Water Deprivation

Abstract

Water deprivation, which MeSH classifies under motivation, is the withholding of water in a structured experimental situation. This article covers thirst as the motivational state that deprivation produces and the drive-reduction account of it, the brain circuits in the lamina terminalis that drive and stop drinking, water restriction as a motivator for trained animals and its welfare limits, and the effects of water deprivation and dehydration on human thirst, cognition and mood.

Keywords: water deprivation, thirst, dehydration

Thirst motivates animals to drink in order to maintain fluid balance, and has conventionally been viewed as a homeostatic response to changes in blood volume or tonicity (Zimmerman et al., 2016). Deprivation or restricted access to either food or fluids is a common research procedure in laboratory animals (Rowland, 2007).

What Water Deprivation Is

MeSH defines Water Deprivation as the withholding of water in a structured experimental situation, files it under motivation and lists no narrower descriptors. The term therefore names a procedure rather than a bodily state: an experimenter removes or rations water, and the resulting thirst is used either as an object of study or as a motivator for other behavior.

The procedure is old in experimental psychology. Stellar and Hill measured the rat's rate of drinking as a function of the amount of water deprivation (Stellar & Hill, 1952), and a long review set out the physiology of thirst that such experiments were probing (Fitzsimons, 1972). In humans, 24 hours of water deprivation caused significant intracellular and extracellular depletions, thirst, and a dry unpleasant tasting mouth (Rolls et al., 1980).

Thirst and Drive

Hull's behavior theory treated deprivation as the source of drive (Hull, 1943). A study of thirst-associated preoptic neurons restates the classical drive reduction hypothesis, that animals learn particular goal-directed behaviors to reduce the level of an aversive drive state, and reports results consistent with it: mice vigorously avoided photoactivation of thirst-associated neurons, and the activity of dehydration-activated neurons in the median preoptic nucleus established a scalable, persistent and aversive internal state that dynamically controlled thirst-motivated behavior (Allen et al., 2017). Drive reduction may not be the whole account of thirst motivation: the same study notes that electrical stimulation experiments indicate that motivational states increase the incentive value of stimuli or behaviors that led to reward, and that the extent to which thirst-motivated behaviors arise from negative-valence drive reduction working with separate positive-valence incentive mechanisms remains to be explored (Allen et al., 2017). A review of motivation concepts in behavioral neuroscience evaluates hydraulic drives and drive reduction alongside incentive motivation, hedonic reactions and reward 'liking' versus 'wanting' (Berridge, 2004). Thirst has long been thought of as a negative homeostatic feedback response to increases in blood solute concentration or decreases in blood volume, but emerging evidence suggests a role for thirst as a feedforward anticipatory response that precedes physiological challenges (Gizowski & Bourque, 2018).

Thirst is important for maintaining body fluid homeostasis and may arise from deficits in either intracellular or extracellular fluid volume; neural signals arising from osmotic and hormonal influences on the lamina terminalis may be integrated within the brain, with afferent information relayed from intrathoracic baroreceptors via the hindbrain (McKinley & Johnson, 2004). Altered body fluid homeostasis produces sensory and perceptional changes, and the brain integrates information from various bodily sources so that thirst and sodium appetite are in a true sense the synthetic products of the nervous system (Johnson, 2007).

Drinking slows markedly before the deficit is repaired. During rehydration after 24 hours of deprivation, subjects drank 65% of their total intake within 2.5 min, and the marked decrease in drinking rate thereafter, and the alleviation of thirst, occurred before plasma dilution had become significant; the authors concluded that presystemic factors may be important for drinking termination in humans (Rolls et al., 1980). Drinking water when thirsty is a pleasant experience associated with activation in the anterior cingulate cortex and orbitofrontal region, whereas the unpleasantness of overdrinking after satiation is associated with activations in the midcingulate cortex, insula, amygdala and periaqueductal grey (Saker et al., 2014).

