Bridging the Gap: Enhancing Gaze-Performance Link in Children with ASD through Dual-Level Visual Guidance in MR-DMT

Weiying Liu, Yanran Yuan, Zhiqiang Sheng, Dandan Lian, Sheng Li, Yufan Zhang, Yulong Bian, Juan Liu

Abstract: Autism Spectrum Disorder (ASD) is marked by action imitation deficits stemming from visuomotor integration impairments, posing challenges to imitation-based learning, such as dance movement therapy in mixed reality (MR-DMT). Previous gaze-guiding interventions in ASD have mainly focused on optimizing gaze in isolation, neglecting the crucial "gaze-performance link". This study investigates enhancing this link in MR-DMT for children with ASD. Initially, we experimentally confirmed the weak link: longer gaze durations didn't translate to better performance. Then, we proposed and validated a novel dual-level visual guidance system that operates on both perceptual and transformational levels: not only directing attention to task-relevant areas but also explicitly scaffolding the translation from gaze perception to performance execution. Our results demonstrate its effectiveness in boosting the gaze-performance link, laying key foundations for more precisely tailored and effective MR-DMT interventions for ASD.

Introduction: The Challenge of Action Imitation in ASD

Action imitation is a fundamental building block of social and cognitive development. For children with Autism Spectrum Disorder (ASD), deficits in this area are a core symptom, often rooted in impaired visuomotor integration (VMI). VMI is the neurological process of translating visual perception (what we see) into motor execution (what we do).

Dance Movement Therapy (DMT) has emerged as a powerful intervention to address these deficits. Mixed Reality (MR) environments offer a scalable way to deliver DMT through virtual agents. However, standard technological interventions often focus solely on "gaze guidance"—ensuring the child looks at the instructor. This paper argues that for children with ASD, looking does not automatically lead to doing. To be effective, interventions must bridge the gap between visual attention and physical performance.

The researchers first set out to empirically validate the "weak gaze-performance link." In a baseline study (Study 1), they tracked the gaze duration of children with ASD as they observed a virtual instructor. They found that longer gaze durations on the instructor did not significantly correlate with better motor imitation performance.

Conceptual overview of the weak gaze-performance link problem (a) and the proposed dual-level solution (b).

This finding suggests that while a child might be paying attention, the neurological pathway required to transform that visual data into a motor plan is disrupted. Simply increasing gaze duration is insufficient if the underlying VMI process is not supported.

The Proposed Solution: Dual-Level Visual Guidance

To address this, the authors developed a Dual-Level Visual Guidance System. This system moves beyond simple attention-grabbing to provide explicit scaffolding for the transformation of perception into action:

  1. Perceptual Level (The "Where"): This layer uses salience modulation to reduce cognitive load. By dimming the brightness and saturation of task-irrelevant background elements, the system creates a subtle "spotlight" on the virtual agent, helping the child focus on the correct visual source.
  2. Transformational Level (The "How"): This layer uses kinetic metaphors to assist in motor planning. The system animates a school of fish that swims along the exact trajectory of the virtual agent's limbs. This provides a predictive, simplified visual cue that helps the child understand the flow and direction of the movement they are meant to imitate.

Experimental Methodology

The research was conducted using a high-fidelity MR-DMT system housed in a CAVE (Cave Automatic Virtual Environment). This setup allows for immersive interaction with a virtual agent while maintaining the child's awareness of their own physical body.

The MR-DMT system setup in a CAVE environment, featuring projectors, RGB cameras, and a virtual agent.

In Study 2, 18 participants were tested using a 2x2 within-subjects design to isolate the effects of the Perceptual and Transformational guidance levels. Performance was rated by experts based on accuracy, rhythm, coordination, and completeness, while gaze was tracked to measure the strength of the link between attention and execution.

Key Findings

The intervention successfully bridged the gap between looking and doing:

  • Strengthened Correlation: In the baseline condition, the correlation between gaze and performance was weak and non-significant ($\rho = 0.392$). With the Dual-Level Guidance, this correlation became strong and highly significant ($\rho = 0.731$).
  • Synergistic Effects: While both individual components (Perceptual only and Transformational only) improved the link, the combined dual-level approach was significantly more effective than either one alone.
  • Performance Gains: Beyond improving the correlation, the system led to absolute improvements in motor imitation scores, suggesting that the scaffolding actually helped children perform the movements more accurately.
  • Responsive Behaviors: The system also maximized the correlation between gaze and target-related responsive behaviors, such as independent participation, reaching a correlation of $ ho = 0.743$.

Conclusion and Future Directions

This research demonstrates that for children with ASD, visual guidance must do more than just direct attention; it must assist in the complex process of motor transformation. By using kinetic metaphors and salience modulation, the Dual-Level Visual Guidance System successfully enhanced the gaze-performance link in MR-DMT.

Future work will look toward longitudinal studies to see if these improvements in VMI transfer to real-world social and physical contexts. Additionally, the authors suggest using neuroimaging to explore the specific neural pathways being activated and strengthened by this dual-level scaffolding approach.

Summary updated

Abstract

Autism Spectrum Disorder (ASD) is marked by action imitation deficits stemming from visuomotor integration impairments, posing challenges to imitation-based learning, such as dance movement therapy in mixed reality (MR-DMT). Previous gaze-guiding interventions in ASD have mainly focused on optimizing gaze in isolation, neglecting the crucial "gaze-performance link". This study investigates enhancing this link in MR-DMT for children with ASD. Initially, we experimentally confirmed the weak link: longer gaze durations didn't translate to better performance. Then, we proposed and validated a novel dual-level visual guidance system that operates on both perceptual and transformational levels: not only directing attention to task-relevant areas but also explicitly scaffolding the translation from gaze perception to performance execution. Our results demonstrate its effectiveness in boosting the gaze-performance link, laying key foundations for more precisely tailored and effective MR-DMT interventions for ASD.
BibTeX
@misc{liu2025bridging,
  author = {Weiying Liu and Yanran Yuan and Zhiqiang Sheng and Dandan Lian and Sheng Li and Yufan Zhang and Yulong Bian and Juan Liu},
  title = {{Bridging the Gap: Enhancing Gaze-Performance Link in Children with ASD through Dual-Level Visual Guidance in MR-DMT}},
  year = {2025},
  eprint = {2510.03724},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2510.03724},
}
Figure