JMIR Publications

ABSTRACT

Background:

Effective preoperative digital health education (DHE) is critical for patients undergoing knee arthroplasty (KA), particularly older adults who face age-related cognitive vulnerabilities. Cognitive Load Theory (CLT) suggests that presentation modality and learning motivation strongly influence cognitive processing, yet their combined effects in clinical education remain underexplored.

Objective:

This study investigated how learning motivation and presentation modality ((text-only, text-graphic composite, and video-based) affect cognitive load and learning performance in elderly patients preparing for KA.

Methods:

A 2 × 3 factorial psychobehavioural experiment was conducted with 62 patients (≥60 years) stratified by learning motivation (high vs low). Each participant completed six DHE modules delivered across the three modalities. Cognitive load was measured using subjective ratings (Chinese NASA-TLX) and objective eye-tracking indicators (average fixation duration, number of fixations, and time to first fixation). Learning performance was assessed through knowledge retention and transfer tests (30 items). Data were analyzed using general linear models, ANCOVA, and correlation analyses, with covariates including knee function, prior knowledge, eHealth literacy, and psychological distress.

Results:

High-motivation learners demonstrated significantly lower cognitive load (NASA-TLX; large effect size) and superior learning performance (medium effect size) compared to low-motivation learners. Video-based materials consistently yielded the lowest extraneous load and supported efficient knowledge acquisition (medium effect size). Text-graphic composites elicited higher cognitive load but facilitated deeper processing and schema construction, particularly for highly motivated learners (medium-to-large effect sizes). Eye-tracking confirmed these dynamics: high-motivation participants showed shorter fixation durations and more efficient allocation of attention across modalities, whereas low-motivation learners displayed scattered fixation patterns, especially in text-graphic conditions. Across the sample, cognitive load was negatively correlated with learning performance (large effect size), reinforcing its role as a key mediator of digital learning outcomes.

Conclusions:

Both learning motivation and modality exert significant, though partly independent, influences on preoperative DHE outcomes in KA patients. Video-based content enhances cognitive efficiency, while text-graphic formats may promote germane load among motivated learners. These findings highlight the importance of motivational scaffolding, adaptive modality selection, and the integration of real-time cognitive monitoring in DHE design. Beyond informing digital education for KA, the study demonstrates the feasibility of applying psychobehavioural experimental methods with older surgical patients, offering a framework for optimising cognitive alignment in clinical education.