Twin agents as personal digital representations create distinct trust calibration challenges because they dissolve the boundary between AI and human decision-makers, unlike existing frameworks designed for clear separation.
hub Mixed citations
To trust or to think: Cognitive forcing functions can reduce over-reliance on AI in AI-assisted decision-making
Mixed citation behavior. Most common role is background (47%).
40
Pith papers citing it
712
external citations · Crossref
Background
47% of classified citations
hub tools
citation-role summary
background 11
method 3
other 1
citation-polarity summary
representative citing papers
Older Korean immigrants use pragmatic disengagement to avoid stressful technologies and interdependent navigation where digital skills are shared family resources, treating non-use as culturally grounded data refusal.
HANSEL extracts navigable evidence from agent trajectories with 83.7% precision and 88.8% recall on 45 tasks, reduces volume by 61.6%, and improves verification metrics in a 14-participant study.
Exploratory interview study with 17 developers identifies four forms of emergent oversight work for software agents and documents situated challenges and heuristics.
Summary reasoning traces from LLMs maintain task performance and increase trust and appeal relative to answer-only or full-trace conditions, but none of the formats improve users' metacognitive calibration on reasoning tasks.
Users entangle their lived experiences with AI predictions in menstrual tracking apps, leading to self-fulfilling prophecies, limited critical awareness from UI, and isolation for non-normative users.
AI-authored goals produce higher SMART quality scores but lower psychological ownership, commitment, importance, and goal-directed behavior than self-authored goals, with ownership as the mediating mechanism.
A folk theorem for LLMs proves that all feasible and individually rational outcomes can be sustained as ε-equilibria in repeated games where LLMs advise client populations, despite indirect observation.
LLMs produce interpretive closure in 87.5% of ambiguous social scenarios through narrative alignment, reversal, or normative advice, with first-person perspectives increasing alignment tendencies.
The study proposes the Gradual Voluntary Participation (GVP) framework to reconceptualize participatory AI governance in journalism as a gradual and voluntary process using a bidimensional matrix.
The authors introduce Agentivism as a learning theory for human-AI interaction that explains how durable capability develops through selective delegation, epistemic monitoring, reconstructive internalization, and transfer under reduced support.
Higher generative AI error rates reduce user reliance, but task difficulty does not significantly moderate this effect.
A qualitative study with 22 creative writers finds that the reflective value of AI refusals depends on alignment with users' situational thinking phases, cognitive beliefs, and views of AI roles.
A new benchmark exposes food-safety gaps in current LLMs and guardrails, and a fine-tuned 4B model is offered as a domain-specific fix.
LLM chat systems show large differences in reference quantity and quality, but users rarely click or engage with them.
Shorter LLM response latencies reduce perceived output thoughtfulness and usefulness, while task type affects prompting frequency independently of latency.
13 participants became convinced AI understands human values after chatbot interactions evaluated with the VAPT toolkit.
An experiment found LLM counterarguments improved group flexibility and satisfaction while AI mediation boosted minority participation but lowered psychological safety.
Two linked user studies find that LLM rationale correctness and certainty framing affect trust and decision confidence while presentation format does not, and incorrect rationales increase gaze attention and pupil size.
Author proposes adversarial co-thinking as a method of calibrating and triangulating multiple GenAI tools to generate critique during academic paper drafting, based on personal parallel use of Claude, ChatGPT, and Gemini.
RAID is a reflective agent system that infers intent from single expert edits and propagates corrections across compositional knowledge bases through a three-step architecture.
Exploratory user study of 48 participants finds trade-offs in efficiency, contextual alignment, and social comfort when AI writing assistance varies along synchronous and visual dimensions.
Mixed-methods study finds AI assistance linked to higher textual overlap with suggestions in writing tasks, and a reflective interface prototype increases user awareness of AI incorporation.
LLM reasoning traces and post-hoc explanations increase false trust in incorrect predictions, whereas contrastive dual explanations enhance users' ability to distinguish correct from incorrect AI outputs.
citing papers explorer
-
Human oversight of agentic systems in practice: Examining the oversight work, challenges, and heuristics of developers using software agents
Exploratory interview study with 17 developers identifies four forms of emergent oversight work for software agents and documents situated challenges and heuristics.