Learner-side gains from curiosity modulation persist even when tutor-side instructional quality remains unchanged or degrades, suggesting curiosity operates as a partially independent interaction-level mechanism¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The dissociation between learner-side and tutor-side outcomes is most visible in the Gemini condition, where learner curiosity and conversational agency improve significantly while instructional quality and cognitive load management degrade. → Ganganath 2026 q2 i? Pearson correlations between learner-side and tutor-side dimensions weaken consistently across nearly all interaction pairs under CURIOBOT compared to Baseline. → Ganganath 2026 (2)

Evidence¶

Ganganath 2026¶

Ganganath, G., Bolonghege, P., Lyu, Q., Varakantham, P., & Kandappu, T. (2026). Curiosity as Linguistic Intervention: Using LLM Tutoring Dialogues to Influence Exploratory Learning Behavior. https://arxiv.org/abs/2606.22349 doi:10.48550/arxiv.2606.22349

q2 · i? · causal · r2

Dissociation analysis (§5.3) drawing on the Table 3 dimension-wise scores. In the Gemini condition, learner-side dimensions improve significantly while tutor-side instructional quality (T1) and cognitive load management (T3) decrease significantly; only p-values printed, no effect sizes.

"From Table 3, we can see that the dissociation between learner-side and tutor-side outcomes is most visible in the Gemini results."

Ganganath 2026 (2)¶

Ganganath, G., Bolonghege, P., Lyu, Q., Varakantham, P., & Kandappu, T. (2026). Curiosity as Linguistic Intervention: Using LLM Tutoring Dialogues to Influence Exploratory Learning Behavior. https://arxiv.org/abs/2606.22349. doi:10.48550/arxiv.2606.22349

q2 · i? · causal · r2

Decoupling analysis (§5.3) using Pearson correlations between learner-side and tutor-side evaluation dimensions under Baseline and CURIOBOT (Figure 5). Correlation values are printed in the figure but no inferential tests on the difference are reported.

"Under CURIOBOT, correlations between learner-side and tutor-side dimensions weaken consistently across nearly all interaction pairs"

Discussion¶

Learner Variables¶

• Interest — predictor: learners who differ on it differ in outcomes

Related Claims¶

• Curiosity-oriented linguistic interventions via CURIOBOT consistently improve all learner-side exploratory dimensions relative to baseline LLM tutoring across model families — related
• Curiosity-modulated tutoring produces roughly 2.4× more conversational turns than baseline under fixed time budgets — related
• Interactive dialogue with the AI tool, available only in some sites, promoted deeper engagement and student agency during revision — related
• Learner-side gains from curiosity modulation generalize across model families, academic domains, and topic complexity levels — related
• Users alter expectations and behavior based on perceived AI capabilities even when actual AI performance is unchanged (the placebo effect of AI) — related
• Curiosity gains depend on operator sequencing: how operators are ordered across turns, not only which operator is applied, shapes curiosity induction — related
• Commercially optimized study-oriented modes achieve higher learner-side scores than baseline, and CURIOBOT shifts general-purpose LLM behavior toward those patterns via prompting alone — related
• Family-side evidence on GenAI guidance is thin, largely descriptive, and centered on parental beliefs and mediation strategies — related
• AI's role in shaping interaction, epistemic agency, and shared regulation in collaborative learning remains under-theorized — related
• Trained students engage in more conversational turns with the GenAI tool — related