A recent study published in Nature Mental Health suggests that autistic and non-autistic individuals use highly similar strategies when learning about other people’s preferences. The findings provide evidence that both groups rely on flexible learning methods, but the unique and varied preferences of autistic individuals can make their specific likes and dislikes harder to predict. These insights help explain common social misunderstandings without assuming that autistic people lack the ability to read social cues.
Navigating social environments requires people to constantly figure out what others are thinking and feeling. One basic building block of social interaction is learning about another person’s preferences, such as their favorite foods or hobbies. Understanding what someone likes makes it easier to predict their behavior, evaluate if they are trustworthy, and establish a meaningful connection.
Recent concepts in psychology offer a new perspective on these social disconnects. The double empathy problem is a theory suggesting that communication breakdowns between autistic and non-autistic people happen because both groups experience the world differently. According to this framework, both groups struggle to understand each other, rather than the social deficit belonging solely to the autistic person.
The current research builds on previous work to explore how social knowledge shapes learning across different groups. “The senior author, Dr. Gabriela Rosenblau, published findings in 2021, using the preference learning paradigm and computational modeling, which found that autistic and non-autistic participants adopted different strategies when inferring peer preferences,” said Shannon Cahalan.
Cahalan is a postdoctoral fellow at the National Institute of Mental Health working in the Section for Social and Developmental Cognitive Neuroscience. She recently earned her doctorate at the George Washington University and served as the lead author of the current study.
“Specifically, non-autistic youth’s learning could be described by a complex model integrating prior knowledge about peers’ preferences, item relationship representations, and trial-by-trial updating while autistic youth’s own preferences informed their inferences about peers,” Cahalan told PsyPost. “While this finding offered a compelling potential underlying mechanism for social challenges in ASD, there were still some gaps to explore.”
The researchers designed the current study to see if people make more accurate guesses about others who share their own diagnostic background. “For example, would these findings hold up in a larger, more representative sample?” Cahalan asked. “How are autistic self-preferences different from non-autistic self-preferences to begin with? Are autistic adolescents actually basing their inferences on other autistic peers? How does variability in autism trait presentation predict social learning performance? We sought to address these open ends in this study.”
The study was broken down into three separate experiments. The first experiment mapped out the personal preferences of a large group of participants. The sample included 228 non-autistic adults, 125 non-autistic adolescents, and 255 autistic adolescents.
Participants completed an online survey where they rated 120 different food and activity items. They used a six-point scale featuring emoji faces to indicate how much they liked or disliked things like sushi, roller skates, and candy. This step allowed the scientists to see exactly how the self-preferences of autistic teens differed from the other two groups.
The data showed that autistic adolescents had a much wider variety of preferences compared to both non-autistic adults and teens. Autistic teenagers showed distinct patterns, tending to give higher ratings to items like candy, writing supplies, and art supplies. They also tended to give lower ratings to items like salads, vegetables, and fitness equipment compared to the non-autistic groups.
Autistic participants also showed signs of behavioral rigidity in their preferences. Behavioral rigidity refers to a strong desire for sameness and a resistance to change, which is a common characteristic of autism. In this context, autistic teens tended to group their preferences strictly by basic categories, meaning that if they strongly liked one type of fast food item, their ratings for all other fast food items followed a highly similar pattern.
In the second experiment, the scientists tested how 191 non-autistic adults learned about the preferences of teenagers. The adults were split into two groups, with 98 adults guessing the preferences of non-autistic teens and 93 guessing the preferences of autistic teens. The adults did not know the diagnostic status of the teenagers they were evaluating.
During the task, participants guessed how much a specific teenager liked an item and then received immediate feedback on the actual rating. The scientists measured prediction errors, which represent the mathematical gap between a participant’s guess and the true answer. Smaller prediction errors indicate higher accuracy and better social learning.
The adults produced smaller prediction errors when guessing the preferences of non-autistic teens. When learning about autistic teens, the adults made more errors initially but steadily improved their guesses over time. Computer models showed that the adults used a fine-grained learning strategy for both groups, meaning they updated their guesses based on how similar the items were to one another rather than clumping everything into broad categories.
