In a 30-participant user study, proactive paradigms are preferred over user-initiated interaction, with lower workload, fewer window switches, shorter editing time, and higher acceptance¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Proactive paradigms (L1-L3) are preferred over user-initiated interaction (L0), with stateful and adaptive settings (L2, L3) achieving the lowest perceived workload, highest usefulness, and strongest adaptability. → Youqing Fang 2026 q2 i? Interaction logs show proactive collaboration reduces window switching and task completion time, while acceptance rates increase from stateless to stateful and adaptive settings. → Youqing Fang 2026 (2)

Evidence¶

Youqing Fang 2026¶

Youqing Fang, Yinhao Tang, Yanan Sun, Jiangning Liu, Ziyi Wang, Xun Zhao, Bin Liu, Weiming Zhang, Kuikun Liu, Wenwei Zhang, Kai Chen. (2026). MindCopilot: Towards Formalizing and Evaluating Granular Human-LLM Co-Writing. arXiv preprint. https://arxiv.org/abs/2605.23535 doi:10.48550/arxiv.2605.23535

q2 · i? · causal · r1

Controlled user study with 30 participants recruited from universities and technology companies, using an online interactive writing system instantiating all four paradigms with a unified GPT-5.1 backend. Likert-scale measures covered perceived workload, usefulness, customization, and adaptability (Figure 6).

"Proactive paradigms (L1–L3) are preferred over user-initiated interac- tion (L0), with stateful and adaptive settings (L2, L3) achiev- ing the lowest perceived workload, highest usefulness, and strongest adaptability."

Youqing Fang 2026 (2)¶

Youqing Fang, Yinhao Tang, Yanan Sun, Jiangning Liu, Ziyi Wang, Xun Zhao, Bin Liu, Weiming Zhang, Kuikun Liu, Wenwei Zhang, Kai Chen. (2026). MindCopilot: Towards Formalizing and Evaluating Granular Human-LLM Co-Writing. arXiv preprint. https://arxiv.org/abs/2605.23535

q2 · i? · causal · r1

Interaction-log analysis in the same user study (Table 5) prints, for example, window switches falling from 15.76 (L0) to 8.15 (L3), editing time from 46.42 to about 31 minutes, and acceptance rate rising from 18.55 (L1) to 27.87 (L3).

"Interaction logs further corroborate these perceptions: proactive collaboration substantially re- duces window switching and task completion time, while ac- ceptance rates increase from stateless to stateful and adaptive settings."

Discussion¶

Related Claims¶

• Stateful proactive collaboration (L2) yields higher acceptance and lower editing effort than stateless proactive collaboration (L1), with lower acceptance in creative than scientific domains — related
• AI assistance reduces subjective mental effort across all tasks even when it does not reduce completion time, dissociating time and effort — related
• Students receiving Adaptive proactive hints based on HelpNeed predictions achieve significantly higher posttest optimality than Control students — related
• The relationship between scaffolding level and students' subsequent provide-context turns is not monotonic — related
• No single control-timing paradigm dominates across subjective measures: eight of ten questionnaire items show no significant differences between paradigms — related
• Proactive feedback produces significant post-intervention increases in Joint Mental Effort of dyads — related
• Proactive multimodal feedback significantly reduces debugging time on task compared to control — related
• Acceptance rates increase with writing progress, with the L2-over-L1 advantage widening in later stages — related
• Proactive flow-aware highlighting helps fiction writers recognize latent knowledge gaps they would not search for on their own — related
• AIfred reduces observed physical-digital context switches by 98% relative to screen-based assistance, while task completion time is longer — related
• Proactive multimodal feedback significantly increases debugging success in pair programming compared to no-feedback control — related