Exploring the Effects of Explainable Artificial Intelligence (XAI) Across AI Tools with Different Levels of Decision Complexity
Year: 2026
Type: dissertation
Abstract: Drawing on Mental Models Theory and Task Complexity Theory, this research examines how visual and textual Explainable Artificial Intelligence (XAI) explanations influence users’ trust and mental model clarity in AI recommendation systems. Using a 2 × 3 mixed factorial experiment, the study compared visual and textual explanations across three decision tasks with different levels of complexity: pizza recommendation, tour planning, and house recommendation. The visual condition presented SHAP feature-importance plots and decision-tree reasoning paths, while the textual condition presented the same explanatory content in natural language. Results showed that visual explanations significantly improved users’ mental model alignment with the AI’s feature-importance structure. In addition, task complexity partially moderated the effect of explanation format on mental model clarity, and higher mental model clarity was positively associated with trust. However, visual explanations did not directly produce significantly higher trust than textual explanations. Overall, the findings suggest that visual XAI is effective in strengthening users’ understanding of AI reasoning, while trust formation depends on broader factors beyond explanation format alone.
Author: Khabaziyan, Pegah
Institution: Concordia University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221492690
