About the Cold Cognition Project

A community-led citizen science study into how cold-water immersion affects real-world cognitive performance.

What we're trying to understand

Cold-water immersion has attracted enormous popular interest, but the evidence on its cognitive effects is thin, inconsistent, and largely based on small, one-off laboratory studies. We want to do something different: collect dense, repeated measurements from real people in real conditions, over months and years.

The central questions driving the project are:

  • Acute effects – does a cold plunge sharpen or slow cognition in the hours immediately after? Is there a sweet spot, or does intensity or duration change the picture?
  • Dose and frequency – does plunging more often lead to different baseline cognitive performance? Is the relationship linear, or U-shaped?
  • Adaptation over time – do effects change as people become more experienced cold-water practitioners?

Because the data are collected by community members from community members, the findings belong to the community. Anonymised results will be shared openly.


The cognitive tasks

Each session takes around 5 minutes and consists of up to five short tasks, presented in a randomised order. All run in your browser – no app download required.

Reaction time & vigilance (PVT)

A 60-second Psychomotor Vigilance Task: respond as fast as you can when a stimulus appears. We measure median reaction time, how much it varies, and how many lapses (responses slower than 500 ms) occur – a sensitive index of mental fatigue and alertness.

Try it now ↗

Sustained attention & impulse control (SART)

Respond to frequent targets but withhold your response to rare non-targets. Errors of commission measure inhibitory control; response-time variability reflects how well attention is maintained over time.

Try it now ↗

Executive control & selective attention (Flanker)

Respond to a central arrow while ignoring flanking arrows that may point in a different direction. The difference in reaction time between congruent and incongruent trials – the conflict effect – is a well-validated measure of executive control.

Try it now ↗

Processing speed (Digit Symbol)

Match digits to symbols as quickly as possible using a key. Correct responses per minute captures processing speed and perceptual-motor coordination – one of the broadest indicators of overall cognitive efficiency.

Try it now ↗

Mood & subjective state

Four quick ratings: valence (how positive you feel), arousal/energy, stress, and mental sharpness. These let us separate objective cognitive performance from how people feel – and how the two relate.

Try it now ↗

What else shapes cognition – and why we capture it

Cognitive performance doesn't exist in a vacuum. Many factors influence how sharp your brain is on any given day. We record these not to be intrusive, but because failing to account for them would make the data scientifically worthless. All fields are optional, and the more context you provide, the more useful your contribution.

Plunge timing & dose

How long ago did you plunge? Water temperature, immersion depth, duration, and context (bath, lake, sea, cryotherapy) all contribute to the dose. Frequency over 7, 14, and 30 days lets us model adaptation.

Sleep

Sleep is the single strongest within-person predictor of reaction time and vigilance. Duration and quality are recorded daily so we can separate cold-water effects from the simple fact that you slept badly last night.

Caffeine & alcohol

Caffeine acutely improves vigilance; alcohol impairs it, with effects that persist well beyond the feeling of intoxication. We record caffeine since your last session and any alcohol in the past 24 hours.

Exercise & meal timing

Acute exercise can boost executive function temporarily; exercising to exhaustion can impair it. We also note time since your last meal, as low blood glucose affects sustained attention.

Illness & gastrointestinal symptoms

Even mild illness meaningfully impairs cognition. GI symptoms are captured weekly because some practitioners report gut-related effects from cold-water immersion – a poorly studied area.

Baseline profile

Age, sex, handedness, height, weight, and years of cold-water experience – collected once at sign-up. These allow us to model individual differences and avoid confounding group-level patterns with personal baseline variation.


How participating works
  1. Create an account and complete a brief baseline questionnaire (age, experience, etc.) and informed consent. No data is collected before consent.
  2. Log your plunges as you do them – water temperature, duration, depth, context. This takes under a minute.
  3. Receive gentle notifications (optional) around 15–30 minutes and again 2–3 hours after each plunge, prompting a short cognitive session. You can also complete sessions at any time independently of a plunge.
  4. Complete a 5-minute session – a few short cognitive tasks and a mood check-in.
  5. See your own data on your personal dashboard: reaction time trends, attention patterns, mood trajectories, and how they relate to your plunge history.

You can participate as much or as little as you like. More frequent sessions mean a richer personal dataset and a stronger contribution to the group analysis, but there is no obligation.


Your data & your rights
  • Consent comes first. No data is collected until you have read and agreed to the study information and given informed consent. You can withdraw at any time.
  • Pseudonymisation. All data is stored and analysed against a random participant ID, not your name or email address. Exports never contain direct identifiers.
  • You own your contribution. You can view your own data on your dashboard at any time. You can request full deletion of your records via your account settings – all raw data will be permanently removed within 30 days.
  • Open results. Aggregate and anonymised findings will be shared openly with the community and, where appropriate, submitted to peer-reviewed journals. Individual-level data is never shared.
  • No commercial use. Your data will never be sold or used for advertising or commercial profiling of any kind.

Aggregate or anonymised results already included in published analyses will not be retroactively altered, in line with standard research ethics practice. This is disclosed in the consent form you sign on joining.


Contact

Questions or concerns about the study? Email the research team at research@coldcognition.org.

Join the study