Paper: "Levels of self-representation and their sociocognitive correlates in late-diagnosed autistic adults"
This paper was published in 2021, in the Journal of Autism and Developmental Disorders.
The authors were Dr Rachel Moseley, Professor Changhong Liu, Dr Nicola Gregory, Ms Paula Smith, Professor Sir Simon Baron-Cohen, and Professor Jie Sui (author links will open in separate tabs).
You can click HERE to download a PDF of the paper.
Keep reading to see a plain English summary, or you can watch my explanatory video!
Why is this an important issue?
Human brains seem to be configured to constantly and automatically look for information that is relevant to you or about you. This ‘self-focus’ or self-bias is reflected, for instance, in the way we pay more attention to objects if they have some association with us, or are faster and more accurate to process self-relevant information. We show a similar processing advantage for information related to the people we associate with, our ‘in-group’, because once we identify with them, our focus on things related to our self rubs off on them! The way we process information about ourselves influences so many of the mental operations we perform every day, and does so in a completely unconscious way.
Little is known about how autistic people process self-relevant information, and whether they show the same bias for self-related information and for information related to their friends (their ‘in-group’). However, differences in the way they process self-relevant information could be highly relevant to autistic people, their social relationships and wellbeing.
Firstly, the self-concept affects how we relate to other people. One of the most fundamental aspects of self-processing is being able to differentiate your own thoughts, feelings, actions, from those of other people. If your ability to do this is impaired, you may struggle to understand other people’s mental states. Furthermore, if you do not preferentially process information about familiar others over information about strangers (the aforementioned ‘in-group bias’, which seems to originate from your self-bias), it is possible that this affects your feelings of connectedness with other people. This may partly be why differences in self-processing are associated with poorer wellbeing and low mood.
Here is a video summary of this study. You can open it up to large screen, and turn captions on and off by clicking the 'CC' button.
What was the purpose of this study, and what did the researchers do?
We examined whether autistic people also show preferential processing for information related to themselves and familiar others. We also looked at whether differences in self-bias and friend-bias were associated with difficulties with understanding other people’s mental states and with the real-life outcome of being highly lonely. To do this, we asked autistic and non-autistic participants to complete two tasks during an online study, and compared their performance.
In Task 1, as an unconscious reflection of self-bias and friend-bias, we asked participants to learn associations between person-labels and geometric shapes. They then needed to respond as quickly as possible, when pairs of words and shapes flashed up on the screen, to tell us if the word and shape belonged together.
In Task 1, participants learnt which person-label belonged with which shape (shown above). The shapes would then appear in quick succession on the screen, paired with one of these person-labels. Participants had to press a button as quickly as possible to indicate whether the person-label matched that shape, or did not match (i.e. did not belong with) that shape.
In Task 2, as a conscious reflection of our preference for self-relevant and friend-relevant information, we asked people to indicate the emotional closeness between individuals by positioning them closer or further away from one another.
In Task 2, participants had to indicate how close two people were emotionally by moving the sliders to indicate the distance between them. Here, if you were very close to your friend (who is positioned at point 0), you might move the slider to place yourself quite close (e.g. point 10). If this "friend" was in fact your mortal enemy, you might place yourself at point 100 to be as far away as possible!
We looked at relationships between performance in these tasks and our two outcome variables:
-
performance in a task about understanding other people's mental states (the Reading the Mind in the Eyes Test [RMET]).
-
feelings of loneliness.
The Reading the Mind in the Eyes Test (RMET): participants have to decide which mental state description matches the expression in the eyes. I think the answer here is 'interested'.
What were the results of the study?
Though both groups were very accurate at Task 1, autistic and non-autistic participants alike were most accurate when it came to self-related items where the word matched the shape it was paired with. Interestingly, though, there was a clear difference when the self-shape was wrongly paired with other labels (‘mismatching’ trials). Autistic people were more accurate on these items, while non-autistic people seemed to trip up on these.
Since we know that non-autistic people are typically automatically drawn to pay attention to self-relevant information even at the detriment of tasks they’re trying to do, we wonder whether this difference reflects that non-autistic people may have found it more difficult than autistic participants to ‘unhook’ their attention from the self-related item. Autistic people might still process self-matching items preferentially, but be slightly less specific in the way that attention is captured by items relating to the self and items relating to other people.
Also in Task 1, autistic people did not show the same degree of friend-bias - that is, they were not significantly better at processing friend-related items than stranger-related items. This suggested that their brain was not more attentive to familiar others, which are usually drawn into the orbit of the self-bias and benefit from that.
This was very consistent with findings from Task 2. Non-autistic people typically move themselves and their friend further away from the stranger. A stranger is a member of an out-group - they might be your rivals, competing for resources, or they might even be a threat to you. Hence we tend to want to create distance between ourselves and a stranger. Non-autistic people wanted to keep themselves AND their friends away from the stranger.
Autistic people kept themselves away from the stranger in Task 2, but did not take their friends with them in the same way. This suggests that they did not differentiate so much between in-group (friend) and out-group (stranger) members. Interestingly, the lonelier participants were, the closer together they placed the friend and stranger. One interpretation of this finding is that autistic participants showed less of the typical non-autistic tendency to form associations with an in-group. If you do not identify with any group of others (i.e. connect yourself with an in-group), and if you unconsciously think of strangers and familiar others in the same way, this could potentially be a very lonely way to be. In light of this, it makes sense that the less difference a person saw between strangers and friends, the lonelier they were. We did not, however, see any relationship between performance on Task 1 and 2 and performance in the RMET.
What are potential weaknesses in the study?
We had quite a small control group, so it is possible that other differences exist which didn’t show up in our findings. We were not able to look at differences between participants that might be based on characteristics like sex and gender.
Our participants were not representative of certain demographics within the autistic community, such as autistic people of colour; autistic trans and non-binary people; and autistic people with intellectual disabilities. People who were diagnosed as children were also under-represented. Our findings are also not necessarily relevant to autistic people in other cultures, since the way people think about the self and others is very affected by the culture they live in.
Finally, it may be that we did not find any relationships between our tasks and the RMET because this task is more a measure of emotion recognition than of understanding the mental states of others, which we had speculated might be affected by differences in the way people process information about self and others.
Thank you
for reading!
How will these findings help autistic adults now or in the future?
Beyond this study, other experiments have told us that differences in the way that people process information about themselves and others affect their ability to form supportive relationships and affect their mood and wellbeing. The ability to identify with an in-group is really important for forming relationships and maintaining your mental health.
These findings hinted at the possibility that autistic people might differ in the way they form and align themselves with in-groups and out-groups. This could have implications for mental health, since loneliness is such a strong factor in mental illness and suicide in autistic people.
It is not clear if these differences merely reflect something that is already occurring - i.e. that differences in self and other processing are a reflection of depression and/or loneliness, or arise as a result of depression and/or loneliness. This could make them quite a useful marker of when someone is struggling. Alternatively, could modifying these very automatic ways that we process information have a positive impact on relationships, mood and wellbeing? We need to know if differences in self and other processing can be changed, and if this doing so would be helpful for wellbeing. We don’t fully know all the potential implications for these findings yet, so really need to do some deeper investigation to find out.
If you found this interesting, you may like to read:
If you are struggling with your mental health, you may like to look at the resources page.