Researchers are figuring out how sense of self develops differently in autistic teens

Teenage woman looking at herself in a mirrorBy guest blogger Dan Carney

Our autobiographical memory is fundamental to the development of our sense of self. However, according to past research, it may be compromised in autism, together with other skills that are also vital for self understanding, such as introspection and the ability to attribute mental states to others (known as mentalising).

For example, experiments involving autistic children have highlighted retrieval difficulties, “impoverished narratives”, and a greater need for prompting, while also suggesting that semantic recall (facts from the past) may be impaired in younger individuals.

Now a UK research team, led by Sally Robinson from London’s St. Thomas’ Hospital, has published the first attempt to assess the nature of – and relationships between – autobiographical memory, mentalising and introspection in autism. Reporting their findings in Autism journal, the group hope their results will shed more light on the way that autistic children and teens develop a sense of self.

The researchers compared the performance of 24 autistic participants (age range: 11-18; 4 girls) and 24 age- and gender-matched neurotypical participants on three tasks. The first, designed to tap semantic and episodic aspects of autobiographical memory, required them to recall what kind of personality they showed in different contexts (at school, with the family, when happy etc) and to describe specific episodes from their lives in which they’d behaved as that kind of person.

A second, interview task measured introspection and mentalising: participants rated both their own ability, and that of a nominated comparison individual (such as a friend or relative), to recognise both their own and the other person’s internal/external traits, such as tiredness (internal trait) or smartness of dress (external). For example, in relation to tiredness, participants rated “How well do you know when you feel tired?”; “How well does X know when you are tired?”; “How well do you know when X feels tired?”; and “How well does X know when he/she feels tired?”.

The final task was the widely used mentalising measure, the Reading The Mind In The Eyes Task for children, which involved participants looking at pictures showing only the eye region of person’s face and inferring their mental state.

Overall, the results demonstrated impaired and preserved abilities in the autistic group, while suggesting atypical relationships between skills. In terms of autobiographical memory, the autistic children and teens: a) produced fewer self-descriptive personality traits than the neurotypical group, b) required more prompting to generate specific episodic memories, c) provided fewer memories containing either emotional or sensory detail, but d) did not differ in terms of the number or type of episodic memories recalled.

That the autistic group recalled a typical number of episodic memories, albeit with greater prompting and atypicality of content, is broadly in line with previous work. The recall of fewer own personality traits, however, is a new finding, and is worth investigating further. This may suggest that autistic individuals’ self-concept is different in some way from non-au.

Meanwhile, on the introspection and mentalising interview, the autistic children and teens (a) rated themselves poorer at knowing about another’s mental state than did the neurotypical group, and b) more surprisingly, they rated others as having more knowledge about their own external/behavioural traits than they did themselves. The researchers suggested that this last finding may betray a lack of confidence in making objective/comparative behavioural judgments – as opposed to subjective internal ones – about the self in relation to others.

Related to this, among the autistic children and teens only, if they rated themselves as having a high level of self-knowledge, they were also more likely to rate others as having a higher knowledge of them. The researchers pointed out that this may highlight the use of introspection as a compensatory strategy in autism, when attributing mental states to others, and that this may play a role in autistic individuals’ documented tendency to over-attribute their own beliefs and knowledge to others.

As expected, the autistic children and teens were poorer than controls on the Eyes Task, another indication of compromised mentalising skills.

There was also a group difference in how the various skills related to each other. In the neurotypical group, performance correlated across the tasks: semantic/episodic autobiographical memory, introspecting and mentalising, which indicates how, in typical development, autobiographical memory has a reciprocal relationship with other aspects of the self-concept. In the autistic children and teens, these correlations were absent, suggesting that this reciprocity between mental processes may be impaired.

Although this study is exploratory and more work is needed, it provides a finer-grained glimpse at how skills associated with self-understanding may interact in autism, and how this may differ from the equivalent processes in typical development.

The new findings may also have implications for autistic children undertaking Cognitive Behavioural Therapy (CBT), an approach where the recall of autobiographical events is important. CBT is commonly used with autistic people to address anxiety and social skill development. In the new research, the autistic group generated as many episodic memories as controls, after having first described their own personality traits. This may suggest that in CBT with autism, accurate episodic retrieval could be supported by first encouraging individuals to reflect on their own personalities in a more general sense.

Personality traits, autobiographical memory and knowledge of self and others: A comparative study in young people with autism spectrum disorder

Dan Carney headshotPost written for BPS Research Digest by Dr. Dan Carney. Dan is a UK researcher specialising in developmental disorders. He undertook his post-doctoral research fellowship at London South Bank University, finishing in 2013. His published work to date has examined cognition, memory, and inner speech processes in Williams syndrome and Down syndrome, as well as savant skills in autism.

2 thoughts on “Researchers are figuring out how sense of self develops differently in autistic teens”

  1. Their research sounds more like language retrieval issues related to talking to strangers, and not knowing exactly what the stranger wants…. most people who are autistic are skittish in those settings. Most autistics actually have phenomenal event and environmental memories stretching back to infancy or even prebirth due to lack of neural pruning — but are hesitant to share because there is nothing more painful than having your actual narrative refused by the recipient, usually with harsh words. All it takes is one or two experiences of that from someone, at some time or other, and you have what the researchers are describing. This is once again a case of non-autistic scientists interpreting the totality of the autistic experience by non-autistic standards applied to externally observed behavior.

  2. Hi Michelle – thanks for the comment!

    I definitely take your point about the interactive nature of the tasks, and I think it taps into a wider issue about task demands influencing the performance of certain diagnostic groups. In my own work with individuals with Williams Syndrome, I’ve considered whether their heightened levels of social motivation – when combined with the one-on-one, highly personal, nature of the testing sessions – may inflate their “natural” level of performance, as it means they try significantly harder than the comparison groups.

    That being said, although personality trait recall was worse, it’s worth remembering that, once more prompting was given, the ASD group did generate as many episodic memories as controls. So it wasn’t always a case of task demands leading to the false impression of poorer performance. Furthermore, when the authors consider the therapeutic implications of this at the end, they’re actually suggesting a way in which to tailor the CBT interaction process to improve the episodic recall of people with ASD i.e. encourage them to think about more general/semantic autobiographical stuff first.

    This seems to me an acknowledgement – albeit an implicit one – that the interactional demands of particular situations may not help individuals with ASD to recall as much as they can, and that this kind of thing can and should be looked at where possible.

    Nevertheless, and although in the piece above I chose to focus on the results/implications as reported by the authors, it would have definitely been interesting to consider the sorts of issues you’ve raised. I’ve often worried that the cognitive literature on “atypical” development simply reads like a long list of deficits.



Comments are closed.