept were given last time. Also, in my own research into mood changes over time, I have had participants tell me, after the study closed, that the diaries they kept helped raise awareness of their moods and also how positive they felt. Although, it should be noted that that stray observation definitely does not count as evidence for mood boosts derived from measuring how we feel.

Moodscope has a novel means of actually recording your mood instead of the traditional make-a-mark-on-a-linear-scale approach, meaning that you have to make a series of clicks in order to manipulate cards until the correct score is showing to match how you feel. Despite the general rule for psychological measures that they should be as quick and simple to use as possible (e.g. requiring as few decisions and mouse clicks as possible), the approach employed at Moodscope works well. However, despite the interesting and engaging means of presenting the measures (in a bid to keep people paying attention and refrain from answering blithely), I leant the process quickly and can near automatically make the appropriate clicks to give the answer I want. I imagine that this would counteract the benefits of an unusual or interesting means of giving answers and so things are back to square one.

In terms of using positive and negative measures, it seems a bit of a waste to then only use a single scale from feeling positive to negative. Also, this scale does seem a bit erratic at times although in general it would approximate any overall rating I would make. There's a base line score built into the black-box calculations for establishing mood, in that if you 'game' the results so that every rating given is the minimum of feeling "0: not at all" you get a score of 20 from the 0 to 100 scale where 50 would be neutral and 100 be most positive mood. This score of 20 is also seen if each rating give is the maximum of feeling "3: extremely", even though the subjective state of an individual answering in this manner is going to be greatly different to that of someone who answers feeling "0: not at all" for every question. Separate positive and negative scales would be a better application of the data that the site collects, while maintaining much of the site's aims and applications.

Each day (more or less), I received an email circular from the founder of the site, with some helpful sounding advice, stories or news. It's a nice personal feel to the way that site is run but, thinking as a person researching into the psychology of emotion regulation, I didn't find the 'self-help' nature of these emails to be either particularly useful or informative. Saying that, I'm sure that others might appreciate the emails and their content but we know so much more backed with evidence to not need to rely on a folksy style of advice. The opt-out option for the email circular was appreciated.

Did the site work for me?

Truth be told, I don't think the site had much of an effect on me. Like anyone else, I experience ups and downs in my mood as part of the everyday ebb and flow of life. At times, I feel in command of my emotions and capable of both coping with negative events and regulating my general positive feelings and expressions; while at other times, I'm overwhelmed and don't feel like there is a way back towards a more comfortable state. Over the course of the PhD, I have learnt much about managing feelings and what different regulation strategies may achieve. Conceiving of emotional regulation as a controllable process or more specifically a controlled process, which seeks to reduce the difference between what we feel and want to feel has placed a certain emphasis on monitoring what I feel anyway. Moodscope as a means to help raise self-awareness of feelings simply wasn't designed for me and so I no longer use it.

In terms of its accuracy, Moodscope does alright generally. My top score sits at about 85% (100% being maximally positive) and my average score is about 65%. This reflects how I would describe my overall trend for how I feel. Further to this, a weekly trend in mood prevails through the general noise of one-time events affecting how I feel. Across a week, mood would slowly drop by a small degree before rising at the weekend - this matches known trends and my subjective experience, which indicates that measurements used are the right ones. Unfortunately, I stopped using the site before I made a significant change to my routine and so I don't have any data to examine the effects of shifting my working schedule to part of the weekend on my mood. Lastly, of note is the days in which I report a low mood (my lowest being about 30%). Looking back at some of the days in which I've left notes alongside a mood score, the lower moods about equally coincide with me feeling a conflict of both positive and negative emotions as well as days in which I have just felt low. This seems to reinforce my comments that the site could be better used with the scores being divided into two dimensions of positive and negative affect.

An aspect of the site I can't review is the mailing of mood results to those close to you. I tested its functionality with an alt account and I can say that the mechanics behind it all works. In terms of it actually doing anything for helping with your mood, I can't say because I didn't have these results sent on to others. I understand its purpose in breaking the self-perpetuating loop associated with social withdrawal and depression as it automatically lets others know how you are feeling. I would believe that this is potentially the site's strongest asset, providing that the process actually encourages support from others.

At the time of writing this, Moodscope has advertised that there are changes ahead for how the site is going to work. I wish them the best of luck for the future and hope that as the site grows, their resources for accessible, effective, and evidence based means of regulating our moods does so too.
Photo links to source

Do we need shoves or just better nudges?

2011-11-18T17:02:00.002Z

Can big changes in our behaviour come from seemingly small changes to our environment? This is the general premise of the book Nudge - an exploration of the applied aspects of the emerging field that is behavioural economics. Nudge has drawn its praises (from political figures eager for a quick fix to society) and criticisms (from political figures who don't think a quick fix is feasible), in the three years since its publication and we can look forward to a whole new back-and-forth in light of other more recent popular science publications. To save everyone time, the Financial Times has wrapped up a collection of book reviews of human behaviour into a neat overview of the state of play of behavioural economics.

The FT article argues that each of the books reviewed makes for compelling evidence that we are not rational mechanical agents, making trade-offs of risk and reward. Alongside this, it also picks out each book's specific criticisms of the 'Nudge' approach, largely along the lines that it is too small an influence. Sometimes those impulsive acts or drives will overcome any influence on our surroundings that a Nudge may have had. Questions arise then, how much do you have to nudge someone before their behaviour changes to a non-trivial degree, and how much of a nudge can give without it turning into the full shove of regulation (something that this government would happily oppose*)?

The second question is a broadly drawn policy matter, which is far beyond my scope and all I would be confident in saying is that regulations are not an a-priori bad that the government paints them to be. My dentist apologised to me, in my last visit, that recent regulations required me to wear plastic glasses so that if the dentist did drop her tools, the tools wouldn't fall and stab me in the eye. I don't look back on the more adventurous times, in which dentists were freer to have a go at some impromptu eye surgery, with any particular fondness. But I am distracting myself.

The first question is the important one. Just what do you have to do to get people to change their behaviours? In the past, I've looked at papers describing our limited capacity for self-control, which is ultimately what the FT article regards as needing to be overcome - picking salads over meatballs and saving for later instead of spending now. There are means of improving self-control capacity, though so far have all been through the exercising of our self-control. This creates a bootstraps problem, in which our capability to improve our self-control relies on existing self-control. It's a similar situation to companies requiring work experience in candidate employees who are applying in order to get work experience.

A paper out this year offers a new insight into improvements in self-control, without the longer term requirements of training and exercising our restraint. Alberts Martijn and Vries (2011) indicate that increasing our own self-awareness can overcome the effect of ego-depletion. Through an implicit priming of our attention to themselves, depleted participants were shown to perform in a second self-control task just as well as those who had not undergone depletion. A neutral prime showed no effects, and the implicit prime of self-awareness did not boost a non-depleted person's performance. The authors postulate that these effects occur because of the priming's implicit nature and that explicit calls to attend to ourselves do not show similar influences. In a similar position to a previous paper, they argue for self-regulation failures to be considered along the motivation intensity theory that if we perceive a goal to be too difficult (say through feeling fatigued) we will ignore it unless the goal is rewarding enough. It's an interesting prospect and one to be taken a closer look at in due course.

It seems that there may be criticisms of nudges still to come, particularly that they are the wrong sort of nudges. Putting clear caloric and health information on food packaging may only go so far, if people aren't considering the food's effects on themselves. I seriously doubt that there is going to be any variant of the quick fix that has a lasting sizable influence on people's behaviour but it's possible that redesigning nudges that appear to have us in mind rather than 'the average adult' could be more effective. Would we reconsider that bag of crisps if it described its contents as 184 of my calories for the day rather than 184 calories - 9% of an adults GDA?

*This is the last week that you can submit your ideas to the Department for Transport for which rail regulations that you want removing. I may know nothing about trains but I do know a lot about how to make this world a less safe place.
Alberts Martijn and Vries (2011) Fighting self-control failure: Overcoming ego depletion by increasing self-awareness. Journal of Experimental Social Psychology 47 (2011) 58 – 62
Photo links to source

The Science of Emotions - Event Video

2011-11-11T10:48:00.000Z

Last Friday, the EROS research group hosted a public event at the Manchester Museum exploring the Why, What and How of Emotions. For the event, each of the research groups in EROS created interactive exhibits and demonstrations looking at different aspects of emotion such as: people's perceptions of emotion, how can emotions help with sport, and how we can change or regulate our emotions. It was an excellent afternoon and a great chance to speak with the public - too often a missing component in scientific research. We look forwards to hosting more of these type of events in the future. Here's the head of the EROS research group, Peter Totterdell, describing the day:

EROS Video from University of Manchester Media on Vimeo.

Liars need good memories*

2011-11-10T11:37:00.004Z

Today sees James Murdoch face the Culture, Media, and Sport Committee (CMSC) for a second time this year, to help clarify some issues that were brought up last time and answer some further questions that have since arisen. Last time around, many people seemed stunned to discover the volume of pertinent events that both Murdoch senior and James did not recall. These ranged from reasons behind signing off large out-of-court settlements and whether they had been briefed or not on the extent of phone hacking. In what has made a remarkably good media narrative, there appears to be a single email, termed the For Neville email, that the CMSC that fundamentally contrasts with some of the assertions made about what has been going on and who knows (or doesn't know) what.

Around the time of the last committee hearing, I started making a simple database of who had worked where, who knew who, and what had happened to people as a way to determine just how close the MET and News Of The World were. I last updated it at the close of the CMSC in which Murdoch senior got a pie in the face. Things have since changed with more revelations, arrests, and resignations so this serves a handy snapshot of information that publicly existed at the time of the last CMSC meeting. Below is a graphical representation of the database, presented using the social network analysis tool *ORA. A red line between people indicates a known personal contact between people, a green line between people and organisations denotes some term of employment there. Green lines between organisations indicate some known communication or ownership (such as News International owns The Sun). Dashed lines represent a partial or historic connection.

Well that's a lot of information there. There's a lot of fun stuff to further examine between the police and News International (particularly that of Neil Wallis), but that can be saved for another day. James Murdoch is tucked comfortably away from any of the more shady characters or goings on, which are situated in the centre and lower middle of the graph. To clarify who people are, we can change the colour of each node based on their attributes known, such as their job or their employment/arrest status. Below are people's job roles.

and now people's status

So this answers the question of who the Neville is in the "For Neville" email Neville Thurlbeck was a journalist at the now defunct News Of The World who was arrested after direct contact with the private eye, Glenn Mulcaire. He also has been investigated for his relationship with police built around a high degree of information sharing. It would probably be inconvenient for James Murdoch if he was aware of what was going on at the News Of The World, particularly if any direct connection between him and some of the less savory goings on. Let's hope his memory improves this time around and we can get some clear answers.

* as upheld by the PCC, the title of an article does not have to reflect the actual content of the article itself. The title of this post is certainly not referring to anyone mentioned in the post.

Psychology to the rescue - in praise of reappraisal.

2011-11-09T22:24:00.003Z

In the past week, BPS research digest has run a special edition looking at how psychology has helped or influenced researchers in their own lives. As my PhD heads ever closer to an actual finishing point, the BPS article serves as a timely reminder for me to look back and ask myself, what have I actually leant? Moreover, what have I actually leant that's useful to me?

The main thing that stands out as a significant learning point for me would be the realisation that our emotional experiences are in some ways controllable. This may seem obvious to some but before the PhD, I had never really given much consideration to my emotions. They were an aspect of my life that just was and I'd not properly thought how emotion and cognition interact, choosing to focus on the latter with little regard to the former. That isn't to say that I wouldn't engage in emotion regulation strategies such as reappraisal, or distraction, I just wasn't aware that these were actual, recognised strategies that can be actively managed.

For this, I'll pinpoint a specific incident in which emotional regulation served an unambiguous role in shaping my behaviour for the better. This example highlights the recursive nature of emotions influencing thought and thought influencing emotions but it also touches on how psychological studies and findings cannot and should not exist in a vacuum, even if we would like to parcel phenomena into neat boxes.

On my walk to the department, I passed by a 40 to 50-year-old man who appeared very obviously drunk, half-sitting half-lying in a heap on the floor in the very inconvenient location for all: an ambulance loading bay at a children's hospital. The situation presented itself almost as an archetypal example of Piliavin's classic study in 1969. As I continued to walk by, I figured that most people would make (and possibly already had made) the exact same calculations I was in the process of - just how high are the costs of helping someone who might be aggressive or even violent towards me? I stopped walking and forced myself to re-appraise the situation and how I felt about the matter. "This person may be in a state of hypoglycaemia and not brought their circumstance upon themselves; even if he is 'just' drunk, drunk people still can need help too; if I still choose not to help after what I know about helping behaviour, how would I regard myself as a psychologist?"

The back and forth conversation with myself must have lasted a couple of seconds but already I had successfully reappraised the situation so that I felt really guilty about my current lack of helping behaviour*. To rectify this, I went over to the man to check that he was ok and was greeted with a "Fuck off you shit!". A good sign: he was conscious and lucid. I felt comfortable in leaving him for a moment to get some proper assistance for him from the staff at the hospital. The A&E receptionist identified him as one of the regulars and that she'd fetch some orderlies to 'help him out', if I could just go back and wait with the man until they arrived.

In the brief moment I had left the man in the ambulance bay, he had attracted the attention of another passer by, who had his own ideas about what to do. A young man had taken offence to the various gestures made by my partner in this story and had decide that the best way to engage in helping behaviour would be with fists raised and shouting "You what? You what?!". I immediately had to intercede and try and diffuse the situation, especially as someone on my team (i.e. a member of the original situation) was in danger. In a neat parallel to my original need for intrapersonal regulation to get things moving, I needed interpersonal regulation to stop things developing. I managed to calm the aggressor with reassurances that I worked at the hospital (not entirely a lie because I did used to), we were dealing with things just fine and that we'd take care of him. I think the ambiguity in the last line satisfied this young man and he went on his way, at which point the orderlies arrived and I was absolved of any further duty.

In this story, there's both a flexibility and rigidity to my feelings and thoughts and that fascinates me. With a little effort I could change my thoughts and feelings about helping someone I had initially categorised as a possible threat. Once I had decided on my course of action though, I didn't perceive an option to change my mind back and just walk away, even when faced with an actual threat.

*being spurred into action to diminish negative affect, as argued by Carver & Scheier's dual process model. I was motivated to act to distance myself from the negative goal of feeling like a bad person, or worse a bad psychologist.

The Mood Altering Effects of Greenfield

2011-08-03T18:14:00.002+01:00

Researchers are showing increasing concern regarding exposure to Greenfield. Greenfield seems to cause, in many, a wide array of severe mood altering experiences. In low doses, mirth and tittering appear to be the main signs of contact with Greenfield, but at higher concentrations, people begin to experience extreme emotional changes including: anger, frustration, and an overwhelming sense of despair. There may also be substantial negative cognitive effects as many people report feeling ‘confused’ and left wondering why they bother with anything. Curiously, despite the insistence that they do not enjoy any part of the experience, participants seem to seek out their next ‘hit’ of Greenfield and are expected to go through this all again at the next opportunity.

While research is still preliminary, that shouldn’t stop me from reporting it anyway. Brain scans of participants who view even just a picture of Greenfield show that brain activity significantly changes in the Fusiform Face Area (FFA) and researchers theorise there would also be potentially irreversible changes in the wiring of the Hippocampus. Researchers highlighted that these changes could well be perfectly normal but could also be made to sound scary if necessary. Activity in the FFA is associated with social interaction and the Hippocampus with memory, both of which are vital in today’s world and can be trained using my own series of computer games (to be released). Repeated exposure over several hours to different pictures of Greenfield leads subjects to show signs of a serious deficit of attention and an increase in fatigue. In addition to this, they develop a restless, irritable, and confrontational manner, demanding that they be left alone and allowed to go home.

Researchers have worked intensively to understand the root cause of the apparent mood and congitive changes in response to Greenfield but still are yet to come to any conclusive decision. Controlled tests of the presentation of other Baronesses has largely shown minimal effects, with most participants failing to show any emotional responses or in many cases, recognition. One notable exception is the severe mood alterations experienced when subjects are presented with the Baroness Thatcher, often forcing researchers to recalibrate their scales for anger.

Keep Calm and Carry on! How your brain can help.

2011-07-08T14:46:00.000+01:00

Glyn Hallam, a PhD student in the neuroscience branch of the EROS research group, presents an overview of the processes involved in the regulation of both our own and other's emotions and the work done to identify the brain regions associated. The talk, titled Keep Calm and Carry On! How your brain can help, was given at the Vitae Yorkshire and North East Hub Public Engagement Competition and won recognition as one of two highly commended talks in science, engineering, maths and medicine. The talk format is known as Pecha Kucha, in which a speaker presents 20 slides each for 20 seconds. Congratulations to Glyn for his award winning talk in what I imagine to be a very demanding style of presentation.

Control the fire of your emotions

2010-10-01T17:12:00.002+01:00

I've just discovered the stats option on the blogger tools and being the vain person I am, I immediately scoured them all looking for numbers of visitors and how people find this site. Unsurprisingly, people come here after searches for social emotions, although a number have chosen to visit via the search: ghosts aren't real. I hope they found confirmation of their hunch. Another means of landing here is by asking: do I deserve to be happy? That unfortunately for those visitors, takes them to this page on a paper on self-esteem and mood improvement motivations, rather than to an answer either way.

What's more interesting than any ego massage I may get from the numbers (and I'll be honest, given the numbers of hits, it's more of a poke in the ribs than a massage) is this video I've stumbled on when searching as others have done: emotions as social control. It's a classic instructional video "Control Your Emotions" from 1950 warning us of the dangers of under regulating our emotional behaviour. We get stern words to camera from a pencil moustached chap in a lab coat, the use of 'wiseguy' as an insult, and wholesome fun at the end. As expected of that era, it's both steeped in behaviourism as an explanation for actions and the presentation of 'traditional' social roles.

The video seems to be simultaneously prescient and yet so wide of the mark at the same time. It describes how emotions (anger) may almost be contagious and that effective means of emotional regulation may be through reappraisal of situations. However, the lens through which emotions are viewed - as an individual's mere responses to stimuli - greatly limits any decent understanding of what's going on. Social emotions and social interaction are hard to conceptualise within a behaviouristic framework because goals, motivations, and felt states can be hidden from observable behaviour.

There are a couple of examples of the complex social interaction beyond behaviourism's reach in the video, both of which are plainly ignored as emotional behaviour. Indeed, the very first instance of any emotional behaviour is that of Jeff attempting to improve his friend's mood. I'm speculating, but, as it doesn't fit the narrative of Jeff controlling (i.e. diminishing?) his emotions, it just acts as set up for the subsequent stimulus/response situation of spilt drinks and anger (also termed mild rage*). Later on, the younger brother is regarded as a stimulus that affects Jeff's emotions alongside the sprinkler or wrench rather than as another person with which social interaction can occur.

Joshua Garrison dissects these types of films (including "Control Your Emotions") highlighting that the overall message to be found is the respect of authority, and 'obey those who know what's best for teenagers'. While the scenes may appear unusual, kitsch and generally unrelatable to us today, the film fails to depict any resemblance of adolescent emotions even within the context of the time. As Garrison notes, Dr Lausen is offering his expertise on a subject not even covered by the film - that of adolescent psychology - but is instead commenting on another adult's fabrication of what teenage behaviour might be like.

While I'm sure nowadays we're still not self and socially aware for our work and ideas to seem incredibly dated 60 years from now, it's this special sort of misunderstanding of humans that will always keep these informative videos looking especially bad.

*It's just too easy to comment on the division of emotion into 3 groups of: rage, fear, and love. I'll let that one slide.
Garrison, J. (2009) The teenage terror in the schools American Educational History Journal 36 (1&2)

Twitter's ever changing moods

2010-07-23T12:20:00.005+01:00

Twitter is a truly massive information sharing platform. While a lot of that can consist of people's minutia of their day, occasionally we can pull some interesting and useful from the noise. As an example of the former, I recently spoke to a researcher who had attempted to grab some data on how the top trends can vary across time. His main finding was that Twitter has such a high noise to signal ratio that it's essentially a pointless tool for people to share anything meaningful. Second, people seem very keen on discussing whatever a Bieber might be.

In contrast, a collaborative effort from some researchers at Northeastern University and Harvard seem to have had a more fruitful experience. They have pulled some 300 million tweets from across the USA in a bid to visually represent mood variation across geographical location across time. Tweets appear to have been analysed using the ANEW system, which is a rating scale for words based on three dimensions of arousal (calm/excited), valence (positive/negative) and dominance (degree of control). For example, fear has a high arousal, low valence, low dominance; fork has neutral arousal, valence and dominance; pride has high arousal, neutral valence and high dominance. The below video shows a time-lapse of a typical day's fluctuations in mood (negative mood, red to positive mood, green). The state geography is represented in cartogram form, displaying tweet density rather than land area.

