AI-assisted journal entries show higher lexical diversity and more emotion-word usage than unassisted entries¶

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

Subclaims¶

q3 i? Entries written with AI assistance exhibited significantly higher lexical diversity and increased negative and positive emotion-word usage than unassisted entries. → Sora Kang 2026

Evidence¶

Sora Kang 2026¶

Sora Kang. (2026). Stop Writing for Me: Generative Refusal in AI Tools for Thought. In Proceedings of CHI 2026 Workshop on Tools for Thought. https://arxiv.org/abs/2607.24751 doi:10.48550/arxiv.2607.24751

q3 · i? · causal · r2

Interaction-log analysis in the same 14-day crossover study compared AI-assisted and unassisted entries. The article reports "significantly higher Lexical Diversity (𝑞=. 025)" and increased emotion-word usage, indicating a broader emotional range articulated.

"entries written with AI assistance exhibited significantly higher Lexical Diversity (𝑞=. 025) and increased usage of both neg- ative (𝑞=. 001) and positive (𝑞=. 025) emotion words compared to unassisted entries"

Discussion¶

Related Claims¶

• Question-driven AI journaling increases intrinsic motivation and acting confidence among actors — the general claim this finding bears on, with the studies behind it
• GPT-assisted writing shows higher lexical diversity, richness, and noun/non-stopword ratios, while independently authored writing shows higher adverb ratio and sentence-length variability — possibly the same claim (merge candidate)
• AI-generated questions instead of draft text significantly reduce actors' cognitive burden during character journaling — related
• Both the course and the assistant increase behavioral engagement, measured as characters written per journaling prompt — related
• AI-assisted peer feedback enhanced lexical diversity only in the first writing task, with the group advantage disappearing by the second task — related
• Type-token ratio showed a non-monotonic relationship with predicted difficulty, with both very low and very high lexical diversity increasing predicted text difficulty — related