Measuring Students' Learning and Task Performance Using Semantic Multi-Modal Aids in Virtual Assembly Environments

Abstract

Learning and performance improvement during procedural task realization in Virtual Reality (VR) environments can be enhanced using different types of cognitive aids. Previous work in literature stressed on task learning (with performance drop) or vice versa. In this paper we propose the concept of semantic aids to be systematically used in virtual environments to increase user performance as well as learning, A Guided Virtual Reality Training System (GVRTS) is used for the assembly of three phase step-down transformer in the experimental study. Group 1 (20 students) used direct cognitive aids (change in color of object candidate for selection, blinking arrow showing final position, a color line bar under the object pointing to its accurate placement). Group2 (20 students) are trained with GVRTS where direct cognitive aids plus visual semantic aids are provided, group3 use direct plus audio semantic aids, and group4 (20 students) use direct aids plus audio and visual semantic aids. A comparison based on task performance, resulted in less performance drop in case of group2, group3 and group4 as compared to group 1. Experimental results also revealed that semantic aids considerably improved learning and knowledge transfer to the real world