Combining small psychological differences predicts a person’s sex with 80 percent accuracy

A recent study published in Scientific Reports provides evidence that combining multiple small psychological differences can accurately predict a person’s sex in 80 percent of cases. The findings suggest that these combined differences in cognition, personality, and interests also help explain why certain occupations tend to be dominated by men or women. This research offers new insights into how subtle psychological variations might shape real-life career trajectories.

In the study of psychology, differences between men and women generally tend to be small. When looking at individual traits, the range of scores for males and females overlaps extensively. For example, while females might score slightly higher on average in tests of verbal memory, many men still outperform many women on these specific tasks. This overlap has led some scholars to propose that psychological sex differences have minimal impact on people’s lives or society as a whole.

At the same time, certain real-world domains show large disparities between the sexes. One prominent example is occupational choice, where specific fields remain highly segregated by gender. The authors wanted to understand how small individual differences might accumulate to produce these larger societal patterns.

“We have long conducted research on sex differences in cognition, aiming not only to identify where such differences exist, but also to understand why they arise,” said Agneta Herlitz, a professor of psychology in the Department of Clinical Neuroscience at Karolinska Institutet. “Our findings typically indicate that most cognitive abilities show small sex differences, so small that some argue that they carry no implications.”

“This observation led us to consider a broader question: if many small sex differences are examined collectively, might their combined effect be more substantial and meaningful?” Herlitz explained. “Hence the research questions: Can we predict individuals’ sex and sex ratio in their occupations, based on their cognition, personality, and interest?”

To explore this, the scientists selected psychological measurements that consistently show minor sex differences across different cultures and age groups. These included cognitive skills like episodic memory, which is the ability to recall specific past events, and spatial awareness, which involves visualizing and manipulating objects in the mind. They also tested verbal fluency, which is the ability to quickly generate words, and emotion recognition.

They also included the five major dimensions of personality. These consist of extraversion, which measures sociability; agreeableness, which reflects warmth and cooperation; conscientiousness, which indicates organization and discipline; neuroticism, which captures tendencies toward negative emotions; and openness to experience, which reflects curiosity and creativity.

Finally, they measured whether participants had a stronger general interest in interacting with people or working with physical things. These basic variables were chosen because they represent well-studied psychological phenomena that are not trivially linked to biological sex. The researchers suspected that evaluating these traits together would reveal a much broader psychological divide than looking at any single trait alone.

The researchers recruited 2,767 participants from Sweden, including 1,465 females and 1,302 males. The participants were between the ages of 35 and 45, a narrow age range chosen to focus on individuals with established careers while reducing age-related variations in cognitive performance. People completed a comprehensive online assessment using their personal computers in a quiet environment.

During the assessment, participants completed 13 specific tasks and questionnaires. To measure verbal episodic memory, they were asked to remember and later recall a list of words. Another memory task tested their ability to recognize photographs of faces they had seen earlier. For spatial abilities, the participants matched the angles of different lines and mentally rotated three-dimensional objects to find matching shapes.

Verbal production was assessed by having participants write down as many words as possible starting with specific letters within a short time limit. The scientists also tested emotion recognition by asking participants to identify people’s feelings based solely on photographs of their eyes. To gauge personality, the participants rated themselves on a standard 60-item questionnaire.

For interests, the participants indicated how much they would enjoy certain everyday situations. These included people-oriented activities, like listening to a conversation in a crowd, and thing-oriented activities, like fixing a toaster. Finally, the participants reported their current or most recent occupation.

The researchers then matched these reported jobs with national registry data in Sweden. This allowed them to determine the typical proportion of men and women working in each specific role. For instance, a preschool teacher would receive a score reflecting the high percentage of women in that profession nationwide.

The authors found expected differences between males and females across the individual tasks. Females performed better on average at verbal fluency, verbal episodic memory, face recognition, and emotion recognition. They also scored higher on all five personality dimensions and showed a stronger interest in people.

Males tended to perform better on the two spatial tasks, which involved judging line angles and mentally rotating objects. In addition, the men showed a much stronger interest in interacting with physical things. Most of these individual differences were relatively small when viewed in isolation.

