Self-reported mental effort increased less across tasks in the ChatGPT condition while perceived difficulty rose equally in both groups¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q3
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? A significant tool×task interaction for self-reported mental effort (χ²(2)=8.88, pHolm=.047) showed mental effort rose further in the No AI group on Tasks 2 and 3. → Bergh 2026 q3 i? There was no significant tool×task interaction for self-reported difficulty; both groups rated subsequent tasks as more difficult (main effect of task p<.001). → Bergh 2026 (2)

Evidence¶

Bergh 2026¶

Bergh, C., Tag, B., Vassar, A., & Renzella, J. (2026). Your Programming Students'Cognition with ChatGPT: Higher Performance, Lower Retention, and Reduced Ownership. https://arxiv.org/abs/2609.21194 doi:10.48550/arxiv.2609.21194

q3 · i? · causal · r2

Confirmatory likelihood-ratio test within a linear mixed model of Paas mental-effort ratings (task×tool, random intercept per participant), Holm-corrected across four cognitive-load indices. Follow-up showed mental effort increased further in the No AI group at Task 2 (beta = 1.31) and Task 3 (beta = 1.13).

"There was a significant tool×task interaction for self-reported mental effort (𝜒 2(2)= 8.88, 𝑝=. 012, 𝑝Holm =. 047)"

Bergh 2026 (2)¶

Bergh, C., Tag, B., Vassar, A., & Renzella, J. (2026). Your Programming Students'Cognition with ChatGPT: Higher Performance, Lower Retention, and Reduced Ownership. https://arxiv.org/abs/2609.21194. doi:10.48550/arxiv.2609.21194

q3 · i? · causal · r2

Same mixed-model confirmatory test applied to Paas difficulty ratings. The interaction was not significant, though follow-up showed a significant main effect of task (F(2, 106.0)=38.71, p<.001): both groups rated the tasks as getting harder.

"There was no significant tool×task interaction for self-reported difficulty (𝜒 2(2)= 2.44,𝑝=. 296,𝑝Holm =. 591) and remains insignificant after Holm correction"

Discussion¶

Learner Variables¶

• Working Memory — outcome: instruction changes it

Related Claims¶

• Confirmatory physiological tests found no significant tool×task interactions for pupil Δ or ln RMSSD, with substantial data loss limiting interpretation — related
• Mental effort and study time increase with a topic's element interactivity level, making them indicators of element interactivity — related
• Prior genAI access and incentives interact on perceived difficulty of the memory test — related
• GenAI access reduces perceived difficulty of the assisted reasoning task — related
• Solving a creative problem-solving task with ChatGPT is perceived as easier and requires less mental effort, but is not more interesting — related
• No differences were found on mental effort during instruction or on transfer across illusion-of-control and instruction-method conditions — related
• Prior knowledge and intervention condition predict repeated cognitive load and interest measures: prior knowledge lowers effort and difficulty and raises interest and understanding, and prompting groups perceive tasks as more difficult — related
• AI assistance reduces subjective mental effort across all tasks even when it does not reduce completion time, dissociating time and effort — related
• Physiological measures have limited validity for effort, especially for distinguishing data-driven from goal-driven effort — related
• Motivational interventions can lower perceived cognitive load and improve retention and transfer, indicating load perception is partly independent of task difficulty — related