The high temporal resolution of the data allows them to examine how universally moods change across days. Further to this, as the data was collected for a whole month, weekly trends can be picked out. Mislove et al. indicate that the moods described on Twitter follow a cyclical pattern here. The results are remarkable in their familiarity: in a typical week our moods are fairly constant from Monday to Friday (with the lowest point coming on a Thursday), mood picks up at the weekend, daily mood can be approximated by a sine wave with peaks around morning & evening and troughs late at night & early afternoon. All of these mood dynamics are well documented (e.g. in the book Changing Moods).

One consideration though, how accurately do people present themselves online? Is it not tempting to present yourself as and idealised version of you who always has a super exciting enjoyable life? Firstly, as the results linked above appear to match typical data so well, it doesn't look like that everyone is being inauthentic. Secondly, recent research indicates that despite the reputation online personas have for being projections of ideal or imaginary selves, there is a pretty decent relationship between our crafted Facebook profiles and our genuine 'real-life' selves.

An interesting, if not immediately plausible followup would be to try to pick out 'follower networks' within the tweets and determine whether this has any impact on moods. Can moods ripple through virtual networks as they are argued to do so though offline interaction?

Reappraisal makes you sweat

2010-07-01T22:09:00.003+01:00

Emotion regulation studies can often face ecological validity issues. These may be in part due to a participant's prior expectation that they are going to have to regulate their emotions. To ensure that participants understand and regulate their emotions appropriately in studies it is usually necessary to give detailed instructions on what to do before any emotional stimuli have been introduced. As a result, participants can be ready and waiting to regulate before the experiment begins and so can potentially regulate emotions in ways that are neither typical nor what the experimenter wanted.

The Modal Model presents emotional experience and regulation as a procedural system (e.g. Gross, 1998). To greatly simplify the model, an emotional stimulus is attended to, appraised, experienced and then responded to. Reappraisal of emotion is considered an antecedent focused regulation strategy because we may change how we view a stimulus before it fully impacts on us emotionally. For example, we may reappraise a frightening scene in a horror movie via the classic phrase "it's only a movie", reminding ourselves that the events unfolding are staged and no one is in any real danger. Antecedent focused emotion regulation is argued to be a highly effective method for managing emotions, both in terms of efficacy of emotional change (Gross, 1998) and in terms of minimal mental resource consumption (Baumeister et al, 2007).

However, in day to day lives, people rarely plan what they are going to feel or how to best do deal with what they will feel*. More often, people may have to manage an emotion while they are experiencing it. This premise forms the rational for the methodology in Sheppes et al. (2009). In their experiment, participants are told to carefully watch a sad film and allow any feelings to arise naturally. They are also told that at some point they will receive instruction to engage in a specific emotion regulation or just to continue watching the film. Depending on the condition, the strategy participants may need to use is either reappraisal (re-interpreting the film in objective, neutral terms) or distraction (thinking of irrelevant, emotionally neutral thoughts). This method attempts to ensure participants attended to the film and experienced a genuine emotional state before engaging in emotional regulation, rather than pre-emptively engaging in regulation before the stimulus has any impact.

Post experiment mood scores indicate that both reappraisal and distraction served to reduce negative emotions at approximately the same amount and both conditions significantly differed in mood score from controls instructed to just watch the film. However, physiological measures indicated that there were differences between the amount of effort expended in the conditions. A contrast analysis indicates that skin conductance levels increased between the film-watching period and the reappraisal period as compared to distraction and control. No differences were found between distraction and control. Finger temperature decreased between film-watching and reappraisal as compared to distraction and control. However, this result only becomes marginally significant when directly comparing reappraisal to distraction.

Skin conductance and finger temperature serve as measures of sympathetic activation, which is a proxy for effort expended. As a consequence, the authors advise that these results should be treated cautiously when discussing the effects of emotional regulation strategies. Although, another paper (Sheppes and Meiran, 2007) indicates that under similar conditions those who engage in reappraisal perform worse at a subsequent Stroop task than those who engage in distraction - a finding argued in the ego-depletion literature to indicate expended effort.

This paper's findings contrast the established view that reappraisal is a relatively efficient method of emotional regulation. This may be due to the use of online reappraisal (reappraising the emotion while it is happening) rather than using reappraisal 'very early on during the emotion generative process', as seen in other studies. Online reappraisal may require self-control to overcome and inhibit an experienced emotion, whereas pre-emptive reappraisal may result in a minimal emotion inhibition self-control challenge and so give findings that appear to support the Modal Model. Distraction's lack of sympathetic activation suggests that it is not merely online regulation that 'costs' but perhaps Reappraisal's requirement to engage with, and inhibit, an active emotional experience.

While by no means perfect, this paper offers a start for more realistic requirements for participants having to regulate their emotions. Perhaps an extention would be to have participants unaware of which condition they are placed in until the instructions are presented. This would probably require some prior training as to what each available strategy is though.

*I imagine few people in England had prepared a plan to regulate their emotions should a goal in the world cup be not given.
Shepps, G., Catran, E., & Mairan, N. (2009) Reappraisal (but not distraction) is going to make you sweat: Physiological evidence for self-control effort. International Journal of Psychophysiology 71(2), 91-96
All other references are cited in the above article.
photo links to source

Like fish oil in a barrel

2010-06-30T18:16:00.006+01:00

BishopBlog has announced a call for nominations for the best examples of journalistic misrepresentation of academic work. The main rules for inclusion are simple and clear:

1. The article must purport to report results of academic research, and judgement will be based on a points scoring system, as follows:

Factual error in the title: 3 points
Factual error in a subtitle: 2 points
Factual error in the body of the article: 1 point

2. Factual errors must be ones that can be judged against publicly available documents – i.e. not just opinions or reports of interviews

Further entry requirements can be found on BishopBlog

It's already been a good year so far for misrepresentation (or if you're feeling generous misinterpretation) of academic work with highlights including the internet rewiring our brains and fish oils improving concentration. The bar has been set with the latter being used as an example entry for a score of 16 points.

I'm searching for an article in the area of social relationships, emotions, or happiness because hot topics like fish oil pills are extensively covered by more prolific bloggers. An afternoon of on-and-off searching has yielded little. This is mostly because any paper's reporting of social psychology seems to be rather superficial or opinion led, which means that there's little chance to get facts wrong. However, an otherwise passable report of levels of life satisfaction across ages offers this gem of a graph, which both crops the Y axis to exaggerate the perception of change in life satisfaction and changes scale 7 times on the X axis.

Moodscope, measuring your daily mood

2010-06-23T20:58:00.016+01:00

Via a series of very interesting and distracting blog posts I've stumbled upon the site Moodscope, which offers a free daily tracking of mood. The site has been created as a tool to improve one's mood through regularly completing a stripped version of the PANAS (positive affect negative affect schedule; Watson, Clark, and Tellegen, 1988). This is a very simple approach, which claims to draw from findings that have gone down in psychology folklore: the Hawthorne Effect.

The Hawthorne effect is the change of behaviour just through observation. During a series of studies, factory workers' productivity increased when the lights were turned up, but productivity also increased when the lights were dimmed. Productivity slumped once the studies had been completed. As a result, the conclusion was drawn that workers' productivity was more closely related to simply being observed as part of a study than any specific change in luminance. However, this effect is repeatedly brought under fire with some arguing against its existence or contribution to effects reported.

In any case, there is something to be said for simply being made aware of one's own mood state (regardless of whether it's due to being observed). A study on trainee teachers indicates that higher mood awareness can relate to an increase in use of engagement mood regulation strategies, which include rationalisation, reappraisal, venting, and seeking social support. Engagement strategies, with the exception of venting, have been linked to greater perception of emotion regulation success and higher self reports of feeling cheerful, calm, and energetic (Totterdell & Parkinson, 1999).

Awareness of moods might not be a panacea though. As with much in life, it's what you do with it that counts - a category of awareness of moods, mood monitoring, has been linked with rumination, a mood-regulation strategy that prolongs negative moods (Swinkels and Giuliano, 1995). Mood monitoring is described in their paper as agreeing with statements such as "I often evaluate my mood" and "I am sensitive to changes in my mood". In contrast, their other defined category, mood labelling, is associated with more adaptive methods of mood-regulation. Mood labelling is described by statements such as "I don't have trouble naming my feelings" and "I am usually tuned in to my emotions".

Swinkels and Giuliano describe the difference between labelling and monitoring in terms of being sick and a hypochondriac. If you are sick you can know what the condition is and how to best treat it. In contrast, hypochondria may include mislead concern about health and a state of vague unease, which has a less clear means of treatment. To return this analogy to affective states it could be the difference between "I'm feeling angry; I need to calm down" and "I don't like how I'm feeling."

Moodscope may encourage more of the latter type of mood awareness (although this is just my own speculation). Cards are presented with 20 affective states found in the PANAS and for each state a rating of how intense you feel it needs to be selected. The selection method by which you do this is really inventive and genuinely causes you to stop and think to what extent you feel each state.

Finally, the site encourages you to share the daily score with those close to you and it even sets up automated emailing to share your score should you wish. I understand that this is in part to really drive home that your results are being observed in order to maximise any Hawthorne effect (if it does exist). On top of this, it provides an immediate link to a supportive social circle - the target audience for this site appears to be those that want to improve their mood. Social support is intimately tied with wellbeing and negative moods can encourage social withdrawal, limiting recovery back to a more positive state. This simple approach may help reduce such a cycle.

I've signed up and will have a look at how well it operates across a longer term, including measures such as relative stability, weekly cycles and face validity of it matching my perceived positive or negative state. I'm sceptical that it is the Hawthorne effect at work & think any changes in mood are more likely to be mood awareness, which may help or hinder depending on individual differences. At this point I won't be sharing my scores.

photo links to source
Swinkels, A., & Giuliano, T. A. (1995). The measurement and conceptualization of mood awareness: Attention directed towards one's mood states. Personality and Social Psychology Bulletin, 21, 934-949.
Totterdell, P., & Parkinson, B. (1999) Use and Effectiveness of Self-Regulation Strategies for Improving Mood in a Group of Trainee Teachers. Journal of Occupational Health Psychology 4(3) 219-232
Watson, D., Clark, L. A., & Tellegen, A. 1988. Development and validation of brief measures of positive and negative affect: The PANAS scale. Journal of Personality and Social Psychology, 54: 1063-1070

Resisting everything but temptation

2010-06-18T00:35:00.015+01:00

The Psychologist is conducting a mass interview of psychology tweeters via the hashtag #psychint this week. The first question what is your favourite study has thrown up a few of the classics: Milgram's obedience, Zimbardo's prison, Asch's conformity and also a few of the more quirky too: such as studies on luck, prayer, & hand-washing.

My choice is probably more likely to be in the latter class of quirky study than paradigm shifting classic, although only time will tell. The area of self-regulation and self control has really grabbed me and the hardest thing to do is picking out my favourite study that best exemplifies this area of research. Do I pick one that shows just how cross domain and far reaching ego-depletion can be? Or possibly the series of studies that shows how exercising self-regulation via... err... exercising... can result in improvements in many areas, including doing more housework and improved social relationships?

Although this might not be my all time favourite study, I picked out Muraven et al.'s (2002) because it's the most surprising finding that I could sum up in 140 characters. The study looks at how ego-depletion will increase one's tendency to drink alcohol in a beer taste test. As expected, those who are ego-depleted (through an unrelated thought suppression task, mind) show less restraint in drinking beer and consuming more units of alcohol overall. So far it seems a fairly by-the-books ego-depletion study, but the study has a twist.

Before the study began, participants were told that that immediately after the beer taste test they would then be taking a driving test and would win a prize if they did well enough. All participants understood that they would need to limit beer consumption in order to win the prize for driving well. Higher alcohol consumption then clearly acts against one's own interests, yet this motivation to not drink can be overcome (or ignored) through ego-depletion.

Further to this, the participants' trait-temptation to drink were recorded. When looking at the individual conditions, trait temptation to drink has no effect. However, there is an interaction between the condition and trait-temptation to drink, as trait-temptation increases alcohol consumption increases but only if the participant is depleted. Another finding indicates that the harder a person tried to suppress thoughts the more they consumed during the self-control task.

So why do I like this study?

First, ego-depletion is a powerful thing. It can lead to poor decisions being made and behaviours that are directly contrasting to that which we may want or expect of ourselves. For example, two depleted participants drank over 1250 ml (over 2 pints) of beer, thereby putting themselves over the drink-drive limit, while anticipating a driving test. Anti drink-drive messages are apparently frequent in US and the message in the experiment that participants should limit alcohol consumption was salient. Essentially, people knew of and then ignored the issues surrounding drinking and driving.

Second, there are examples of strong self restraint and people behaving in ways above what would be stereotypically expected. The paper plots the interaction between behavioural trait and experimental condition to map out alcohol consumption. As trait-temptation to drink increases it appears that those in the control condition actually drank less than those controls who do not consider themselves tempted by alcohol (although the individual findings are non-significant). Muraven et al. suggest that people can (over)compensate for their trait tendencies and that depletion may undermine this.

This general argument forms the basis for a later review-paper by Baumeister et al. (2006) Self-control can modify how we behave and can ultimately change how we present who we are. "... if your self-regulation is powerful enough, then regardless of your inclinations, past experiences, or neuroses, you can always do the adaptive or right thing. Self-regulation can be the trump card of personality".

We are walking paradoxes: we are both weak and strong. Our 'true' uninhibited selves are kept in check by our self-regulation efforts, which in turn are no worse a reflection of who we really are.

socialemotions is 1 this week.
photo links to source
Muraven, M., Collins, R.L., Neinhaus, K. (2002) Self-control and alcohol restraint: An initial application of the Self-Control Strength Model. Psychology of Addictive Behaviors, 16(2), 113-120.
Baumeister, R.F., Gailliot, M., DeWall, N., Oaten, M. (2006) Self-Regulation and Personality: How Interventions Increase Regulatory Success, and How Depletion Moderates the Effects of Traits on Behavior. Journal of Personality 74(6), 1773-1802

Bipolar disorder and goal striving

2010-06-16T20:47:00.011+01:00

A new paper in psychological science puts to the test an interesting model on the origin of emotional states and their purpose. The model comes from the field of control theory, which has been mentioned before at socialemotions.

Carver and Scheier's 1998 book On the Self-Regulation of Behaviour details a hierarchical framework of goals we pursue. Within each of these tiers we have goal states that we pursue* via means of discrepancy reduction - in that we see a difference between the our current situation and our goal and then attempt to minimise this difference by working towards the goal. Emotions are said to arise from the derivative of this discrepancy reduction effort.

To clarify, our emotions are predicted to be measures of the rate of progress towards a goal and signals to adjust effort accordingly. If we are falling behind our expected rate of progress we feel negative and motivated to increase effort, if we are progressing faster than expected we feel positive and are inclined to slack off a bit. Hypothetically then, a mood disorder could influence how a person alters their speed of progress towards goals.

Coincidentally, this is the hypothesis of the paper. Potentially, a person with bipolar disorder (i.e. they experience states of mania) may not keep their progress towards goals in check as much as a control. The specific prediction made is that those with bipolar disorder (BD) may be less responsive to unexpectedly high progress towards a goal and so would be less likely to reduce the rate of progress than a control.

To measure this, Fulford et al. gave 12 persons diagnosed with BD and 12 controls diaries to fill out 3 times a day for 21 consecutive days. They were asked to record progress towards 3 personally meaningful goals such as "Spend at least 2 hours on a specific work project" or "spend at least 1 hour of 'quality time' with my son."

Their results indicate that there is a significant difference between controls and BD when it comes to slacking off if experiencing faster than expected progress towards a goal. It should be noted that both groups did coast when ahead but BD to a smaller extent. On top of this, results indicate that overall if falling behind, people will make more effort towards goals. Their results support the general hypothesis that people engage in goal/state discrepancy reduction and also that the rate of progress may affect subsequent effort.

It should be noted that these subjects were in remission for the duration of the study and that a total absence of coasting was not to be expected. The authors speculate that a severely attenuated coasting phase when ahead in goal progress may act with other prodromal symptoms to contribute to a full manic episode.

A final note by the authors indicate that BD participants reported a higher expectancy for goal progress, which is consistent with higher life goals in BD than in controls (Johnson et al. 2009, cited in Fulford et al.). This brings up another study looking at Carver and Scheier's 1998 model, although this is less supporting of the specific predictions made by the model.

In the stunningly well named Effect of Affect in Organizations is a study looks at the goal progress of 15 social workers over the course of four weeks (Holman, Totterdell & Rogelberg, 2005). The overall findings support the idea that discrepancy reduction plays a role in the control of behaviour and that affective state is related to effort. However, Holman et al. also find that goal distance also has an impact on affect, contrasting with Carver & Scheier's model that bases affect solely on rate of goal progress.

While the study does not indicate any direct relationship between affect and effort, it does indicate that affect does moderate the strong relationship between expectancy and effort. Resultantly, it seems that affect can help to jolt effort back on track if expectancies are low but high expectancies appear to override the effect of affect. This specific finding by Holman et al. complements Fulford et al.'s results, indicating that BD have higher expectancies and a diminished response to positive affect.

*or avoid if the goal is negative but it makes for unclear writing.
photo links to source - an individual's management of a manic episode
D. Holman and P. Totterdell are members of the Eros project and my project supervisors
Holman, Totterdell & Rogelberg (2005) A daily diary study of goal striving: the relationship between goal distance, goal velocity, affect, expectancies and effort. In Effect of Affect in Organizational settings. Ed: Ashkanasy.
Fulford, Johnson, Llabre and Carver (in press) Pushing and Coasting in Dynamic Goal Pursuit: Coasting is Attenuated in Bipolar Disorder, Psychological Science

Smile like you mean it

2010-06-15T18:34:00.000+01:00

The Duchenne smile has widely been believed to be a key indicator that the person smiling is genuinely experiencing a positive state. This smile includes the movement of upper facial muscles, such as the orbicularis oculi, pars lateralis, which cause the crinkles round the eyes seen in a genuine smile (also known as crows feet) seen here. The general understanding being that, while people can consciously move the mouth muscles (zygomaticus major) to form a smile, the upper muscles are not so much under conscious control and are more likely to betray a person's 'true' feelings.

However, a recent paper highlighted by Psyblogs indicates that the Duchenne may not be the silver bullet for identifying a person's felt state. The paper indicates that, contrary to long held expectations, the vast majority of people can create the Duchenne smile on demand. Moreover, photographs of this smile can fool other into thinking it is a genuine expression.

A further experiment in the paper indicates that people can determine if a smile is "Duchenne" even when the top half of a face is masked, and the marker of Duchenne-ness is not present. This indicates that there are other indicators that we use to identify between smile types. The final study indicates that the dynamic process of a smile may also be relevant to someone working out if it is genuine. Indicators of a non-genuine smile include asymmetry and a shorter peak intensity alongside more obvious, non-positive affect indicators such as frowning or lip pressing.

The Duchenne smile then does not seem to be this foolproof indicator that a person's smile is genuine. Even some of the spontaneous smiles (and presumably genuine) shown by participants lacked the key orbicularis oculi, pars lateralis movements.

One consideration not mentioned in the paper though is just how people who were required to fake smiles approached the task. In the occupational psychology literature, there is a distinction made between surface acting and deep acting. The former is to just fake an expression and the latter is the changing of one's own emotional state, which then may feed-forward to naturally affect expressions. If participants engaged in deep acting because they wanted to present convincing smiles, this may trigger spontaneous smiling.

Even if this is the case though, the smiles still didn't get past observers when they're watching the smile on video. Perhaps deep-acting is too slow of a process to manage felt and expressed states to give that all important genuine expression.

photo links to source

The world revolves around me.

2010-06-14T15:32:00.003+01:00

An ego-net is a term from social network analysis that indicates who an individual knows. On top of this it also highlights which of the ego's contacts also know each other. This creates a network of interconnected people, rather than just a list of who an individual knows. One of Facebook's popular applications is the Circle of Friends, which can present a picture of our own networks, although it leaves no room for any actual interpretation of what it means. Also, it seems to be broken at the moment.

The social network analysis program *ORA developed by Casos has a function to import Facebook data using the data mining tool Cemap. To develop the ego net the tool picks up all your contact's names (you can only use it on your account) and any other information that your contacts make available to you, including your mutual friends list. With the mutual friends list you can identify know who knows who in your social circle. It will then graphically display the network nodes to form a spider web of contacts.

