Are Blonds Really Dumb?

Blonds are dumb, foreigners lazy, women can't do math. We continuously encounter such statements in our every-day lives – even if most people obviously wouldn’t take them seriously. Nevertheless, we often act towards others as if they were members of a particular group and nothing more. The bases for such behaviors are stereotypes and prejudices. Stereotypes are subjective views about the characteristics, attitudes, and behaviors of the members of a particular group. They are overgeneralizations, whereby the members of a group are assigned certain characteristics – merely based on their group membership.

Of course, we don’t like to think of ourselves as prejudiced. We believe in equality between genders and races. We might joke about the dumb blond, but do we really believe it? Do we really think women can’t do math? Do we think Black Americans are more aggressive than White Americans? Most of us would probably answer no. While we hold egalitarian beliefs of equality and equity, and while we think we’re not judgmental and that we act in a fair manner, research has indicated that the “evil” stereotypes lie hidden deep inside of us all. People activate such stereotypes in many situations and based on minimal cues – even the presence of a member of the particular group is enough. In other words, when encountering a member of a particular group, we automatically recall information about this person’s group, such as typical attitudes and behaviors.

Patricia Devine from the University of Wisconsin-Madison (1989) investigated which elements of stereotyping are automatic and which can be consciously controlled. In a series of experiments about the (North American) cultural stereotype of Blacks, she first demonstrated that both people who show high levels of prejudice towards Blacks and those who show comparatively low levels are equally knowledgeable of the stereotype contents. Accordingly, Blacks are, for example, believed to be aggressive and criminal-like, but also athletic and rhythmic. In order to investigate whether the mere knowledge of such a stereotype can influence our thoughts and behavior – irrespective of whether we consciously believe in it – Devine designed a study in which people did not realize that their stereotypes had been activated. She then assessed how aggressive people judged an ambiguous behavior. Participants in the study saw brief flashes and had to indicate whether these were in their left or right visual field. In reality, these flashes were words, which were so briefly presented (80 milliseconds), that the human brain could not consciously recognize them. Nevertheless, under these conditions our brains are still able to process their semantic content. Such a procedure is called subliminal priming. For half the participants, these priming words referred mostly to the stereotype of Black Americans (e.g., ‘Blacks’, ‘Negros’, ‘niggers’, ‘poor’, ‘lazy’, ‘athletic’) – however, none of the words referred to aggressiveness and hostility. For the other half of the participants, the presented words were mostly neutral in content (e.g., ‘water’, ‘television’, ‘number’, ‘however’). After participants had seen 100 such words, they were asked to proceed to an allegedly unrelated second experiment about impression formation. Participants read a paragraph about a person, Donald, who engages in several ambiguously aggressive behaviors. For example, Donald refuses to pay his rent until his apartment is repainted. Participants then judged how hostile they considered Donald’s behavior.

The striking result of Devine’s research is that if participants had been primed with words related to Black Americans, they judged Donald’s behaviors to be more aggressive than if they had been primed with neutral words. This effect was equally strong for people with high and low prejudice levels. Thus, even though participants could not consciously read the priming words, the stereotype had been activated and influenced the subsequent judgment of another person, Donald, whose race was not mentioned. Devine hence found out that in situations where people cannot easily recognize that they are stereotyping, they tend to evaluate others based on stereotype-congruent attributions – especially if the observed behaviors are ambiguous and allow multiple explanations. Even though we can consciously control and counteract stereotypic thoughts, we are not always aware that there is something to counteract and hence don’t.

Researchers from New York University (Bargh, Chen, & Burrows, 1996) took this line of research even further. They demonstrated that the activation of a particular stereotype does not only lead to congruent judgments about others, but it also influences one’s own behavior. In their experiment, participants completed a supraliminal priming task, the so-called Scrambled Sentence Task, where participants have to form a grammatically correct four-word sentence from five given words. For a third of the participants, these sentences contained content related to the trait of rudeness (e.g., aggressively, rude, disturb), for another third politeness was included (e.g, respect, honor, considerate), and the remaining participants received sentences with neutral content. Participants were asked to complete the Scrambled Sentence Task and then find the experimenter in an adjacent room. However, when the participant had finished, the experimenter in the other room was in a conversation with someone. The object of the experiment was to find out if the participant would interrupt the experimenter. The researchers hypothesized that activated categories, such as rudeness and politeness, would affect it accordingly. This is exactly what they found: about 65% of the participants who had unscrambled sentences relating to rudeness, interrupted the conversation, whereas only about 40% of participants with neutral sentences did so and less than 20% of participants who had unscrambled politeness-related sentences. In a similar second experiment, the researchers showed that the activation of the elderly stereotype with words as ‘Florida’, ‘old’, or ‘lonely’, significantly decreased the speed at which participants walked after the experiment – even though the priming had not explicitly referred to the stereotypic content of ‘slow’. A third experiment had a more direct link to the work of Devine: for half of the participants the stereotype of African-Americans was subliminally primed. In comparison to a neutrally primed control group, these participants reacted with more overt hostility when the experimenter told them that because of a computer failure they had to do a tedious and annoying task again. In other words, the work by Bargh, Chen and Burrows demonstrates that the activation of a stereotype, e.g., simply constructing sentences which refer to a particular trait can influence a person’s subsequent behavior.