However, when the researchers combined the scores from all 13 tasks and questionnaires into a single predictive model, a distinct pattern emerged. By evaluating an individual’s overall psychological profile, the model correctly predicted whether the participant was male or female in 80 percent of the cases. The model showed no bias, predicting male and female identities with similar accuracy.

“Although the pattern was expected, we were somewhat surprised by the rather high accuracy,” Herlitz told PsyPost. “We could, however, have increased the prediction accuracy further by including predictors that are strongly and directly tied to sex, such as sexual attraction and body morphology, or by incorporating many other psychological variables with small sex differences. Or by using more fine-grained predictors, for instance, individual test items rather than domain-level composites.”

The most accurate individual predictors of a person’s sex were their performance on the mental rotation task, their verbal episodic memory, their levels of neuroticism, their performance on the line angle task, and their interest in things. For instance, higher scores on the interest in things questionnaire were associated with a much lower likelihood of the participant being female. Even when the scientists removed the three strongest predictors from their statistical model, they could still correctly predict the participant’s sex with 70 percent accuracy.

The combined psychological differences also provided evidence regarding real-world career choices. The researchers found that a person’s cognitive, personality, and interest profile accounted for about 22 percent of the variance in the gender segregation of their occupation. For example, an individual with a strong interest in physical objects was highly predicted to work in a male-dominated field, regardless of their actual sex.

To understand this impact, the authors looked at hypothetically typical male and female psychological profiles. They calculated that a man with a typical female psychological profile would work in a job with about ten percent more women than a man with a typical male profile. This suggests that combined psychological differences play a notable role in guiding people toward certain types of work.

“We think that the most important point is that it gives us a better understanding of why women and men seem to be more different in everyday life than the rather small sex differences we typically find in basic psychological attributes would indicate,” Herlitz said. “The association between sex differences in cognition, personality, and interest and sex ratio in occupational choices is one way of anchoring the abstract question about predicting sex into something more consequential, making it more than a statistical exercise.”

While the study provides robust data, the authors noted several limitations to their approach. The cross-sectional design means that the study only looks at a single point in time. Because of this, it is not possible to prove that these psychological traits directly cause people to choose specific jobs.

“It is good to remember that this is an observational study, thus, we cannot claim that the sex differences in psychological attributes cause men and women to select male- or female-dominated occupations,” Herlitz explained. This highlights a common limitation in survey-based psychological research.

While it is unlikely that a person’s job completely changes their basic cognitive abilities or personality, the relationship between interests and career paths is likely complex and interactive. Another limitation involves the online testing environment. Because participants took the tests at home, the researchers could not control for distractions or varying levels of effort.

However, this lack of control would affect both male and female participants equally. Because of this, it likely did not alter the overall pattern of the findings. The authors also highlighted that their study relied on a binary view of sex, matching legal sex at birth with current legal sex.

They excluded a small number of participants who had transitioned or reported a different gender identity. This decision was made primarily because hormone treatments could independently influence cognitive and behavioral results. Future research using much larger samples could include non-binary individuals and those with different gender identities to provide a more comprehensive view of human psychology.

It is also easy to misinterpret these findings as evidence for a strictly male brain or female brain. The researchers explicitly caution against this view. Because individual traits still show massive overlap between the sexes, no single characteristic can definitively categorize a person as male or female.

Instead, the study suggests that human behavior is shaped by complex constellations of many small characteristics working together. By looking at these traits in combination, we can better understand how subtle differences influence life choices. The findings provide a framework for exploring how basic psychological variations contribute to larger societal patterns.

Moving forward, the scientists plan to build on these findings to better understand workplace disparities. Herlitz intends “to continue to explore the explanation for psychological sex differences, but also in more detail try to address the question of why men and women tend to work in different occupations.”

The study, “The power of many small sex differences in cognition, personality, and interests,” was authored by Agneta Herlitz, Joakim K. E. Frostegård, Martin Asperholm, Richard Bränström, Elizabeth Guest, Joakim Martinsen, Hedda Sonnegård, Kimmo Sorjonen, Lisa B. Thorell, and Björn N. Persson.