Four types of LLM reliance (Strategic 34.3%, Instrumental 30.9%, Dialogic 30.4%, Dependent 4.5%) were identified among undergraduates, with AI literacy predicting type and value/cost beliefs predicting intensity.
hub Canonical reference
Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task
Canonical reference. 86% of citing Pith papers cite this work as background.
42
Pith papers citing it
Background
86% of classified citations
abstract
This study explores the neural and behavioral consequences of LLM-assisted essay writing. Participants were divided into three groups: LLM, Search Engine, and Brain-only (no tools). Each completed three sessions under the same condition. In a fourth session, LLM users were reassigned to Brain-only group (LLM-to-Brain), and Brain-only users were reassigned to LLM condition (Brain-to-LLM). A total of 54 participants took part in Sessions 1-3, with 18 completing session 4. We used electroencephalography (EEG) to assess cognitive load during essay writing, and analyzed essays using NLP, as well as scoring essays with the help from human teachers and an AI judge. Across groups, NERs, n-gram patterns, and topic ontology showed within-group homogeneity. EEG revealed significant differences in brain connectivity: Brain-only participants exhibited the strongest, most distributed networks; Search Engine users showed moderate engagement; and LLM users displayed the weakest connectivity. Cognitive activity scaled down in relation to external tool use. In session 4, LLM-to-Brain participants showed reduced alpha and beta connectivity, indicating under-engagement. Brain-to-LLM users exhibited higher memory recall and activation of occipito-parietal and prefrontal areas, similar to Search Engine users. Self-reported ownership of essays was the lowest in the LLM group and the highest in the Brain-only group. LLM users also struggled to accurately quote their own work. While LLMs offer immediate convenience, our findings highlight potential cognitive costs. Over four months, LLM users consistently underperformed at neural, linguistic, and behavioral levels. These results raise concerns about the long-term educational implications of LLM reliance and underscore the need for deeper inquiry into AI's role in learning.
hub tools
citation-role summary
background 7
citation-polarity summary
roles
background 7representative citing papers
Developers using AI showed the same core problem-solving behaviors as those without but differed in how they became stuck and recovered, with AI helping or hindering in specific cases.
NIRVANA supplies keystroke-level logs, complete ChatGPT dialogues, and copied content from 77 students to reconstruct AI-assisted essay writing and classify students into four behavioral profiles: Lead Authors, Collaborators, Drafters, and Vibe Writers.
Proposes the CoRe-3 (FJS) competency model separating Framing, Judging, and Steering for generative AI use, with preliminary validation via simulations on an open platform showing skill dissociation and validity.
The paper identifies a Reasoning-Sycophancy Paradox and introduces the EduFrameTrap benchmark to evaluate how LLMs handle epistemic pressure in tutoring across multiple subjects.
AI alignment must move beyond assuming users have fully formed goals and instead provide active cognitive support to help form and refine intent over time.
Critical Inker scaffolds critical reflection during AI-assisted writing via Socratic questioning and visual logical-error feedback, reporting 91.2% argument overlap with ground truth and 87% validity accuracy in a pilot evaluation.
RelianceScope is a new analytical framework that maps AI reliance into nine engagement patterns across help-seeking and response-use, situated in students' prior knowledge and instructional context, validated on programming course logs.
TaskLens uses LLMs to generate task-specific scaffolded interfaces that reduce perceived workload and improve performance and concept learning for novices using professional 3D software.
Introduces the Techno-Supremacy Doctrine as an analytical framework and finds that AI executive discourse shows polarization with a general increase in pro-technology-solution narratives after ChatGPT, often acknowledging risks only to advocate further tech development.
Ethnographic study of 51 AI chatbot users finds that perceived gains in individual agency shape sustained usage patterns more than accuracy or reliability concerns.
A multisite biometric study finds lower cognitive engagement under AI assistance via EEG and blink rate, with physiological-performance links present only in the non-AI condition.
Large-scale topic modeling of 270k Reddit posts shows GenAI discourse in education shifting from detection-evasion to enforcement, with K-12 teachers emphasizing cognitive dependency, academics focusing on detection, students on career anxiety, and adversarial themes driving engagement and cross-sta
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.
A minimal three-variable dynamical model of human-AI feedback predicts that increasing reliance on AI induces a transition to a low-diversity suboptimal equilibrium, interpreted as an emergent information bottleneck.
Interviews with 22 developers produced a preliminary reliance-control framework that uses levels of control over AI to identify appropriate reliance in software engineering.
Agentic entropy names the systemic drift in AI coding agents away from architectural intent; a new framework using conformity seeding, reasoning monitoring, and causal graph interfaces supplies process-level oversight to complement existing review methods.
Trust-driven routine use of generative AI is linked to reduced cognitive engagement in STEM students, with higher technophilic traits increasing vulnerability.
