You can try to deny it, but we all have stereotypes. Some of them turn out to be correct and others false. Nevertheless, we use these stereotypes as shortcuts to interpret the information around us, and in particular, people. Interestingly enough, we all think we are right about our stereotypes. This happens with all thoughts; we think there is truth in our opinions, politics, values, etc. If you think all black people are criminals, you are obviously wrong, but in your own mind you are not. In fact, it is contradictory to believe in something that we do not believe to be correct. If you think a man deserves a CEO position more than a woman, are you right? Where does the truth lie, if not in your own thoughts?


Nowadays, women are equally as qualified as men for jobs that were previously held by mostly men. For instance, Sexton and Bowman-Upton (1990) have shown that female and male entrepreneurs possess nearly equal characteristics such as conformity, interpersonal affect, social adroitness, harm avoidance, and succorance. In short, they found no evidence in gender-related intellectual traits that would give grounds for sexual stereotyping.

And yet, we stereotype.

The study conducted by Uhlmann and Cohen (2007) enlightens us on what processes our minds follow in the making of our thoughts, and in particular gender stereotypes.

Sixty-five participants took part in the first experiment. Participants were primed with a sense of personal objectivity that allowed them to believe that their judgments and decisions were objective. To reach this goal, the researchers asked them to fill out a questionnaire about how rational, impartial, and objective they are when making a decision. Participants were told that this was a preliminary task, having nothing to do with the next part of the experiment.

Then the researchers told the participants that it was time to start the real experiment (in fact, it already started a while ago). They asked them to imagine being a company executive. The company they work for asked them to hire a new factory manager. This is a very important position and the future of the company depends on the person they are going to hire. For the sake of their own career, it would be best to make a good (and objective) decision.

Half the participants were presented with the description of a male applicant, and the other half with a female applicant and were asked to evaluate three things. First of all, they had to rate the applicant on a list of traits concerning their technical, organizational and interpersonal skills. In other words, after reading the description, they had to evaluate how good they thought this person was with paperwork and people. Second, they had to say how important they thought each of these traits were for the position of factory manager. And third, they had to rate their overall impression of the applicant for the job at hand: would he or she be a successful factory manager or not?


Interestingly, the descriptions of the male and the female applicants were exactly the same. Moreover, they were written in a way to convey as unambiguously as possible an image of the applicant as having strong technical and organizational skills, but lacking interpersonal skills. What the researchers wanted to know was if the participants would give different evaluations depending on the applicant’s gender. More precisely, if they would discriminate against the female applicants.

Nonetheless, there could be different ways to discriminate. Maybe they could just give lower scores to women. But they could also use a subtler form of discrimination by constructing hiring criteria in a biased way. For example, they could give a female applicant a fair score on ability to build good work relationships (an interpersonal trait), but at the same time judge this trait as less important for the position of factory manager.

Take for instance, Monin and Miller’s (2001) study where they found that participants who had the opportunity to assert their non-sexist beliefs only facilitated their discrimination against women more. This “credentials effect” is another example of a subtle form of discrimination, where one establishes their non- sexist identity before behaving in a sexist way.

So would the participants prefer the male over the female? And if so, which of the two forms of discrimination would they use?

The participants primed with a sense of personal objectivity rated the male applicants better than the female applicants with a statistically significant difference (on average 5,06 vs 3,75), meaning a difference not due to chance. Furthermore, using the first form of discrimination described above, they discriminated against female applicants by giving them lower scores.

On the other hand, a control group that was primed at the end of the experiment (after applicant evaluations) rated the male and female applicants in a non-statistically different way.

What’s interesting about this experiment is that all the participants were men. In fact, most people in charge of hiring applicants for high-status jobs in the real world, are men. Out of curiosity, the researchers conducted another identical experiment, but this time with only female participants. In this experiment, not surprisingly, no discrimination occurred.

A shortcoming of this experiment is that the researchers didn’t check prior stereotypical beliefs of the participants. They assumed in fact, that everyone from the Occident would hold at least some gender stereotypes. But even if this was the case, there are surely differences among people in how strongly they endorse gender stereotypes. Uhlmann and Cohen then conducted a second experiment in which they took this aspect into account.

In experiment 2, they demonstrated that the effect seen in experiment 1 takes place only when people already hold gender stereotypes. The researchers used the same methodology seen in experiment 1, but this time they made the participants fill out a questionnaire that checked their prior gender stereotypes. Participants were told they had just been elected as the mayor of the town and during the election process they had promised that the new police chief would reduce crime and misconduct. If the promise was not kept, it would seriously affect their re-election.

The job applicant was either Brian (male, for half of the participants) or Karen (female, for the other half) and they were described as well-educated and skilled in administration, although had little experience making arrests and chasing criminals.

As in experiment 1, participants were asked to rate the applicants. The participants had to give a numerical value on how important they considered each of these traits were in becoming a police chief (1 = makes success as a police chief much less likely, 11 = essential to success as a police chief). They later responded to three questions which valued if they considered the applicant good enough.

The results were intriguing. The participants who had been primed with a sense of personal objectivity (giving a numerical value on how important they considered each of the initial traits) valued the man better than the woman but only if they had stereotypic beliefs, even though ‘Karen’ and ‘Brian’ were identical.

What could the explanation for this be?

The most probable explanation is that we are rationalizing our evaluation. For example, if a manager evaluates a male applicant better than a female one, the manager structured his thoughts in “there are some jobs which are better suited for men”. They therefore see themselves as objective even though they clearly are not. A final experiment was conducted to further explore this notion.

