Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 3, 2026Journal of Management Information SystemsOpen Access

How Reflection Enhances Task Factuality in the Use of Large Language Models

View Full Paper
Ask AI
Bookmark
Share

Authors

MTMonideepa TarafdarMAMartin AdamLNLong The Nguyen

Discussion

Loading...

Member takes

Overview

Randomized experiment reveals that human reflection enhances task factuality during text co-creation with AI, suggesting interaction designs should actively foster critical cognitive states.

Key Points

  • To examine how human reflection—conceptualized through interaction modes and cognitive states—enhances output factuality during collaborative text-generation tasks with large language models.
  • Conducted a randomized experiment with N=280 large language model users who wrote a short essay on a designated topic in partnership with an AI system.
  • Manipulated interaction mode (adversarial vs. conversational) and evaluated user cognitive states (shallow, dialogic, and critical) using survey measures and objective assessments.
  • Adversarial and conversational interaction modes differentially enhance task factuality during human-LLM co-creation tasks.
  • User cognitive states (shallow, dialogic, and critical) mediate the effect of the interaction mode on overall task factuality.

Cite This Study

Tarafdar et al. (2026) studied this question.

synapsesocial.com/papers/6a8a63d251cd0cd12ac14722https://doi.org/10.1080/07421222.2026.2692283
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Context-Grounded Factuality Enhancement in LLM Responses via Multi-Stage Critique and Refinement2025
  2. 2Supporting Self-Reflection at Scale with Large Language Models: Insights from Randomized Field Experiments in Classrooms2024 · 50 citations
  3. 3Facilitating Human-LLM Collaboration through Factuality Scores and Source Attributions2024 · 3 citations
  4. 4Supporting Self-Reflection at Scale with Large Language Models: Insights from Randomized Field Experiments in Classrooms2024 · 1 citations
  5. 5Self-Alignment for Factuality: Mitigating Hallucinations in LLMs via Self-Evaluation2024 · 2 citations