Abstract
<p>Increasing evidence shows a co-occurrence of autism spectrum disorder and aphantasia (a lack of visual imagery) with both cohorts showing reductions in visual working memory (WM). Here we explored whether visual imagery ability (measured with the Vividness of Visual Imagery Questionnaire, VVIQ, and Spontaneous Use of Imagery Scale, SUIS) mediated the relationship between autistic traits (measured with the Autism Quotient, AQ-50) and visuospatial WM in the general population. Participants who reported being more detail-oriented (higher ‘attention to detail’) and who reported fewer difficulties with communication (lower ‘communication’ scores) performed better on a WM task. Participants who reported preoccupation with one task and difficulties switching tasks (higher ‘attention switching’ scores) on the AQ-50 made more binding errors, selecting a colour associated with a different shape during encoding. These relationships were not mediated by visual imagery scores, suggesting that visual imagery does not underpin WM difficulties associated with autistic traits in our study. Despite neither VVIQ nor SUIS scores predicted WM performance separately, within a multiple regression model both higher imagery vividness and, surprisingly, less frequent use of imagery significantly predicted better WM performance. While vivid imagery may support WM performance, spontaneous, task-irrelevant imagery may instead distract from task performance.</p>