As an example here is my Facebook ego net with all names removed. I'm the red dot in the centre.

This image has been coloured according to Newman grouping, an algorithm that picks out clusters of similar people. It divides up groups based on numbers of connections, building on the assumption that a number of people who all share connections are more likely to be part of an identifiable group than those who have few connections between them. The Newman grouping is a strong exploratory method of identifying groups in a network because it doesn't require specification of the number of groups. Other methods require us to determine how many groups the network should be split into, which becomes self-defeating if we are using an algorithm to find that out.

The approach here has picked out 8 distinct groups, and has overall been very successful at representing how I would have divided the social network. The red cluster seems to have hoovered up a few people who do not belong there but, as it has also placed me in the red group it's not surprising to have others connecting to me, and only to me, to also be in the group. One of the groups is surprising to see though, and demonstrates just how powerful the approach can be.

First, we have the large red group (college friends), the turquoise above that are friends from an old job, the orange are friends from undergraduate, the yellow: postgrad friends, the dark blue are those from a more recent job, and the two light blue groups at the bottom are my family and my partners family. Finally, the sickly green group in the top right are an unusual bunch that have been grouped together mostly because they have some weak ties with my university friends but not enough to be put into either group. These are people I've added from classes in undergraduate, who I never really associated with and have added them 'just because'.

The network looks quite an intricate mess like this (often termed a Persian rug). If we selectively remove the node representing me, we can start to see how the groups interact.

Having hidden all my connections we can begin to see how these groups connect. First, we have a couple of isolates, a dyad and triad. These small groups are unconnected to any other and appear to have been placed in red group because I was. In contrast to this, the isolated group of 6 (dark blue) seems to be interconnected enough to be recognised as the independent cluster it is. More generally, the network is still strongly held together by a very central node. As this network is built around my data it's no surprise that the next most central is my partner of 7 years.

If we hide her connections, the network beings to spread further.

A common measure used in SNA is a node's centrality. The above picture indicates that once my connections had been hidden, my partner was a bridging point between the red, blues and the super-group of uni students. If we consider how information could spread through connections (say if my cousin had a baby) the only way that my uni or college friends could have heard about this through the network (and I'm sure that they would have wanted to know) is via my partner. She is said to have high betweenness centrality (as she is between lots of other nodes). With the two central figures now hidden, who steps in to become the most central node? For this we'll only focus on the uni network because the above picture shows that the network has now fractured into multiple networks.

Running some analysis on the change in betweenness before and after the removal of the most central nodes indicates that a few people become massively more central in the network. With the exception of the highly connected yellow node by the turquoise, these nodes are relatively sparsely connected (low degree centrality) but seem to have the right connections bridging across the groups. For example, the yellow node in the middle has jumped from being the 19th in betweenness centrality to 3rd because its few connections are made at important locations (i.e. is the only one to connect to each group).

Should you want to try out your own ego-net and find out just how well your social groups could get on without you, the analysis and data mining tools can be downloaded from the Casos website. *ORA is a bit fiddly to use but the network visualisations, mini simulations of network dynamics and plain English explanations of network measures far outstrip any other SNA program I've seen.

There may be trouble ahead

2010-05-31T20:09:00.007+01:00

Have you ever had the feeling that there's trouble brewing ahead? Perhaps you've noticed that the atmosphere in the pub has changed slightly and you just know that someone's about to start a fight. Maybe you go to nice places where this sort of thing doesn't happen.

In any case, usually before there is any actual trouble, there is an opportunity to engage in some interpersonal emotional regulation to try to calm things down a bit. New research suggests that this might not be so special to humans (although we can still lay claim as kings of kicking off for no reason in the pub) as capuchin monkeys have been seen grooming each other with greater frequency just before trouble is expected.

Grooming behaviour has important roles in monkey societies and, as the New Scientist article explains, less dominant males use this as means of asking permission to share food in advance. It might be a bit of a push to immediately assume that this is emotional regulation behaviour as there's always a danger of projecting human behaviours on to animals to explain their actions. An opposing view in the short article highlights that the monkeys may be grooming each other to alleviate their own anxiety about predicted trouble rather than to change other's feelings.

I don't really mind whose view may be correct because it's still quite exciting that animals may be engaging in emotional regulation.

photo links to source

Sleeping is giving in...

2010-05-26T19:15:00.000+01:00

Life hacker has a few practical tips on how to improve the quality of your sleep. The article acknowledges that for some, a lack of sleep can be described with a sense of pride (look at how hard I must be working) but then draws the parallel with eating habits and that few would boast if they were malnourishing themselves.

The advice ranges from cutting down on the late night computer or television activity as the bright screens late at night could interfere with circadian cycles (bright light has been used as a treatment for SAD) to cutting out distractions such as early morning noise & light to a coffee ban in the evening. While these may seem a bit of the quick-fix variety, the article also offers ideas on improving habits and changing your routines to get more useful sleep.

What am I feeling today?

2010-05-14T17:47:00.002+01:00

Dr. Deb has a neat little article on alexithymia, which is essentially the inability to recognise or understand one's own emotions. It is also thought to impact on the ability to understand the emotional experience of others. Given that it can be difficult to verbalise what's being felt, some may engage in other means of understanding feelings - such as cutting oneself.

Rather than it being recognised as a psychological condition, it is regarded alongside personality traits. This is largely because it appears to be very stable, as opposed to a 'state-dependent phenomenon'. However, the paper Deb links to argues that the relationship between personality and alexithymia may be mediated by current states, such as mood.

Their results indicate that higher scores of alexithymia are associated with an increase in state anxiety suggesting that it may not be as 'stable' as something like personality. Although personality itself, while commonly thought to be this unchanging feature of who we are, has a decent amount of state variability too.

Lee, Y. el al (2010) Direct and indirect effects of the temperament and character on alexithymia: A pathway analysis with mood and anxiety Comprehensive Psychiatry 51,(2) 201-206 (Article linked on Dr. Debs site)

Headache contagion

2010-05-12T16:49:00.004+01:00

New TED talk out today by Nicholas Christakis on the topic of social networks and their influence on our behaviour. Despite the methods of analysis and subsequently the remarkable findings coming under question by Cohen-Cole & Fletcher (2009), I have a tremendous amount of interest and enthusiasm for the overall concept of Fowler and Christakis' work. This is largely due to their popular paper on the spread of happiness throughout a social network being released in the week I started the PhD and like a duckling I have imprinted on the first thing I saw.

I've yet to find a direct reply by either Fowler or Christakis to the claims made in Cohen-Cole & Fletcher (2009) - namely that using the same data set and network analysis methods they managed to find contagion effects for headaches, acne and even height. All of these we can safely assume aren't going to be transmitted from person to person. They argue that not enough weight is placed on situational effects, which may appear as contagion effects in the analysis.

Nevertheless, the talk is well presented and it really does feel like there is some substance to the work - we are social creatures that gather a large amount of information about what is normal behaviour from our peers who in turn refer to their peers for reference, and so on...

Cohen-Cole, E. and Fletcher, J.M. (2008) Detecting implausible social network effects in acne, height, and headaches: longitudinal analysis British Medical Journal 337

A stitch in time

2010-05-10T16:26:00.010+01:00

I'm sure very few people find themselves at the altar saying I do while thinking but I probably won't in a few years. However, records indicate that divorce or separation are becoming more commonplace, with some estimates to be over half of marriages ending but those means. Rey (2010) describes this as the failure paradox, "how is it that a sentimental relationship designed to last will very probably break down?" and proposes a dynamic model to describe the process of a relationship break up.

Rey collects a handful of simple rules to generate a model of satisfaction within a long term relationship and indicates what is necessary to ensure a sustained and happy time together. Firstly, and to keep things simple, the couple is designated as a single entity (by assuming that long term partners are similar to each other). This reduces the number of actual parameters necessary but lends the model's outcome to be considered as mutual relationship satisfaction.

Secondly, relationships are considered to deteriorate in quality over time from the high point of starting out in the relationship. Given that Rey cites a major reason for divorce as, "gradually growing apart and losing a sense of closeness, maybe staying together but emotionally detached until their loneliness is not longer bearable*", it is a plausible but bleak description of relationship progression.

Finally, the second law of thermodynamics (for relationships) is invoked to argue that this continual decay can be counteracted by conscious effort made in the relationship. Specifically, Rey states, "things fall apart unless energy is supplied to keep the relationship alive and well". There is a caveat to this though: beyond a certain point relationship repair becomes too effortful relative to reward to be of any value.

Using these rules, we can work out a couple's projected trajectory for their relationship. The state space (that is a map of effort in and current satisfaction) indicates that there is a point of stable equilibrium where effort and satisfaction are both constant and within plausible bounds. Other points on the map indicate instability:** a couple with a very high initial satisfaction may separate if effort is not put in; a couple who begin below the marked threshold of satisfaction may improve their relationship only if substantial effort is made.

The dynamics of the model exist as a saddle point, which means that there is as such an ideal target trajectory for couples to follow in terms of effort and satisfaction. Rey indicates that in principle this could be maintained if couples identify and correct for any perturbations in the relationship that may reduce effort invested. The key outcome of this set up though is the need to quickly identify when effort is not being invested and deviation from the optimal trajectory begins. Insufficient effort invested in the relationship in an attempt to get things 'back on track' may only be a stall and the situation could become irreversible given the too high an effort is deemed not worth it.

All in all the paper doesn't make for happy reading; continual talk of inevitable separations and typical patterns in a deteriorating relationships can't inspire much confidence. However, this is a necessary issue to discuss because a better understanding of how things can go wrong can help us to identify how to make things go right. The paper confirms that it may be better to attend to relationship issues while they are small and manageable rather than letting them accumulate beyond any point of reasonable effort.

On that note, I've just remembered that I still haven't done the washing up.

Rey, J. (2010) A Mathematical Model of Sentimental Dynamics Accounting for Marital Dissolution PLoS ONE 5(3)
*From the California Divorce Mediation Project
** This just means change rather than an unstable or rocky relationship per-se
Today's paper comes courtesy of @keepstherainoff
photo links to source

FPTP as a feedback loop

2010-05-06T09:49:00.005+01:00

Today's the day when we all head out to the polls, do our bit for this great nation, and then get to sit back for the next five years complaining how the county is going to the dogs. The hardest part of the day is deciding which person you fancy enough to put a little kiss by their name, hoping that they might finally notice you. Fortunately, the first-past-the-post system really narrows down those people you have to choose from, because every one knows you only ever really have a choice between two people.

To alleviate any potential bias I may have in writing I'll be talking about favourite fruits rather than political parties. In my local area, it is an absolute shoe-in that everyone's favourite fruit is Banana and I would literally eat my hat if Blueberry or Strawberry were voted most delicious. The results for 2005 show that Banana won 50% of the vote Blueberry got 30% and Strawberry got 12.5%. On top of this, Banana has had a lot of positive media exposure in the past few weeks.

My partner says she will vote Banana for two reasons: they are pretty tasty, not her favourite but good enough; she really dislikes Blueberries and believes a vote for anything other than the Banana (e.g. voting for Lime) just helps out Blueberry. However, I will vote for Strawberry because it really is my favourite and I don't approve of tactical voting.

Under the current system, I will waste my vote because it affects neither of the frontrunners. However, if I were to vote for either of the frontrunners, it only brings out the worst of the FPTP system. People voting tactically, such as my partner will, helps create an almost unchecked positive feedback loop (one behaviour encourages more of the same). If people decide to vote for one of the top two fruits because no other fruit is currently within a decent chance of winning, the gap between the top two and the rest can only increase.

This tactical voting then becomes even more necessary as people may be tempted to switch from Strawberry to Banana because they really don't want risk a Blueberry win. A result of this is the leading two candidates take more votes than they are actually worth thereby falsely presenting the situation as a two-horse race*. So my vote is what it is not just because Strawberry is the best but also as an opposition to the inevitable trend of the leading two hoovering up all the votes.

Could FPTP work if people genuinely voted for their favourite candidate? Probably not as we would then get more three-way splits** in which people can legitimately complain that the majority of people don't even like the 'favourite' fruit.

photo links to source
*the candidates are now horses, not fruit.
**candidates are back to fruit and now there's ice-cream involved.

Are sports drinks so effective you don't even need to drink them?

2010-04-30T16:23:00.005+01:00

Yesterday, New Scientist reported that a sports -energy-type-drink can boost your physical performance before the drink has been metabolised and reached the muscles. I immediately jumped to the conclusion that some sort of placebo effect is going on; however, this appears to be not the case. Ganta, Stinearb, & Byblow (In Press) were one step ahead as in the second experiment the participants didn't even drink anything but used the solution as mouthwash. Those that had the carbohydrate solution experienced significantly larger motor evoked potentials (essentially instructions telling the muscle to work) than those that had a placebo solution. There's no way participants would think that not drinking the solution would help and even if they did the control solution should have an effect too.

The authors present this as the first evidence of this effect and suggest that it provides evidence of a link from oral receptors with motor commands. They argue that the detection of carbohydrate in the drink sends a signal which integrates with the descending motor commands from the brain to facilitate increased output at the muscle site. As such, there is a message to keep on working for just a bit more because help is on its way.

However, this isn't the first experiment to attempt to examine this link. Whitham and Mckinney (2007) found that there "was no positive effect of a carbohydrate mouthwash on running performance of ~1h duration". They also found no differences when measuring across a variety of physiological measures. This of course may mean that the effect found by Ganta et al. may just be a really small effect overall compared to what would be a much bigger effect of increased endurance. Alternatively, a difference may be due the different type of exercise used to induce fatigue (running vs. bicep curls).

Slightly at a tangent, but I've recently seen adverts for a popular sports drink that now comes in low calorie form. To do this, they have cut the sugar content (and thus the calories) by two-thirds. I'd love to know how this now works as an energy drink because as far as I understood, glucose delivery was the point of these things. For example, Gailiot et al (2007) use a glucose drink to demonstrate resource replenishment and a low sugar drink as a control to demonstrate no effect. Perhaps the company's R&D department are looking the wrong direction to find the low calorie approach and should be going in to mouthwash. I doubt we'll be seeing spittoons at the gym any time soon though.

New Scientist helpfully link to the paper but it is really densely packed with acronyms and neuroscience terminology that I have long since forgotten, including some measure of activity requiring TMS.

Ganta, N., Stinearb, C.M., & Byblow W.D., (In Press) Carbohydrate in the mouth immediately facilitates motor output Brain Research
Whitham, M., & Mckinney J. (2007) Effect of a carbohydrate mouthwash on running time-trial performance Journal of Sports Sciences, 25(12), 1385-1392
Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., Plant, E. A., Tice, D. M., et al. (2007). Self-Control Relies on Glucose as a Limited Energy Source: Willpower Is More Than a Metaphor. Journal of Personality and Social Psychology, 92(2), 325-336.

A man's wandering eye affects a woman's perception

2010-04-28T20:45:00.008+01:00

I could talk today about our PM's display of self-regulation by holding back what he really thought about a voter's question on immigration until he was safely out of earshot yet still miked up. However, everyone in the country seems to have caught bigotgate* fever so I have little to add that hasn't already been said.

Instead, there's an imaginative paper in the recent issue of Emotion proposing that "fluctuations in security regarding one's romantic relationship can literally affect how one sees the world". The theory being that given one may look to their partner for support and reassurance in the face of negative affective stimuli, then an induced lack of support or security in the relationship may heighten anxiety, which in turns increase attention orienting towards threats.

To test this, Most et al. (2010) invited 25 heterosexual couples to take part in an attention / emotive distractor task and a simple picture attractiveness rating task. Each couple was split so the woman in the couple was given the emotive distractor task (identifying the orientation of a picture appearing 300ms after a neutral or aversive image is flashed up) and the man rated the attractiveness of landscapes and later other women. The critical part of this is that both experiments were happening in the same room at the same time so the woman clearly knew that her partner was rating "single women - some of whom are on campus". The woman's unease with the situation was recorded via self-report at the end of the experiment.

Results indicate that the women reported greater unease during the rating females condition. Across both the above experiment and a similar, which contained erotic and aversive images as a distractor for the woman, that self-reported uneasiness with the situation increased emotion induced blindness when presented with aversive images only. Unease did not cause a general decrease in performance in the attentional task (e.g. possibly through a generalised increase in anxiety). Individual differences in the woman's sensitivity to relationship threat appear to also be a factor in orientation towards a threat, but Most et al. leave that to be followed up. They also note that there's no theoretical reason as to why they picked women to do the main task and men may be equally or even more so affected by the induced unease.

From a priming, or spreading activation model (e.g. think of a fire-truck, now name the first colour that springs to mind) this all makes sense. A subtle but nevertheless raised state of anxiety about one's relationship may well be enough to prime or activate the threat orientation behaviour seen. It might be interesting to see if the man's attractiveness ratings have any bearing on the woman's unease responses and in turn threat orientation. Could a stream of 'aesthetically challenging' people to rate actually increase relationship security and task performance?

photo links to source
Most SB, Laurenceau JP, Graber E, Belcher A, Smith CV. (2010) Blind jealousy? Romantic insecurity increases emotion-induced failures of visual perception. Emotion, 10, 2, 250-256
*I guess that means it happened in the Bigotgate hotel (A Bed&Breakfast that turns away same sex couples)

Good bloggers remix. Great bloggers re-post.

2010-04-26T09:34:00.005+01:00

Short post today as it's been my turn to do the monthly blog over at EROS (Emotional regulation of others and self).

Recently, I've given a handful of people a diary study to complete after taking a pre-test EROS scale. Each participant had expressed surprise that we had bothered to include negative emotional regulation strategies in the proposed EROS scale (along the lines of I have intentionally tried to make myself feel worse) because they thought that they were the only person who ever did that.

I've used this as a launch for essentially remixing this prior post with a few others from here. It's interesting to see that even with completely different results in page counts the overall ratios remain the same: more people talk about emotions as things they are rather than things they feel. Also, there is still a very large number of sites for those not wanting to be happy, while only a few for those who want to be sad.

Also, it's a little disarming to see my own site crop up in the first few results when I ran the searches this time.

Everything's better with music

2010-04-23T16:41:00.004+01:00

My project supervisor has just sent me a link to a really useful site. stereomood.com categorises pieces by emotion most closely associated or relevant activity planned. I've only have a quick look but it appears to based on user input, not on any algorithm or pattern matching software as I've seen developed elsewhere. This is really handy for me as A I rely on half yearly updates from my brother to let me know what's new in the music world and B This is how I would organise my music if I really was inclined to put a bit of effort in.

I quite like how there's: romantic, candlelit dinner, sexy, foreplay, make love as distinct emotional styles / activities, as if you're going to temporarily halt the evening's goings on make the all important playlist switch.

Facebook knows all and sees all.

2010-04-21T17:42:00.006+01:00

There's a paper from the recent Conference on Human Factors in Computing Systems (Atlanta 10-15th April) which demonstrates how researchers can begin to use the sheer volume of data that we as a collective choose to create. Adam Kramer has used the LIWC model (described in an earlier post) to harvest data from publicly accessible Facebook status updates. The LIWC model can give an index of the general positivity and negativity of an individual statement. Once you have the system analysing the status updates for approximately 100 million people you have something very powerful.

Kramer chose to use Facebook status updates over wall posts, tweets or blog posts for three reasons. 1. They are largely undirected - people typically use them to announce to the world what's going on over initiating conversation, whereas, twitter and blogs can be directed. 2. They are typically self-descriptions containing affective or emotional content; in contrast, Twitter is used both for personal updates and for information sharing. 3. Status updates are short and so the presence or absence of emotive words has more impact.

The resulting data can then be used to estimate satisfaction with life, a component of subjective well-being, using the plausible assumption that those who have a higher satisfaction are more likely to express themselves in a positive way and less in a negative way. The results aren't recorded in a straight counting system but given as a percentage of emotive words averaged across each status update in the day and then z-corrected against the results for all days. This way instead of measuring an absolute change Kramer is looking at the relative change for people.

Using a survey on life satisfaction a prediction can be made as to whether one's life satisfaction can predict how positive one's status updates. Despite the incredibly varied nature and open ended potential for a status message (particularly given that the person has no idea that the message is being monitored), a hierarchical linear model picks out a small correlation (r=.17). The result suggests "those more satisfied with life do indeed score higher on the metric [Kramer's corrected LIWC score], relative to other users, for a given day."

