PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 6, 2026Open Research Europe0 citationsOpen Access

Feasibility of a theory-driven dynamic difficulty adjustment algorithm integrating individual performance and normative data for adaptive modulation of spatial and distractor inhibition attention in a non-immersive virtual reality task: Evidence from individuals following stroke

View Full Paper
GSGregorio SorrentinoGEGauthier EverardTLThierry Lejeune

Key Points

  • Investigate the feasibility of a dynamic difficulty adjustment algorithm for attention tasks in stroke patients.
  • Conducted a single-group repeated-measures feasibility study in inpatient and outpatient rehabilitation settings.
  • Engaged ten stroke patients in three 15-minute sessions of a VR task with DDA.
  • The algorithm adjusted difficulty based on real-time performance to keep engagement within 65-85% success rate.
  • All participants completed the protocol without any adverse events.
  • Performance of participants converged toward the target engagement range over the sessions.

Abstract

Background Post-stroke attentional deficits are heterogeneous and may involve both spatial attention impairments, such as unilateral spatial neglect, and non-spatial deficits, including distractor inhibition. Conventional cognitive neurorehabilitation tasks often rely on fixed or manually adjusted difficulty levels to match the level of the patient, which may not adequately accommodate individual variability. Dynamic Difficulty Adjustment (DDA) algorithms offer a performance-driven approach to modulating task demands. The present study examined the feasibility of a theory-driven DDA algorithm integrating individual performance and normative data within a non-immersive virtual reality environment task. Methods A single-group repeated-measures feasibility study was conducted in inpatient and outpatient rehabilitation settings. Ten individuals with stroke completed three 15-minute sessions of DDA-REAsmash. The multi-parameter DDA algorithm continuously adjusted stimulus presentation time, distractor inhibition demands, and spatial configuration based on real-time individual performance, aiming to maintain an optimal engagement range of 65–85% success rate. Normative reaction-time percentiles derived from an independent mean age-matched sample were implemented as lower algorithmic constraints. Feasibility was defined as stabilization of performance within predefined engagement boundaries, and adaptive behavior was examined in relation to baseline attentional profiles. Results All participants completed the protocol without adverse events. Across sessions, performance converged toward the target engagement range. Conclusions These findings indicate the feasibility of a theory-driven multi-parameter DDA algorithm integrating individual performance and normative data to maintain individualized engagement in a heterogeneous post-stroke sample. Further controlled studies are required to determine clinical efficacy and functional outcomes.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Sorrentino et al. (2026) studied this question.

synapsesocial.com/papers/69fa986a04f884e66b53228chttps://doi.org/10.12688/openreseurope.23432.1
Ask AI
Helpful
Bookmark
Share
View Full Paper