Neural Circuits of Thirst

Thirst is the basic instinct to drink water (Oka et al., 2015). Two genetically separable neural populations in the subfornical organ trigger or suppress it: optogenetic activation of excitatory neurons marked by the transcription factor ETV-1 evoked intense drinking behavior even in fully water-satiated animals, whereas activation of a second population marked by the vesicular GABA transporter VGAT drastically suppressed drinking even in water-craving thirsty animals (Oka et al., 2015).

The subfornical organ also looks ahead. Thirst-promoting neurons there respond to inputs from the oral cavity during eating and drinking and integrate these inputs with information about the composition of the blood, which allows them to predict how ongoing food and water consumption will alter fluid balance and to adjust behavior pre-emptively (Zimmerman et al., 2016). In dehydration-activated neurons of the median preoptic nucleus, free water consumption rapidly decreased activity within about a minute, similar to subfornical organ neurons, whereas during lever-pressing for water activity decreased slowly with progressive water intake, and appears to integrate water intake over time (Allen et al., 2017). Excitatory populations in the lamina terminalis form a hierarchical circuit: nitric oxide synthase-expressing neurons in the median preoptic nucleus integrate signals from the thirst-driving neurons of the subfornical organ, while inhibitory neurons in the median preoptic nucleus that express the glucagon-like peptide 1 receptor are activated immediately upon drinking and monosynaptically inhibit subfornical organ thirst neurons (Augustine et al., 2018). The perception of thirst is thought to occur in higher-order centres such as the anterior cingulate cortex and insular cortex, which receive information from midline thalamic relay nuclei (Gizowski & Bourque, 2018).

Figure 1

Thirst-Driving and Thirst-Stopping Signals in the Lamina Terminalis

Figure 1 summarises the populations. Small populations of hypothalamic neurons sufficient to evoke a complex motivated behavior such as drinking may be used as entry points to identify circuits that energize and direct behavior to specific goals (Sternson, 2013). In a primer on the field, the problem is framed as how the brain monitors the water content of the body and then transforms that physical information into the motivation to drink (Leib et al., 2016).

Water Restriction in Animal Research

Water deprivation is also a tool. Toth and Gardiner reviewed the physiological and behavioral considerations raised by food and water restriction protocols (Toth & Gardiner, 2000), and a later review addressed how to balance welfare considerations with scientific inquiry when food or fluid is restricted in common laboratory animals (Rowland, 2007).

In mice trained on a head-fixed two-choice visual discrimination task, behavioral activity was reduced during food and water restriction, but average discomfort scores remained in the sub-threshold and mild categories, and the type of restriction influenced the number of sessions to the learning criterion and the number of trials performed per session but did not affect maximum learning curve performance (Goltstein et al., 2018).

Water restriction requires rigorous monitoring of animals' health and hydration status, and an alternative is free home-cage access to water made slightly sour with citric acid: across multiple laboratories, mice reduced their intake of citric acid water while maintaining healthy weights, continuous access only subtly impacted their willingness to perform a decision-making task, and using citric acid water instead of restriction only on weekends left learning and decision-making behavior unaffected (Urai et al., 2021).

Dehydration and Human Cognition

Age changes the response. After 24 hours of water deprivation, healthy elderly men had greater increases in plasma osmolality, sodium concentration and vasopressin than young men, yet they were less thirsty and drank less, so that their plasma and urine were not diluted to predeprivation levels (Phillips et al., 1984).

The cognitive evidence is mixed. In 16 young volunteers, 24 hours of water deprivation produced a 2.6% decrease in body weight but did not significantly change cognitive-motor function on a paced auditory serial addition task, a choice reaction time test, a manual tracking test or a Stroop test, although several reaction time-based responses showed significant interactions between gender and dehydration, while subjective ratings shifted toward increased tiredness, reduced alertness and greater perceived effort (Szinnai et al., 2005). A review of behavioral and neuroimaging studies identified methodological weaknesses, including inconsistent measurements in cognitive assessment and the lack of objective hydration state measurements (Masento et al., 2014).

A meta-analysis of 33 studies, providing 280 effect size estimates from 413 subjects, found that impairment of cognitive performance with dehydration was small but significant (ES = −0.21), with tasks of executive function, attention and motor coordination significantly impaired and greater impairment in studies reporting more than 2% body mass loss (Table 1; Wittbrodt & Millard-Stafford, 2018).