In the third experiment, Uhlmann and Cohen tested the power of people’s thoughts on their behaviour and decision making by inducing an “I think it, therefore it’s true” mindset in participants. They hypothesized that subjects would act on thoughts that are available to them, or cognitively accessible. 21 male and 17 female undergraduate participants completed a sentence unscrambling task in which they had to unscramble 5 words, then drop 1 word, to make a meaningful sentence with the 4 remaining words. To make gender stereotypes cognitively accessible, half of the participants unscrambled nonsense sentences that hid gender stereotypic words such as pink, nurse, and Barbie, while others in a control condition only saw neutral words such as curtain, building, and train.

Next, half the participants were primed with a sense of personal objectivity by answering a questionnaire about self-perceived objectivity (as in experiment 1 and 2), while others were assigned to the control objectivity prime condition. Results showed that those primed with gender stereotypes but not primed with their personal objectivity had no effect on their applicant evaluations.

The same participants were then presented with the hiring scenario ‘Hiring a Company Representative’ (like in experiment 1) in which the subject was responsible for hiring one of two applicants to be the company representative of a manufacturing company whose success or failure would greatly impact the company. The applicants were Lisa and Gary whose descriptions made them appear as unambiguous and identical in merit as possible. Participants selected which applicant they would hire by choosing a number on a scale where 1 = Lisa is better suited, 6 = Lisa and Gary are equally suited, and 11 = Gary is better suited.

Participants then had to list all their thoughts on why Lisa and Gary should be hired. The researchers assessed participants’ depth of processing by measuring the number of positive thoughts participants had toward each applicant while making their decision. Uhlmann and Cohen found that participants listed more positive thoughts toward the male applicant than the female when primed with the gender stereotype regardless of the objectivity prime. What is most interesting is the effect produced by the gender stereotype prime when coupled with the personal objectivity prime. Its result strongly suggests that participants simply holding positive thoughts were extremely likely to hire the male rather than female applicant. No other pairing of conditions were as significant.


In addition, there seemed to be a gender difference between female and male participants. Tentatively speaking, male participants seemed more inclined to uphold gender stereotypes compared to female participants when primed with both personal objectivity and gender stereotypes.

Contrastingly, a study by Rashotte and Webster Jr. (2005) showed that both genders can stereotype against females. Their collegiate participants evaluated other female and male college students in terms of desirability and skill in certain activities (changing a tire, changing a diaper, and growing vegetables). In all participants, this produced status beliefs, which are beliefs in one gender’s superiority and competence over the other. As expected, the status beliefs of both female and male participants favoured males over females, thus displaying stereotypical behaviour. However, Rashotte and Webster Jr. caution that the continued production of these status beliefs can lead to status generalization which would provide more advantageous results for men in many ways.

From Uhlmann and Cohen’s experiment, we learned that the human brain is far capable of absorbing and being influenced by environmental stimuli than we consciously believe it to. Even when we are made consciously aware of our thoughts (for instance, believing that one is objective) as shown by this study, they have the power to alter how we make important decisions. Recall that when participants both endorsed gender stereotypes and felt objective, they were most likely to discriminate. This effect appeared only in men, and in fact, women did not discriminate. Thus, it is important to recognize these consequences in real life, as hiring decisions for high-status jobs are often ambiguous and usually made by men.

Like Pronin and Kugler (2007) demonstrated in their study on the introspection illusion, people tend to overvalue their inner thoughts and feelings, while undermining those of others. In addition, we display a “bias blind spot” where we believe we are immune to biases, but think others are not. Interestingly, unlike the “I think it, therefore it’s true” effect, the bias blind spot can be eliminated by educating participants on the significance of non-conscious processes.

Uhlmann and Cohen also established that the belief in one’s own personal objectivity allowed participants to use their thoughts to make their hiring decision – however discriminatory, subjective, or brief they had been. Practically speaking, their experiment sheds some light onto how profoundly we may be influenced by particular environmental factors when in conjunction with certain thoughts, whether temporary and insubstantial or deeply rooted in our beliefs. This is not to say that just one condition such as the environment or our cognitive content drives us to behave unpredictably. Rather, it is their coupling that pushes us to act on sometimes illogical and momentary thoughts which then, may influence how we make important decisions (such as hiring an applicant for a substantial company based solely on prejudicial thoughts toward their gender). These experiments show how vulnerable humans can be to subtle manipulations and stimuli which impact not only ourselves but those whom we interact with. More shockingly, we realize from these results that perceiving oneself to be objective actually makes you more subjective. Corporations today encourage employees in hiring positions to view themselves as objective and rational. Fortunately, we can now see how dangerous this can be since it creates environments that prime a sense of personal objectivity. This begs the question, if we aren’t aware of our subjectivity, when are we truly objective?


  • Monin, B. & Miller D. T., (2001). Moral Credentials and the Expression of Prejudice. Journal of Personality and Social Psychology, 81, 33-43.
  • Pronin, E. & Kugler, M. B., (2007). Valuing thoughts, ignoring behavior: The introspection illusion as a source of the bias blind spot. Journal of Experimental Social Psychology, 43, 565-578.
  • Rashotte, L. S. & Webster Jr., M., (2005). Gender status beliefs. Social Science Research, 34, 618-633.
  • Sexton, D. L., & Bowman-Upton, N., (1990). Female and male entrepreneurs: Psychological characteristics and their role in gender-related discrimination. Journal of Business Venturing, 5, 29-36.
  • Uhlmann, E. L., & Cohen, G. L. (2007). “I think it, therefore it’s true”: Effects of self-perceived objectivity on hiring discrimination. Organizational Behaviour and Human Decision Processes, 104, 207-223.