Also of note, is the rise in positive expression around significant dates in the American calendar (e.g. Halloween, Thanksgiving) and there are two pronounced dips in positivity on the days Heath Ledger and Michael Jackson died. Finally, Kramer credits Mother's day 2009 being a much happier day than in 2008 due to the rise of the older demographic in Facebook's usergroup. It's noted that a limitation of this approach exists in unintentional recording of a potentially emotionally neutral status (e.g. People entering "Happy Holidays" as a status is picked up as a positive state). The author, however, regards this as being emotionally positive in itself.

For me, the most significant outcome of the paper is the demonstration of how we are generating masses of data just waiting to be found if only people care to look for it. The other day Psy-Blog highlighted a recent study that shows just how open and accessible people's Facebook profiles are. Of the 400 people randomly selected 98% have their mini-feed (general Facebook activity) available and a further 83% had their wall (person to person communication) on display. For better or for worse, be it for a modern day anthropologist or for more targeted marketing the data is there and ready to be used.

Kramer, A.D.J (2010) An Unobtrusive Behavioural Model of "Gross National Happiness" Conference on Human Factors in Computing Systems, Atlanta.
Photo links to source.

That much?

2010-04-20T15:42:00.005+01:00

Once again, The Onion's News in Brief comes worryingly close to reality.

If I could talk with the animals.

2010-04-14T16:11:00.010+01:00

Yesterday, my supervisor sent round a recent article in Psychological Science (free access), which indicates that humans may not be the only animals to experience the effects of self-control. Dogs who have participated in a self-control task will then persist at a difficult puzzle for a diminished amount of time when compared to those who have not had to exert self-control. Just as humans would.

Participants were given a tug-a-jug toy to play with over the week prior to testing to familiarise the dogs with the toy and to build an association of receiving food with playing with it. To induce a depleted state, dogs were instructed to 'sit and stay' in the middle of an empty room for 15 minutes. If they moved from their position the owner would enter the room and put the dog back. Controls were kept in a dog cage for 15 minutes. The tug-a-jug with food and a block of wood inside that were too large to fall out the opening was then introduced and the dogs' persistence at retrieving the treats were recorded. Dog's exhibiting self-control before the second task spent significantly less time on the toy before ignoring it.

In a brilliant twist, the researchers have decided to test not just whether dogs experience ego-depletion (which would surely be a paper in its own right) but also to test Galliot et al.'s findings that consumption of glucose will restore the ability to engage in self-control. Again, results were comparable to humans, depleted dogs that had a sugar drink would persist longer at the task than those depleted dogs that didn't. As such the 'self' been restored as it would in humans.

Miller et al. highlight that one view of how and why humans engage in self-regulation might not be so compatible with their current findings. It is often argued that humans differ from animals because we have a sense of self. "This sense of self allows them to compare their actual state of self with their standard or ideal self. It is the desire to behave in consonance with the ideal self that fosters better self-control." So either dogs also hold ideal concepts of their dog-selves or perhaps ego-depletion just comes from over use of executive controls that co-ordinate thought and run goal-directed behaviour.

So far my supervisor hasn't quite warmed to my idea for an extension to the study, which may begin to answer important question we've all wondered at some time or another: Do dogs experience emotions? Other studies have demonstrated that an induction of positive affect restores depleted self-control resources. Potentially then, restored resource could act as a proxy measure for investigating if making a fuss over the dog improves affect. I'm taking inter-personal emotional regulation to new levels.

photo links to source
Miller, H.C., Pattison, K.F., DeWall, C.N., Rayburn-Reeves, R., Zentall, T.R. (in press) Self-Control Without a “Self”? Psychological Science

Dyadic disaster

2010-04-12T15:57:00.005+01:00

The 3-minute mini lecture series that has taught me so much presents an excellent example of emotion regulation of both oneself and others. In this example, we see how repeated attempts to upregulate another's emotions in a dyadic interaction result in neither party experiencing positive emotions. This appears to have been caused by the second party in the dyad also attempting to regulate the other's emotions, resulting in a compromise that suits no one.

Others in this series can be found via here, here and here

Computer models superior to human participants

2010-03-08T15:46:00.004Z

The Onion has a good piece today on the hazards of working with human participants - particularly if you divide them into a control group and an out-of-control group.

This too shall pass

2010-03-03T18:58:00.004Z

A stunningly ambitious video from Ok Go has been sweeping across the web for a few days now and the guys deserve all the acclaim they get. This too shall pass video is a continuous camera shot of an oversized Rube Goldberg Machine.

This too shall pass as a phrase has a lot of history and as the story goes is an enduring emotional regulation technique. As with many legends, lots of origin stories exist but all seem to centre round the same premise. An eastern king (often regarded as Soloman) wanted a magic ring that could make happy men sad and sad men happy. After much deliberation a ring is delivered with the phrase inscribed. This happy king is saddened to realise his wisdom and fabulous wealth and tremendous power were but fleeting things, for one day he would be nothing but dust.

Abraham Lincoln puts his own twist on the phrase, with the king in his story asking for a phrase that will be true and appropriate in all situations

Live out of the moment

2010-03-01T17:05:00.008Z

Thought provoking TED talk today by Daniel Kahneman on happiness and memory. Specifically, he argues that there is a distinction between being happy with your life and being happy about your life. The first is the enjoyment we derive from life on a moment-by-moment basis and the second is the satisfaction gained from a retrospective look at our life.

The premise is illustrated well with examples of a painful medical procedure and a relaxing vacation. Research indicates that the recollection of pain during a colonoscopy is affected by how the experience ends. Even if the procedure for person A takes twice as long as for person B (and so has a greater overall moment-by-moment discomfort), providing the discomfort gradually decreases with duration, person A will remember the experience as less painful.

Similarly, a two-week holiday is not twice as nice as a one-week holiday in our recollection if the second week is no more special than the first. I've noted before that my enjoyment of a holiday is always affected by the hassle of travelling on the return, particularly having to navigate confusing terminals at unusually early times in the morning. Ideally there should be a way of moving the travelling hassle to some point in the middle of the holiday so it can end on a nice relaxing moment.*

Kahneman puts this down to how we edit our memory. An astonishingly large amount of our lives is just forgotten; most of the moment-by-moment experiences count for nothing when we look back at our carefully sculpted version of our past. There may be no point to 'living in the moment' if you don't remember any of it, or worse, have unsatisfying memories. He argues that the Gallup poll shows that the two concepts of experience and memory are only weakly related at a 0.5 correlation (I guess you don't win the Nobel prize by being happy with a 0.5 correlation).

I can see how our moments of happiness in the present don't always relate to our memories of happiness. Self-control failures may result in a temporary pleasant experience (such as over indulging on cake while on a diet) but it's unlikely that in a week one would look back and think 'I am really glad I ate so much'. In contrast, persevering with tasks you don't enjoy (staying in and studying) can lead to rewards that are well worth the effort expended and greater satisfaction in life. Perhaps a time when the two states of experience and memory align is the experience of 'true happiness'.

The take home message of the talk for researchers into happiness and well-being is to ensure that what you claim to measure is what you actually are measuring. Happiness and well-being can frequently be interchanged in our minds (and our study's participants' minds!) and some clarity needs to be brought to the subject.

*actually the ideal would not have to travel at all and just live somewhere nice and warm - Kahneman has views on the effect of this on happiness too though.
photo links to source

Seeking justification for napping at work

2010-02-24T15:07:00.007Z

Continuing on from Monday's post, a new article on afternoon napping is in press at Biological Psychology. Zhao et al. examine how the type of afternoon napping may have effects on mood, sleepiness, and alertness. 30 habitual nappers were divided into 3 categories of control (no napping), afternoon napping in a bed and afternoon napping in a chair.

Research of people's sleeping while in a chair has so far focused on a reclining posture but this paper looks at the effects of the forward bending posture one would use to nap at a desk. According to the current paper, past results imply that this posture may inhibit the benefits of taking a nap. Even so, it remains a popular posture for napping amongst students that want a nap but can't make it back to their dorm in time. Furthermore, the surveyed students consider this an effective approach to napping.

Participants were surveyed on their subjective fatigue, sleepiness and mood before and a half hour after a 30 minute nap (or 30 mins of just sitting if they were controls). As a measure of alertness, they also completed a tone identity task where they had to quickly identify a high pitch tone, ignoring a more frequent lower tone. Finally, sleep condition participants were wired to an EEG throughout the study as a measure of alertness in terms of brain activity.

In both sleep conditions participants reported feeling less sleepy and fatigued after a nap and all conditions reported being in a better mood in the post-test session. Both sleep conditions reported being less sleepy and fatigued that those in the waking condition and those that napped in the bed reported a bigger improvement than those in the chair. Finally, the paper suggests that those that slept had a better mood than the no nap group but fails to provide a significance level for this statement so we'll disregard that (all the others are reported and are at least p=0.007).

No significant effects for the response time in the audio task were found; although, the napping in a bed condition showed significantly improved accuracy. The napping in bed condition also raised their post test accuracy to significantly higher than both the no-nap and chair-nap group, which in turn were better than the no-nap group.

The EEG results indicate that while napping participants only showed stage 1 & 2 of sleep (not surprising given that the nap was only 20 mins). Without getting all technical about the EEG readings it looks like the bed-napping group show a higher p300 reading (the brain's electrical response seen in the 'oddball task' design - the surprising tone in the group of regular tones) than the chair-napping group.

In summary, it looks like the napping in the chair will improve the subjective experience of sleepiness and fatigue. However, only napping in the bed actually objectively improves one's alertness. What would be an interesting followup is looking at performance on not just rigid tasks measuring alertness but tasks where one's perceived fatigue state may have an effect on performance (e.g persistence at problem solving tasks). It may result in improved persistence despite a lack of any objective effect on restoring depleted energy.

photo links to source
Zhao, D., Zhang, Q., Fu, M., Tang, Y., Zhao, Y. (In Press) Effects of physical on sleep architectures and post-nap functions among habitual nappers. Biological Psychology

Research to lose sleep over.

2010-02-22T13:00:00.008Z

There are a couple of presentations at the AAAS that look rather exciting for me on a personal and professional levels. Firstly, Matthew Walker and co. highlight that a post lunch nap may enhance later performance on memory tests. Previous studies have indicated that performance at tasks such as pattern production and memory improve after sleeping. Their results support the body of evidence that a rest during our afternoon lull in energy may be overall a beneficial move to take. Still, it's hard to convince others in the office that this is the case.

The idea that sleeping (and therefore not doing anything) is associated with laziness in our culture is a real shame. Yes, there are certain restrictions on when you have to be awake, especially if you like having a job. Even so, time for sleep should be a priority when we allocate our daily routine but it often gets pushed by the wayside for other more useful activities. Insufficient sleep is detrimental to our ability to concentrate on tasks and may affect our persistence at goals. If anything, sleep appears to be the antidote to laziness.

The presentation by James Fowler indicates that, who we associate with may affect how much sleep we get. In a further bid by the world to complicate my research, sleep duration can be added to the list of things that are contagious. So, of the five general areas of my work: expressions, emotions, goals, self-control, and sleep, each one is affected by social networks.

Fowler's social network analysis study (a method that he has previously used to indicate that happiness, smoking and obesity are contagious) indicates that adolescents with high centrality, lots of important connections in a network, were more likely to experience lack of sleep. Effects can be seen at the fourth degree removed: a person's behaviour can end up 'influencing a friend of a friend of a friend’s friend'. Bear in mind that Travers & Milgram's (1969) letter study indicates that 6 degrees of separation can be enough to connect any two people.

Further results indicate that those who are likely to experience a lack of sleep were also more likely to smoke pot and that both lack of sleep and drug use radiate out through the network. An intervention is suggested that getting those who are short on sleep to have a nap in the afternoon may result in decreased drug use (which may improve memory on 2 accounts then).

All I need now is to convince my supervisor that napping at the desk is a valid use of my time.
photo links to source

Hurt Feelings

2010-02-19T15:41:00.010Z

SciAm Mind has an article on why we can feel physically hurt when we are emotionally hurt. The idea that our feelings are associated with our hearts is commonly accepted (also cards with cartoon limbic systems on them seem less cute). What is the nature of this link though, why do we feel 'heartache' and have 'gut-wrenching' experiences if it's all happening in the brain?

'According to a 2009 study from the University of Arizona and the University of Maryland, activity in a brain region that regulates emotional reactions called the anterior cingulate cortex helps to explain how an emotional insult can trigger a biological cascade. During a particularly stressful experience, the anterior cingulate cortex may respond by increasing the activity of the vagus nerve—the nerve that starts in the brain stem and connects to the neck, chest and abdomen. When the vagus nerve is overstimulated, it can cause pain and nausea.'

The anterior cingulate cortex (ACC) is involved in a wide range of activities and phenomena. It's associated with emotion, empathy, 'high level, rational' processes like error correction, alongside a host of other lower autonomic functions. One of its strangest roles is that mentioned in the SciAm article: the experience and control of pain. Some suffers of chronic and unexplained pain have shown substantial activity in the ACC. However, patients who have undergone a bilateral anterior cingulotomy (lobotomy of both brain hemisphere's ACCs) still experience the pain, they just say that it doesn't bother them as much or that they don't seem to notice it anymore (Cohen et al., 1999). It's a tad unethical but it would be interesting to see how those patients feel in response to events causing emotional pain as described above.

Cohen, R.A., Kaplan, R.F., Zuffante, P. et al. (1999) Alteration of intention and self-initiated action associated with bilateral anterior cingulotomy. Journal of Neuropsychiatry and Clinical Neuroscience 11, 444–453.
photo links to source

Hiding Data

2010-02-15T22:00:00.010Z

I have recently applied for ethical approval of a study on people's changing emotional states. As noted in the application, recording people's day-to-day emotions and moods could start intruding on their lives. Asking people to note down their day-to-day behaviours and how it makes them feel could uncover rather personal details that they may not want others to know about (e.g. Employees aren't going to want their boss to know how they really feel about the company's new mandatory skant dress code).

Data confidentiality is one key ethical concern in social psychology; one of the easiest ways to keep data safe away from prying eyes is to delete it after its use has expired. However, when does data expire? And if the original data isn't there any more, then how can we tell that the research is any good? UEA is currently finding this out for us all with the ongoing, crappily named, climategate situation. So far, the popular answers appear to be: data never expires and predictions are worthless without the original data.

In the future, I may be obliged to delete sensitive data. As a result, the computer models that were built on the data would apparently become highly questionable. I believe that this is clearly not true.

The beautiful thing about science is that it never stops. Data exist all around us begging to be looked at and questioned. Similarly, models, like all other hypotheses, beg to be tested. So, if an original data set for a model has gone that is no reason to question its predictions; the very existence of a model should be enough to do that. If the model is robust, it should cope with new data; if not, the model needs changing. So the question people should be asking UEA is not 'Where is the old data?' but 'Where can we find new data?'

However, those who revel in the UEA situation aren't asking that question. With just a flat dismissal of the present ideas because of an incomplete past we stagnate; for some this is good enough because not doing anything is considered a victory. A popular argument against global warming is that it stopped (or was disproven) in 1998, as if that is the end of the matter. No emissions need to be capped, resource consumption doesn't need to be checked, and we can all buy SUVs. More importantly, no more data needs to be examined because we don't want to disrupt the status quo.

The social influence of the chameleon

2010-02-12T10:01:00.013Z

Bps research digest has just listed the top evidence based tips for improving your* chance at romance this Valentine's weekend. Some of the papers listed are rather strange but generally that's just the product of having an interesting idea.**

One of the papers that caught my eye was the use of mimicry to improve women's speed dating success. We unconsciously mimic behaviours, postures and emotional expression of those we engage with all the time, known as the Chameleon Effect. It appears that on top of this the active mimicry of verbal and/or non-verbal behaviour such as: repeating back phases, ('You really do this?', 'yes, I really do this.') or touching the face shortly after the unknowing participant does so, will result in a greater desire in men to meet them again.

I don't have access to the research paper linked above but Gueguen, Jacob, & Martin's (2009) review highlights the myriad of social interactions affected by the use of mimicry. Participants report liking confederates more if they have been mimicked and report that interaction in their co-operative task went more smoothly and was more harmonious. Even just reading back the table order was enough for a waitress to significantly increase her tips.

Gueguen et al. pose two possible explanations for why mimicry could be so useful in social interaction. Firstly from an evolutionary viewpoint, success in social interaction is a necessary requirement for survival. Humans are required to co-operate in order to cope with the demands of the world; no man is an island, etc. Alternatively, the familiarity theory suggests that mimicry is associated with greater feelings of familiarity and closeness. Familiarity in turn is associated with liking and so mimicry may indirectly enhance the likeability of a person.

In Hatfield, Cacioppo & Rapson's (1994) book Emotional Contagion they describe how liking may impact on mimicry. People who are close tend to be more likely to mimic each other in posture and gesture. They sample from previous narratives that describe how as intimate conversations progress couples are more likely to engage in synchronisation of behaviour. It may be possible that the use of mimicry may help along with the development of intimacy.

Each of the interactions described in the review are positive in nature. Possibly some of the effects of enhanced liking are due to a positive feedback: as one person is mimicked they feel closer to their partner and so increase mimicry themselves. How would a negative interaction be affected by mimicry? Would the feelings of familiarity reduce the negativity between persons? In the review counsellor mimicry of clients arm and leg position enhances client perception of empathy (of course assuming that the client is in a negative affective state). Or could a continual feedback of negative posture and interaction only escalate tensions or negative feelings?

*Particularly if you're male.
**Perhaps one of the best examples of an interesting idea: Langer, Blank, & Chanowitz’s (1978) 'Copy machine study'. People are more likely to comply with a request, if given a reason. Even if it's nonsense.
Gueguen, N., Jacob, C., & Martin, A. (2009) Mimicry in social interaction: Its effect on human judgement and behaviour. European Journal Of Social Sciences, 8(2), 253-259
photo links to source

More than turning a frown upside down.

2010-02-08T15:08:00.010Z

SciAm Mind has a piece today on an inventive means of testing the facial feedback hypothesis (FFH). Facial feedback is the term given to the effects of our expression on our experience of emotions and was perhaps best demonstrated in Strack, Martin and Stepper's (1988) easily replicable experiment. If you hold a pen between your teeth, you partially contract muscles associated with smiling. As a result of this, your experience of amusing cartoons is heightened; in contrast, a pen held between the lips diminishes your experienced amusement to the same cartoons.*

Up until recently experiments on the FFH have mostly been based on people enhancing or suppressing their displayed emotions, which has its complications if you are trying to measure emotional experience. People may try to enhance their current emotion by means such as remembering past events unrelated to the experiment and so start introducing errors into the design.

A recent trend of injecting poison into one's face to become beautiful has enabled a more controlled experimental design to come about. Botox injections paralyse muscles and so should prevent any facial feedback. The particular muscles that are targeted in such treatments are those used for frowning so the FFH for anger or sadness can be examined. The current study looks at how a botox injection can interfere with participants' ability to read emotive sentences.

Previous research indicates that while expressing one emotional state (again through use of the pen in teeth technique) it takes longer to evaluate sentences that are incongruous to how you feel (Havas, Glenberg & Rinck (2007). In today's study participants were unable to display sadness or anger, and presumably therefore feel less angry or sad, in response to the sentences and took longer to read these than controls. Those with botox read sentences that were more positive in content just as easily as controls.

This is not the first time research has been conducted using botox to affect emotions, Henlotter et al (2009): scanned participants while they were viewing pictures of angry or sad faces. A region closely associated with emotional experience, the Amygdala, showed a decrease in activation in botoxed participants. This again points to the existence of a feedback loop, which we may use to help infer our emotional experiences.

* It helps if you aren't aware that this is the purpose of the experiment.
photo links to source

Personality analysis via your tweets.

2010-02-05T15:09:00.007Z

I've somehow found a cool page that will analyse your twitter posts for emotional content. Analyze Words reportedly uses the 'junk' words in sentences such as; pronouns, articles, and prepositions to map out a person's psychological state. This method contrasts a more common sense approach of looking at emotive language or phrases, this process used seems to be more concerned with how people say things rather than what they say.

Across dozens of studies, junk words have proven to be powerful markers of peoples psychological states. When individuals use the word I, for example, they are briefly paying attention to themselves. People experiencing high levels of physical or mental pain automatically orient towards themselves and begin using I-words at higher rates. I-use, then, can reflect signs of depression, stress or insecurity.

There's a whole bank of papers on the site in supporting evidence that I'm still yet to really investigate but as a gut reaction I feel quite sceptical about the whole thing. I tried it out on Monday before and after tweeting and this seems to have greatly shifted my scores. I didn't have the foresight to save the results but I'm fairly sure that my scores on the Upbeat and Personable scales have shot up. This may be due to the use of the word 'we': a word they seem to associate with both scales, so at least the model is consistent.

