A novel image data base for social concepts reveals preference biases in autistic spectrum in adults and children

Abstract

Human beings display the extraordinary ability of grasping and communicating abstract concepts. Yet, no standardized instruments exist to assess this ability. Developing these tools is paramount for understanding abstract representations such as socialconcepts,withramificationsineducationalandclinicalsettings.Here,wedevelopedanimagedatabasedepictingabstract social concepts varying in social desirability. We first validated the image database in a sample of neurotypical participants. Then, we applied the database to test different hypotheses regarding how social concepts are represented across samples of adults and children with autism spectrum condition (ASC). Relative to the neurotypicals, we did not observe differences related to ASC in identification performance of the social desirability of the concepts, nor differences in metacognitive ability. However, we observed a preference bias away from prosocial concepts that was linked to individual autistic traits in the neurotypicals, and higher in ASC relative to the neurotypicals both in adults and children. These results indicate that abstract representations such as social concepts are dependent on individual neurodevelopmental traits. The image database thus provides a standardized assessment tool for investigating the representation of abstract social concepts in the fields of psycholinguistics, neuropsychology, neuropsychiatry, and cognitive neuroscience, across different cultures and languages.