Metacognition is the capacity to monitor your own thinking — to notice the difference between understanding something and merely recognizing it. It is the faculty that tells you when to slow down, check a claim, or admit you are out of your depth.
Fluent AI output is unusually good at producing the feeling of understanding without its substance. An answer arrives complete, confident, and well-formed, and the reader's sense of having grasped it can run well ahead of any actual grasp. A 2026 study in Social Sciences validating a scale for epistemic trust in AI found that people's self-reported reflectiveness predicted how much authority they granted AI far better than their measured capacity for cognitive reflection — a gap between believing you are thinking critically and doing so.
This hub gathers work on that gap: how AI use reshapes self-monitoring, calibration, and the ability to know what you actually know.
Global reading scores just hit their lowest point on record. The more interesting finding isn't that AI is involved — it's that AI reading aids help almost nobody equally, and the readers they help least are the ones who were already good at reading.
By Joe Jurczyk
The debate has fixated on jobs. The more consequential question is what happens to thinking itself.
By Joe Jurczyk
APA Monitor • Jul 2026
The American Psychological Association surveys psychologists on how routine AI assistance is changing skill development and everyday reasoning.
Review of Education (Wiley/BERA) • Jan 2026
Maps the existing research literature on how LLM-based chatbots affect students' social and emotional learning, synthesizing findings across studies to surface recurring themes and open questions for future research.
MIT Media Lab • Jun 2025
An EEG study finding that heavy reliance on AI writing assistance is associated with reduced neural connectivity and weaker recall of one's own work.
Societies (MDPI) • Jan 2025
A peer-reviewed survey study linking frequent AI tool reliance to lower critical thinking scores, especially among users who never built the underlying skill unassisted.
Dalhousie University • August 14, 2026
An interdisciplinary one-day symposium in Halifax on how advances in AI, cognitive science, and neuroscience inform one another, with keynotes on tool use, creativity in humans vs. foundation models, and reasoning.
University of Porto • May 4, 2026
A hybrid academic conference (online May 4-5, in-person May 6-8 in Porto) on what AI reveals about understanding, intentionality, agency and consciousness, including how AI reshapes human cognition through extended and scaffolded practices like search, writing, and memory.