09/01/2015 05:03 pm ET Updated Sep 01, 2016

Ingroup/Outgroup Biases at Play in Police-Community Relations

Our ancestors evolved in a world where all travel was on foot. This sharply curtailed the geographical radius of their social universe, and meant that virtually all of their social encounters were with people who looked like they did. Thus, there was little opportunity or need to distinguish individuals based on race. Consequently, it is highly unlikely that the mind evolved perceptual mechanisms that encode for race per se. Yet from Staten Island to Ferguson to Prairie View, we see evidence to the contrary. Why such a mismatch between theory and experience?

The answer was broadcast across movie screens this summer, in The Stanford Prison Experiment. Philip Zimbardo's 1971 study is infamous for its disturbing demonstration of the alacrity and vigor with which biases in person perception are activated. With little encouragement, subjects assigned to be guards rapidly began dehumanizing and abusing subjects assigned to be prisoners. Prisoners who complained about civil rights violations were accused of insubordination, which was used to justify further abuse. The guards could inflict torture with impunity because they were armed with wooden batons (although instructed not to use them) and the authority of Dr. Zimbardo. The dynamic that emerged between guards and prisoners is similar to that infecting police-community relations in many U.S. cities, with one critical difference: nearly all the study participants were white. Clearly, the enmity between guards and prisoners was not racially motivated.

Biases in person perception--a.k.a. ingroup/outgroup psychology--refers to the universal human tendency to classify people according to whether or not they are a "member of my group." The criterion for inclusion can be a shared experience, interest, or attribute. Race is but one potential marker of group membership; each of us belongs to many social groups, some of which overlap in a Venn-diagram-like manner. For example, I am a member of the group that identifies as Oregonians, the group that identifies as educators, and the group that identifies as cat lovers. The tendency to experience affiliative emotions toward those who are similar to us in some way emerges early in development: by 12 months, infants prefer an animal puppet who likes the same food they do to one who does not. Unfortunately, this tendency has a dark side: cross-culturally, humans exhibit positive biases toward ingroup members and negative biases toward outgroup members. In extreme cases, these biases are expressed as a belief that the ingroup is virtuous and superior, while the outgroup is immoral, inferior, and even sub-human. The Zimbardo experiment triggered these mechanisms by immediately dividing participants into two coalitions with opposing interests.

A more recent study addressed the question implied by Zimbardo's findings: Can race be overridden as a visual cue of group membership? Subjects were shown photographs of four African-American and four Euro-American men, told that the men were members of two rival basketball teams, and asked to sort the men into their respective groups. In addition to verbal cues (statements made by the players in the context of their rivalry), subjects were provided with a second visual cue: four men wore gray shirts and four wore yellow shirts. Because shirt color did not correspond to race, the study tested whether encoding by race is automatic and inevitable. The results: subjects overwhelmingly used shirt color to sort the players, suggesting that the use of race as a cue of group membership is context-dependent.

In the context of police-community relations, another criterion of group membership at play is place. Humans are territorial animals, and ingroup/outgroup biases may have evolved in tandem with this trait. During our long history as foragers, life depended on the right to hunt and gather resources in a delimited region--a right acquired through birth and social ties. A person who used resources to which he was not entitled--for example, a hunter who crossed a border in pursuit of game--was in effect committing theft. Thus, the ability to distinguish those entitled to use one's homeland from those who were not had implications for survival. Accordingly, in many forager groups, interlopers were treated with suspicion. Among the Iñupiat, for example, outsiders who used land without authorization were regarded as trespassers and customarily killed on sight. In ancestral environments, visual cues were a fairly reliable means of ascertaining affiliation: slight departures from the physical appearance of the local population (e.g., in dress, hair style, or weaponry) indicated that a person was "not from around here" and thus "not a member of my group."

Every time the police enter a community to which they do not belong, they cross a border. No matter how well-intentioned, this action is bound to trigger some uneasiness on both sides. The less the two parties have in common, the greater this tension is likely to be. A case in point: the harassment of the predominantly black citizenry of Miami Gardens by their predominantly white police department, recounted in a recent episode of This American Life. A contributing factor was the department's recruitment of personnel from outside the state, many of whom had no experience with African-American communities.

One way to manage ingroup/outgroup biases is to shift group allegiance by muting certain membership cues and highlighting others. This was the logic of Skateland's "NO CAPS--NO COLORS" policy aimed at preventing violence on the premises by discouraging patrons from identifying as members of rival gangs. At the same time, owner Craig Schweisinger made the rink a prominent venue for rap music, which in effect encouraged patrons to identify as rap fans--a group that included everyone present. Community policing--the assignment of officers to specific neighborhoods so they get to know the inhabitants--is a step in the same direction. This policy calls for partnerships between police departments and communities, but unfortunately doesn't specify what form they should take. The research reviewed here suggests that cultural exchange might be the answer: from babies to basketball teams to skaters, "what people like" (e.g., food, sports, music) consistently trumps "what people look like" as a cue of group membership.