Image Map

guy curlingby: Andrew Novick
Evidence in animals (1) and Alzheimer’s patients (2) suggests that androgens possess cognitive enhancing properties. However, very few studies have investigated the role of testosterone on cognition in healthy individuals. The difficulty in assessing these effects is probably rooted in the idea that testosterone is a “social hormone,” influencing behavior only when status is challenged or threatened (3). Phrased another way by Wood, “testosterone amplifies the reinforcing or aversive value of social behaviors, thereby serving as an aid to discriminating successful and unsuccessful social interactions.” (4) Taken together, testosterone should only affect cognitive performance in healthy individuals when social status is at stake or when the task is competitive in nature. Recent studies have investigated this very concept.

When the Alpha Male Loses: Studies in Testosterone and Cognitive Performance

“Where is the poet who has sung of that most lacerating of all human emotions, the cut that never heals-male humiliation?”

-Tom Wolfe, I am Charlotte Simmons
A group at the University of Texas (3) published a 2005 study investigating the relationship between testosterone levels, cognitive performance, and social status. 88 college students (52 females, 36 males) were given salivary testosterone assays and told that they would complete a task demonstrating “the effects of individual variables on group functioning.” Before completing verbal fluency and spatial ability tests, the students were primed for status. One-third were told that they were picked to be the “leader” of their group (high status), one-third the “follower” (low status) and the remainder were given no information on status to act as a control. To analyze the results, the researchers divided males and females into “High T” and “low T” individuals.

The results showed no difference in performance between high T and low T individuals when controlled for status. However, high testosterone and status had significant consequences on cognitive performance. Among high T individuals, leaders performed the best, followed by controls and then followers. High T followers also performed significantly worse compared to low T followers. These results suggest that high T individuals have an increased sensitivity to status. High status increases cognitive performance in high T individuals, while low status impairs it. There was no relationship between status and performance in the low T group, suggesting that individuals with low testosterone levels do not share such status sensitivity.

The authors also observed that high T individuals are impaired more by low status than they are enhanced by high status. Couple this with the earlier observation that low T/low status performed significantly better than high T/low status, the question is raised, “What is it about high testosterone and low status that’s such a big deal?” It appears that high testosterone individuals are threatened when placed in low status situations and have a desire to re-gain high status. Their attention to status ends up distracting them and wasting cognitive resources. The authors conclude, “Ironically, then, the desire to regain high status may be the biggest obstacle to regaining it – at least on a cognitive task.”

The same research group (5) also investigated the interaction of “stereotype threat” and “stereotype challenge” with testosterone levels and intellectual performance. Stereotypes are often statements related to status in different areas (white men can’t dance, black men are good at sports, etc). In the first study, a group of male and female college students who scored high in “Math Identification” were divided up into two groups. The first group was asked to answer a questionnaire designed to prime the stereotype that females possess weak math abilities. The second group received a control questionnaire. As expected, only high testosterone females exposed to the stereotype performed significantly worse than the control group. The stereotype conferred a threat to status on the high testosterone females, similar to being placed in the “follower” group in the previous study. The decrease in performance in high testosterone females also corresponded with an increase in anxiety levels.

The second study gave male participants the opportunity to confirm a positive stereotype and improve their status (rather than having their current status threatened by refuting a negative stereotype). Half of the participants were told that they were taking a test that would only identify people of exceptional math ability. The second group was told that it was taking a test that would only identify people of weak math ability. High testosterone males performed significantly better on the “exceptional ability” test than on the “weak ability test.” Here, high testosterone coupled with an opportunity to enhance status (exceptional ability test) enabled individuals to “rise to the challenge.” In the absence of a status enhancing opportunity (weak ability test), high testosterone individuals seemed unconcerned with their performance.

Wolf et al (6) treated elderly men with 250mg of testosterone enanthate and gave them both verbal and spatial cognitive tests both before and 5 days after the injection. The testosterone treatment blocked the practice effect on verbal fluency without affecting other measures. In other words, the testosterone treated males didn’t improve as much as expected from taking a test twice.

The relationship between testosterone and learning from a repeated task was also investigated by Schultheiss, et al, (7) who utilized another variable called “implicit power motive.” Implicit power motive is the “nonconscious disposition to experience having impact on others as rewarding.” Basically, there are two types of power motives: personalized vs. socialized, with no difference in baseline testosterone levels between the two. Individuals with highly personalized implicit power motive receive reward from asserting direct dominance over others (e.g. a boxer). Those with highly socialized implicit power motive receive reward from having impact through pro-social means (e.g. an artist).