This week, I have been reading Supercrunchers, which is about the power of mining massive data sets. Ayres claims that in this modern age we leave a digital trail that's so comprehensive our Visa bill can be a reasonable predictor of our likelihood to get divorced. Casinos can interrupt a particularly painful losing streak with a complementary meal at the restaurant so that you leave feeling content with your £600 loss. What's more, they can apparently calculate the maximum amount that you can lose and still come away feeling like you had a positive gambling experience thanks to a 'free' meal.

There is a possibility then that Analyze Words has it figured out and through the mining of data these seemingly insignificant words are indicators to our psychological states. If the program were trained on large enough sets of data some underlying and previously ignored effects may come through. It still might be worth taking your twitter results as an indication of your personality with massive fistfuls of salt though, particularly as one of the social dimensions examined is Spacey/Valley Girl.

photo links to source

Research is exhausting work

2010-02-03T17:35:00.005Z

Not a real post today because I'm rather exhausted from helping out with a study on the effects of witnessing emotional regulation. This part of the study, described in the final paragraph of the linked blog, is the result of a collaboration with a recent departmental visitor who researches into workplace aggression and employee wellbeing.

Recent research indicates that witnessing self-control may have depleting effects on the observer if they imagine themselves in that situation. An earlier post describes that witnessing people exhibiting strong self-restraint may even prime you to be better at self-control. This current research aims to investigate these possible effects in a workplace environment.

The research is being carried out in what would be typically be regarded as a high stress environment and there have been moments where the whole place seems to be buzzing with energy. This, and the rather cramped nature of the work environment, presents situations where emotional regulation is likely to be both frequent and observable.

The workplace structure divides people into distinct group identities, and each group has its own level of authority and power at work. A key part of the study is the how one identifies with those engaging in either workplace aggression or emotional regulation. High identification with those regulating emotions or those on the receiving side of aggression may predict greater levels of exhaustion. Possibly, the perceived power structure affects who each person is more likely to identify with or who is likely to regulate emotions.

Having worked in a similar environment before, I can attest to the draining nature of just being present there. As a low power employee, I was often required to regulate my own and others' emotions to encourage co-operation with my requests. I would frequently see others in my position having to do the same and even more frequently hear about it in the office.

The familiar feeling of exhaustion was back today after spending the day with participants encouraging survey response completion and reimbursing those who had completed theirs. Perhaps I was unconsciously revisiting my prior draining experiences and heavily identifying with my past self. Or maybe just being nice requires a lot of effort from me.

photo links to source.

Use of music and media to regulate emotion

2010-02-01T19:21:00.006Z

One of the key aspects of music is its ability to inspire emotion in us. Appropriate selection of music can be used in experiments to help manipulate our emotional state and in our day-to-day life we can use music to help get us into a desired emotional state. Are there patterns for the use of music to regulate emotions though? Are some uses of music to regulate emotions more effective than others and do they always have the intended effect?

Greeenwood and Long (2009) identify how our mood and our ability to regulate emotions can influence the specific media we use for emotional regulation. It appears that there are three broad categories for media usage: being in a positive mood, being in a negative mood, and feeling bored. In both positive and negative moods music was most frequently used whereas when bored people are more likely to turn to television. In terms of emotional regulation strategy music use is correlated with both positive strategies (e.g. reflecting on events) and negative strategies (e.g. rumination) whereas television use was only associated with poor emotional regulation approaches.

Miranda and Claes (2009) also indicate that music can be used as a facilitator of both positive and negative regulation approaches. In a study looking at the relation between adolescent music use, peer affiliation, and depression, they indicate that levels of depression are linked with use of emotional regulation through music. Their results indicate that problem-oriented coping (e.g. reappraisal of situation) through music listening is associated with lower levels of depression in females. In contrast, using music to avoid the situation is associated with higher levels of female depression.

Greenwood & Long suggest that future research could 'examine what kinds of music selections in negative moods have emotion-regulatory utility and what kinds sustain or exacerbate ruminative thinking'. Miranda and Claes indicate distinctions between five major music clusters that peer networks seem to relate to, one that they examine in detail is the Metal music group. Metal music has previously been linked with depression in adolescents, and can occasionally crop up in the news as a thing for parents to worry about. Given this, it could be seen as a strong candidate as a genre for maladaptive emotional regulation use.

Miranda and Claes report that those in the Metal with peers rated as having high levels of depression are likely to have higher levels of depression themselves. However, those with peers of low levels of depression Metal music preference did not predict depression. They further indicate that peer depression moderates the link between Metal music and female depression and that there is no moderating effect by the above mentioned maladaptive coping.

Music can be used as both an adaptive and maladaptive means of regulating emotions. As the behaviours noted (e.g. re-appraisal, rumination) are well known as adaptive and maladaptive emotional regulation methods, there are no surprises to see that using music to facilitate them changes little. Experimental manipulation of mood could be used to better understand why people seek out particular media for emotional regulation. Awareness of how we use music and media to regulate our emotion could help us identify our personal maladaptive and adaptive strategies.

Greenwood, D.N., & Long, C.R. (2009) Mood specific media use and emotion regulation: Patterns and individual differences. Personality and individual differences 46, 616-621
Miranda, D., & Claes, M. (2009) Music Listening, coping, peer affiliation and depression in adolescence, Psychology of Music, 37(2) 215-233
photo links to source

What the world wants

2010-01-30T18:46:00.012Z

The paper on Thursday indicates that wanting to feel bad, or at least not wanting to feel happier once in a negative mood, is not that uncommon. Running an in-no-way-scientific study of my own I've dug up some numbers to consider.

A Google search of the phrase 'I want to feel happy' finds 1.68M pages, 112K of which are on Blogger. In contrast, the phrase 'I don't want to feel happy' returns 13.1M pages with only 5 on Blogger. These figures change substantially if we slightly tweak the phase. For example, 'I want to be happy' returns 10x as many pages: 17.1M overall with 1.31M from Blogger. What's more 'I don't want to be happy' gives us a massive increase from 5 to 187K on Blogger and the truly staggering figure of 809M sites overall.

Using the same approach we get the following figures for feeling sad:
I want to feel sad 1.28M overall, 115K Blogger.
I want to be sad 23.6M overall, 5.58M Blogger.
I don't want to feel sad 12.2M overall, 2.57M Blogger.
I don't want to be sad 54.6M overall, 14.6M Blogger.

Other permutations of the phrase such as 'I want to be more happy' and 'I don't want to feel sadder' hardly return any sites - possibly because it's beginning to mangle the English language. The only variations with any sizable amount of hits are: 'I want to feel/be happier', again with be far outstripping feel in page-count.

So what does this all tell us? For a start, there is a sizable community out there who don't want to improve their mood (and also a lot of these people go to yahoo answers for psychological advice). Secondly, while people may not want to be happy, a much smaller proportion actually want to be sad - perhaps they are aiming for a more neutral emotional state, such as feeling calm. Finally, and in quite a surprising result, people often talk about being happy or sad rather than feeling happy or sad.

The use of the term being strikes me as a much more permanent state than feeling. In my mind, being implies a central trait of a person, perhaps viewed alongside personality 'I am an extraverted, happy person' rather that transitive experiences that you feel 'I feel hungry and sad'. There are possible implications for emotional regulation here. If people do veiw emotions as relatively stable states, they may believe it is difficult to regulate their emotions and so be unlikely to try and change their feelings, possibly reinforcing that belief.

If we widen the search to look at others' emotions we see some more patterns (e.g. 'I want him to feel happy', 'I don't want her to feel sad'). Firstly, there are 6.35M sites for 'I want him to be happy' and 3.96M for 'I want her to be happy'. Second, there are almost no hits for 'I want him/her to feel/be sad', which is nice, I guess. The phrase 'I don't want him/her to feel happy' is also a dead end but the phrase, 'I don't want him/her to be happy' returns 4.82M opposing his happiness and 3.11M opposing her happiness. Finally, there are a few hits for 'I don't want him/her to be sad', with 400K for him and 212K for her.

Again, we have this contrast between people not wanting others to be happy, while also not specifically wishing them to be sad. Also, the difference between being and feeling emotions is still present. For almost every search, more people want to change males' emotions than females’, although more people want females to feel happy than males (2.62M vs. 129K). This is reversed when looking at wanting people to be happy though.

Of course there are several key flaws with using a Google search as a data set as it lacks any intelligent sifting of the data. Firstly, we have song lyrics included, which millions of sites seem to be dedicated to. Secondly, even if we did find a relevant site, the context of the phrase used is important. The phrase 'I don't want to be happy' would be logged here as a desire for a negative state; however, the phrase could only be a part of a more positive statement: 'I don't want to be happy with a low paying job, which is why I'm enrolling in college.'

In sum, this has been a fun little exercise, which possibly tells you more about my definition of fun than anything about emotions. There's definitely a distinction in numbers between people who regard emotions things you are and those who view emotions as things you feel. This may indicate that people's perceptions and beliefs about emotions are worth considering when looking at emotion regulation. Overall, this exercise best shows that data is really everywhere and with a bit of work there's something interesting just waiting to be found.

photo links to source

Do we deserve to be happy?

2010-01-28T16:42:00.007Z

One assumption that seems very pervasive in the emotional regulation literature is that people want to improve their mood if they are feeling gloomy. However, this general belief may be steering research away from some interesting behaviour. Sometimes people in a gloomy mood might not want to feel better; sometimes people may even want to feel worse than they currently do. For today, we'll look at the first point, highlighted by Wood et al.'s (2009) vignette:

'Holly and Lucy, who share an apartment in Toronto, are each feeling down in the dumps after a bad day at work. To lift her spirits, Holly calls a friend to arrange a night out. She encourages Lucy to join them, pointing out that eating at their favourite restaurant and watching a new comedy movie should make Lucy feel better. Lucy knows that this is true but declines, saying she is "just not in the mood."'

This behaviour directly contradicts the accepted wisdom but doesn't seem like an artificially contrived, implausible scenario. Perhaps there is merit to investigating if our goal of mood improvement is a flexible one. Wood et al. suggest that our motivation for mood improvement may be related to our self-esteem.

Participants of high self-esteem (HSE) and low self-esteem (LSE) underwent a negative mood induction task (thinking of pre-determined negative scenarios). Following this, they then answered questions on mood repair motivation, deserving of the mood, and mood typicality. They found that when in a sad mood, LSEs feel less motivated to improve their mood than HSEs. This is in part due to LSEs feeling less deserving of a more positive mood and also partly because LSEs regard a sad mood to be more typical of themselves. In a control condition of neutral mood induction, there appears to be no difference between LSEs and HSEs motivation for mood regulation.

Subsequent studies in their paper elaborate on the concept of participants deserving a good mood. Deservingness of a good mood can be affected by remembering rejection; while this can affect both groups it affects LSEs more than HSEs. Despite this change in deservingness HSEs will still be motivated to repair mood.

The final study aims to tease apart being in a sad mood and feeling undeserving of a better mood. The prior studies induced a sad mood as a product of the exercise in remembering rejection and so it cannot easily be determined how much just being in a sad mood may affect deservingness. LSE participants were randomly divided in to the sad mood induction condition from study 1 or the sad mood induction and recalling rejection condition. Participants were then asked to rate the desire to watch various mood affecting video clips, which acts as an indication of mood improvement motivation.

While participants in the control and recalled rejection condition both expected that a comedy clips video would elevate their mood*, the recalled rejection group expressed less of a desire to watch the video than those who were just in a sad mood. Wood et al. indicate that the motivation to watch mood improving videos is affected by one's sense of deserving to be in a good mood.

This paper helps in the case to redirect emotional research. If research is built on a narrow presumption, such as people are motivated to feel good, so many valuable ideas aren't even considered. This paper opens up many new considerations, such as: how easily can our motivation to improve our mood be changed, are our goals for mood regulation relatively constant, how effectively can motivation to improve mood be restored?

*with no difference between the groups as to how positive they would feel watching it.
Photo links to source
Wood, J.V., Heimpel, S.A., Manwell, L.A., Whittington, E.J. (2009) This Mood is Familiar and I Don't Deserve to Feel Better Anyway: Mechanisms Underlying Self-Esteem Differences in Motivation to Repair Sad Moods, Journal of Personality and Social Psychology, 96,2,363-380

Don't have nightmares, do sleep well.

2010-01-25T20:39:00.007Z

Dreams are strange things and people seem to place a lot of meaning in them, (is this a cause or effect of the prevalence of psychoanalysis in our popular culture?) Sci-Am Mind highlight a paper by Roberts, Lennings and Heard that offers an understanding of how our dreams could affect us. Popular theory suggests that we dream and even have nightmares as an adaptive means of coping with stress, although, their paper suggests otherwise.

Nightmares have been proposed to act as tension reduction mechanisms that help us 'minimize stress and retain normal [brain] functioning while assimilating emotional reactions to stressful or traumatic life circumstances.' There's evidence to suggest that dreams and nightmares could be a way of processing negative affect; Barrett & Loeffler (1992) found in a sample of 20 depressed and non-depressed women higher levels of depression are associated with lower levels of dreaming.

However, there is a body of evidence to suggest that nightmares are linked to anxiety and may even result in the induction of tension. To test this against the classic approach the researchers measured in 624 children and adolescents: life event distress (e.g. parental divorce), anxiety, and nightmare distress (e.g. fear of falling asleep). A positive relationship between life events and anxiety coupled with a negative relationship between anxiety and nightmare distress would be supportive of the classic approach. In contrast, a positive relationship between nightmare distress and anxiety alongside the expected positive relationship between life events and anxiety would support their hypothesis.

Their results indicate there is a positive relationship between anxiety and nightmare distress. Furthermore, the relationship between nightmare distress and anxiety was stronger than life distress and anxiety. In sum there seems to be no support for the classic tension reduction hypothesis. Further examination of the data indicates that it's not merely the frequency of nightmares that connects them to anxiety but the amount of distress they induce.

If we look at the questionnaire used to identify nightmare distress (NQD, Belicki, 1992) it has such questions as 'Are you ever afraid to fall asleep for fear of having a nightmare?' and 'Do nightmares interfere with the quality of your sleep?' The nightmare distress may be related to elevated anxiety through sleep disturbance. 'People who experience nightmares may be prone to additional problems such as anxiety, depression and social inhibition, ... fatigue, ... cognitive impairments'. Poor sleep or sleep deprivation can interfere with cognitive tasks such as the ability to regulate affect. Nightmares may induce anxiety both by being a source of anxiety and by even impairing mechanisms for coping.

Roberts, J., Lennings, C.J, & Heard, R. (2009) Nightmares, Life Stress and Anxiety: An Examination of Tension Reduction. Dreaming 19(1),17-29
Other citations are found in the above paper
photo links to source

They don't write them like they used to

2010-01-22T19:08:00.004Z

Due to a surplus of emotion, Friday's post has been cancelled. Instead here's by far the best song with 'Emotion' in the title.

Self-control may be contagious

2010-01-21T09:16:00.006Z

Today we come full circle in our look at self-control and contagion, as a new study indicates that self-control itself is contagious. Previous studies along a vaguely similar line have alluded to this (e.g. we are more likely to over-eat when in a group of people doing so). However, this appears to be the first example of actual, and perhaps retrospectively obvious, contagion effects. As a former barman I would often overhear the words to the effect of: Well if you're having another drink I may as well too.

The take away message from these sorts of studies is that our surrounding environment is much more influential on how we act that we would like to admit. While there are important studies that look at self-control in terms our physical state (e.g. Gailliot et al.’s work on the glycogen / self-control relationship), the understanding of our behaviour in the context of the social world is vital too. For example, a student’s performance at university is be dependent on the lottery that is freshmen flatmate allocation (E.g. Nudge p.59).

Vandellen & Hoyle use 5 studies to illustrate how other’s self control can affect our own behaviour; today we’ll just look at studies 1,2, & 5. In the first, participants wrote about either a friend with good or bad self-control and any change between pre & post handgrip duration was recorded. Those who wrote about poor self-control performed significantly worse in the second grip task, whereas there was no change when writing about good self-control.

The second study builds on this by getting participants to observe self-control. Those who watched another participant eat cookies and refrain from eating carrots (low-self control condition) performed worse on a Stroop task in comparison to participants who had watched another eat carrots and refrain from eating cookies (high self control). A surprising result in this study is that those who had to eat the carrots and actually exhibit self-control did not show any difference in Stroop performance from those who got to eat cookies. While this contrasts previous research on depletion of a shared resource, public displays of self-control are considered to require less effort than private self-control.

In the fifth study participants wrote about either a friend with good self-control, poor self-control or one who is moderately extraverted. Participants then were required complete a lexical decision task deciding if a string of letters were words or non-words. The words in the task used were either related to exerting self-control, failing at self-control or neutral in the context. Finally, their persistence on a difficult Remote Associates Test (see 3 words and think of a fourth that links them) was recorded as an indicator of self-control.

As expected from study 1 the writing task predicted persistence at the RAT task. Also, participants who wrote about good self-control were better at recognising words related to exertion of self-control (e.g. achieve) when compared to failure words (e.g. indulge) in the lexical task. Participants who wrote about poor self-control did not show any difference between recognition times of the words. Finally, accessibility of high self-control words predicted performance in the RAT task.

These results suggest that self-control can be dependent on self-regulation accessibility. Social factors may influence the accessibility of regulation – seeing others successfully exhibit restraint can make it easier for us to self-regulate. I imagine this could have useful implications for support groups (from fitness / sporting clubs to addict support groups). 'By exerting self-control, people can increase the likelihood that others around them who take note of this will in turn be more able to exert self-control themselves'. Careful management could possibly result in a positive feedback of self-control within a group.

photo links to source
vanDellen, M. & Hoyle, R. (In Press) Regulatory Accessibility and Social Influences on State Self-Control Personality and Social Psychology Bulletin

Revenge of the Blue Monday

2010-01-18T16:51:00.007Z

Today marks the fifth anniversary of the invention of Blue Monday: a national holiday for science journalists. According to Cliff Arnall and the folks at Sky Travel, a very complicated (and therefore correct!) formula shows that the 3rd Monday of January is the most depressing day of the year. Better people than I have had a go at the formula before, and there's research to indicate that our mood on a Monday is no different from that on any other weekday.

The defence for releasing thinly veiled adverts in the form of science news (e.g. Boffins have worked out the perfect way to spread butter - use this particular brand) is that it helps raise the general public's awareness of science. But if the science is bad, what's the point of making people aware of it? This displaces genuinely interesting science from the headlines. For example: "A sine wave with a period of 7 days accounted for 40% of the variance in the daily mood data".

Our ups and downs in life can be partially accounted for by a very simple formula but because the researchers don't go on to say, and the remaining 60% of the variance depends on what brand of butter you buy, we never hear of it.

Larsen & Kasimatis (1990) Individual differences in entrainment of mood to the weekly calendar, Journal of Personality and Social Psychology. 58(1), 164-171.
Photo links to source - say no to adverts in science. Adverts next to science are ok by me.

Contagion of blame, more than passing the buck

2010-01-16T09:44:00.010Z

The mere observation of one person blaming another can be enough to trigger blaming someone else in an unrelated personal failing. This is thought to occur because observing blame encourages us to form a goal of self-image protection.

As a rule we like to have a positive self-image. While we may see someone fail and assume that's part of who they are*, often we will externalise our own failings to our situation to protect our self-image. This blaming of circumstances or others conveniently protects us from having to acknowledge our own contribution to an outcome we don't like.

Fast and Tiedens argue that the formation of self-protection goals are the mechanism by which blame can become contagious. As goals can be primed by the environment and have been shown to be contagious the causal attributions that arise from these goals may also follow a pattern of contagion. Essentially, we witness one person blaming another, infer their self-protection goal, form our own self-protection goal and are more likely to blame others later. While there has been lots of research on contagion, Fast and Tidens' paper is the first example of causal attributions being contagious.

Their first and second experiments highlight that blame is contagious. Participants read either a news article of The Governator blaming others or taking full responsibility for a proposition failing. Those who read about blame were more likely to make external attributions when writing about a personal failing of their own when compared to those who read about Arnie taking full responsibility. Similar results were found in when reading about students blaming their university for poor job opportunities.

The third and fourth experiments again required participants to read about instances of blame or accepting responsibility. However, rather than looking at tendency to blame others these experiments focus on self-image protection goals. Experiment 3 highlights that people were more likely to develop a self-image protection goal if they had read a passage featuring blame, this was found to be independent of mood and any social learning (e.g. on reading the passage, I feel it is more acceptable to blame others). Finally, experiment 4 indicates that self-affirmation reduces the tendency to blame others. This is possibly because self-affirmation is an alternative means of protecting self-image so participants did not need to protect self-image by blaming others.