Worked Example

The 24-hour deprivation study can be set against the meta-analytic threshold. Szinnai and colleagues' volunteers lost 2.6% of body weight after 24 hours without water (Szinnai et al., 2005); for a 70 kg adult that is 70 × 0.026 = 1.82 kg. The meta-analysis put the threshold at 2% body mass loss, 1.4 kg for the same adult, and found an effect size of −0.28 above it against −0.14 at or below it, an estimate whose 95% confidence interval reached 0.00 (Wittbrodt & Millard-Stafford, 2018), so the point estimate doubled (0.28 / 0.14 = 2.0). On those numbers the deprivation in the 2005 study crossed the threshold, yet the objective tests in that study showed no significant overall change, although several reaction time-based responses showed significant interactions between gender and dehydration (Szinnai et al., 2005).

Discussion

Three conclusions follow. First, the deprivation procedure produces a motivational state with a measurable neural basis: activity of dehydration-activated median preoptic neurons established an aversive internal state that controlled thirst-motivated behavior (Allen et al., 2017), and distinct subfornical populations turn drinking on and off (Oka et al., 2015).

Second, thirst is not only corrective. Subfornical organ thirst neurons anticipate how eating and drinking will alter fluid balance (Zimmerman et al., 2016), and dehydration-activated median preoptic neurons appear to integrate water intake over time (Allen et al., 2017), and the marked decrease in drinking rate, and the alleviation of thirst, occur before plasma dilution becomes significant (Rolls et al., 1980).

Third, the human effects of a day without water are modest. Objective cognitive-motor function was preserved at a 2.6% loss of body weight (Szinnai et al., 2005), while pooled studies found a small overall impairment that was larger beyond 2% body mass loss (Wittbrodt & Millard-Stafford, 2018).

Current Directions

Three lines of work are active. Anticipation: emerging evidence suggests that thirst is a feedforward response promoted by rises in core body temperature, food intake and signals from the circadian clock (Gizowski & Bourque, 2018). In mice, the sleep period is preceded by an increase in water intake promoted entirely by the central clock and not motivated by physiological need, an effect that relies on suprachiasmatic vasopressin neurons projecting to thirst neurons in the organum vasculosum lamina terminalis (Gizowski et al., 2016). Satiety: inhibitory median preoptic neurons monitor real-time fluid ingestion, and loss-of-function manipulations of them lead to overdrinking (Augustine et al., 2018). Refinement: citric acid water is being tested as an alternative to water restriction for training mice (Urai et al., 2021).

Common Misconceptions

“People stop drinking when their blood is back to normal.”

After 24 hours of deprivation, the decrease in drinking rate and the alleviation of thirst occurred before plasma dilution had become significant (Rolls et al., 1980).

“Older people feel thirst as strongly as younger people.”

After 24 hours of water deprivation, healthy elderly men were less thirsty and drank less than young men (Phillips et al., 1984).

“Water restriction is the only way to motivate trained mice.”

Mice given free access to citric acid water instead of restriction on weekends showed unaffected learning and decision-making behavior (Urai et al., 2021).

Glossary

Anticipatory thirst.

Thirst regulated ahead of changes in blood composition, as when thirst neurons respond to eating and drinking before fluid balance has changed.

Body mass loss.

The water deficit expressed as a percentage of body mass, used to grade dehydration; impairment of cognition was greater above 2%.

Citric acid water.

Water made slightly sour with a small amount of citric acid, offered freely as an alternative to water restriction.

Dehydration.

A water deficit of the body, graded in studies by the percentage of body mass lost.

Drive reduction.

The hypothesis that animals learn goal-directed behaviors that reduce the level of an aversive drive state.

Effect size.

A standardized mean difference between conditions; negative values in the dehydration meta-analysis mean worse performance when dehydrated.

Lamina terminalis.

The forebrain region whose osmotic and hormonal inputs may be integrated to generate thirst, containing the subfornical organ and the median preoptic nucleus.

Median preoptic nucleus.

