Understanding the Cognitive and Neural Mechanisms Underlying Social Evaluation

Danny Glover, the stage and movie actor from the Lethal Weapon series, recently filed a complaint with the New York City Taxi and Limousine Commission (TLC) charging that local cab drivers repeatedly refused to pick him up because of his race. He claims that although he is a movie star in Hollywood, on streets of New York City he is, first and foremost, a Black man. This social group membership, he contends, is his primary identification in society and because of negative attitudes towards his social group, he is treated unfairly.

In this millenial year it is somewhat shocking, yet at the same time not surprising, that incidents like those alleged by Mr. Glover still occur. Although at one time there were social norms and laws that dictated the attitudes towards racial groups, since the civil rights movement of the 1960's these social norms have changed. One result of these changes is that the endorsement of prejudicial attitudes has dramatically declined over the last several decades. However as the incident described above illustrates, this dramatic decline in stated racial attitudes has not accompanied a similar decline in societal racial problems.

Social psychologists have suggested this discrepancy between reported attitudes and behavior may be due to the fact that the influence of attitudes is much more subtle and pervasive than is suggested by measures of self-report. There are two reasons why these self-report measures may not mirror behavioral responses. First, individuals may be reluctant to admit that they endorse certain prejudicial attitudes. It is likely that the cab drivers mentioned above would not express negative attitudes towards Blacks (especially to the TLC), yet they still intentionally avoided Mr. Glover. Second, it has more recently been suggested that we may not always be consciously aware of our biased behavioral responses. This type of unintentional social group bias is reflected in the recent Massachusetts Institute of Technology (MIT) report on gender discrimination, which found a "pervasive and unintentional" pattern of discrimination against female faculty. In his commentary, MIT Dean Robert Birgeneau, stated: "I believe that in no case was this discrimination conscious or deliberate. Indeed it was totally unconscious and unknowing."

Using indirect behavioral measures, social psychologists are beginning to differentiate between social responses and attitudes that are purposeful and direct (explicit) and those that are unintentional and indirect (implicit). Explicit attitudes are those we consciously express and often believe to be true. Implicit attitudes are expressed through our behavior, without our conscious awareness and/or control. While our explicit attitudes may reflect the ideals that we endorse, our implicit attitudes may reflect our underlying knowledge of cultural norms, our own social group membership, and our personal experience. These implicit attitudes may or may not be consistent with our explicit attitudes. For instance, in contrast to studies showing relatively unbiased explicit evaluations of Black Americans by White Americans, there is robust evidence for negative evaluations using implicit or indirect measures.

This research provides scientific confirmation of the sentiments long expressed by members of underrepresented social groups. Even though we now have laws against prejudicial acts, and even though many individuals may be well intentioned, cultural stereotypes of social groups exert a subtle but real influence on behavior. Social psychologists have approached this problem by identifying personal and social group factors related to the direct and indirect expression of social attitudes. I propose another approach. It is equally important to understand the specific mechanisms of learning and expression that underlie these behaviors. Are we endowed with special cognitive and neural mechanisms to learn and express our social attitudes? Or do we learn and express social attitudes using the same systems we use to evaluate non-social, emotional stimuli? In other words, can we show that social learning and evaluation are rooted in the ordinary mechanics of mind and brain.

Recent research on the basic cognitive and neural mechanisms of emotional learning has suggested some links to studies of social attitudes. Much in the same way that social psychologists differentiate direct and indirect means of social evaluation, cognitive neuroscientists have identified brain mechanisms related to the direct or indirect expression of learned emotional responses. Studies conducted across a range of species have shown that the amygdala is a critical brain structure in emotional learning. Recent research conducted in my lab and others has shown that in humans, the amygdala's role in emotional learning is often limited to the indirect expression of the emotional response. For example, it is a classic finding that when people are startled (for instance, by a loud noise) in the presence of negative stimuli (e.g, a dark, empty street) this startle response will be enhanced or potentiated. This startle potentiation indirectly indicates the emotional evaluation of the stimulus. We have shown that patients with amygdala damage, in contrast to normal controls, do not exhibit this startle potentiation in the presence of negative, non-social stimuli. At the same time, these patients rate these stimuli as equally arousing and negative as control subjects. In other words, they are able to explicitly report the emotional evaluation of the stimuli, even though they do not show an implicit or indirect emotional response. The performance of these patients is in contrast to the performance of patients with damage to a neighboring brain structure, the hippocampus. Hippocampal damage results in the opposite pattern; patients are able to indirectly express an emotional response to a stimulus through behavior, even though they cannot explicitly report the emotional significance of the stimulus. Thus, different cognitive and neural systems underlie the direct and indirect expression of affective evaluation.

Investigations into both social attitudes and the cognitive neuroscience of emotional learning have reported dissociations between indirect and direct affective evaluation. These common results led to me to wonder if similar mechanisms underlie these two types of findings. Could amygdala activity be related to the indirect evaluation of social groups? In order to test this hypothesis, we examined activity in the amygdala in White American subjects while they viewed pictures of unfamiliar White and Black male faces. Subjects were given three measures of social evaluation. One was an explicit, self-report of racial attitudes. The other two were indirect, behavioral assessments of racial evaluation.

Although the majority of White subjects showed greater amygdala activity to the Black compared to White faces, there was significant variability among the subjects. When we examined the relationship between amygdala response and measures of social evaluation, we found that the White subjects who showed greater amygdala activity to the Black vs. White faces also showed stronger biases in social responses on our indirect measures. This same relationship was not found with our direct measure of social attitudes. In other words, the amount of amygdala activity to the Black vs. White faces was specifically related to indirect social responses. These results suggest that the same neural systems that underlie the indirect expression of emotional learning to non-social stimuli are also related to the indirect expression of social responses.

In an effort to understand some of the variables related to these results, we performed a second study identical to the first, with one exception. The Black and White faces belonged to well known and positively regarded individuals (e.g., Martin Luther King, John F. Kennedy). In this study, there was no consistent pattern of amygdala activity to Black vs. White faces in the White subjects and no relation between amygdala activity and any of our behavioral measures of evaluation. These results suggest that the amygdala's response to Black faces in White subjects is diminished with specific experience, familiarity, and positive evaluation.

These are the first studies to link the evaluation of social groups to brain activity. By creating this link we add what we know about the cognitive function of these neural systems for emotional learning to our understanding of social evaluation and attitudes. These preliminary studies start to delineate mechanisms involved in social evaluation, but they also raise several crucial questions. For instance, what factors are related to the variability observed among the White subjects? What kind of responses would we expect from members of other racial groups, such as Black Americans and Asian Americans? What other brain regions are related to social evaluation and what are their behavioral roles? Precisely how does familiarity, positive information, or specific experience with members of different social groups alter the neural systems of social evaluation? In the proposed research we aim to address these questions. By identifying the cognitive and neural mechanisms of social responses we can initiate discovery of the means by which they are learned and modulated.

In his complaint to the Taxi and Limousine Commission, Mr. Glover requested that New York City cab drivers receive diversity training. How can we be sure this diversity training changes the explicit and implicit mechanisms of social evaluation? Given what we know about learned emotional responses in general, what types of training might be most effective at altering both direct and indirect social evaluations? Understanding the complex mechanisms underlying social responses will help us answer these questions.