ThinkReps
Evidence, not hype
Every exercise is based on published research. Here is what the studies found — including their limits.
Summaries are our own, in plain language. Follow the links to read the original papers.
When a tool does a mental job for us, we tend to practise that skill less. Here is what researchers have found about offloading thinking to technology.
Science
What they found: When people expected information to be saved on a computer, they remembered the information itself less well — but remembered where to find it.
How we use it: Why you can never look anything up mid-exercise.
Read the paper →Scientific Reports
What they found: People with more lifetime GPS use had worse spatial memory when finding their own way. In a three-year follow-up, heavier GPS use went with a steeper decline.
How we use it: Handing a skill to a machine can weaken it — the core idea behind ThinkReps.
Read the paper →Human Factors
What they found: Across aviation, medicine and industry, reliable automation leads people to monitor less and miss its errors (complacency and automation bias).
How we use it: Why AI never does the reps here; it only coaches.
Read the paper →Societies
What they found: In a survey of 666 people, more frequent AI tool use went with lower critical-thinking scores, partly explained by cognitive offloading. This shows a correlation, not cause and effect.
How we use it: A warning sign worth taking seriously — not proof.
Read the paper →CHI 2025
What they found: In a survey of 319 knowledge workers, more confidence in generative AI went with less critical thinking, while more self-confidence went with more.
How we use it: Our coach scores you and asks questions instead of doing the work.
Read the paper →arXiv
What they found: In an EEG study of 54 people writing essays, those using ChatGPT showed the weakest brain connectivity and struggled to quote their own essays.
How we use it: In Expression you write every word yourself, with pasting disabled.
Read the paper →Neuron
What they found: Six weeks of daily memory-palace training more than doubled how many words beginners could recall, and shifted their brain networks toward those of memory champions.
How we use it: The Memory exercise teaches the same technique.
Read the paper →Psychological Science
What they found: Practising recall (being tested) led to better retention a week later than spending the same time re-reading.
How we use it: Every exercise makes you retrieve, not re-read.
Read the paper →Psychological Bulletin
What they found: A review of hundreds of experiments: spreading practice over time beats cramming it into one session for long-term memory.
How we use it: Short daily workouts instead of rare long ones.
Read the paper →Journal of the Association for Consumer Research
What they found: The mere presence of your own smartphone — even switched off and face down — reduced available working memory. Leaving it in another room helped most.
How we use it: Focus asks you to put your phone away first.
Read the paper →CHI 2008
What they found: After interruptions, people made up for lost time by working faster, but paid for it with more stress, frustration and time pressure.
How we use it: Leaving the page during Focus costs points.
Read the paper →Educational Research Review
What they found: A meta-analysis of 54 studies with over 170,000 participants found better reading comprehension on paper than on screens, especially for informational texts.
How we use it: Screens make deep reading harder, so we train it on purpose.
Read the paper →Psychological Science
What they found: People with stronger number skills were less swayed by how numbers were framed and drew more precise meaning from them when deciding.
How we use it: Reasoning keeps your number sense in shape.
Read the paper →Review of Educational Research
What they found: A meta-analysis of 48 writing-to-learn programmes found a small positive effect on achievement — larger when learners reflected on their own understanding.
How we use it: Expression pairs writing with rating your own clarity.
Read the paper →Journal of Personality and Social Psychology
What they found: People with weaker skills tended to overestimate their performance; training improved both their skill and how accurately they judged themselves.
How we use it: We show how your self-rating compares with the coach's.
Read the paper →Behavioral and Brain Sciences
What they found: Argues that reasoning evolved to make and evaluate arguments in conversation: alone we tend to confirm what we already think, and we reason better when arguing with others.
How we use it: Debate puts an opponent in front of you.
Read the paper →Journal of Personality and Social Psychology
What they found: Asking people to “consider the opposite” reduced biased reading of evidence more than simply asking them to be unbiased.
How we use it: You argue a side you didn't choose, then make the best case for the other.
Read the paper →Why a mentally active life may matter beyond today's workout.
The Lancet Neurology
What they found: People with more education, mentally demanding jobs and stimulating hobbies tend to show fewer symptoms for the same amount of brain damage (“cognitive reserve”). The evidence is mostly observational.
How we use it: A reason to keep your mind active for life — not a promise.
Read the paper →Brain-training games have been oversold. ThinkReps is not a medical product, has not been tested in a clinical trial and will not raise your IQ. What the evidence does support: practising real skills keeps them sharp, recall and spacing make learning stick, and arguing well can be trained. Several AI studies above are early, correlational or not yet peer-reviewed — we will update this page as better evidence arrives.
Psychological Science in the Public Interest
What they found: A major review: brain-training games improve the trained task and sometimes similar tasks, but there is little evidence they improve everyday thinking.
How we use it: Why we train real-world skills and never promise a higher IQ.
Read the paper →Paraphrased from their published work.
Be sceptical of claims that brain games make you smarter in general: practice mostly improves what you practise.
— Daniel Simons et al. (2016)
Reasoning works best as a social activity. We spot flaws in other people's arguments far more easily than in our own, so arguing with others makes us think better.
— Hugo Mercier & Dan Sperber (2011)
Complacency with automation is not laziness: when a system is usually right, even experts pay less attention to it and miss its rare mistakes.
— Raja Parasuraman & Dietrich Manzey (2010)
Testing is not just a way to measure learning. The act of recalling something is one of the most powerful ways to learn it.
— Henry Roediger & Jeffrey Karpicke (2006)
A mentally engaged life is linked to more resilience against brain ageing, though proving cause and effect is still hard.
— Yaakov Stern (2012)