A lamina terminalis region whose neurons integrate thirst-driving signals and, in an inhibitory population, are activated by drinking.

Optogenetic activation.

Driving a genetically defined population of neurons with light, as used to switch subfornical organ thirst neurons on.

Plasma osmolality.

The concentration of solutes in blood plasma, which rose more in elderly than in young men after 24 hours without water.

Presystemic satiety.

The termination of drinking before plasma dilution becomes significant; in humans the attenuation of drinking was subjectively attributed to stomach fullness.

Subfornical organ.

A circumventricular organ containing distinct neural populations that trigger or suppress thirst.

Vasopressin.

A hormone that rose more in elderly than in young men after 24 hours of water deprivation.

Water restriction.

Limiting an animal's water intake and rewarding it with water during a behavioral task.

Key Researchers

Charles W. Bourque

Neuroscientist at McGill University and the Research Institute of the McGill University Health Centre; showed with colleagues that clock-driven vasopressin neurotransmission mediates anticipatory thirst prior to sleep. Faculty Page - Lab

Clark L. Hull

(1884-1952). Behaviorist at Yale University whose drive-reduction theory treated deprivation as the source of drive. Wikipedia - Wikidata

Zachary A. Knight

Professor of Physiology at the University of California, San Francisco, and Howard Hughes Medical Institute Investigator; showed that thirst neurons anticipate the homeostatic consequences of eating and drinking. ORCID - Faculty Page - Lab

Michael J. McKinley

Physiologist at the Florey Institute of Neuroscience and Mental Health and the University of Melbourne; studies the lamina terminalis and the regulation of thirst and fluid intake. Faculty Page

Yuki Oka

Professor of Biology at the California Institute of Technology; identified subfornical organ populations that drive and suppress thirst. Faculty Page - Lab

Neil E. Rowland

(1947-2024). Professor of Psychology at the University of Florida; reviewed the welfare of food and fluid restriction in laboratory animals. Faculty Page - Memorial

Frequently Asked Questions

What is water deprivation in psychology?

MeSH defines it as the withholding of water in a structured experimental situation; deprivation or restricted access to food or fluids is a common research procedure in laboratory animals (Rowland, 2007).

What causes thirst?

Thirst may arise from deficits in either intracellular or extracellular fluid volume, with osmotic and hormonal influences on the lamina terminalis integrated within the brain (McKinley & Johnson, 2004).

Which brain cells control thirst?

Distinct subfornical organ populations trigger or suppress thirst (Oka et al., 2015), and inhibitory median preoptic neurons activated by drinking inhibit subfornical thirst neurons (Augustine et al., 2018).

Why does drinking feel satisfying before the body is rehydrated?

Thirst-promoting subfornical neurons respond to inputs from the oral cavity during drinking and adjust behavior pre-emptively (Zimmerman et al., 2016), and in humans the alleviation of thirst occurred before plasma dilution had become significant (Rolls et al., 1980).

Does dehydration affect thinking?

A meta-analysis found a small but significant impairment overall, larger for attention, executive function and motor coordination and beyond 2% body mass loss (Wittbrodt & Millard-Stafford, 2018), while a 24-hour deprivation study found objective cognitive-motor function preserved at a 2.6% loss of body weight (Szinnai et al., 2005).

What is drive-reduction theory?

The classical drive reduction hypothesis posits that animals learn particular goal-directed behaviors to reduce the level of an aversive drive state, and thirst-associated preoptic neurons have been shown to encode such an aversive drive, although the extent to which thirst-motivated behaviors arise from drive reduction working with separate positive-valence incentive mechanisms remains to be explored (Allen et al., 2017).

Why are older adults at risk of dehydration?

After 24 hours of water deprivation, healthy elderly men were less thirsty and drank less than young men, so that their plasma and urine were not diluted to predeprivation levels (Phillips et al., 1984).

Is water restriction harmful to laboratory mice?

In one training study average discomfort scores remained in the sub-threshold and mild categories under food or water restriction (Goltstein et al., 2018), and water restriction requires rigorous monitoring of animals' health and hydration status (Urai et al., 2021).

References