Schultheiss found that when individuals with highly personalized implicit power motive won a cognitive task against an opponent, not only did their testosterone levels increase, but so did their implicit learning: they unconsciously “learned” how to do the test faster for the next time as if to assure future dominance. Those with highly personalized implicit power motives who lost didn’t experience any change in testosterone but their implicit learning dropped significantly. It was as if defeat caused them to give-up, or obsession with re-gaining status was too distracting. On the other hand for those with highly socialized implicit power motive, winning had little effect on testosterone or learning. But losing did induce a significant drop in testosterone which correlated with an increase in learning. Here, high testosterone appears facilitative to learning when one is in high status (“winner”), while low testosterone is facilitative to learning when one is in low status (“loser”). However, both of these observations correspond to different personality traits.

This research can be summarized by the following points:

1. Individuals with high testosterone demonstrate sensitivity to status.

2. High testosterone facilitates cognitive performance when the individual is in a high status position.

3. High testosterone facilitates cognitive performance when the individual is given the opportunity to enhance status significantly (in a “challenging” rather than “threatening” situation)

4. High testosterone impairs cognitive performance when the individual is in a low status position.

5. Personality traits, such as amount and type of “implicit power motive” affect both testosterone and learning in response to cognitive competition.

#4 is perhaps the most interesting in that it makes a case for the importance of testosterone plasticity instead of chronically elevated levels. Most of us must endure various periods of subordination and possibly some stereotype threat in order to achieve our goals. Academia is one situation where the student is often placed in a position of low status and expected to perform cognitively. Paulo Freire, in his landmark work, Pedagogy of the Oppressed , details how modern educational institutions set up a dominant-subordinate relationship between teacher and student. And while most college students probably wouldn’t complain of overwhelming feelings of low status, being a successful student often implies being an obedient student: following rules, meeting deadlines, and giving a professor what he or she wants. Thus, it’s possible that chronically high testosterone levels might lead to frustration with low status rather than academic performance. Instead, moderate testosterone levels that increase slightly with academic mastery and decrease slightly with academic failure would be ideal.

This puts anabolic steroid use in a new light. Supraphysiological levels of androgens could impair cognitive performance in certain scenarios and possibly prevent learning from failure, especially in individuals who are already status sensitive and are motivated by personalized implicit power. Similarly, the “nootropic” use of androgens (short acting DHT derivatives and intranasal preparations) may only work when the individual is in a high status position. For example, a high-achieving, prepared student might benefit from intranasal 5-alpha-androstanediol before an exam, but it might actually be detrimental to a student with a history of poor performance who hasn’t studied or been to class and is trying to “wing” the exam.

There is no Spoon: The Relevance of Pre-Competition Appraisal

“Do not try and bend the spoon. That’s impossible. Instead…only try to realize the truth.”

“What truth?”

“There is no spoon.”

“There is no spoon?”

“Then you will see that it is not the spoon that bends, it is only yourself.”

-The Matrix

Astute readers might be thinking that this idea of “status” seems very subjective. Shouldn’t an individual be able to conceive of oneself as having “high status” despite external feedback to the contrary (stereotype threat, being labeled a “follower”)? If the individual had conceptual control over status, the idea of high testosterone impairing cognitive ability would be moot. A more controllable view of competitive appraisal is exactly the direction that Alicia Salvador in her review “Coping with competitive situations in humans” takes us (8).

For the past 15 years, Salvador’s research team has been investigating hormonal and psychological responses to competition and its outcome. There are some fundamental differences between her studies and the cognitive studies above. An advantageous difference is that Salvador’s team uses a reciprocal hormonal model instead of a baseline one. They measure how certain behaviors and outcomes affect testosterone levels and how testosterone levels affect certain behaviors and outcomes. A not so advantageous difference (for relating the findings to the above studies) is that the research is done on physical sports competitions rather than cognitive ones. Nonetheless, Salvador’s conclusions suggest that the cognitive impairing effect of high testosterone might be prevented by certain coping strategies. Her conclusions also suggest that testosterone levels in response to competition can also be manipulated by coping strategy.

Mazur, one of the founders of testosterone biosocial research, presented the original reciprocal relationship involving testosterone and agonistic encounters. In a competition between two individuals, the winner would most likely be the one who had the higher testosterone levels to begin with. In response to winning, his testosterone levels would rise further ensuring future dominance. Similarly, the loser’s testosterone levels would drop, facilitating future submission.

