Abstract: Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by impairments in social communication, restricted interests or activities, and repetitive stereotyped behaviors. In addition to these core features, children with ASD often exhibit deficits in memory function (Griffin et al., 2021). Summers and Craik (1994) demonstrated that action encoding can enhance memory in children with ASD. However, other studies have shown that action encoding may not be effective in certain contexts (Wang et al., 2022; Xie et al., 2022), suggesting that the efficacy of action encoding as a memory enhancement method for children with ASD may depend on specific conditions. Identifying the factors and mechanisms that influence the effectiveness of action encoding in promoting memory could alleviate the stress on parents and schools in raising children with ASD. Additionally, it could provide insight into the memory characteristics of children with ASD, offering valuable guidance for research on the pathogenesis or diagnosis of ASD. Given these implications, this study reviews existing research on the effects of action encoding on memory in children with ASD and addresses the following questions: (1) What factors influence the effectiveness of action encoding in enhancing memory in children with ASD? (2) What theoretical contributions can be made by exploring these influencing factors to improve the predictive power of theories regarding the efficacy of action encoding in children with ASD? (3) What are the limitations of current research, and how can future studies address these gaps to deepen our understanding of the relationship between action encoding and memory in children with ASD.
Previous research revealed that the efficacy of action encoding in promoting memory in children with ASD is influenced by the type of action encoding and the type of ASD. Specifically, action encoding involving self-performed and experimenter-performed actions enhanced memory in children with ASD who had no comorbidities with other disorders and exhibited lower-middle intelligence (SPM scores of 10% and 25%) as well as upper-middle intelligence (SPM scores of 50% and 75%). In contrast, self-performed action encoding did not facilitate memory in children with ASD who had no comorbidities and demonstrated lower-middle intelligence (SPM score of 5%). Additionally, the effects of imagining present-performed actions and imagining future-performed actions on memory differed significantly in children with ASD who had no comorbidities and exhibited upper-middle intelligence.
Previous research on the effects of action encoding on memory in children with ASD has predominantly contributed to theories of self-performed actions, including the non-strategic encoding theory, multimodal theory, motor encoding theory, the ‘four-component theory’, and the ‘five-component theory’. The non-strategic encoding, multimodal, and motor encoding theories emphasize that the motor component, elicited by the execution of an action, is the primary driver of the self-performed effect. In contrast, the ‘four-component’ and ‘five-component’ theories propose that, in addition to the motor component, semantic and imagery components are also critical to the self-performed effect. Following the discovery that self-performed action encoding enhances memory in children with ASD, previous studies have generally concluded that, similar to typically developing (TD) children, children with ASD exhibit self-performed effects due to the motor component, albeit with weaker activation. However, unlike TD children, the self-performed effects in children with ASD do not arise from the imagery component. These similarities and differences underscore the importance of prioritizing the motor component over the imagery component when applying self-performed action encoding theories to predict whether children with ASD will demonstrate self-performed effects.
Despite these valuable findings, previous studies have identified only two types of factors influencing the effectiveness of action encoding in promoting memory in children with ASD—namely, the type of action encoding and intelligence level—without elucidating their specific mechanisms. Moreover, these studies have offered an incomplete exploration of the mechanisms underlying the memory-enhancing effects of action encoding in children with ASD. Specifically, they have predominantly focused on applying self-performed action encoding theory to the ASD population, while neglecting theoretical advancements in understanding experimenter-performed and imagery-performed action encoding. To facilitate a more comprehensive and effective application of action encoding in interventions and to strengthen the theoretical foundation guiding such interventions, future research should: (1) broaden the investigation of factors influencing action encoding in memory enhancement for children with ASD, including action encoding modality, type of ASD, and type of action; and (2) refine existing theories by incorporating identified moderators, particularly through the development of theoretical frameworks that explain the memory-enhancing effects of experimenter-performed and imagery-performed actions in children with ASD, thereby enhancing the predictive accuracy of these theories; and (3) design and implement a targeted memory intervention program specifically tailored to the unique needs of children with ASD.

Key words: action encoding, autism spectrum disorder, memory, type of action encoding, intelligence