The paper points to a new direction for contagion research and also for better understanding what factors affect how we make attributions. It also gives indication of how group leaders ought to manage blame, which would be useful as blame can have the destructive effects of helping us not learn from our mistakes. While accepting responsibility doesn’t seem to have contagious effects it does reduce the likelihood of blame spreading, which could be damaging to an organisation. The study could be further built on by staged events of blame rather than having participants read examples or comparing different organisational cultures against each other.

photo links to source
*Fundamental attribution error
Fast, N.J & Tiedens, L.Z (2010) Blame Contagion: The automatic transfer of self-serving atributions. Journal of Experimental Social Psychology 46, 97-106

"I don't even tell my husband... since my job would put him to shame'

2010-01-13T14:21:00.018Z

How do people cope with working in some of societies least respected jobs? From the mid-1990's onwards China saw a huge growth in a numbers of life insurance agents, a job that ranks in reputation alongside female escort there (Li 2000, cited in Sheng, 2009).

First off, let's look at why life insurance agents get so little respect. The marketing tactics encouraged are aggressive, such as badgering customers until they finally agree to buy insurance. Within the company agents may need to abuse relationships (e.g. steal clients) to meet sales targets. Finally, to continue growth in this sector and as part of mass hiring strategies used, little qualification is needed. In sum, it is a frustrating job having to sell to customers that not only don't want the product but also can be rather hostile on a personal level.

One day last month, I tried to persuade a housewife to buy my life insurance while her son was doing homework. She replied to my persuasion in this way:
"Do not disturb me please! Now I need to help my son to study.
If he cannot study well, then he has to sell life insurance".

As a means of generating a positive working atmosphere that dispels new trainees' negative attitudes towards the job, companies engage in brainwashing (Sheng's description). Activities staff participate in include: regular morning assembly, morning synchronous exercises, and rhythmic applause designed to encourage positive emotions. Coupled with stories of success as a life insurance agent new employees develop high expectations and that they too could be a success. The day-to-day grind of continual rejection from customers and the unglamorous reality of work is a stark contrast from the image presented by the company, this discrepancy can lead to emotional dissonance.

So how do they cope with this discrepancy of having to embrace the positive ethos of the company and a job that can even cause isolation from one's social network? Sheng notes several cognitive strategies used such as: re-appraisal, taking a pragmatic approach, and an idealistic approach to work. Constant, and often rude, rejection by customers can be re-interpreted as an indication of the good nature and persistence of the agent. Pragmatism emphasises the gains from having the job (everything will be better if you become rich) and that negative events are just ignored. Last, idealism gives the job a sense of purpose, unlike pragmatism the focus is on the belief that this job improves customers’ lives and is ultimately altruistic.

However, each of these approaches has its limitations, agents are encouraged to sell through their social network to meet targets (despite this inevitably causing problems when close friends or family refuse) so re-appraisal doesn't repair any lingering tension between parties. A pragmatist would be less concerned by this (I can use the money to make new friends) but of those surveyed just over 20% were satisfied with their income. Finally, it's hard to maintain an idealistic, altruistic view of the career when insurance is only affordable by the richer in society, (idealism is illusionary).

Behavioural means of coping highlighted include: hiding one's profession, consumption, and exploiting weak ties. To avoid the public shame of being a life insurance agent it may simply be easier to lie. Coping can be also achieved by engaging in materialistic consumption. If every success is celebrated by meals and shopping, positive affect will be raised, as will incentive to continue. Last, the exploitation of weak ties* enables a remote social network to be tapped, possibly without the conflict arising seen when dealing with close social contacts.

Again each of these methods have their limitations. Hiding one's profession is a temporary strategy at best; while it may reduce negative feelings associate with the shame of the profession, it is a poor solution given that agents are encouraged to sell to friends and family. Material consumption could serve to destabilise one's affect further and heighten emotional dissonance. As emotional regulation becomes tied to success and material purchases a bad spell may cause prolonged negative emotions that can't be improved without a big sale. Finally, while the use of weak ties greatly expand one's social reach, substantial effort is necessary with no guaranteed return.

Sheng highlights a final, and highly effective, behavioural method of reducing emotional dissonance. 80% of agents quit within 3 months of starting. Rather than having to expend effort coping with the emotional dissonance, agents target the primary cause of the dissonance. The paper's end quote sums the whole issue of agents' well-being rather vividly:

'In much the same manner that firms were able until recently to pollute the air and the water without paying costs of using up these resources, they (firms) continue to be able to pollute our mental health with impunity.'

Photo links to source
*while our friends (strong ties) all know each other forming a shared social network our acquaintances (weak ties) are unlikely to so they are all links to unique social networks. Granovetter (1983) Sociological theory
Sheng (2009) Emotional conflicts and coping strategies: the case of life insurance agents in china, The Copenhagen Journal of Asian Studies, 27(2) 6-39

Offence? No! I was just stirring up debate!

2010-01-11T17:27:00.009Z

A rather unusual billboard advert outside my window has been pulled today, ironically as a proof of its own effectiveness. The ad state quite clearly:

CAREER WOMEN MAKE
BAD MOTHERS have your say

I was angry on two counts - firstly the message and delivery is obviously offensive. Perhaps an equally controversial statement could be Stay at home fathers bankrupt families. Archaic gender roles are still there in the message but now a different group is being slated - one that hasn't had to spend years fighting for equality and is still waiting for it.

Secondly, the context of arguments like this is designed to get me angry. The message is put forward often under the banner of we're just asking questions or we're trying to encourage debate. With the establishing of this as a launch pad, in response to reasonable arguments of why the message is not valid, the message is changed to well there must be some merit to our side otherwise we wouldn't be having this debate. A foothold in public discourse is secured out of a non-existent argument without having to do any tedious work like checking facts or presenting a cogent theory (see intelligent design, climate change 'sceptics').

We were offered to voice our opinion on the billboards statement at a particular website. The site is a poll for opinions on all sorts of matters ranging from the above to the likelihood of England winning the world cup. It's part of a campaign to demonstrate the power of outdoor advertising on online activity. I got angry enough about the message to check out the site so it worked on me. A few others also got angry and have had the ad pulled. It seems that it has proven that getting people angry will drive them online to complain, but not on your website, on ones that matter.

photo links to source

Social Emotions is written before a live studio audience

2010-01-08T15:33:00.010Z

A question has always bothered me as a fan of rubbish TV programs: why do terrible sitcoms always come with an (annoying) laughter track? Presumably it must work; the laughter track must improve the enjoyment of a show otherwise producers wouldn't waste their time with it. That's just the way things are done is a bad, but strangely pervasive, reason when it comes to explaining how the world works. So, is there actual evidence for it working or is the reason for its existence much like this thought experiment?

Leventhal & Mace's work (Mace's undergrad dissertation) indicates that in adolescents the laughter track has effects on both expression and opinion of the film watched. Firstly, participants laughed more with the canned laughter than in a pre-test, as expected. Also, those who showed little expression with the pre-test film later became more favourable to the film with a laughter track. However, 'In every case those subjects who smiled and laughed a great deal on the pre-test became less favourable to the film when they were exposed to canned laughter', they also reported smiling less.

So the results indicate that although participants laughed more with a laughter track some reported finding the film less funny than the control pre-test film. Their explanation for this involved quite a complex interaction. Canned laughter may focus attention to strong responses (laughter) at the expense of weak responses (smiling), those who are highly expressive actively discount their own strong responses. They are laughing more (irrelevant to them) and smiling less (relevant) and so rate the film as worse.

Chapman (1973) indicates that canned laughter increased overt mirth responses in adults but had no effect on the actual ratings of how humorous the audio clips were.

Finally, Porterfield et al. (1988) looks at self-consciousness and response to a laughter track. A higher rating of self-consciousness predicts lower ratings of funniness of clips when accompanied with canned laughter but not without. Self-focus correlated positively with expressed amusement.

As mentioned earlier Mace suggests that '[People with] High control over the onset and termination of their own expressive behaviour... tend to discount their laughter when making cognitive evaluations' (quoted in Chapman '73). Assuming high expressive control and self-focus are related then Porterfield et al's findings complement this well.

So it looks like overall the laughter track has an effect by increasing laughter and in some improving the opinion of the show. This comes at the price of others finding the show less humorous and becoming bitter about having to pay the TV licence, but at least they're laughing too.

photo links to source
Leventhal, H. & Mace, W. M. (1970). The effect of laughter on evaluation of a slapstick movie Journal of Personality
Chapman, A. (1973) Funniness of Jokes, Canned Laughter and Recall Performance Sociometry
Porterfield, A.L., et al. (1988) Private self-consciousness, canned laughter, and responses to humorous stimuli. Journal of Research in Personality. 22(4),409-423. 36(4) 569-578

Ghosts aren't real / I'll use a sword to fight the dragon

2010-01-06T20:28:00.005Z

Bps research digest has a cool developmental piece today looking at the emotional regulation behaviours of young children for coping with fear inducing scenarios. 4, 5 & 7 year olds were asked to imagine themselves in a given scenario, indicate how fearful they (and sometimes others) would be, and how they would cope with the fear.

Children were given short pictorial stories of scary scenarios such as 'You are having a picnic at a campsite with mummy when you see something behind a tree. You think it's a bear.' or 'You are in your house, next to the stairs. You see something near the ceiling and think it is a ghost.'

They were able to identify that peoples fear reaction varies with age and gender: adults are generally less scared than children and daddy would be even less scared than mummy. This recognition of a difference in reactions to the same scary object becomes more pronounced as children age. (Parallel with a developing understanding of theory of mind?)

Coping strategies broadly divided into two categories of cognitive (Think "it’s just a scare-crow") and behavioural (Go see what [threatening silhouette] really is) as a means of reducing fear. Behavioural strategy was further divided into avoidance (Look for my mom and dad) and engagement (Slice the alligator to pieces). Avoidance was preferred by girls while the approach and often violent engagement of the object preferred by boys.

An interesting and key part of the paper is another division of regulation categories into reality affirmation (Ghosts aren't real) and positive pretence (I’ll use a sword to ﬁght the dragon). Older children were more likely to pick reality affirmation than positive pretence while younger children preferred the latter. Even though the younger children chose to interact with the imaginary creatures further questions indicated that they knew that they were only imaginary.

A measure of the effectiveness of the various coping strategies across the ages would be rather neat as different regulation strategies vary in effectiveness in adults (e.g. Gross's division of antecedent vs response focused). Sayfan & Lagattuta pose this thought as further research, although how you ethically get the children to test the effectiveness of fear regulation methods I don't know. They also tentatively offer effectiveness as explanation as to why younger children use cognitive or reality affirming strategies less often: younger children just view these methods as less effective.

In my own experience I can remember not being able to sleep after watching Jurassic Park and my parents' reassurances that dinosaurs weren't around anymore was of no help. However, my piling of stuffed toys against the door was enough to allay my fears as the raptors wouldn't be able to get in. I think this had more to do with the risk of my parents being wrong as those dinosaurs were pretty convincing.

Photo links to source
Sayfan L., & Lagattuta K.H. (2009). Scaring the monster away: what children know about managing fears of real and imaginary creatures. Child development, 80 (6), 1756-74

Resolution: Be perfect at everthing

2010-01-04T15:03:00.006Z

The New Year is upon us and with it come the annual stern warnings that resolutions are only ever made to be broken and make us feel bad. The Guardian had a piece last week on Robert Wiseman's research that pins down the reasons for failing when it comes to our annual self-improvement ritual.

He points out the main pitfalls that most people experience, such as focussing on the downside of failure or failing to effectively plan out how to achieve these goals. In contrast the secrets to success are to break the goals into small steps, reward achievement and tell people about the resolutions.

Wiseman's statement, "Failing to achieve your ambitions is often psychologically harmful because it can rob people of a sense of self control." highlights the importance of making sure that resolutions work. Not only does a broken resolution undo any work put in but can also lead to a situation where one is worse off than if they hadn't made any resolutions.

One of the key problems with resolutions that the Guardian barely touches upon is the grouping of several resolutions together under the guise of a super resolution. The resolution of improving health may require changing diet, exercising, cutting down on alcohol and quitting smoking. With so much on one plate (except the dinner plate) failure is almost inevitable as self-control is a limited resource.

Often termed the "what the hell" effect the first chink in the armour leads to the whole dam bursting*. When we finally give in to temptation and eat one tiny slice of cheesecake, we think 'Ok the diet restarts properly tomorrow' and before we know it the whole cheesecake has gone because somehow those calories don't count. In terms of getting healthy we may find ourselves overindulging in beer and smoking again because we forgot to go to the gym.

*Mix those metaphors!
Photo links to source

The cookie task

2009-12-18T12:18:00.005Z

Another quick video tailored for the lay audience in the mini-lecture series on managing emotions. This time, a hero of mine demonstrates the strength-model of self-regulation. The effortful task of delayed gratification of a relatively small reward leads to spectacular failure at self-restraint when facing a big delayed gratification task later on. The whole process is described as "very emotionally draining".

Others in the series are mentioned here and here

Internet hero worship

2009-12-02T16:38:00.004Z

While I'm on the topic of ancient web sites about emotion I have to bring up an old favourite that has been around as long as I can remember. The site was founded in 1998 and it seems that Eric still updates occasionally. It's quite impressive that an amazingly simple premise has such longevity.

What the world feels

2009-11-30T11:09:00.005Z

I've just found a really cool site* that trawls blogs for the words 'I feel ...' and arranges these in various stylish means. The applet lets you view a scrolling news feed of peoples' various feelings, and thanks to the need to attach a personal profile to every site you ever join the site can map out blogger demographics such as: the location, age, gender, and even the weather at the time of the feeling. This site has apparently been around for a very long time so it may be old news to most.

An interesting quirk of the program is that it collects any sentence with 'I feel ...' in it and tries to fill in the next word for the 'mounds' graphs (giant jellies proportional to the frequency of use). So while the number one feeling for today is 'better', apparently 20 people feel 'umbrella'. This presumably exists as a combination of people writing in metaphors and the webbot trying to pick out something to attach to 'I feel ...' (e.g. I did feel a bit like Mary Poppins without an umbrella)

This is is probably a dangerous waste of time for me because each line links back to the original blog entry and I love being nosy. Also, I feel a bit like I'm snooping on other people's diaries**.

*click the we feel fine applet link in the 1st paragraph
**I feel that I may be trying too hard to get a line on the site.

I know you're happy, but what am I?

2009-11-20T13:20:00.008Z

I've just been sent a link to yesterday's article in THE reporting that a recent ESRC, MRC & publicly funded project on subjective well-being has stirred up some controversy. The project is a response to the rising influence of subjective well-being on public policy and services and calls for "research to underpin public policy development that will enhance the well-being of citizens". As this research falls in times of recession, THE state that this work will look at how "the nature of work - "good jobs" versus "bad jobs", different types of unemployment and job security - affects individuals' personal satisfaction

It seems that academics have been anonymously* objecting to the project, and the promotion of subjective well-being in general, on the grounds that it encourages complacency and self-delusion in the masses (quite how serious this complaint is, I don't know). In any case, the THE brings up Brave New World in which the intellectually inferior Epsilons self-affirm that they are glad they are the lowest class because the highest Alphas have so much complicated work to do. Only by being on the bottom rung of society doing menial tasks are Epsilons truly happy.

But is the improvement of well-being really sinister enough to necessitate bringing in Huxley? What's the alternative to improving peoples' subjective well-being - leave things as they are and hope that losing a job makes a man feel bad enough that he goes out looking for another to improve his well-being? It's certainly how a rational actor would behave but mere humans may have maladaptive mechanisms for improving perceived happiness forsaking their 'objective' well-being (it's why credit cards exist).

It comes back to questions far beyond my capacity of answering: Am I really happy? Do I just think that I'm happy? Would I know if there was a difference?

* Note: If the THE quote a statement made under ones full name on a public message board they are not exactly protecting anonymity
photo links to source
The ESRC pay me just enough money for me to never question their motives for anything.

Panic! at the disco

2009-10-23T11:00:00.010+01:00

I've just bought a cool book on the exceptionally unusual behaviour that groups can generate. Outbreak! The encyclopaedia of extraordinary social behaviour is a weighty tome cataloguing the reported incidents of strange goings on in groups throughout history ranging from the Salem witch trials, to the famous War of the Worlds broadcast that caused widespread panic in the streets, to doomsday prophesies. I first saw the book on the Mind Hacks page and Vaughn Bell's review is on the back cover.

An encyclopaedia is always going to be a very dry read, but some of the shorter entries lend to a rather cool 'mystery novel' feel:

"Derby, England: 1905
Over five days during May of 1905, 45 pupils at a girl's school in Derby, England, exhibited fits of screaming and falling unconscious. They were described as weak and had to be carried home to rest. Suspecting a noxious gas or vapor was responsible, authorities placed mice in the classrooms, but they were unaffected. No further details could be ascertained."

Of course, this book had the potential to compare the past to the present and define the unusual behaviour in terms of what we know now to be 'normal behaviour'. The authors have done a much more thorough job than that by citing newspaper and police reports from the time to indicate that even within the context of the time these social behaviours were unusual. And even if we were feeling complacent in these enlightened times the book catalogues the 'Mad Cow' scare (1985 onwards), irradiated mail scare (2002, usa) and many modern conspiracy theories, highlighting that strange behaviour in collectives is not a relic of the past.

My personal favourite example of the contagion of unusual behaviour is seen in a small Romanian town that witnessed the beginnings of an alien invasion on the horizon. The strange coloured flashing lights that had caused mass evacuation, turned out to be a disco from a nearby town. This occurred in 2004.

Detecting Emotion from Music

2009-10-21T13:06:00.003+01:00

Another paper from the Scacsnet workshop at Pro-Ve conference today. This time we are looking at a proposed program which can detect the emotional content from music.

I don't think I need to tell you that music can be incredibly influential on our moods. An appropriate piece of music can really make a scene in a film stand out - a slow swell in the strings section can bring us hope, can raise suspense, or can break our hearts depending on the subtle differences between pieces. It serves as a useful aim then, to be able to reverse engineer the process and be able to map feelings to tailor a music score to guarantee the desired reaction.

This is most likely a way off but the process described in Wang, Lv, and Shao's paper has potential. Their work appears to be incredibly reductionist, in an attempt to pick out emotional responses to particular musical features (such as key, time-signature, instrument, pitch, intensity, speed - 31 in all). These features are picked out from short excerpts from classical midi pieces. In the era of broadband it may be worth reminding you what arguably one of the greatest pieces of music ever as a midi file sounds like. It's not entirely clear but it seems that the music was then further reduced to the main melody playing the main theme of the piece (the bit every one recognises) .

Participants rated each of the 224 themes as either depressive, tender, calming, happy, or passionate and then a variety of algorithms are compared in their results for picking out common features for the themes (for example if the majority of the themes rated as tender share a particular key we can start to associate the two). Results indicate that of the 5 states, the algorithms are better at picking out commonalities in passionate and calming music and that the features that seem to be carrying the most information are the speed and intensity.

It's not exactly a shocking finding - my miscellaneous music folders are divided into upbeat and relaxing music; this categorisation is broadly based on speed and intensity. Nevertheless, this may be a promising line of work. I only hope for their participants' sake they switch from midi to mp3 soon.

Wang, Lv, Shao (2009) Detect Emotion from Music, a Comparative Study. Presented at SCACSNET workshop, PRO-VE'09, Thessaloniki
photo links to source

You may already be a winner!

2009-10-19T14:07:00.011+01:00

Science takes an unusual turn this week with the presentation of a conference paper on the emotion elicited by popup advertising. Without doubt popup advertising is annoying and it seems an unusual way of getting people to click the link. The last thing I would want is to have people associate with any work of mine is being annoyed.

One key underlying mechanism advertising relies on is the mere-exposure-effect. Quite simply, the more you see a stimulus the more you like it, even if your initial reaction to the stimulus was negative*. However, just seeing a stimulus all the time doesn't necessarily mean we will grow to like it. If we are attending to one stimulus and having to actively ignore a distractor our liking for the distractor drops - known as the distractor-devaluation process.

The paper reports that the repetitive presentation of one stimulus as a distractor does not improve a person's liking of it, implying that the distractor-devaluation process has a greater effect than mere-exposure.

This can be tested in the naturalistic environment of participants surfing the internet and having pop-up adverts interfere, participants' liking of the advertised products are then assessed against controls. Or rather this could have been tested had pop-ups not been regarded as so annoying that browsers now have built in blockers. Does that still count as evidence for the distractor-devaluation process?