If we see the latter experiments in the context of Devine’s work, our problem with the innocent joke about the poor and aggressive black guy or the Dumb Blonde becomes apparent. Even if we don’t specifically believe in a stereotype, we still know about it and an encounter with a member of a stereotyped group, such as Blacks, will lead to the activation of that stereotype. As such activations go mostly unnoticed, we have no chance of counteracting the stereotypic content. We might see the person across the street as aggressive, simply because he is Black. Moreover, as the activation of such stereotypic thoughts also affects our own behavior, we might – unconsciously – act towards him in a more hostile manner than normal. It becomes obvious that such circumstances most likely lead to a self-fulfilling prophecy: the Black person, who we see as aggressive simply responds in the same manner as we treat him – aggressively.

But what about other stereotypes? Women can’t do math, but are good at languages. If it’s simply a gender difference, why complain? Ryan Brown and Robert Josephs from the University of Texas at Austin (1999) shed new light on these apparent gender differences. In an experiment, they framed a math test either as being indicative of weak abilities or of exceptionally strong abilities. Women performed particularly badly only when the test was framed as indicative of weak abilities, whereas the reverse was true for men, who scored worse when they believed the test to be indicative of strong math abilities. According to Claude Steele’s and Joshua Aronson’s  Stereotype Threat Theory (1995), the thought of confirming a a negative stereotype about oneself – or of being treated in terms of it – poses a threat. The experience of such a threat in turn causes malfunctioning and underperformance. A vicious circle! The reasons for such underperformance caused by the stereotype threat are twofold. First, the constant thought of not doing something, such as scoring low on a math test, might prime exactly that behavior. Second, the thoughts spent on trying not to confirm the stereotype take away cognitive capacities from the actual task. Obviously, this once again creates a source for self-fulfilling prophecies. So next time when you want to make a blonde joke before your (blond) girlfriend takes her math exam, you might just reconsider.

Taken together, the research on stereotypes suggests that there are many processes by which stereotypes enter our everyday lives. Stereotyping, that is, the activation of stereotype-congruent information, seems inevitable in daily situations. Through the activation of such category-based knowledge, our cognition and behavior is influenced. This may lead to a continuous confirmation of the stereotypic content, because by expecting certain behaviors from another, we almost automatically elicit such behaviors. Furthermore, if members of a stereotyped group are afraid of confirming negative stereotypes about themselves, they might fall pray to the stereotype threat and thereby underperform. On a more positive note, stereotypes are of course not always bad. In many situations, category-based knowledge really simplifies the world for us. We do not have to process every tiny piece of input we receive individually. By applying categories and stereotypes, our information processing is greatly facilitated. Some – even negative – stereotypes can also be good for us and we might not want to get rid of them. For example, when we see a group of men with shaved heads ahead of us, it might be a good idea to change to the other side of the road and avoid a potential situation of danger with a group of skinheads. Nevertheless, the research teaches us to be particularly aware of the stereotypes we have and encounter. Just thinking, “I don’t believe in them anyway” doesn’t do the trick; we have to constantly remind ourselves to counteract them. You don’t believe me? You are convinced you only stereotype others when you want to? Have a look at Hidden Bias and take a test from Project Implicit to explore your hidden biases. You’ll be surprised to find out that you are more prejudiced than you thought!


Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230-244.

Brown, R. P. & Josephs, R. A. (1999). A burden of proof: Stereotype relevance and gender differences in math performance.Journal of Personality and Social Psychology, 76, 246-257.

Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-18.

General Protection Fault (March 7, 2005). Comic Retrieved June 11, 2007 from GPF Comics.

Steele, C. M. & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797-811.

article author(s)

article keywords