Mobile Augmented Reality as a Remediation Tool for Assembly Drawing Decoding: A Mixed-Methods Study

Abstract

Aim/Purpose: Many learners struggle to decode complex mechanical assembly drawings because decoding requires coordinating 2D interpretation, inferred 3D structure, and kinematic reasoning. This study examines whether a single-session mobile augmented reality (AR) intervention may support below-proficiency learners in reaching a curriculum-relevant competence threshold, how they perceive the AR tool, and what experiential themes characterize AR-supported decoding practice. Background: Most AR studies in engineering graphics address introductory visualization rather than the higher-order task of interpreting complete assembly diagrams, and few offer theory-grounded explanations of mechanism. This study frames assembly decoding as a representational integration task. It draws on Cognitive Load Theory, the Cognitive Theory of Multimedia Learning, and scaffolding theory to explain which specific AR affordances may reduce extraneous reconstruction demands and support schema formation for struggling learners. Methodology: A quasi-experimental design with an embedded mixed-methods component was used with 60 second-year baccalaureate students (Sciences et Technologies Mécaniques track, Moroccan technical high school). Students scoring below a diagnostic cutoff (n = 25) received a single-session AR remediation using the eDrawings app; the remaining 35 followed the standard pathway without AR. Data sources included pretest and posttest scores, a Technology Acceptance Model questionnaire, and open-ended comments and classroom observation notes. Contribution: This study contributes theoretically by advancing an integrated framework of cognitive load, multimedia learning, and scaffolding theory to assembly decoding; empirically by providing mixed-method evidence from a classroom remediation context; and practically by translating findings into actionable guidance for implementing AR as a remediation tool. Findings: All AR remediation learners reached or exceeded the curriculum proficiency threshold on the full-decoding posttest (minimum score = 66/100). On a parallel identification measure, the AR group showed a large pre-to-post improvement (ΔM = 49.75 percentage points, Cohen’s dz = 3.62). Technology acceptance was high across all four TAM constructs, with perceived ease of use significantly predicting both perceived usefulness and attitude toward use. Four qualitative themes were identified, spanning representational understanding, engagement, peer collaboration, and technical implementation constraints. Recommendations for Practitioners: Instructors should structure AR use around explicit 2D-to-3D mapping prompts, sequenced to rubric indicators; assign collaborative roles to formalize peer reasoning; and proactively plan for technical readiness through device specifications and backup provisions. Recommendation for Researchers: Research should examine which specific AR affordances benefit different learner profiles and clarify the underlying cognitive mechanisms using process-level measures. Impact on Society: Mobile AR, embedded in well-designed instructional sequences, may help struggling learners reach competency thresholds in technical drawing, with potential implications for workforce readiness in the engineering and manufacturing sectors. Future Research: Future studies should employ randomized or matched, multi-session designs with delayed posttests and transfer tasks, compare AR with alternative approaches such as CAD or VR, and validate the framework in neighboring domains such as geometric dimensioning and hydraulic schematic interpretation.