# Forty-two men faced three minutes of ice water or a calm control, and acute stress made them more likely to predict that another person would repeat an earlier action

> Men placed under acute stress became more likely to predict that another person would repeat an earlier action, according to a controlled experiment involving 42 healthy young adults. Their ability to track what the other person believed remained similar to that of...

Canonical URL: https://www.argo.net/forty-two-men-faced-three-minutes-of-ice-water-or-a-calm-control-and-acute-stress-made-them-more-likely-to-predict-that-another-person-would-repeat-an-earlier-action/
Byline: ARGO.net Editorial Team
Published: 2026-08-19T22:20:02+00:00
Categories: Explainer, Humans

![Participant_observed_during_a_psychology_test](https://www.argo.net/wp-content/uploads/2026/08/participant_observed_during_a_psychology_test.jpg)

Men placed under **acute stress** became more likely to predict that another person would repeat an earlier action, according to a controlled experiment involving 42 healthy young adults. Their ability to track what the other person believed remained similar to that of men in a calm control group.

The findings suggest that pressure can influence how people forecast someone else's next move. Stress may draw attention toward visible past behavior, giving it more weight than a person's current goal. Such a shift could affect social judgments during arguments, emergencies, tense meetings and other situations where people must think quickly.

The [study](https://doi.org/10.1371/journal.pone.0283782), published in **PLOS ONE** in April 2023, was led by **Sarah Witt**, Sabine Seehagen and [Norbert Zmyj](https://ifp.ep.tu-dortmund.de/institut/personen/norbert-zmyj/). Witt and Zmyj were affiliated with **TU Dortmund University**, while Seehagen was based at Ruhr University Bochum.

## Acute stress favors predictions based on past behavior

People often predict behavior by combining several clues. They consider what someone knows, what the person wants and what the person has done before. Under pressure, however, the mind may rely more heavily on information that is easy to notice and remember.

The experiment found that stressed participants gave greater weight to **past behavior**. When asked where a character would search for a violin, they were more likely to choose a place connected to where he had gone earlier. Men in the control group placed less weight on that former location.

The shift resembles earlier findings about stress and habitual action. Research has suggested that people under stress sometimes repeat familiar responses instead of carefully adjusting their actions to a new goal. The 2023 study extended that idea into social prediction, where participants were judging another person's likely behavior.

## How researchers raised cortisol levels

Researchers randomly assigned 21 men to a stress condition and 21 to a control condition. All participants were between 19 and 33 years old. Four additional men began the stress procedure but removed their hands early, so their results were excluded from the main analysis.

The stress group completed a **socially evaluated cold pressor test**. Each man kept his right hand in ice water measuring between 0 and 2 degrees Celsius for three minutes. A reserved female experimenter watched him and a video camera recorded the procedure. Physical discomfort combined with social observation created a brief laboratory stress response.

Men in the control group placed their hands in warm water measuring between 35 and 37 degrees Celsius. They were neither watched by an experimenter nor recorded. The same three-minute period therefore lacked the cold and social pressure used in the stress condition.

Participants later rated how painful, difficult, unpleasant and stressful the experience felt. Men exposed to the ice water gave higher ratings in every area. Their responses confirmed that the procedure produced a much stronger subjective feeling of stress.

## The violin test measured social reasoning

Saliva samples allowed the team to measure **cortisol**, a hormone released as part of [the body's stress response](https://www.nimh.nih.gov/health/publications/so-stressed-out-fact-sheet). Samples were collected before the water test, then 20 minutes and 35 minutes after it began. Cortisol levels started at similar points in both groups and later rose more strongly among the stressed men.

Twenty minutes after the water procedure began, each participant completed a modified **false-belief task**. Such tasks test whether a person can follow someone else's view of a situation, even when that view differs from reality.

Participants heard a story about a boy named Niklas, who placed his violin inside a blue container. Niklas then left the room. While he was away, his brother moved the violin into a yellow container. The brother also changed the positions of four colored containers around the room.

By the end of the story, the violin was inside the yellow container. The blue container had moved and a black container now occupied the blue container's original location. Participants divided 100 percentage points among the four containers to show how likely Niklas was to search each one first.

## What changed under stress

The black container gave researchers their main test of prediction based on earlier action. It stood where Niklas had originally gone when he put away the violin. Choosing it meant that a participant expected Niklas to return to a familiar location, even though the container at that spot had changed.

Stressed men assigned a significantly higher probability to the black container than men in the control group. The one-sided statistical test produced a p-value of 0.038, with a moderate difference between the groups. A separate Bayesian calculation gave modest support to the researchers' prediction, finding the result about 2.2 times more likely under their proposed effect than under a no-effect model.

