Autistic Traits Affect Reward Anticipation but not Reception

Persons with autism may be experiencing troubles interacting socially with others because of a decreased sensitivity of their brains to social stimuli (like faces, speech, gestures, etc.). Because autism is a spectrum reaching from neurotypical persons with little or no autistic traits on one end and low-functioning persons with autism on the other, we measured brain responses to social and non-social rewards in 50+ neurotypical (i.e. not diagnosed with autism) participants differing in their levels of autistic traits. Our results show that autistic traits even in neurotypical participants influence how their brains process rewards!

Image by alteredego from Pixabay 

You can find the original article here (open access).

Why is social reward so important?

Think about how we all learned what is socially appropriate. It all started with us as babies and children, trying out different behaviours and waiting for our mums, dads, and other caregivers to react. Sometimes they smiled, clapped their hands and said “very good!”, and sometimes they frowned, shook their finger and said “do not do it again”. Based on their reaction, we probably either liked the feedback (when getting a smile and encouraging words) or disliked it. This liking is a crucial word here: Most of our brains are wired in such a way, that receiving social rewards (like smiles and prises) gives us pleasure and increases the chance that we will repeat the same or similar behaviour in the future. This makes sense: We want the reward and we like it. This seems trivial, but it’s an incredibly important mechanism in the human brain. Wanting and liking rewards (and also learning based on rewards) are all parts of reward processing.

Think about what would happen if we didn’t like our mum’s smile or if we didn’t see the connection between the appropriate behaviour and her smile. Well, likely the chances that we would repeat the behaviour that led to the smile in the first place would be low. In turn, we wouldn’t learn so easily what’s appropriate and what’s not, which would make social interactions hard and confusing. Because of that, we would probably not like spending time with people, because it would be stressful to figure out what’s happening, and in the end – we just wouldn’t excel in social interactions, which would be frustrating and simply hard to understand.

This vicious loop is exactly what one theory, called the social motivation hypothesis, proposes as a possible reason for social impairments in autism:

Autism as a spectrum

So the social motivation hypothesis proposes that persons with autism have problems with interactions in social settings, because social stimuli are not as rewarding for them, as they are for this without autism (often called neurotypicals). However, autism is a spectrum, which means that those diagnosed with autism range from high- to low-functioning (respectively: those, who can function in the society on their own, have normal intelligence and language skills; and those who do not), and that there are also gradually increasing levels of autistic traits in the general population (neurotypicals):

Hence, we can measure the levels of autistic traits in people without autism, and see whether these levels are linked to any cognitive processes which we would also link to diagnosed autism. This is what we did in this study: We measured the autistic traits in neurotypicals and linked these traits to processing of social and monetary rewards.

The study

In our study, we measured the brain responses to different rewards via EEG (electroencephalogram). In this method we measure the electrical patterns of the brain’s activity with electrodes placed on a cap we put on participants’ heads. It’s non-invasive and pain-free.

We asked the participants to play a game, in which first they saw a cue which told them what kind of a reward they would get if their guess was correct, and then they had to guess the colour of the next card drawn from a deck of blue and purple cards. If they were correct (e.g. indicated the next card would be blue and it was), they received one of three types of rewards:

  • a social reward: this was a picture of the smiling face of the main experimenter present in the lab during the study;
  • a monetary reward: a picture of a 5-cent coin (and later participants indeed got their winnings on top of their participation reimbursement);
  • both: a picture of the smiling experimenter and also 5 cents.

We looked at three stages of reward processing: early wanting (how much I want the reward I know I can get if I perform well), late wanting (how much I want the reward now that I have already performed) and liking (how much I enjoy getting the reward).

In the early wanting, we found enhanced brain responses to all of these rewards in those participants, who had higher levels of autistic traits than in those who had less of these traits. Moreover, all participants on average wanted the combined (smile plus 5 cents) reward the most. However, in late wanting all of these effects disappeared, even though the brain showed even more activation than in the early stages. Finally, when participants received the reward, we observed a clear preference for the monetary outcomes with or without the smile (the brain responses were the strongest for the 5 cents alone and the 5 cents with the picture of a smiling face, compared to just the smiling face).


Contrary to what the social motivation hypothesis suggests, we found no signs of diminished processing of the social reward in higher levels of autistic traits. What is more, we found even enhanced early wanting in the participants with higher autistic traits, which suggests that, contrary to the hypothesis, the autistic traits can be linked to more wanting and anticipation of reward, regardless of the type.

Why did we observe such surprising results? We propose that the difference between our study and other similar experiments lies in the type of the social reward we used. Typically in psychological research, we would show a picture of a an unknown person smiling to a participant in a successful trial. However, an unfamiliar person, who is not related to the study at all, is likely to be less rewarding than the experimenter who leads the participant through the study, is present, and has a direct link to the task. Thus, we think that by using a picture of a socially relevant person, like the experimenter, we increased the rewarding value of the social rewards for the participants with higher autistic traits. This is supported by previous works showing that person with autism have problems with processing unfamiliar faces, but process familiar faces similarly to neurotypicals (see Pankert et al., 2014).

Of course, we cannot generalise these results to the population of persons diagnosed with autism. However, our results inspired another study, this time addressing reward processing in autistic traits and in autism, which will be described soon and shared in the scientific community. Stay tuned for the updates in our News & Blog too!

Would you like to learn about this study in more details?

Check the original article here (open access).


Chevallier, C., Kohls, G., Troiani, V., Brodkin, E. S. & Schultz, R. T. The social motivation theory of autism. Trends in Cognitive Sciences 16, 231–238 (2012).

Pankert, A., Pankert, K., Herpertz-Dahlmann, B., Konrad, K. & Kohls, G. Responsivity to familiar versus unfamiliar social reward in children with autism. J. Neural Transm. 121, 1199–1210 (2014).

Leave a Reply

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

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

Facebook photo

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

Connecting to %s