How well can we communicate emotions purely through touch?

Romantic couples outperformed pairs of strangers

Whether it’s a raised eyebrow or curl of the lip, we usually think of emotions as conveyed through facial expressions and body language. Science too has focused on these forms of emotional communication, finding that there’s a high degree of consistency across cultures. It’s only in the last few years that psychologists have looked at whether and how the emotions can be communicated purely through touch.

A 2006 study by Matthew Hertenstein demonstrated that strangers could accurately communicate the ‘universal’ emotions of anger, fear, disgust, love, gratitude, and sympathy, purely through touches to the forearm, but not the ‘prosocial’ emotions of surprise, happiness and sadness, nor the ‘self-focused’ emotions of embarrassment, envy and pride. Now Erin Thompson and James Hampton have added to this nascent literature by comparing the accuracy of touch-based emotional communication between strangers and between those who are romantically involved.

Thirty romantic couples (the vast majority were heterosexual) based in London took part. One partner in each romantic pair attempted to communicate 12 different emotions, one at a time, to their partner. They sat at opposite sides of a table divided by a curtained screen. The emotional ‘decoder’ slid their forearm through the curtain for the ‘encoder’ to touch, after which the ‘decoder’ attempted to identify which of the 12 emotions had been communicated. The participants were filmed throughout.

After this, the romantic couples were split up and participants paired up with a stranger to repeat the exercise (encoders and decoders kept whichever role they’d had first time around). Strangers were usually formed into same-sex pairs, to avoid the social awkwardness of touching an opposite-sex partner. This created an unfortunate confound, acknowledged by the researchers, which is that most romantic couples were opposite-sex whereas most stranger pairs were same-sex. However, focusing only on results from same-sex pairs versus opposite-sex pairs suggested gender was not an important factor.

The key finding is that although strangers performed well for most emotions, romantic couples tended to be superior, especially for the self-focused emotions of embarrassment, envy and pride. Thompson and Hampton calculated that chance performance (i.e. merely guessing) would produce an accuracy rate of 25 per cent. Although there were 12 emotions to select from, the rationale here is that some are far more similar to each other than others, so even a guesser would perform better than 1/12 accuracy. Romantic partners communicated universal emotions, prosocial and self-focused emotions with an accuracy of 53 per cent, 60 per cent and 39 per cent, respectively – in each case, far better than chance performance. In contrast, strangers achieved accuracy rates of 39 per cent, 56 per cent and 17 per cent, for universal, prosocial, and self-focused emotions respectively, with the last considered as no better than chance performance.

How did the romantic couples achieve their greater accuracy? They touched for longer, but this wasn’t correlated with accuracy. Using footage of the experiment, the researchers coded the types of touch used (a wide range of discrete touch types were identified, from trembling and scratching to slapping and squeezing), and for each emotion it was clear that strangers were using similar kinds of touch as were romantic couples. This means that there were either subtle differences in the touching used by romantic couples, which the experimenters had failed to detect, or the ‘decoders’ were interpreting the same touch cues differently when they were delivered by an intimate partner.

This topic is ripe for further investigation – for example, does the touch advantage shown by romantic couples extend to non-emotional communication? Would other long-term, but non-sexual, relationship partners such as siblings, show a similar advantage? And would romantic partners still display an advantage if they didn’t know who was doing the touching? ‘Our findings extend the literature on the communication of emotion,’ the researchers said. ‘The nature of particular relationships appears to have the ability to diminish the ambiguity of emotional expression via touch.’

ResearchBlogging.orgThompson, E., and Hampton, J. (2011). The effect of relationship status on communicating emotions through touch. Cognition and Emotion, 25 (2), 295-306 DOI: 10.1080/02699931.2010.492957

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

7 thoughts on “How well can we communicate emotions purely through touch?”

  1. Interesting, but I honestly don't understand how something like this can be published.

    “This created an unfortunate confound” … sorry, every student would have told you that with this design the results are not interpretable. You could have just as well not done this study.

    Again, the idea is interesting, but this is such a major factor!

    Oh boy. And there were so many ways around this! Let the people touch each other anonymously in a group for 2 hours, or give them thin gloves, etc. pp.


  2. Dear Anonymous,

    It is always easier to criticise with hindsight.
    Please credit the researchers for, a) conceiving the study, b) conducting the study, c) acknowledging its limitations and d) informing us of their study.

    Also, I would like to acknowledge the publishers for publishing the study in spite of its limitations so as to take the field forward.

    I would be interested in knowing the potential practical implications of such studies. Development of a tactile language for non-verbal expression of emotions for the blind is one that comes to mind.

    Many thanks.

Comments are closed.