*The Lady Gaga effect
Panagiotidou & Vivas (2009) Visual attention and Emotion: The distractor devaluation effect in online advertising. Presented at SCACSNET workshop, PRO-VE'09, Thessaloniki

A journey into the uncanny valley

2009-10-14T10:46:00.006+01:00

Today's TED talk briefly covers the new generation of emotional robots, which incorporate accurate recording of a human's expression with faithful mechanical replication of the facial muscle movements.

Previous robots such as Kismet look quite unrealistic but still convey emotion well enough to really engage human partners. The new generation of robots have been wrapped in a rubber like material -frubber- that makes them look very human. The live demo of the Einstein model is somewhat unnerving and the model looks like it belongs in a horror movie set in Madame Tussauds where the waxworks come alive.

The goal of the project is to build robots that can empathise with humans*, this is obviously a long way off but successful and convincing mimicry of emotion is a good first step.

*and therefore halt their inevitable uprising
photo links to source

Calm Down, Calm Down

2009-10-12T16:14:00.002+01:00

New research reported in The Onion indicates that successful interpersonal emotion regulation may be more difficult to achieve than expected. Specifically, use of the phrases "take it easy" or "just mellow out" are only likely to exacerbate the situation. The majority of uses of "calm down" were considered ineffective (although this has been known for sometime)

Photo links to source. Alright!?

Hot headed arguments

2009-09-28T16:04:00.010+01:00

Today's report in the Guardian on climate change highlights a very worrying trend that seems almost unstoppable - bizarre arguments in the comments sections. Such 'controversial topics' as these spark off massive and circular debates about whether empirically observable events are real or part of a vast, evil, left-wing conspiracy to get us all eating healthily and taking public transport.

I think it comes down to this: people make a general assumption that anything they haven't tried before can't be all that hard to do. Everyone can run the England football team better than who ever may have just led us to an embarrassing defeat on penalties. The man on the street could run the country better than those clowns in the government because he would actually use some common sense for once. Editorials argue if global warming is real why is it cold outside?. People know that the latest warnings from the met office are probably wrong because it couldn't even predict that this summer was going to be crap.

I'm not going to venture an opinion about the reality of global warming because, in comparison to all the data and research done by others far more qualified than I, I know basically nothing about it. However, I do know a bit about making predictions using computer models and that in some cases a long term prediction could actually be far more accurate than a medium term prediction. Instead of global warming, let's get all warm and fuzzy and talk about our feelings.

For the purpose of this exercise, we have 3 main categories of feeling: our emotions, our moods and our overall affective wellbeing (sometimes described as general disposition). These have a great deal of influence and overlap on each other and many people have tried separating these terms out. In this example we shall say: emotions are short-term, moods are kind of middling, and affective wellbeing is long-term.

With a few hourly measurements of emotion (& barring any unforeseen circumstances) it is quite possible to make some accurate predictions of that person's emotional state for the remainder of the day and the next and possibly even the next. However, things start to break down after a few days because underlying fluctuations in mood are going start colouring emotional responses to events and messing things up. We can't use short term data to accurately predict medium term data this way; we can no better predict our specific emotional state will be on Monday 19th October 11:00 than we could predict the weather for that time.

If we were to space out our regular measurements of emotion across days rather than hours we can start to do a little better. If we fit our short term model to this longer term data, it still won't be able to predict the person's specific emotional state for very much outside the range of data because moods start messing things up again. But using the longer term data we can start to pick out patterns in mood changes and we can gather a general indication of the underlying trend of how the person's feelings change. In this case the underlying trend is affective wellbeing and unlikely to vary much (according to setpoint theory).*

We can start to make reasonable predictions about how the person will generally be feeling in a years time (again assuming no drastic life changing events). It won't be spot on but the more general long term prediction would probably be a safer bet than any specific medium term prediction.

This all seems very common sense and quite a pointless exercise but replace emotion, mood and affective wellbeing with weather, seasons and climate and stretch the timescale then suddenly computer modelling no longer becomes an obvious means of predicting future trends but propaganda from the shadowy cabal run by Al Gore.

*perhaps a bad example to highlight climate change predictions but its all I've got
photo links to source

So very Alone in the Wild

2009-09-25T09:48:00.010+01:00

Alone in the wild, an intimate and somewhat alarming documentary about Ed Wardle's journey in the Yukon wilderness, just finished this week. After 50 days in the wilderness solitude itself looks as dangerous to Ed as the bears.

Firstly there are the highly distorted sleep and eating schedules that impact on the body: deregulated sleep patterns cause exhaustion, can play havoc with emotional stability and the ability to cope with stress. The minimal amounts of food available to Ed only further impact on this as well as dropping his heart rate to worrying low levels. Even after a supply drop and Ed is no longer in a physically dangerous situation the solitude has become too much and he calls in the rescue plane.

We are social creatures and without contact things can start to get a bit unusual, such as hearing voices or having imaginary conversations (Ed describes having a long and touching conversation with his girlfriend on an imaginary phone). The final scenes show Ed back in a hotel, readapting to an environment more suited to humans. With the stress and exhaustion his emotions come to the surface and he just starts laughing at himself in the mirror. He probably just needs some time alone to reflect on what he's been through.

Lots of pieces on solitude at the channel 4 site, with people retelling their experiences in some of the most extreme examples of being alone.

Photo of the Yukon river links to source

Fear of the Dark

2009-09-23T11:17:00.005+01:00

Sci am report that scary music is more effective when we listen with our eyes shut. We take in so much information about the world through our eyes.

As a quick demonstration, try walking in an open space with your eyes shut*; within a few paces your walking speed dramatically slows and your arms will be out in front of you trying to feel for obstacles, even though you know that all's clear.

Our brains are evidence gatherers and when one key source of evidence is absent we try to gain as much as we can from other sources. I suspect that this is what's happening in the above article: our brains hear scary music but see that everything is safe and so put relatively little weight on the sound input. When we no longer have that information that everything is ok, the scary music becomes much more important to the brain and is regarded to be more scary.

As the paper is an fMRI study, it mostly focuses on the difference in frontal and amygdala activity under the varying conditions of neutral or scary music and eyes open, closed or watching a scrambled film. Emotional evaluation responses were also recorded a while after the study, which shows that this is not just extrapolation of higher amygdala activation must mean greater emotional experience. Participants' felt responses to the negative music was dependent on eyes being opened or closed.

*Safety first! Try in an open field.
photo links to source

Smiles save lives

2009-09-21T14:35:00.000+01:00

District 9 is a rather intense film that is definitely worth watching. What struck me is how well the premise is presented: aliens (described as stupid prawns by most humans) live on earth as refugees, they have done for a while now and it's up to a contractor to forcibly relocate them further from the city - there's something very everyday about the whole thing and so it retains its believability.

Because I never switch off from work mode and actually enjoy anything I caught a neat little throwaway line as the relocation team prepare to enter the alien slums. The armed team have been briefed about potential violent encounters during the operation and a tannoy announces "Remember employees - a smile is cheaper than a bullet". It's a good piece of detail that just adds to the realism of the situation - the company expects agression so it should be met with a polite calm smile as to not further any problems. In contrast, if they were expecting a relatively calm situation a bit of aggression may be encouraged to try and induce anxiety and thus cooperation or at least conformity with demands.

photo links to source

Running on pure emotion

2009-09-18T11:15:00.006+01:00

Once again emotion regulation is back in the news. Adebayor's insight into emotion regulation has spawned quite a bit of chatter about the office regarding theories of emotion regulation. He does make a good point that emotions shouldn't be considered as singular events and they may well last longer than 10 seconds. He goes on to say "People say I had time to think because I ran the length of the pitch but that is not true" and essentially argues that he was unable to self-regulate because he was caught up in the emotion.

Anyway, it does raise an interesting question - does regulation ability drop as the minutes tick by in a match? Viewing self-regulation as a limited resource, which is consumed throughout the match, would predict that this is the case. A relatively simple measure of this would be to check the distribution of yellow and red cards in a match and uses this as a proxy measure for self control. A negative skew on a density plot would indicate that players are more likely to be have committed a bookable offence - possibly as a result of diminished self-control.

Despite the muddled plotting, Nunes and Sousa indicate that there is a definite skew towards later in the match for bookable offences. The solid line for red cards shows a substantial peak in the dying minutes. I think the dotted line (yellows) and thick dot dash (2 yellows) are mislabeled as they state that double yellows also peak at the end of the match while remaining less likely than the straight red in the first half. Also the tendency to make half time substitutions is picked out quite nicely in the data.

Of course there are many more factors to this than just lapses in self-control such as increased pressure towards the end of the game, physical exhaustion leading to clumsy tackles etc. My officemate suggested that refs could go into a match with a quota of cards to be awarded in mind. If they haven't met their quota by the 85th minute then they card people for minor infractions that wouldn't be worth stopping play for in the 20th minute. Quite a creative idea for someone who just asked me what a yellow card means.

Nunes, S. and Sousa, M. (2006) Applying Data Mining Techniques to Football Data from European Championships. 1st Conference of Methodologies of Scientific Enquiry, Portugal
photo links to original

I'm real happy for you Joe Wilson but Kanye had the best emotional outburst of the month

2009-09-16T17:24:00.009+01:00

Emotion regulation is back in the news again this week with Kanye West rushing the stage to interrupt Taylor Swift's award for best female video. The Guardian have helpfully collected his recent emotional outbursts so why people are shocked that he interrupted a speech is a mystery.

A blog at the Guardian yesterday claims that the whole thing may have been a set up to promote the involved artists. Although, given his history and the emotional interview with Leno, it may just be that he's poor at holding back. I still can't see how Beyonce's video was the best of all time or how anyone could get so worked up about it.

In any case the incident has spawned a fantastic youtube effort combining recent health-care reform disruptions and people reportedly behaving like a jackass.

*A bad week for photos

"You lie"

2009-09-14T10:47:00.008+01:00

I may have missed the boat with this post about a recent emotional outburst making the news but it's worth exploring anyway. On Wednesday president Obama gave a speech to congress which has made the headlines for not so much what he said but what Rep. Joe Wilson, South Carolina(R) said. In what appears to an unprecedented event, Wilson interrupted the speech with a clear and impassioned "You lie!". This was specifically in response to the claim that illegal immigrants will not be covered for treatment at the tax payers' expense under the new system.

What has sparked my interest is the aftermath of the event and the various statements Wilson has made. In a statement made shortly after the speech: "I let my emotions get the best of me" which I can understand. If he is listening to a speech about a topic he is passionate about and feels that the facts are being misrepresented he could quite easily snap and heckle. However, in a statement reported on the 11th he claims not to have snapped. "... if I'd snapped I would have kept on objecting, but I didn't." I'm not quite sure how these two statements are compatible.

So did he snap or not? Did his emotions get the better of him and he couldn't keep quiet any longer? Or, as those reading his twitter feed have suggested, did he plan to interrupt? The reported tweet: "Happy Labor Day! Wonderful parade at Chapin, many people called out to oppose Obamacare which I assured them would be relayed tomorrow to DC" would certainly imply some premeditation. Obviously there's no way of knowing; perhaps he knew what he was going to say but left it up to the moment to decide if he was going through with it.

In any case, the media seems to have focussed on him rather than the nuances of the bill, and a massive fundraising effort for him is in full swing. (However, there's an even more impressive fundraising effort for his Democratic opponent). If it was premeditated the plan went off (almost) without a hitch, if it was an outburst it shows just how costly self regulation failures can be.

Can't think of a photo / can't use the photo of the incident.
*Yes all the links are to CNN but would you have preferred foxnews?

If early birds catch worms, what do night owls catch?

2009-09-11T16:11:00.008+01:00

There's a quick post in Sci Am today, which highlights some of the differences between early birds and night owls. As well as the obvious differences (one of the two groups prefer to wake up much earlier than the other) the article describes that night owls, better perform at attention / alertness tasks later on in their sleep / wake cycle. Although, I can't find a statement in the original paper to this effect; in fact they state that they didn't find any difference in Stroop performance.

The paper does mention subjective feelings of alertness higher in night owls than early birds after 10.5 hours awake. This is not, as most of the comments seem to argue, merely that night owls are more alert later in the day (obvious) but that they are more alert relative to early birds when time since sleep is accounted for. A person sleeping from 3am to 11am is more alert at 9pm than a person sleeping from 10pm to 6am is at 4pm.

Our alertness can be seen as a function of multiple processes, such as our circadian cycle, and the time since sleep. As time since sleep increases then our desire to sleep, in the article termed sleep pressure, also increases. On top of this the circadian and ultradian cycles affect our alertness, such as experiencing an afternoon lull in energy.

What makes this new paper interesting is the possible explanations for why late sleepers feel more alert 10.5 hours after waking. The authors suggest that this may be due to "previously reported chronotypical differences in the vulnerability to increasing time spent awake and/or in the strength of circadian arousal signal in the evening". Essentially, it could be an effect of desynchronised circadian and waking cycles or, perhaps more interestingly, that night owls just get less tired throughout the day. Next question is can this extended alertness be developed / trained?

Photo links to original
Schmidt, C. et al. (2009) NeuroImage, S187 Organization for Human Brain Mapping 2009 Annual Meeting

The Myth of Blue Monday

2009-09-07T12:33:00.006+01:00

Much like the USA's 20th president, I am no fan of Mondays and definitely consider them the worst day of the week. I am wrong on this matter and apparently have been wrong for quite some time. Stone et al. (1985) conducted a diary study to test a hypothesis that's so 'obvious' no one else would have considered to doubt - Monday is the worst day of the week.

To test this they asked 46 married couples to keep mood diaries recording for 90 consecutive days. Husbands were asked to fill out how closely a list of adjectives applied and wives were asked to report their perception of their husband's moods. The data indicates that our moods are no worse on Monday than any other weekday. Significant variations found in the mean mood are only due to our moods improving with the arrival of the weekend.

To check that they hadn't picked out a bizarre pocket of the population that eagerly anticipates Mondays, Stone et al followed up with a survey asking participants on which day of the week their mood was worst. 65% reported that they believed it was Monday. To further check that individual responses weren't being washed out in the mean scores they examined the individual results of those who believed Monday was the worst day. The definitive worst day for the 65% was pretty evenly split across the weekdays.

Stone et al. close with a very diplomatic "the blue Monday phenomenon would appear to be the result of non-empirical evaluations".

photo links to source
Stone, A., Hedges, S., Neale, J. & Satin, M. (1985) Prospective and cross-sectional mood reports offer no evidence of a "Blue Monday" phenomenon

Emotions, Risk & Money

2009-09-04T17:20:00.007+01:00

Yesterday Slate published an interesting article describing how the contagion of moods may affect our decision-making and risk-taking behaviour. As Slate remind us, risk-taking behaviour can be linked with the current fluctuations in the economy. In good times we are all feeling positive we are more likely to take chances than during tougher times when we become risk averse.

The column features insights from two figures in the emotion contagion field, Peter Totterdell explains the processes by which emotional contagion can occur and Sigal Barsade describes how this could have an impact on the economy. Emotion contagion, at it's simplest, can be demonstrated by two people sitting silently facing each other in a room. Over time their moods will converge, usually closer towards the more expressive of the two people (cite). This is thought to be a function of our tendency to mimic each other and the existence of facial feedback (our expressions, as well as reflecting our emotions, can influence them).

“Bubbles come from a type of emotional contagion,” Barsade explains. “Everyone said, ‘This is great, let’s go buy these houses, let’s do whatever!’ And they didn’t pay enough attention to their personal situations.”

This can be seen as a positive feedback loop, where one particular behaviour encourages more of the same. The article also highlights the effects of the media on people's mood and their subsequent behaviour. As we kept receiving messages that the economy was shrinking we became cautious and collectively tightened our belts. This reduced the flow of money and so the credit crunch effectively became a self-fulfilling prophecy. Of course I'm not suggesting that the media caused the credit crunch, and neither does the article, but it could be seen as a contributing process.

Photo links to source
Disclosure: Peter Totterdell is head of the EROS research group. I am a PhD student also part of EROS

R.E.M. Rapid Emotion Modification

2009-09-02T16:28:00.010+01:00

A quick video demonstrating that our emotions can rapidly shift in line with the mood of the group. This is part of the informal mini lecture series described in an earlier post.

End of Summer

2009-08-26T16:25:00.006+01:00

500 days of summer is a rather good film that is chopped up and mixed around to create a very good film. The story follows a young lad Tom's infatuation and relationship with Summer across a selection of key moments in their lives, lurching from one day to the next in no particular chronology. Without spoiling too much, the relationship does not work out and much of the film is spent on Tom's attempted winning of Summer back.

Several of the scenes are shown multiple times in different contexts, which really highlights how we can look back on our past and see what we want to see. In a sense this indicates the cognitive appraisal process of emotion. How we choose to interpret a situation can affect how we feel about it. Of course, this is an insufficient means of explaining all emotive responses (e.g. initial fear reaction to harmless spiders) but appraisal of the situation can modify response to the stimulus (e.g. catching and releasing the spider rather than running away). There are a couple of moments in the film where re-appraisal of situations dramatically alter Tom's emotional response to the same emotion eliciting stimulus.

The film has been getting quite mixed reviews with many of them highlighting that it's a rehashing of quite a tired formula but I thought their relationship was handled in a more interesting way than most traditional rom-coms. (Note: The Proposal wasn't worth writing about - the Film 4 review is dead on)

Photo links to source
*I do watch other sorts of films, honest

Losing Control

2009-08-24T16:20:00.008+01:00

I often find myself going through phases of high productivity and phases of less than optimum work output. I am reasonably sure that I am not alone in this. Low points in working are often accompanied by poor monitoring and poor goal setting - ultimately poor self-regulation. Baumeister, Heatherton & Tice's book Losing Control neatly describes this behaviour:

Each violation of standards brings negative affect, which makes it unpleasant to be self-aware so the person avoids monitoring his or her own behaviour, which makes further violations possible. The longer this goes on, the more unpleasant it is to resume monitoring oneself, because one must recognize that one has severely violated one's desired patterns of behaviour.

The cycle is presented as self-perpetuating leading to ever increasing disparity between desired behaviour and actual behaviour. A person feels blue because they have failed their diet and so resorts to comfort eating - behaviour substantially differs from the standard. The above effect could also be achieved by shifting the standard rather than letting the behaviour slip. Belief of failing to keep to a diet may lead to the self-imposition of an even stricter diet - one that would require even more willpower to sustain. Failings would become more frequent and the diet is ineffective.*

photo links to source
*replace diet with work and comfort eating with computer games and you have a good estimate of last week
Baumeister Heatherton & Tice (1994) Losing control: How and why people fail at self-regulation.

Running on Empty

2009-08-21T16:42:00.013+01:00

So far we have mainly looked at self-regulation as a function of an internal energy state. Repeated exercise of self-control can lead to poorer performance in later tasks involving regulation. However, the ability to regulate may depend on more than just mental resource described in literature on ego-depletion. A 3-part study by Alberts et al. indicates that implicit self-control beliefs may be able to overcome some effects of ego depletion.

They argue that self-control performance may be affected by the thinking of self-control in terms of a limited resource or as motivation. Essentially, if people believe they will get tired after exerting self-control then they will experience this effect. They propose thatself-control performance might not be a matter of energy alone, but also of implicit theories and expectancies about self-control. Depleted participants primed with perseverance cues outperform those who experience a neutral implicit prime when it comes to another self-regulation task. Likewise, a negative prime will reduce subsequent self-regulation ability.

Their findings may augment the ego-depletion model but first it raises lots of questions. Does priming discourage conservation behaviour so that we use that last bit of resource we were saving (pushing on regardless of what the fuel light says)? Does perseverance priming increase the amount of resource (Refuelling)? How much resource do we actually have?

Alberts et al. make suggestions for these questions. Firstly, it doesn't appear to be a case of using up your last remaining resource reserves as studies 2 (persevere) and 3 (give up) combine to indicate that those who persevere in an early task in a series do not differ in behaviour on the very last task from those who were primed to give up on the early task. If reserve resources were used up it would be expected that those who initially persevered are going to run dry before those that gave up.