VizCopilot integrates topic modeling with document visualization to support user oversight of retrieved context in enterprise chatbots, enabling detection of misalignments and adaptation of prompting strategies.
Interviews reveal a four-stage vibe coding workflow that accelerates prototyping while introducing tensions between quick efficiency and reflective design intention, plus asymmetries in trust and ownership.
AI argumentative feedback on community notes produces larger quality improvements than supportive or neutral feedback in a hybrid moderation experiment.
The paper presents a proof-of-concept closed-loop system using consumer EEG to detect high cognitive engagement and defer multi-agent robotic communications until lower workload.
Copa is a theory-guided multimodal LLM agent that supports high school computational modeling through adaptive feedback, shown in a 33-dyad study to increase student confidence and conceptual verbalization without fostering dependence.
This position paper advocates shifting AI education in materials discovery from basic tool access to a workflow-aligned literacy model that builds scientific judgment and equitable outcomes.
citing papers explorer
-
Four Types of LLM Reliance and Their Predictors Among Undergraduate Writers: A Mixed-Methods Study at a Minority-Serving R1 University
Four types of LLM reliance (Strategic 34.3%, Instrumental 30.9%, Dialogic 30.4%, Dependent 4.5%) were identified among undergraduates, with AI literacy predicting type and value/cost beliefs predicting intensity.
-
ChatGPT: Friend or Foe When Comprehending and Changing Unfamiliar Code
Developers using AI showed the same core problem-solving behaviors as those without but differed in how they became stuck and recovered, with AI helping or hindering in specific cases.
-
NIRVANA: A Comprehensive Dataset for Reproducing How Students Use Generative AI for Essay Writing
NIRVANA supplies keystroke-level logs, complete ChatGPT dialogues, and copied content from 77 students to reconstruct AI-assisted essay writing and classify students into four behavioral profiles: Lead Authors, Collaborators, Drafters, and Vibe Writers.
-
Framing, Judging, Steering: An Assessable Competency Model for Teach-ing Students to Reason With Generative AI
Proposes the CoRe-3 (FJS) competency model separating Framing, Judging, and Steering for generative AI use, with preliminary validation via simulations on an open platform showing skill dissociation and validity.
-
Sycophancy is an Educational Safety Risk: Why LLM Tutors Need Sycophancy Benchmarks
The paper identifies a Reasoning-Sycophancy Paradox and introduces the EduFrameTrap benchmark to evaluate how LLMs handle epistemic pressure in tutoring across multiple subjects.
-
Alignment has a Fantasia Problem
AI alignment must move beyond assuming users have fully formed goals and instead provide active cognitive support to help form and refine intent over time.
-
Critical Inker: Scaffolding Critical Thinking in AI-Assisted Writing Through Socratic Questioning
Critical Inker scaffolds critical reflection during AI-assisted writing via Socratic questioning and visual logical-error feedback, reporting 91.2% argument overlap with ground truth and 87% validity accuracy in a pilot evaluation.
-
RelianceScope: An Analytical Framework for Examining Students' Reliance on Generative AI Chatbots in Problem Solving
RelianceScope is a new analytical framework that maps AI reliance into nine engagement patterns across help-seeking and response-use, situated in students' prior knowledge and instructional context, validated on programming course logs.
-
AI usage patterns are shaped by perceived gains in human agency
Ethnographic study of 51 AI chatbot users finds that perceived gains in individual agency shape sustained usage patterns more than accuracy or reliability concerns.
-
Using Biometrics to Understand AI-Assisted Coding Performance and its Perception
A multisite biometric study finds lower cognitive engagement under AI assistance via EEG and blink rate, with physiological-performance links present only in the non-AI condition.
-
ChatGPT vs Teachers vs Students: Large-Scale Analysis of Generative AI Discourse in Education Communities on Reddit
Large-scale topic modeling of 270k Reddit posts shows GenAI discourse in education shifting from detection-evasion to enforcement, with K-12 teachers emphasizing cognitive dependency, academics focusing on detection, students on career anxiety, and adversarial themes driving engagement and cross-sta
-
Overreliance in Writing Tasks: Exploring Similarity-Based Measures of AI Influence on Writing and Proposing a Reflective Writing Interface Intervention
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.
-
Human-AI Co-Evolution and Epistemic Collapse: A Dynamical Systems Perspective
A minimal three-variable dynamical model of human-AI feedback predicts that increasing reliance on AI induces a transition to a low-diversity suboptimal equilibrium, interpreted as an emergent information bottleneck.
-
Towards an Appropriate Level of Reliance on AI: A Preliminary Reliance-Control Framework for AI in Software Engineering
Interviews with 22 developers produced a preliminary reliance-control framework that uses levels of control over AI to identify appropriate reliance in software engineering.
-
Beyond the 'Diff': Addressing Agentic Entropy in Agentic Software Development
Agentic entropy names the systemic drift in AI coding agents away from architectural intent; a new framework using conformity seeding, reasoning monitoring, and causal graph interfaces supplies process-level oversight to complement existing review methods.