The third experiment shifted the focus to the autistic adolescents. A group of 83 autistic teens guessed the preferences of non-autistic teens. A separate group of 119 autistic teens guessed the preferences of other autistic teens.
The authors expected the autistic teens to be more accurate when judging peers from their own diagnostic group. The data actually showed the opposite pattern. Autistic adolescents made more accurate guesses when predicting the preferences of non-autistic teens.
Just like the adults, the autistic teens actively used the feedback to improve their guesses over time. They also used the exact same fine-grained learning strategy, which contradicts older ideas that autistic teens only rely on their own likes and dislikes to judge others. The authors noted that autistic teens did use general knowledge about other autistic people to form their starting expectations, but this strategy did not lead to better accuracy.
“I was quite surprised as to how integral ‘variability’ was to the interpretation of our results,” Cahalan said. “I went in with a strong expectation that our results would be clear-cut and autistic youth would present with signature and distinct social learning strategies.”
“Instead, our insights were based more on how autistic youth themselves are distinct from one another,” Cahalan explained. “Working on this project fundamentally changed how I approach research focused on autism and related neurodevelopmental disorders.”
The researchers suspect that the wide variety in autistic preferences makes individual autistic teenagers harder to predict. Because autistic likes and dislikes are so diverse, relying on an average group profile does not help much when judging a specific person. Both adults and autistic teens found it easier to predict the non-autistic profiles because those preferences were much less varied and fit a more standard pattern.
“Variability is a ‘feature’ of autism and not a ‘bug,’” Cahalan noted. “The impact of variability, that is, was a core theme of our findings. On one hand, autistic individuals may be more difficult, in general, to learn about because there is variability in their preferences.”
“The ‘average’ autistic preferences are less likely to capture the preferences of a given autistic individual and this has broad ramifications for social learning,” Cahalan added. “On the other hand, variability in autistic traits, like rigidity for example, were especially predictive of autistic teen’s preference learning performance.”
In addition to those broad patterns, the scientists looked at how individual personality traits affected learning within the autistic group. Participants who reported higher levels of overall autistic traits tended to integrate new feedback more slowly. Those who reported higher levels of behavioral rigidity tended to make more overall errors during the guessing game.
Several limitations provide context for how to interpret these findings. The study compared the learning styles of young adults with those of teenagers. Because the two groups were not the same age, the scientists could not perfectly separate the effects of age from the effects of the autism diagnosis.
The complex nature of the guessing task also created challenges for some participants. A small number of autistic participants were unable to finish the experiment because the rules were too demanding. The authors suggest that future studies should create simpler activities to include people across the entire autism spectrum.
The sample of autistic girls was relatively small, with only 55 female participants included in the main tasks. This imbalance made it difficult to detect any reliable differences between how boys and girls learn about social preferences. Future work will need to include more balanced groups to see how gender might interact with social learning.
The study did not tell participants whether the profiles they were evaluating belonged to autistic or non-autistic people. People might change their learning strategies if they explicitly know the diagnostic background of their partner. Future research could explore whether revealing this information alters how people approach social guessing games.
The rating scale used in the study also had a restricted range of only six options. This limited how much the prediction errors could fluctuate during the mathematical analysis. Researchers plan to use continuous rating scales with a wider range of options in future versions of the experiment.
Moving forward, the team plans to expand on these findings by looking at brain activity. “We are continuing to examine social learning at the neural level in autistic and non-autistic youth,” Cahalan said. “Critically, we have acquired extensive neuropsychological, demographic, and behavioral data on our participants which will allow us to further explore how individual differences predict preference learning performance and related functional neural activity in autistic vs. non-autistic youth.”
The study, “Modeling how autistic and non-autistic groups learn about their own and each other’s preferences,” was authored by Shannon Cahalan, Raphael Perla, Sophia Block, Mikaila Loughlin, Christoph W. Korn, and Gabriela Rosenblau.
Leave a comment
You must be logged in to post a comment.