Reserve resources could have always not been fully depleted for the experiments, but when are resources fully used up? Alberts. et al indicates that trouble of actually falsifying the nature of the resource without the aid of specific values to test. As for priming as refuelling resource levels - this is left to be tested later, if someone has the motivation.

photo links to source
Alberts et al. Carrying on or giving in: The role of implicit cognition in overcoming ego-depletion

Emotions prime out-group bias

2009-08-19T16:39:00.005+01:00

Today we have another paper from the recent issue of Emotion. Dasgupta, Desteno and Williams run a neat series of 3 implicit association tests (IATs) to highlight how emotional states can influence prejudices held against minority groups. This work builds on previous research on how people use heuristics to process information about out-groups. They hypothesise that while previous research has shown that some emotions (anger, happiness) promote the use of heuristics, emotions will only have an effect on evaulations of minority out-groups if they are relevant to perceived expectations of the group.

The minority out-groups used in the IAT were for study 2, homosexuals, and for study 3, Arabs and with the emotional priming of disgust and anger. Dasgupta et al. suggest that for many Americans, homosexuality is perceived as violating moral values and therefore associated with disgust. They also state that 'several past studies show that Arabs as a group consistently elicit anger-based emotional responses'.

Priming is powerful stuff, participants primed to think of professors have performed better on general knowledge quizzes, people who have been primed to think of the elderly will walk slower down a hall after the experiment. In the paper's experiments participants were primed with either emotion (or as a control no emotion) and then given the IAT for one of the two minority groups. For both studies there is a significant effect on the participants' performance on the IAT under the congruent emotional priming. With the incongruent emotional priming (e.g. anger and homosexuality) there is no significant effect.

Study 1 is particularly illuminating as it uses the minimum groups paradigm, essentially creating simple and arbitrary groups of 'us' and 'them'. With no prior stereotype or heuristic to work from I would have expected the emotion priming to have no effect. However, both disgust and anger show effects on IAT performance indicating a created bias against the out-group where none existed before. It's as if participants erred on the side of caution towards the out-group. Not knowing why they are biased against the out-group, both emotions primed bias 'we don't know anything about them but we do know that they disgust and anger us'

Dasgupta et al conclude that 'emotions may focus attention on semantically applicable features of out-groups'. This increased attention to the stereotypical negative features may increase negative evaluations of the group. Can stereotypes disappear? The semantic linking could possibly be weakened - the paper highlights that in an analysis of 900 Hollywood films the core stereotype of Arabs involves aggression and fanaticism. Portrayal of a group in this manner is clearly not going to foster anything but use of stereotype and anger towards the group.

Photo links to source
Dasgupta, N., Desteno, D. & Williams, L. (2009) Fanning the flames of Prejudice: The influence of specific incidental emotions on implicit prejudice. Emotion 9(4) 585-591

If only we were less impulsive

2009-08-17T11:59:00.003+01:00

A study in the new issue of Emotion indicates that impulsivity may affect the quality and duration of a night's sleep. Previous research (Schmidt et al. 2008) has linked the two before but this paper starts to uncover the mechanisms by which the two may be linked. Counterfactual thinking (thinking how our lives would be different if we had made different decisions) is suggested as a mediating mechanism for impulsivity and sleep quality. In summary, Schmidt & Van der Linden (2009) present 3 hypotheses:

1. Impulsive urgency is linked to the frequency of counterfactual thoughts and emotions
2. Frequency of counterfactual thoughts and emotions at bedtime is linked to insomnia severity
3. The previously found link between impulsive urgency and insomnia is at least partially mediated by these counterfactual thoughts.

While the findings match the hypotheses, no other factors regarding impulsivity (lack of premeditation, lack of perseverance, and sensation seeking behaviour) are linked to either counterfactual thinking at bed-time or insomnia.

The authors suggest that counterfactual thinking is linked to insomnia through sleep-interfering dreams. Because counterfactual thinking is associated with aversive emotions (shame, guilt etc) we may actively avoid them throughout the day by thought suppression. Once we sleep though, our prefrontal cortex - the area of the brain responsible, amongst other things, for executive control and self regulation - deactivates. This effectively stops thought suppression and enables the counterfactual thoughts to rebound and affect our dreams. It's a cool idea but is it plausible?

They later go on to suggest that prolonged poor quality sleep may lead to resource depletion, decreasing self-regulation, increasing impulsivity. One question arises, if poor self regulation is associated with impulsivity how do impulsive people successfully suppress their counterfactual thoughts throughout the day? Possibly the cognitive load of suppressing the counterfactual thoughts leaves no resources available for regulation of other behaviour, which leads to impulsive behaviour, which leads to more counterfactual thoughts to suppress...

photo links to source
Schmidt, R.E & Van der Linden, M. (2009) The aftermath of rash action: sleep-interfering counterfactual thoughts and emotions, Emotion, 9(4), 549-553

The eyes have it (or not)

2009-08-14T11:25:00.008+01:00

Sci-Am Mind & Brain posted a transcript on cultural effects on recognising emotional expression, which connects rather nicely with the topic of Monday's post. Using the facial action coding system (a popular system that is assumed to communicate expression universally), Jack et al. indicate that there is significantly lower recognition in facial expressions among east asian participants than western caucasian participants. Results come from a small sample (13 in each category) and effects seem to be limited to mis-recognition of fear and disgust for surprise or anger.

The cause of this is attributed to the gaze of participants; while western participants were recording using eyes and mouth as a guide, east asian participants had a systematic bias towards the eye region. I like Jack et al.'s use of emoticons as an indication the differences between cultures, with eastern representations being eye-oriented for smiling (^.^) and surprise (O.O) vs the mouth-oriented western representations of :) and :0

The mis-recognition is given to be a combined effect of ambiguous information and the comparatively reduced emotional expression in eastern culture. Errors were made towards a less socially threatening emotion (surprise rather than fear) possibly because of the higher value of social cohesion associated with eastern culture. It would be interesting to see western caucasian responses to the ambiguous information that east asians may use to decode emotional expression. Would they be as likely to make errors towards less socially negative emotions?

Jack, R.E et al (in press) Cultural confusions show that facial expressions are not universal, Current Biology
photo links to source

Science of Happiness

2009-08-12T12:46:00.011+01:00

Today the Guardian is reporting how we can be happier in life through simple emotion regulation techniques such as smiling, counting your blessings and reliving happy memories. The report is based on research by Richard Wiseman and can be found here. So far, it seems that the research is not published, or even submitted to a journal for publication, meaning that these results need to be taken with a substantial amount of salt. The claim that these simple exercises can affect happiness across an entire week being one of them.

The implication for findings like this, or certainly the finding generally forwarded by the media, is that we can be happier just by wanting to. The ultimate extension of the argument being If only those who were depressed just tried to cheer up a bit then maybe they could snap out of it. The problems with this being: A regulation is effortful and a just-do-it mentality isn't always enough and B we are not quite the masters of our circumstances as we like to think (as indicated by a great TED talk on success by Alain de Botton).

Another consideration would be if this is a short-term solution or a lasting effect. The hedonic treadmill argument suggests that we adapt to events that make us happy (such as winning the lottery) leaving us no happier over a long period of time than if nothing had changed. The boosts from the tasks may be useful while they are still novel but once we acclimatise to doing the tasks what happens?

This post isn't a dig a Wiseman or his work - he's done some interesting work on parapsychology and suggestibility, is a promoter of applied psychology in self-help, and Quirkology is a rather fun read. It is always annoying though when the media gladly reports unpublished research and presents it as quick fix solutions.

photo links to source

Cultural effects on happiness

2009-08-10T20:00:00.008+01:00

Mind Hacks has just put up a link to an interesting study in the new issue of Emotion. The paper indicates that our cultural folk beliefs of happiness and unhappiness affect our understanding of these experiences.

American students report that happiness is a near completely positive experience, whereas Japanese students were more likely to report happiness as an ambivalent experience. The achievement of happiness, when viewed as 'an end result of personal pursuit', seems like a uniquely positive experience. Uchida & Kitayama argue ambivalence arises from a more holistic model of experience, 'If it [happiness] should invite envy of others, the sense of accomplishment might be compromised'.

Positive hedonic experience also differs in experience across cultures; in American culture it is more closely associated with personal achievement, while in Japanese culture it more closely associated with social harmony. This can be tied back to the idea that in western cultures the self is regarded as independent, whereas, in eastern cultures the self is seen as interdependent.

Uchida & Kitayama close with a careful reminder that 'findings should not be taken to imply that Japanese and Americans fail to understand each other's versions of happiness or unhappiness'. Their work highlights that theories of culture need to be better understood as they have influencing factors on a person's psychological processes.

Uchida & Kitayama (2009) Happiness and Unhappiness in East and West: Themes and Variations, Emotion 9(4) 441-456
Photo links to source

Two great minds discuss emotions

2009-07-31T15:43:00.004+01:00

In this 3 minute video from 2008 we find out how easily our emotional state can be affected by others' decisions. It's good to see that these two giants of education in the field of emotions, and in many more areas, have made their content freely available through mini lectures.

Moon review (ish)

2009-07-27T16:15:00.014+01:00

With its sparse cast, one of which an AI, and it setting of a futuristic yet plausible isolated space station, Moon is obviously going to be compared to 2001. Moon is good enough to be favourably compared to such films and is one of the better films I've recently seen. I may be a little biased as I like sci-fi that doesn't require me to learn the history of planet Xyzzy, films that are a bit weird, and characters that seem like they are losing grip on reality.

The main premise is Sam, the lone operator of a base on the far side of the moon, is finally coming home after 3 years of social isolation save for occasional transmissions from earth and an AI named Gerty. 3 years of surrogate social contact has taken its toll on Sam and he begins to see things that aren't there, crashes his lunar rover and is rescued by his younger self.

Gerty doesn't quite have the cold sophistication and all-seeing-eye of HAL; he does however appear to have emotions, presented visually by some cute, but still slightly creepy, emoticons. For someone interested in both computational models and emotions the obvious question is, 'Is Gerty really experiencing emotions or is it just following instructions to display emoticons at programmed appropriate times?' For others, the obvious question is 'Why did we invite Dave if he's just going to talk over the film?'

I feel my question is resolved in one of Gerty's briefest displays of emotion when it is caught doing something it shouldn't*. When Sam 'surprises' Gerty the emoticons change briefly from the default smile to a worried thinking expression. Gerty covers the unwanted expression with some quick thinking, a plausible lie about the events, and displaying the smile emoticon again. A robot programmed to simply present an emotion in response to Sam's emotional state wouldn't experience conflicting emotions or need to regulate itself.

Anyway, enough dissection of what must be 1 second of footage. The film is really worth watching and it's a shame most will miss out because it's not the typical summer film offering.

*avoiding spoilers makes for convoluted and vague writing
photo links to source

Control your emotions

2009-07-24T22:53:00.017+01:00

Today's paper is loosely connected to my commentary on Monday's paper. Dieffendorf and Gosserand (2003) propose a framework for emotional labour within the field of control theory.

Control theory is a largely an engineering term to do with error reduction; the classic example being a thermostat having a reference point for how warm a room should be, the difference between what the room temperature currently is and its reference value instructs the heating to come on until the error between room temperature and the reference is reduced. How can emotions be considered in terms of such cold and mechanical processes?

Emotional labour is proposed as a regulation of the current emotional state to meet a reference value set by the job's display rules. In terms of Monday's commentary it is finding where you ought to be on the emotional map and correcting where you are until you meet that goal. Control theory and psychology have crossed numerous times before; most attributions are made to Carver and Scheier (1982), though the first control-theory/psychology cross, known as PCT, appears to belong to Powers et al (1960).

A key premise of control theory is the layers of control. Higher, and more abstract, levels of control set reference points for lower, more specific and mechanical, levels of control. Multiple high level goals can come into conflict with each other as the both could ultimately share the same final output. The paper suggests that high level goals, such as 'be true to ones feelings', 'be a good employee' can come into conflict when they both try to change the output level of emotional expressions. This conflict is resolved by shifting the relative importance of these two high level demands so that one temporarily has overall influence on the emotional output.

Based on the principles of control theory Diefendorff and Gosserand outline some propositions emotional labour. Some are rather common sense - 'specific display rules lead to less variance in display rules'. Some transcend multiple levels of control - 'valence of an emotional display rule is a function of the strength and direction of the relationship between performing the emotional display and the attainment of higher-order goals'. Finally, some specifically require a control theory approach to appreciate - 'Speciﬁc display rules lead to low intrinsic motivation and job satisfaction and high burnout'.

Monday's paper argues that specific display rules would be unrelated to job satisfaction. In contrast, Diefendorff and Gosserand argue that greater control would be needed to keep emotions within an acceptable error margin. This greater level of self-monitoring can lead to a greater chance of burnout (Grandy, 2000) and a decrease in intrinsic motivation for the job (Ryan and Deci, 2000). More importantly, their model provides more than just simple propositions as it starts to explore the processes by which we decide to (or not to) regulate our emotions.

photo links to source
Diefendorff and Gosserand (2003) Understanding the emotional labour process: a control theory perspective. Journal of organisational behaviour 24,945-959

Lost in your emotions

2009-07-20T22:47:00.007+01:00

Another classic today. Morris & Feldman's 1996 paper 'The dimensions, antecedents and consequences of emotion labour' is a clear description of how emotional labour (regulating emotions as part of a job) should be broken down into its elements and how circumstances affect each element.

They conceptualise emotional labour in four dimensions: Frequency of display, attentiveness paid to display rules, variety of emotions required to display, and the resulting emotion dissonance (difference between what you actually feel and what you express). I would have included duration and emotional intensity as dimensions, although these come under attentiveness in the paper.

For each of the dimensions various antecedents are proposed, such as explicit display rules and close monitoring effecting the frequency of emotional display. A workplace example is the clear rules and close supervision given to the front of house staff at Disney World ensures that positive emotion display is very frequent.

Power of role receiver (importance of client) is argued to both positively affect attentiveness and negatively affect the variety of emotions displayed. Flight attendants engage in more emotional labour in first class areas than in economy (Hoschild 1983). Morris and Feldman argue that in presence of high status clients emotions expressed become much more restricted (specifically to the more positive expressions). In contrast variety of emotional display is proposed to be positively associated with task variety.

Emotional dissonance is predicted by the nature of the work; face to face interaction with clients is likely to result in more dissonance than when clients are not present. Finally a person's job autonomy will reduce the dissonance experienced as they may not be inclined to rigidly meet the display standards. Each of the dimensions is associated with emotional exhaustion.

Emotional dissonance is the only dimension associated with job satisfaction (obviously negatively). Consider your current emotional state as a point on a map of all emotional states. Emotional labour could be regarded as the travelling from your current state to another. High frequency journeys and greater distances travelled will result in higher levels of exhaustion, as would having to travel to multiple locations.

Each of the above dimensions points to various states, or places on the map, that need to be visited. Emotional dissonance points to a state that needs to be visited but also indicates a separate place of where you want to be. This discrepancy between where you have to travel and where you want to be is an exhausting and unsatisfying journey once you get there.

Photo links to source
Morris & Feldman (1996) The dimensions, antecedents and consequences of emotional labor, Academy of Management Review 21(4), 986-1010

Emotions are Social

2009-07-17T13:04:00.001+01:00

Social emotions has been going for a month now and somehow we're yet to look at a classic paper called 'Emotions are Social'. The general held doctrine at the time of writing ('96) is that emotions are essentially internal and personal reactions. Parkinson questions this and argues that emotions may be better viewed in terms of their communicative properties.

When viewing emotions as internal events, it lends to a procedural interpretive process. I experience an emotion and then express it to the best of my ability; you then receive this information, decode it and then react to the content, in turn shaping a new emotion for expression back to me. This system seems like a laboratory-controlled conversation and in everyday life this might not suffice.

In certain social situations a person's expressive behaviours may be mimicked by others, this is done without specific deliberate interpretation. (Indeed, deliberate imitation of expressive behaviours, as taught in some NLP type seminars, can come across as forced and out of step with the flow of a conversation). The natural imitation of a person's gestures and expression is hypothesised to change one own internal state via facial and bodily feedback (Hatfield, Cacioppo, Rapson, 1992). As such, we experience emotional contagion, bypassing the need to procedurally observe, interpret, and modify expression.

If emotions were tools of communication rather than merely internalised states, it would be expected that greater expression is seen in the presence of others. Parkinson highlights a number of examples of this occurring, Kraut & Johnston's (1979) efforts being particularly inventive. People were observed during a game of 10 pin bowling. Facial displays of bowlers observed were much more pronounced when turning around to an audience than when facing the result of the bowl. It wasn't the emotive event of knocking down pins that elicited expression but the aim of sharing this event with others.

Emotions are used to impact on our environment. We use them as social tools to communicate our needs and wants from others. Anger, for example, may communicate the message 'Take me seriously and give me the respect I deserve!' whereas sadness is characterised as a request to 'Comfort / reassure me!'. Emotions in this context become more relational rather than personal expressions. Social emotions, if you will.

Parkinson, (1996) Emotions are social, British Journal of Psychology, 87, 663-683.
Disclosure: Brian Parkinson is a co-investigator in the EROS research group. I am a PhD student also part of EROS.
Photo links to source.

Put your head on my shoulder

2009-07-15T11:32:00.002+01:00

We have already looked at the concept of how we can regulate our behaviour by use of the internal 'resource'. In that instance resource is regarded as the mental energy we have to consume to complete a task. Resources come in far more shapes and sizes.

Hobfoll (2002) brings together many of the theories of resources to highlight some underlying and connecting ideas of what resources are and how they can be used. Resources are considered to be entities of value to us (e.g. close attachments, self-esteem) or means to obtain valued entities (e.g. money, social network).

These general resources are gathered and protected to help sustain our well-being. In times of stress we can use these resources to shield ourselves and cope with the stressful situation. However, this may result in resource deterioration (much like ego-depletion in the strength model). For example, when facing a crisis a person may mobilise resources such as social support. The support deterioration deterrence model (Kaniasty & Norris, 1996) indicates that after the initial gains from this, support often wanes - effectively the resource has been consumed.

Baltes (1997) suggests that there is a positive trade off though; successfully addressing the crisis can result in an increase in resources such as self-efficacy and even an improved social structure. It seems that consumption of resources is useful if the goal can plausibly be met and is worth achieving.

Over accumulating resources has its drawbacks too; striving for money or fame is negatively related to well-being (Ryan & Deci 2000). Even though being rich and famous could clearly be seen as substantial resources, the expenditure or neglect of other, more stable, resources (such as close-attachment or autonomy) may have a strong negative impact. Investing resources in more achievable goals (such as improving close relationships) could be far more beneficial and useful for coping with stress.

Hobfoll concludes with the universal agreements between the various theories of resource. Firstly, people strive to accumulate and keep social, cognitive and biological resources. Those with resources are less likely to encounter stressful situations, and those that do are more likely to cope well. Resources are linked and there is a general tendency for one resource to enable developing others. Finally, resources are long term, especially when compared to (usually) transitory periods of stress.

Hobfoll, (2002) Social and Psychological Resources and Adaptation, Review of General Psychology, 6(4), 307-324.
All others are cited in the above article
Photo links to source

Control your choices

2009-07-08T09:25:00.010+01:00

Another book today, 'The Paradox of Choice' by Barry Schwartz. The related TED talk covers the topics in the book very well and, like the book, is filled with New Yorker cartoons to illustrate the points.

In this modern age of abundant choices for everything we buy, do, and can be, we could potentially all be endlessly happy. Out there somewhere could be the perfect, stereo system, career, and life partner for us. Our satisfaction of the choices we make are affected by all the other choices we didn't take, and for those Schwartz calls maximizers settling for 'good enough' is never good enough.

Recently, I went to the Big Session - a small folk festival in Leicester. There were three stages, one of which doubled as the beer tent. I think I got more overall satisfaction from this small festival than going to Glastonbury festival 2003. There were certainly bigger and better bands at Glastonbury - there were also far fewer terrible poetry sessions for me to endure. But there was always the nagging sensation that while I was enjoying one band, I was missing out on 10 other potentially great bands.

Schwartz, would suggest that my Glastonbury experience was marred by the missed opportunities and by regret of the decisions I did make (How and why did I end up watching Macy Gray's full set?). At the Big Session there were so few options it was possible to make sure that no good band was left unheard and that there is no room for regret.

How can we over come the dissatisfaction caused by too much choice? There will always be an increase in the number of products available for us to choose from and as mere consumers we have little influence of that. We can however start to choose what we choose*. The second order choice sounds like it could just complicate things, but by restricting our lower level choices 'we will be able to choose less and feel better'. A concrete example of this would be to make a rule when clothes shopping to only visit two stores. By choosing at a higher level it greatly limits the amount of clothes to make a decision on and so derive greater satisfaction from the purchase.

photo links to source
*control theorists would argue that in meeting this higher level goal a new standard is set for the lower level goal. This 2nd tier choice goal is set by a higher standard to meet - such as be happy with decisions - and so on...