How Much Dehydration Does It Take? Part One: What the Experiment Shows

Abstract

"Drink more water" is among the most repeated pieces of cognitive advice in circulation, and among the least examined. The useful question is not whether severe dehydration impairs thinking — it does — but at what deficit anything measurable begins.

This series covers three papers. Part One takes the experimental end: a trial in which twelve young men went thirty-six hours without water, were tested, then drank a litre and a half and were tested again.

It shows clear effects on memory, attention and mood, and recovery after drinking. It is also small, unblinded, and studies a deficit almost nobody encounters — and each of those limits shapes what the result can be used for.

1.The Design

Na Zhang, Songming Du, Jianfen Zhang and Guansheng Ma, working between Peking University and the Chinese Nutrition Society, published this self-controlled trial in the International Journal of Environmental Research and Public Health in 2019.

Twelve male students from a medical college in Cangzhou completed it. After an overnight fast, baseline measurements were taken: first morning urine and blood osmolality to establish hydration state, blood pressure, a visual analogue scale for thirst, a mood questionnaire, and a battery of cognitive tests.

The battery covered digit span forward and backward, digit-symbol substitution, a sustained arithmetic task, and a Stroop test — short-term memory, processing speed and attentional control.

Participants then went thirty-six hours without any water, while still eating three meals a day. They were retested, then drank 1,500 mL of purified water over fifteen minutes, waited an hour, and were tested a third time.

"Self-controlled" means each participant served as his own comparison. There was no separate control group, and no attempt at blinding was possible: a man who has not drunk for a day and a half knows it.

2.What Dehydration Did

Against their own baseline, participants after thirty-six hours showed lower vigour and lower esteem-related affect on the mood inventory, a lower total digit span score, and a higher error rate on the sustained arithmetic task.

So: short-term memory down, attention less accurate, mood flatter. The direction is unsurprising and the consistency across measures is worth something.

The magnitudes deserve attention rather than the p-values. Total digit span fell from 14.3 to 13.3 — one item. Error rate on the arithmetic task rose from 0.01 to 0.16. The first is a small change on a noisy measure; the second is a sixteen-fold rise on a measure that started near the floor, which is a different kind of statement and easy to misread as the same one.

3.What Drinking Did

One hour after 1,500 mL of water, fatigue and total mood disturbance had improved against the dehydrated state. Digit span rose. So did the number of correct digit-symbol substitutions, reading speed and a composite measure of mental work capacity, while reaction time fell.

Reading speed changed most dramatically, from 339.3 to 486.4 units per minute. A forty-three per cent improvement in one hour is larger than dehydration research usually produces, and in a repeated-testing design with no control group, practice on a task performed for the third time in three days is the obvious competing explanation.

One reported figure does not add up. Forward digit span after rehydration is given as 8.6 and the total as 15.4, which leaves 6.8 for the backward score; the paper reports 1.2. The forward and total figures are internally consistent with each other and with the dehydration values, so the backward figure appears to be a transcription error rather than a finding — but it is printed as a result, and a reader should know not to build on it.

4.What Thirty-Six Hours Is

This is the central limit, and it is not a criticism of the study — it is a boundary on what the study answers.

Thirty-six hours without any fluid is not a state that office workers, students or athletes ordinarily reach. The everyday question is whether a one or two per cent body mass deficit — the range produced by a warm afternoon and a forgotten water bottle — does anything measurable.

This design cannot answer that, because it did not test it. Establishing that a severe deficit impairs cognition tells you the relationship exists; it does not tell you where along the curve it becomes detectable, and the shape of that curve is the whole practical question.

Twelve participants, all male, all young, all students at one medical college. The authors are describing a mechanism, not a population.

5.The Unblindable Problem

The deeper difficulty is that this design cannot separate the physiology of dehydration from the experience of being thirsty.

A participant who has been deprived of water for a day and a half knows he is dehydrated, expects to perform worse, and is being asked to report his own mood. Vigour and esteem-related affect are self-reported constructs. The cognitive measures are objective, but effort is not.

This is not a flaw the authors could have designed away in an oral-deprivation protocol. It is intrinsic — and Part Three covers the studies that found a way around it, along with what happened when they did.

Editorial Comment

MindHeaven® makes no claim relating to hydration, and sells nothing that substitutes for drinking water.

One safety point, stated plainly because this article describes a protocol that sounds easy to imitate. Thirty-six hours without fluid was conducted under supervision with blood osmolality monitoring in healthy young men. It is not a self-experiment. Fluid deprivation is genuinely dangerous in older adults, in anyone taking diuretics, and in kidney or heart conditions, and there is nothing to be learned from repeating it at home.

The reason this topic is in a cognitive performance library is that hydration is one of the few interventions in this space that is free, immediate and available to everyone — which makes it worth stating accurately rather than either dismissing or inflating.

Part Two leaves the laboratory for a cohort of nearly two thousand older adults, and asks the question this trial cannot: whether how much you drink predicts how your cognition changes.