The result suggests that acute stress increased attention to a person's earlier movement through the room. A former action became a stronger clue for forecasting the next action. The experiment therefore measured a change in prediction, rather than observing the character's real behavior.

The task involved a single earlier action. Habits usually develop through repeated behavior, so the finding is best described as increased reliance on former behavior. Future experiments will need repeated actions to test whether stress also changes how people judge another person's established habits.

## False-belief reasoning remained stable

The blue container tested whether participants understood Niklas's false belief. Niklas had placed the violin there and had never seen his brother move it. Someone following Niklas's knowledge could therefore expect him to search the blue container, even though the violin was elsewhere.

The study abstract states, "Stress did not affect participants' attribution of false beliefs." Men in both conditions gave similar probability ratings to the blue container. Their responses suggest that the brief stress procedure preserved their ability to follow what Niklas believed about the violin.

Ratings for the yellow and white containers were also similar between groups. The yellow container held the violin at the end, so choosing it could reflect the participant's own knowledge of the real location. The white container served as another possible search location without the same direct link to the violin's history.

Psychologists often call the ability to track another person's beliefs **Theory of Mind**. An earlier [stress experiment](https://pubmed.ncbi.nlm.nih.gov/19470361/) using different social tests also found a complex picture, with results varying according to sex and the kind of task. Social reasoning includes several mental skills and stress may affect each one differently.

## Why mental shortcuts may grow under pressure

Stress prepares the body to respond quickly. Heart activity changes, stress hormones enter the bloodstream and attention can narrow toward information that appears useful for immediate action. Careful mental planning may become harder when several demands compete for limited attention.

Predicting a return to an earlier location provides a simple rule: people often go where they went before. Following that rule requires less mental updating than comparing the character's knowledge with the new positions of several containers. The shortcut may help explain why stressed participants placed greater weight on Niklas's former movement.

The **prefrontal cortex**, a brain region involved in planning and control, is sensitive to stress hormones. It also supports social reasoning and helps people suppress distracting information. Reduced mental flexibility under pressure could therefore favor familiar clues, although the experiment measured behavior and cortisol rather than brain activity.

**Goal-directed reasoning** uses current aims and expected results to select an action. Habit-like reasoning gives more weight to earlier responses. The balance can be useful during danger, when fast reactions save time, yet rigid predictions may create problems during social conflict. Someone under pressure might assume that another person will repeat an old action even after the person's goals have changed.

## Limits of the 42-man study

The research used a **small experimental sample**, with 21 participants in each main group. A single unusual response can have a larger influence in a sample of this size. The Bayesian analysis also provided modest evidence, which calls for replication with more participants.

All participants were healthy young men. The researchers chose this design because menstrual cycle phases and oral contraceptives can influence cortisol responses, which would have required additional controls. The narrow sample means the result applies most directly to men within the age range studied.

Earlier research has reported sex-related differences in how stress influences social thinking, although findings have varied between experiments. Larger studies that include women and carefully track hormonal factors could reveal whether the same prediction shift appears across sexes.

The researchers also used a one-trial social task. A participant made one set of probability judgments after hearing one violin story. Repeating the test with different characters and situations would reveal whether the effect is stable across many social decisions.

Laboratory stress has another limit. Three minutes of cold water and observation produced a clear short-term response, while chronic stress can last for weeks or years. The study supports a focused conclusion about temporary stress under controlled conditions. Everyday social pressure may involve longer timelines and personal relationships that a laboratory story cannot fully reproduce.

## Questions for future research

Future experiments could test larger groups with a wider range of ages and backgrounds. Researchers could also compare physical stress, social evaluation and time pressure to see whether each one changes social prediction through the same process.

Tasks with repeated actions could examine true habits. A character might follow the same route many times, then receive a new goal that requires a different route. Researchers could measure whether stressed observers continue expecting the familiar response after circumstances change.

Brain imaging and tests of attention could help explain the mechanism. The 2023 experiment did not directly measure inhibitory control or other executive skills. Adding such measures could show whether reduced flexibility, stronger memory for earlier behavior, or faster detection of regular patterns drives the result.

Newer evidence also urges caution around broad claims about stress and habit. A [2025 PLOS ONE study](https://doi.org/10.1371/journal.pone.0327807) found no general increase in habitual action after acute stress in two experiments. Differences in tasks, timing and individual cortisol responses may help explain why studies reach different results.

For now, the violin experiment offers a specific clue about social judgment: acute stress may lead people to forecast another person's next move by leaning more heavily on what that person did before. Further research can test when this shortcut helps people react quickly and when it produces rigid assumptions about changing human behavior.