While Mazur’s hypothesis sometimes holds true for humans engaged in sports, it’s difficult to re-create. Salvador argues that the mere outcome (winning or losing) has little relation to testosterone levels. Hormonal responses are instead mediated by complex psychological processes. Testosterone levels before a competition are correlated with rank, previous wins, and desire to win (status, motivation). An increase in testosterone in response to winning is largely dependent on how positively the winner appraised himself. Was the win due to his performance or an external factor? Only winners who had high levels of internal attribution (“blaming” themselves for good performance) experienced significant increases in testosterone.

Salvador concludes that victory and increases in testosterone are most likely if the individual positively appraises the competition and takes an active coping strategy. In the appraisal, the competitor asks, “Is this competition important to me? What is the relevance to my sense of status? How prepared am I? Do the demands of this situation meet my resources?” The answers to these questions will determine whether the competition appears as a “challenge” (important to status, resources are greater than demands) or a “threat” (important to status but demands are greater than resources). Viewing the competition as a challenge will ultimately enable a positive stress response with active coping, increases in testosterone, and a controlled cortisol response. In Salvador’s own words:

Conscious or unconscious thoughts of the individual will ultimately determine the coping behavior. If the individual ‘appraises’ the situation as important for him/her and dependent on him/her, and he/she has resources to control it, the probabilities of employing an active strategy increase. This coping response includes increases in T (especially in an aggressive/competitive situation) and SNS activation.

So the million dollar question is: if the high testosterone “followers” in the original study and the stereotyped high testosterone females in the second study had employed a positive appraisal, could they have avoided impairment? It’s possible, although Salvador does admit that both unconscious and conscious thoughts control appraisal and coping. But if the high testosterone female subjected to stereotype threat were to consciously appraise the situation such that she ended up feeling challenged instead of threatened, it’s possible she might have overcome some cognitive impairment. Reminding herself of superior math preparation, previous success and expertise could secure feelings of high status. To convince herself of the invalidity of the presented stereotype, she might try conjuring up images of successful female mathematicians and inferior males. However, some might argue that the unconscious effects of stereotype threat would trump such efforts.

In conclusion, testosterone has a dual effect on cognitive performance. It increases performance when the individual is in a high status position or given the opportunity to enhance status. Similarly, testosterone can impair cognitive performance when the individual is forced into a low status position or has his/her status threatened. Both of these effects are due to the relationship between testosterone and status sensitivity. High testosterone individuals have a need for high status and experience anxiety when that status is threatened or taken away. Such anxiety, while evolutionarily beneficial for re-gaining status in nature, can be distracting for re-gaining status through cognitive means. Some of the effects of testosterone on status sensitivity and cognitive performance are personality type dependent and might be manipulated by consciously appraising the situation and employing different coping strategies.


Frye CA, Seliga AM. Testosterone increases analgesia, anxiolysis, and cognitive performance of male rats. Cogn Affect Behav Neurosci. 2001 Dec;1(4):371-81.

1. Cherrier MM, Matsumoto AM, Amory JK, Asthana S, Bremner W, Peskind ER, Raskind MA, Craft S. Testosterone improves spatial memory in men with Alzheimer disease and mild cognitive impairment. Neurology. 2005 Jun 28;64(12):2063-8.

2. Newman ML, Sellers JG, Josephs RA. Testosterone, cognition, and social status. Horm Behav. 2005 Feb;47(2):205-11. Epub 2004 Dec 19

3. Wood RI. Reinforcing aspects of androgens. Physiol Behav. 2004 Nov 15;83(2):279-89.

4. Josephs RA, Newman ML, Brown RP, Beer JM. Status, testosterone, and human intellectual performance: stereotype threat as status concern. Psychol Sci. 2003 Mar;14(2):158-63.

5. Wolf OT, Preut R, Hellhammer DH, Kudielka BM, Schurmeyer TH, Kirschbaum C. Testosterone and cognition in elderly men: a single testosterone injection blocks the practice effect in verbal fluency, but has no effect on spatial or verbal memory. Biol Psychiatry. 2000 Apr 1;47(7):650-4.

6. Schultheiss OC, Rohde W. Implicit power motivation predicts men’s testosterone changes and implicit learning in a contest situation. Horm Behav. 2002 Mar;41(2):195-202.

8. Salvador A. Coping with competitive situations in humans. Neurosci Biobehav

Rev. 2005 Feb;29(1):195-205. Epub 2004 Dec 13.


Cognitive and Competitive Aspects of Testosterone