Last look

Uncomfortable truths

By Brian Nosek
This is an image of Brian Nosek

Nosek.
Photo by Stephanie Gross.

Beliefs help humans make sense of the world, allowing us to understand our past and present and to predict and control our future. Our most elaborate beliefs are explanations of ourselves — who we are, what we are like and why we do the things we do. Forty years’ worth of psychological research reveals that humans are superb at generating explanations for their behavior but not accurate explanations. Self-explanations appear to say more about human capacity for telling a good story than the quality of self-insight. But identifying and giving up false beliefs is not so simple. Like possessions, we hold on to beliefs about ourselves long past their usefulness.  

One persistent belief that defies evidence is the conviction that all of our behavior comes from conscious intentions. Much of human behavior is governed by mental processes that exist outside of conscious awareness and conscious control. As examined by my colleague Tim Wilson in his book “Strangers to Ourselves,” we may have little more insight into explaining our own behavior than we do explaining the behavior of others. Conscious experience provides a compelling, but false, sense of knowing ourselves. The resulting beliefs might just be good stories that feel right, even if they are not.

The fact that much of human behavior is managed by automatic or nonconscious processes suggests a disconcerting conclusion: our conscious intentions and values may not be directing our behavior as much as we believe they are. Recent psychological research is marked by a proliferation of methods designed to assess implicit biases — thoughts and feelings that exist outside of conscious awareness or conscious control. One such method, the Implicit Association Test (IAT), assesses the strength of associations between concepts, such as racial groups, and attributes, such as good or bad. The task doesn’t involve an act of introspection to decide one’s feelings. It just measures how quickly concepts can be paired, and this is interpreted as an indication of how strongly those concepts are associated in one’s mind. The surprising result is that most respondents, even those who endorse egalitarian values, find it easier to associate white faces with “good” and black faces with “bad” than the opposite, indicating an implicit pro-white bias. What’s more, the strength of this implicit bias is related to behavior. Studies have shown, for example, that people with stronger implicit pro-white biases exhibit less friendliness with a black compared with a white interaction partner. The link between implicit bias and behavior appears to be particularly strong with behaviors that are difficult to control, such as nonverbal behavior. Sometimes, people are more biased than they think they are.

There is a phony comfort in writing “people are more biased than they think” because it fosters an illusion that I am not one of “them.” Measures like the IAT, however, can be administered on oneself, even knowing what they are intended to measure. Doing so forced me to confront a firm belief that I am unbiased with clear evidence to the contrary. Like that of many others, my performance revealed biases that are inconsistent with my conscious values. Uncomfortably, I face evidence that the scientific revelations about implicit biases in the human mind generally also apply to my own.

Implicit biases can contradict values, escape detection and influence action. For me, the uncomfortable truth is a lesson in humility. My mind, like all minds, is shaped both by processes that I endorse as my own and by processes that I am unaware of and might even dislike. Having convictions is not sufficient to guarantee that my actions will live up to their prescriptions.

Scientific evidence can clarify how the mind works but not what we should do about it. Accepting that implicit biases exist does not call for conceding mental authority to them. I maintain one belief with little doubt: I intend to live up to my values. The veracity of this belief is self-evident by having it. Whether I am successful at living up to my values, however, is a claim calling for evidence. Consideration of the evidence will be more accurate if some humility can temper the feeling that I know, with just a moment of self-reflection, why I do the things I do.

Gaining insight into the mental operations that lead behavior astray of values is a precondition for predicting, understanding and controlling implicit biases. That insight moves us toward the purpose and promise of having beliefs in the first place.

You can try out the IAT by visiting http://implicit.harvard.edu/.