Adaptive virtual reality systems employed for autism spectrum disorder interventions: a systematic review (Preprint)

Abstract

Adaptive systems serve to personalize interventions or training based on the user's needs and performance. The adaptation techniques rely on an underlying engine responsible for processing incoming data and generating tailored responses. Adaptive systems in virtual reality (VR) have proven to be efficient in data monitoring and manipulation, as well as in their ability to transfer learning outcomes to the real world. In recent years, there has been significant interest in applying these systems to treat deficits in autism spectrum disorder (ASD). This is driven by the heterogeneity of symptoms among the affected population, which leads to the need for an early customized intervention that targets specific symptom configurations for each individual.

Recognizing these technology-driven therapeutic tools as efficient solutions, this systematic review aims to explore the application of VR adaptive systems in interventions for young individuals with ASD.

A systematic review of the past ten years literature was conducted using three different databases. Overall, a total of 10 articles were included. Relevant information extracted from studies was the sample size and mean age, the study's objectives, the skill trained, the implemented device, the adaptive strategy employed, the engine techniques, and the signal utilized to adapt the systems

Studies have included level switching and/or adaptive feedback strategies, weighing the choice between a machine learning–adaptive engine (ML) and a non-machine learning–adaptive engine (non-ML). Adaptation signals ranged from explicit behavioral indicators like task performance to implicit biosignals such as motor movements, eye gaze, speech, and peripheral physiological responses.

Findings reveal promising trends in the field, suggesting that VR-automated systems leveraging real-time regression switching levels and/or multimodal feedback driven by machine learning (ML) techniques on embodied signal processing have the potential to enhance interventions for young individuals with ASD. Limitations and future directions are also discussed.