In a recent study published in Scientific Reports, researchers investigated whether autistic characteristics affected the strength of color-shape associations (CSAs) based on the binding errors (BEs) in the incongruent and congruent trials.
Background
Non-synesthetes relate particular shapes to specific colors (i.e., blue-square, yellow-triangle, and red-circle). The CSAs could impact the binding between shapes and colors, and increase BEs in incongruent color-shape pair identification instead of the congruent ones. Autism spectrum disorders (ASDs) are characterized by multisensory consolidation impairments and atypical-type sensory processing.
About the study
In the present study, researchers determined the association between CSAs and the extent of autistic characteristics among Japanese individuals.
For the analysis, 51 individuals residing in Japan (24 men with a mean age of 23.0 years) were recruited from the National Rehabilitation Centre for Persons with Disabilities participant pool. The included individuals did not experience synesthesia and self-reported normal color vision and corrected-to-normal ability to precisely recognize small details. CSA strength was evaluated by assessing CSA-induced BEs, and autistic characteristics were evaluated using the Japanese version of the Autism Spectrum Quotient (AQ-50) questionnaire.
Japanese versions of the Wechsler Adult Intelligence Scale-III and the second edition of the Autism Diagnostic Observation Schedule Component (ADOS-2) were provided to 10 individuals who were diagnosed with ASDs by healthcare professionals. The behavioral experiment involved three assessments. First, the individuals had to identify if the two alphabets presented were different or the same. Second, they had to identify the color of either of the two shapes presented.
Finally, they had to report their confidence level in choosing the correct color. Two black letters and nine colored shapes were utilized as visual stimuli. Three colors (i.e., blue, yellow, and red) and shapes (i.e., square, triangle, and circle) generated nine color-shape combinations, with the three shapes displayed as black-line sketches. Differences in BEs between incongruent and congruent colored-shaped conditions were calculated as BEs induced by the CSAs.
Following the experiment, participants filled out questionnaire surveys assessing their explicit CSA, wherein three shapes and a four-color wheel were presented, and the participants were required to select the color that matched best with each of the shapes. The mean percentage of BEs, feature errors (FEs), and hits for all individuals with color-shape congruency factors (congruent CSAs: square-blue, circle-red, and triangle-yellow). The team excluded two individuals who made >50.0% errors in the alphabet task; thus, 49 participants were considered for the final analysis.
Results
The team observed a significant association between AQ-50 scores and BE occurrence when the red-circle and yellow-triangle congruent CSAs were presented. Participants with greater autistic characteristics made more BEs in identifying incongruent versus congruent color-shape associations, indicating stronger associations between the yellow color and the triangle and the red color and the circle.
The accuracy rates on the color and letter tasks were 73% and 94%, respectively. A significant difference was observed in binding errors in the incongruent and congruent conditions, with more Bes reported for incongruent CSAs (15%) than the congruent ones (9.2%). In contrast, more hits were reported for congruent CSAs (39%) rather than incongruent CSAs (34%).
However, there were no significant differences in FEs for the incongruent and congruent CSAs. For all color-shape pairs, the binding errors for incongruent CSAs were significantly greater than those for the congruent CSAs (5.7% versus 3.1% for the circle-square pair, 4.2% versus 3.0% for the circle-triangle pair, and 4.6% versus 3.2% for the triangle-square pairs), following previous studies.
Further, more hits were observed in the medium (16%) and high (18%) confidence trials than in the low confidence ones (9.0%). However, the differences between the high versus medium confidence levels and the low versus medium confidence ratings were not statistically significant. As per previous studies, the participants selected blue for squares, red for circles, and yellow for triangles more frequently than the other colors.
Conclusions
Overall, the study findings showed that autistic characteristics significantly correlated with the binding errors inscribed by the yellow-triangle and red-circle CSAs, and individuals with greater autistic characteristics showed stronger associations between the triangle shape and yellow color and the circle shape and the red color. Therefore, autistic characteristics influenced CSA preservation.
Particular CSAs may be inherited or acquired in the initial life periods, with robust binding being maintained among individuals with greater autistic characteristics, slightly influenced by statistical learning in later life. Future research is required concerning color-shape associations by assessing CSAs among blind individuals and infants.