TRAINING WITH AI: STUDYING THE EFFECTS OF AI FEEDBACK ON REACTIONS AND SKILL ACQUISITION IN OCCUPATIONAL TRAINING

The rapid adoption of artificial intelligence (AI) technology in recent years has transformed many organizational processes. Among the processes that have integrated AI is occupational training. Yet, AI as a design component of occupational training remains underexplored in research. This dissertation investigated the effects of receiving AI feedback compared to human feedback on training reactions and learning. Drawing on feedback intervention theory (FIT) and human-AI interaction literature, AI feedback was hypothesized to benefit skill acquisition by mitigating training anxiety and strengthening task engagement. Trait goal orientation was predicted to moderate the relationships between feedback source, training anxiety, and task engagement. In post-hoc analyses, the effect of a hybrid human-AI feedback source on training reactions and learning outcomes was explored.

An experimental design with repeated measures was used in which undergraduate students (n = 302) participated in a brief virtual reality welding training and were assigned to receive AI, human, or human-AI feedback. Performance was captured on six welding trials to assess trainees’ skill acquisition. Growth curve modeling and path analysis were used to examine the theoretical model. Findings showed that receiving AI feedback does not directly benefit skill acquisition; however, it did so indirectly by reducing training anxiety which in turn enhanced skill acquisition. Performance-prove goal orientation negated the anxiety-reducing effect of AI feedback, while performance-avoid orientation amplified it. AI feedback was neither beneficial nor harmful for task engagement regardless of trainees’ goal orientation.

The findings of this study contribute to the literature on occupational training, workplace feedback, human-AI interaction, and AI in personnel management processes, each of which are discussed in detail. Practical implications are also offered including the affordances of AI for training feedback and the practical benefits of using AI for this purpose. Future research should focus on dynamic relationships between feedback source and training outcomes, other ways in which AI can be used as a design component of occupational training, and field studies.