Synaesthesia in Adults with Autism

Background: Case studies and anecdotal reports have suggested a link between autism spectrum disorders (ASDs) and synaesthesia, an idiopathic condition in which stimulation of one sensory modality automatically evokes a perception in an unstimulated modality (e.g., a bell ringing triggers the perception of the color blue). The prevailing neurobiological models of ASD and synaesthesia are similar in their emphasis on atypical neural connectivity, and a genetic study also alluded to an association, linking synaesthesia to a chromosomal region associated with autism. A pilot study by Baron-Cohen et al. found a greater rate of self-reported synaesthesia in autistic people compared to non-autistic people (12.7 and 4%, respectively). The difference between the groups was non-significant, likely due to an underpowered sample.

Objectives:  To elucidate the relationship between ASD and synaesthesia through estimating the prevalence of synaesthesia in a larger sample of high-functioning autistic adults (Study 1), and comparing the nature of synaesthesia in synaesthetes with and without autism (Study 2). 

Methods:  Study 1- 2291 adults registered on two volunteer databases were invited to join the study via email. 172 autistic adults and 123 typically developing (typical) adults participated. There were no differences in age. The autistic group was 46% female and the typical group was 70% female. Mean AQ, EQ, and SQ-R scores of both groups fell within the normal ranges for their respective groups. Eight were excluded for reporting a self-diagnosed ASD. Participants completed online and paper questionnaires that defined synaesthesia, and then assessed experiences through a series of screening items. Conservative criteria were used to determine status. Participants who denied experiencing synaesthesia or had a history of drug use or neurological conditions were considered non-synaesthetes. Study 2- Self-reported autistic and typical synaesthetes (n=31 and n=7, respectively) from Study 1 were included. 29 previously confirmed typical synaesthetes were randomly selected from our database and added to the typical synaesthete group to increase power. 54% of autistic synaesthetes were female compared to 81% of typical synaesthetes. There were no differences in age. Questionnaire responses from both groups were compared.

Results:  Study 1- The current estimate of synaesthesia prevalence is 4.4%. Based on self-report, we obtained significantly different rates of 5.7 and 18.9% for the typical and autistic groups, respectively (X2 (1, 287) = 10.68, p < .01, Φ = .19). Study 2- Both groups reported instantaneous percepts, unlearned pairings, and lifelong synaesthesia. Consistent with the literature, colors triggered by sounds and written linguistic stimuli were the most common variants reported by both groups. Autistic synaesthetes reported more changes in percept strength, greater influence of emotions on synaesthesia, and greater visual interference from synaesthesia.

Conclusions: Future studies should confirm these prevalence estimates by verifying self-reports with objective measures (e.g., tests of internal consistency). Discovering a greater rate of synaesthesia in the autistic population might suggest the conditions share neurobiological underpinnings. Even if synaesthesia occurs at the same rate in the typical population, understanding how synaesthetic experiences are manifested in behavior could aid understanding of autistic symptomatology.