![]()
Carmen Miranda is probably best-known today as the former spokesperson for Chiquita bananas, but she was equally famous — and outrageous — as an actress, singer, and dancer in the 1940s and 1950s.
Cognitive dissonance is a psychological phenomenon that occurs when people’s actions contradict strongly-held beliefs. It’s such a distasteful feeling that people will often invent convoluted justifications to account for their actions. For example, if a white employer who believes herself not to be a racist decides not to hire an African American job applicant, she might justify her decision by convincing herself that the applicant was unqualified for the job.
So what does this have to do with Carmen Miranda? Researchers have found that cognitive dissonance can affect a variety of human judgments, from estimates of statistical likelihood to social assessments, to self-image. What researchers Emily Balcetis and David Dunning wanted to know is whether it could affect the human perceptual system. That’s where Carmen Miranda came into play. The researchers needed to come up with an unappealing way to measure distance — in this case, the distance across a university quadrangle. So what they did was ask students to walk that distance, in full daylight, while dressed as Carmen Miranda: wearing a grass skirt, a coconut brassiere, and, of course, a giant fruit-laden headdress.
The key to the study was the students’ belief that they had chosen to do this. Some of the students were simply told that they were participating in an experiment about their reaction embarrassment and then told to put on the costume and walk across the quad, after which they’d be quizzed about their reaction. Others were told they could either do the Carmen Miranda walk or some other (unnamed) embarrassing task. However, they were strongly encouraged to dress up in the fruity hat, and in fact, none of them chose the other task. So does dressing up as Carmen Miranda make you believe that the quads are wider than they really are? Here are the results:
The students who believed they chose to dress up as Carmen Miranda estimated the length of the quads as significantly shorter than those who were simply told to wear the costume. Another group of students was simply asked to estimate and walk the distance without the costume, and again, they believed the distance was significantly longer than those who thought they had chosen to wear the costume.
Balcetis and Dunning say that this misperception of the distance (the quads were actually more 365 feet long) is caused by cognitive dissonance. Since they chose to walk in the costume, these students convinced themselves their decision must be rational: after all, it’s not a very long distance to walk.
In a second experiment, the researchers used a completely different activity to see if cognitive dissonance affects perception. As before, some students were led to believe they had a choice of activity to perform, while others were simply told to do it. This time, the activity was to kneel on a skateboard and push themselves up a hill using their hands. Before they started, they were asked to estimate how steep the hill was using two different methods: drawing the angle on a sheet of paper, and using a large protractor with a movable arm. Once again, the students who believed they had chosen the task estimated the hill as significantly less steep than those who were simply told to do it.
So whether or not you think you choose to do something can affect your perception of physical properties of the world.
This could be a handy parenting tool: if you want to take your kids on a day hike but you’re not sure whether they’ll enjoy walking the distance, give them a choice of two similar hikes. Since they “picked” the hike, they’ll believe it’s an easier task. Or at dinner time, you could offer your kids the choice of two vegetables. The possibilities are endless. And it’s all because of Carmen Miranda!
Emily Balcetis, David Dunning (2007). Cognitive Dissonance and the Perception of Natural Environments Psychological Science, 18 (10), 917-921 DOI: 10.1111/j.1467-9280.2007.02000.x