New research explores how understanding the relationship between empathy and reward may help our approach to autism

‘The term sympathy was very popular all the way from the philosopher David Hume in the 18th Century onwards,’ explains Bhismadev Chakrabarti, Associate Professor at Reading University’s Centre for Integrative Neuroscience and Neurodynamics. ‘Then the use of the word sympathy keeps going down all through the centuries and the word empathy is a late 20th Century phenomenon.’

Dr Chakrabarti recently gave a paper around his current interest on the relationship between empathy and reward at a British Academy conference dedicated to empathy, neuroscience and conflict resolution. The conference theme is a sign of the wave of interest in the notion of empathy in the scientific community, though as Dr Chakrabarti explains, empathy appeared, ‘within the discourses of philosophy and the humanities but didn’t enter science discourse until recently. A lot of it has to do with the work Simon Baron-Cohen has done with autism.’

Dr Chakrabarti did his doctorate with Professor Simon Baron-Cohen, Director of Cambridge University’s Autism Centre where Dr Chakrabarti holds an honorary position. Chakrabarti’s training was in chemistry then neurobiology, but his interest began to widen from the area of pure brain chemistry to look at broader questions around understanding human emotions. ‘Then I did my PhD with Simon-Baron Cohen and I told him I want to work on emotions and he said, “well if you want to find how emotions work in the typical brain, then it is worth finding out when the brain is not typical any more such as in autism, where there is a deficit in recognising other people’s emotions and responding to them.”’


He describes his scientific journey as one going from the bottom up – from molecules to brains to the bigger construct. At the heart of this is a fascination with the rich dynamic of emotions, how for example we ascribe complex emotions to what we perceive through our eyes such as minor changes in someone’s facial expression or modulation of voice. ‘I’m interested in the processes that can influence how I understand another person’s emotion or how I respond to that person,’ says Dr Chakrabarti.

How I respond to that person may depend on sympathy or empathy. Sociologist Richard Sennett writes in his book Together: The Rituals, Pleasures and Politics of Co-operation that, ‘Both sympathy and empathy convey recognition and both forge a bond, but the one is an embrace, the other a encounter. Sympathy overcomes differences through imaginative acts of identification; empathy attends to another person on his or her own terms.’

Dr Chakrabarti’s recent experiments and research papers show that this empathic relationship is driven by the mechanism of reward, an idea that has been put forward by social psychologists. So for example we respond empathically to people from our own ethnicity or to a friend. ‘Friendship is a beautiful psychological construct,’ says Dr Chakrabarti, ‘but at the end of the day what you are changing about the friend is the extent of reward-response you have to the friend. So reward is the signal, the final downstream signal that alters how much you will empathise with or not empathise with another person.’


So he and his team did an experiment exploring this question of reward and empathy, measuring people’s responses to a game using three different techniques – facial electromyography, functional Magnetic Resonance Imaging and electroencephalography. The electromyography recorded the degree of spontaneous imitation via the muscles used for smiling in response to a positively conditioned face and a negatively conditioned face. ‘What we found is that people had a much greater spontaneous imitation of positively conditioned faces compared to the ones that are negatively conditioned,’ explains Dr Chakrabarti.

The fMRI study looked at the connection between mimicry and reward, and discovered that people looking at more rewarding faces have a much stronger coupling between the brain regions involved in imitation and reward. The Electroencephalography results confirmed the fMRI study.

But what’s at the heart of this work for Dr Chakrabarti is how it plays out in relation to autism. His studies suggest that this link of reward and empathy seems to be weaker in individuals with autism. One area of evidence being the social skills training done with autistic children, much of which is based on reward-learning – success at a task will be rewarded with an activity they enjoy doing.  But says Dr Chakrabarti, ‘a lot of these children going through this training regime do not actually improve in social behaviour much. That might possibly because the learnt reward is not translating into empathy.’

He thinks that a weaker connection between the brain system involved in reward-processing and the brain systems involved in empathising, might underly some of the features that we see in autism. ‘Rather than a specific deficit in reward-processing or a specific deficit in empathy as such, it’s more really the connection between reward and empathy.’

Next steps will be to conduct further experiments which reverse the act of imitation – subjects will interact with some people who will mimic them more and others less. It’s based on social psychological studies showing that for example in a restaurant if there are waiters mimicking your behaviour and there are waiters not mimicking you, you are more likely to tip the waiter mimicking you. ‘We are systematically changing the extent of mimicry,’ says Dr Chakrabarti, ‘and seeing how that changes the reward associated with social stimuli and whether this link, going the other direction from mimicry to reward is weaker in autism.’