-
Thinking Less, Trusting More: GenAI's Impacts on Students' Cognitive Habits
Trust-driven routine use of generative AI is linked to reduced cognitive engagement in STEM students, with higher technophilic traits increasing vulnerability.
-
Multi-Modal Multi-Agent Robotic Cognitive Alignment enabled by Non-Invasive Consumer Brain Computer Interfaces: A Proof of Concept Exploration
The paper presents a proof-of-concept closed-loop system using consumer EEG to detect high cognitive engagement and defer multi-agent robotic communications until lower workload.
-
A Theory-Guided LLM Pedagogical Agent for STEM+C Scaffolding Without Over-Reliance
Copa is a theory-guided multimodal LLM agent that supports high school computational modeling through adaptive feedback, shown in a 33-dyad study to increase student confidence and conceptual verbalization without fostering dependence.
-
Preparing Students for AI-Powered Materials Discovery: A Workflow-Aligned Framework for AI Literacy, Equity, and Scientific Judgment
This position paper advocates shifting AI education in materials discovery from basic tool access to a workflow-aligned literacy model that builds scientific judgment and equitable outcomes.
-
Prober.ai: Gated Inquiry-Based Feedback via LLM-Constrained Personas for Argumentative Writing Development
Prober.ai constrains LLMs via personas and JSON schemas to deliver gated, inquiry-based questions on argumentative writing weaknesses, aiming to reduce cognitive debt from AI overuse.
-
The Crutch or the Ceiling? How Different Generations of LLMs Shape EFL Student Writings
Advanced LLMs improve EFL writing scores and diversity for lower-proficiency students but correlate with lower expert ratings on deep coherence, acting more as crutches than scaffolds.
-
The Epidemiology of Artificial Intelligence
AI functions as a determinant of health with ambient and personal exposure types, requiring new epidemiological study designs beyond current experiments.
-
Participatory, not Punitive: Student-Driven AI Policy Recommendations in a Design Classroom
Student-facilitated workshops in one design class produced AI policies highlighting double standards in disclosure requirements between students and faculty, demonstrating value in participatory governance.
-
Measuring Changes in Instructor Class Design and Student Learning After the Release of Large Language Models (LLMs)
A pilot mixed-methods study at one university uses surveys and pre/post-LLM grade data to document patterns in faculty course design and student learning outcomes after generative AI release.
-
AI Scientists as Engines of Discovery: A Case for Development within Reformed Institutions
Position paper arguing that multi-agent AI systems can become AI scientists and calling for reformed scientific institutions to support their development with emphasis on verification and dual-use safety.
-
Before You Think: System 0, AI-Mediated Cognition and Cognitive Colonization
System 0 is positioned as theoretically distinct from Tri-System Theory and Thinkframes, with cognitive colonization as a key mechanism of invisible AI influence on cognition.
-
Exploratory Responsiveness and Adaptive Rigidity under AI-Assisted Optimization
A dynamical framework is proposed where AI assistance substitutes for or amplifies exploratory search on rugged landscapes depending on the level of adaptive responsiveness.
-
Position: Adopting AI in Practice Does Not Guarantee the Productivity Boost
Adopting AI does not guarantee productivity boosts due to five moderating factors (human resource composition, baseline capability, learning curve, incentives for fair use, and objective flexibility), by revising an existing economic model.
-
Security, Privacy, and Ethical Risks in OpenClaw
The paper analyzes security, privacy, and ethical risks in the OpenClaw AI agent system arising from its architecture, storage, tool use, and integrations, arguing these form major barriers to trustworthy adoption.
-
What if AI systems weren't chatbots?
Chatbot AI systems often fail complex needs while projecting authority, contributing to deskilling, labor displacement, economic concentration, and high environmental costs, so alternative pluralistic and task-specific designs are needed.
-
Counterargument for Critical Thinking as Judged by AI and Humans
Student-written counterarguments to AI-generated thesis statements demonstrate logical reasoning as a component of critical thinking, and LLMs can assess such writing at scale with moderate agreement to human raters (Gwet's AC2 ~0.33).
-
Position: the Stochastic Parrot in the Coal Mine. Model Collapse is a Threat to Low-Resource Communities
Model collapse threatens AI democratization by disproportionately impacting low-resource and marginalized communities through reduced training efficiency and data distributions skewed away from distribution tails.
-
Brainrot: Deskilling and Addiction are Overlooked AI Risks
AI safety literature overlooks cognitive deskilling and addiction risks from generative AI despite public concern about them.
-
Report on CHIIR 2026 Workshop on Generative AI and Academic Search (GAI&AS)
A summary of the CHIIR 2026 workshop on Generative AI and Academic Search, covering themes of foundations, applications, and search-as-learning.
- Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems