We’re unable to read our own body language

A fascinating study has shown that we’re unable to read insights into ourselves from watching a video of our own body language. It’s as if we have an egocentric blind spot. Outside observers, by contrast, can watch the same video and make revealing insights into our personality.

The premise of the new study is the tip-of-the-iceberg idea that what we know consciously about ourselves is fairly limited, with much of our self-knowledge lying beyond conscious access. The researchers wondered whether people would be able to form a truer picture of themselves when presented with a video of their own body language.

In an initial study, Wilhelm Hofmann and colleagues first had dozens of undergrad students rate how much of an extrovert they are, using both explicit and implicit measures. The explicit measure simply required the students to say whether they agreed that they were talkative, shy and so on. The implicit measure used was the Implicit Association Test, and was intended to tap into subconscious self-knowledge. Briefly, this test reveals how much people associate ideas in their mind (such as ‘self’ and ‘shy’), by seeing whether they are quicker or slower to respond when two ideas are allocated the same response key on a keyboard.

Next, the participants recorded a one minute television commercial for a beauty product (they’d been told the study was about personality and advertising). The participants then watched back the video of themselves, having been given guidance on non-verbal cues that can reveal how extraverted or introverted a person is. Based on their observation of the video, they were then asked to rate their own personality again, using the explicit measure.

The key question was whether seeing their non-verbal behaviour on video would allow the participants to rate their personality in a way that was consistent with their earlier scores on the implicit test.

Long story short – they weren’t able to. The participants’ extraversion scores on the implicit test showed no association with their subsequent explicit ratings of themselves, and there was no evidence either that they’d used their non-verbal behaviours (such as amount of eye contact with the camera) to inform their self-ratings.

In striking contrast, outside observers who watched the videos made ratings of the participants’ personalities that did correlate with those same participants’ implicit personality scores, and it was clear that it was the participants’ non-verbal behaviours that mediated this correlation (that is, the observers had used the participants’ non-verbal behaviours to inform their judgements about the participants’ personalities).

Two further experiments showed that this general pattern of findings held even when participants were given a financial incentive to rate their own personality accurately, as if from an outside observer’s perspective, and also when the task involved anxiety personality ratings following the delivery of a short speech.

What was going on? Why can’t we use a video of ourselves to improve the accuracy of our self-perception? One answer could lie in cognitive dissonance – the need for us to hold consistent beliefs about ourselves. People may well be extremely reluctant to revise their self-perceptions, even in the face of powerful objective evidence. A detail in the final experiment supports this idea. Participants seemed able to use the videos to inform their ratings of their “state” anxiety (their anxiety “in the moment”) even while leaving their scores for their “trait” anxiety unchanged.

“When applied to the question of how people may gain knowledge about their unconscious self, the present set of studies demonstrates that self-perceivers do not appear to pay as much attention to and make as much use of available behavioural information as neutral observers,” the researchers said.

ResearchBlogging.orgHofmann, W., Gschwendner, T., & Schmitt, M. (2009). The road to the unconscious self not taken: Discrepancies between self- and observer-inferences about implicit dispositions from nonverbal behavioural cues. European Journal of Personality, 23 (4), 343-366 DOI: 10.1002/per.722

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

9 thoughts on “We’re unable to read our own body language”

  1. I like your study and you have an interesting approach that holds much promise, keep up the good work. I would like to attempt to answer a question that you posed, “Why can't we use a video of ourselves to improve the accuracy of our self-perception?” The subjects that you used where not sufficiently skilled, they did not have the knowledge of behavior or body language to participate if you changed your subjects you will have a different outcome. I ran into this same problem when I was doing my study on “self”. I ended up using graduate students you can read the results of this study, get a copy of The Power of Self Separation you will enjoy it.


  2. It's hard to explain why a poor read on one's own body language could possibly be fitness-enhancing.

    I wonder whether, for each individual, there might be certain gestures/positions/sequences whose significance he happens to perceive more poorly than the average individual does.

    So much human communication is fake, or at least managed and sanatized relative to what we really feel – because we usually want to display ourselves as the ideal ally or mate, and rarely want to display what we actually are like. It may be, then, that much of our insight into others pertains to stuff those others are feeling but /not/ trying to communicate to us: a person “leaks” certain true emotions against his will, via whatever gestures he happens to intuit the meaning of less well than the average person. In other words, each person bluffs using those gestures he happens to intuit well, and leaks truths when he inadvertantly makes those gestures that he poorly understands. And so, most people agree that the salient gestures of person A are X1, X2, and X3 – but these are precisely the gestures whose import person A does not understand and feel as well as most poeple do. He “cannot read himself.”


  3. I think it's perfectly reasonable not to adjust your assessment of your personality traits one the basis of a short video. For example, I'm an introvert. If I had an extrovert few hours, sure I would say I was high on “state extroversion”, but of course I shouldn't adjust my assessment of my (trait) introversion-extroversion notably. My assessment of my personality is based on tons and tons of evidence and it would be foolish of me if I'd let a few minutes or hours make a dent in my assessment.

    As for implicit and explicit measures, I've seen two explanations when they disagree. One (which you seem to subscribe to) is that the explicit measures give a wrong result (for whatever reason), another is that the two measure genuinely different phenomena.


  4. This seems a very difficult study to control. The subjects knew they were looking at a video of themselves.

    Hard to know how to control for this but it seems an enormous confounding effect. What would they have said if they had seen an unknown other behaving in precisely the same way?

    Did the authors attempt to control for this, assess baseline performance on ability to determine the personalities of others etc?


  5. I think people need to be more careful about making such generalised statements. Certain types of people would be more likely to apply to a study about personality and advertising than others. What about people who are overly self-conscious? Those types of people are less likely to sign up for such studies, but almost by definition, they are the more likely to take notice of their own body language and analyse what it may say about themselves.

    I don't know about the original article, but this post is bad science.


  6. I think it would be pretty bizarre to reevaluate somebody's self perception because their hands fidgeted in a video or whatever. People have an enourmous amount of data points on their own personality that is so far beyond the number of data points available to the people who ran the study that the whole exercise seems incredibly silly.


  7. Anonymous, you said: “People have an enourmous amount of data points on their own personality that is so far beyond the number of data points available to the people who ran the study”. I fear you have misunderstood the study. There was a mismatch between people's implicit and explicit ratings of their own personalities. The question was whether seeing a video of themselves would bring their explicit self-rating in line with their implicit self-rating. It didn't. Observers, by contrast, were able to use information in the videos to make ratings of the participants that were correlated with those participants' own implicit self-ratings. The implication is that we can't or won't, even with a financial incentive, use the material in the video to update our explicit rating of our own personality. At the same time, we're happy to use that kind of information to inform our rating of other people's personality.


  8. I am a communication consultant and coach people in overcoming their fear of public speaking. Time and again my clients over-estimate the extent to which they are portraying negative non-verbals such as blushing. (They think it is worse than it really is.) They also misjudge the extent that such non-verbals matter to an audience. Could this add further support to the notion that we are not very good at judging our own body language? I would suggest that it is how we feel about ourselves that influences our judgment of ourselves rather than seeing “objective” recordings of our body language.
    It is a very helpful debate to have … and in the workplace such research could influence how we conduct performance reviews, for example.
    Rachel Green


  9. Could it be pssible that the person is distracted with the general self image of himself?. Like when some people hear themself on a recording, they pay more attention to the way they sound rather than what they are saing. Maybe allowing subjects to se the playback a few times may yeild diferet results.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s