Do We Know What They Know We Know? Calibrating Student Trust in AI and Human Responses Through Mutual Theory of Mind

TL;DR

This study reveals that in educational settings, students' trust in AI and humans are influenced by different factors, leading to high trust but low reliance on humans and vice versa for AI, challenging the assumption that trust and reliance are always coupled.

Contribution

It introduces the use of Mutual Theory of Mind to distinguish between trust and reliance, showing they are independently influenced by epistemic and social factors in student-AI interactions.

Findings

Students trust humans but rely less due to social barriers.

Students trust AI less but rely more due to social affordances.

Trust and reliance are driven by different social and epistemic factors.

Abstract

Trust and reliance are often treated as coupled constructs in human-AI interaction research, with the assumption that calibrating trust will lead to appropriate reliance. We challenge this assumption in educational contexts, where students increasingly turn to AI for learning support. Through semi-structured interviews with graduate students (N=8) comparing AI-generated and human-generated responses, we find a systematic dissociation: students exhibit high trust but low reliance on human experts due to social barriers (fear of judgment, help-seeking anxiety), while showing low trust but high reliance on AI systems due to social affordances (accessibility, anonymity, judgment-free interaction). Using Mutual Theory of Mind as an analytical lens, we demonstrate that trust is shaped by epistemic evaluations while reliance is driven by social factors -- and these may operate independently.