Although categorizing people is much like categorizing objects,

Consequences of Categorization

Us and Them: Social Categorization

Although categorizing people is much like categorizing objects,

there is a key difference. When it comes to categorizing people, we

notice whether the person we’re categorizing is a member of our group

or not. We have a very strong tendency to carve the world into “us”

and “them”. Groups that you identify with – your country, your

religion, your political party, even your hometown sports team – are

called ingroups, whereas groups other than your own are outgroups.

Just how easily ingroups and outgroups can be created may

surprise you. Imagine that you are a participant in a study. You are

alone in a dark room and are estimating the number of dots on a

screen. After you have completed this task, the experimenter explains

that some people consistently overestimate the number of dots,

whereas others consistently underestimate the number of dots. He

tells you that you are a dot overestimator. Do you think you would

treat dot overestimators differently from dot underestimators? You’re

probably thinking that you would not because in addition to dot

estimation being an unimportant classification, you haven’t met

anyone from the other group of dot underestimators.

You are then told that you have been chosen randomly to award

payment to two other people. All that you know about these people is

that one is a dot overestimator like you and the other is a dot

underestimator. Would you give more money to the person who is like

you – the dot overestimator – or would you divide the money equally

between the two people? When Tajfel conducted this study in order to

see if this minimal distinction between groups could influence

behavior, the results were strikingly clear. Participants awarded more

money to ingroup members than to outgroup members. This happened

2

even though the participants did not receive any money themselves.

Based on no information other than knowledge of group membership,

participants used the category “my group” and “not my group” in order

to give more money to the person who was like them, instead of

dividing the money equally. This pattern of discrimination is known as

Ingroup Favoritism or Outgroup Discrimination and it has been

replicated many times, in many countries.

What Causes Ingroup Favoritism?: The Social Identity Theory

Explanation

Why should mere membership in a group lead to consistent

ingroup favoritism? Tajfel and Turner proposed Social Identity Theory,

which has been the subject of much debate, in order to answer this

question. According to this theory, our identities have two

components: a personal part and a social part. Our personal identities

contain specific and personal information about us, such as our

interests, preferences, abilities and traits, while our social identities

are based on the groups to which we belong.

Social Identity Theory also assumes that self-esteem is very

important to us. As a result, we continually strive to enhance our

feelings of self-worth. One way to increase self-esteem is through

personal achievement. Getting an “A” in a course or getting a

promotion at work for example, are good ways to boost self-esteem.

Self-esteem can also be enhanced though, through group

membership. Thinking that our group is better than other groups is a

quick and nice way to get a big self-esteem boost.

According to Tajfel and his associate Turner, Ingroup Favoritism

may occur because treating the ingroup more favorably is a way of

saying that our group is better than other groups. Giving more money

Mobile User

3

to people like us is a way to pat ourselves on the back. We will return

to the issue of self-esteem and its relationship to prejudice later.

Consider the ramifications of this basic social process. We may

share many connections with others such as race, ethnicity, age,

religion, even a shared birthday, school, etc., and this can result in a

favorable judgment and actions towards others. This can be a potent

source of discrimination – not based on dislike – based on a preference

for people like me.

The Outgroup Homogeneity Effect: “They’re all alike but my group is

diverse”

One consequence of ingroup/outgroup categorization is a

phenomenon known as the Outgroup Homogeneity Effect. We assume

that there is greater similarity among members of the outgroup than

among members of our own group.

Can you think of any examples of this effect? If someone asked you

to describe the elderly for example, what would come to mind? You

would probably think about them as one big mass of people who drive

too slowly, walk too slowly, can’t hear, are frail and get easily annoyed

(Oh no! That’s my stereotype…). Since you are relatively young, the

Outgroup Homogeneity Effect tells us that you probably don’t think

about this outgroup as a diverse group of people (which it is). You don’t

think about the elderly grandmother who is kind and helpful and bakes

cookies for her neighbors; you don’t think about the elder statesman

who is wise, learned and knowledgeable about many things; and you

don’t think about the elderly senior citizen who is lean, fit and athletic.

In short, you miss the diversity in a group if it is not your group. Older

people do this too when they talk about “young people today.” They

don’t recognize that there are many different types of young people.

4

Similarly, people from China, Korea, Taiwan and Vietnam see

themselves as different from one another but to non-Asians, they may

be seen as simply Asians.

The Cross Racial Identification Bias

Research has shown that outgroup members even seem to look

alike to us: People are less accurate in distinguishing and recognizing

the faces of members of racial outgroups, especially if they are

unfamiliar with those other groups. This is The Cross Racial

Identification Bias. Below is an example of this effect. This remark was

made by John Motson, a white football commentator during an

interview:

There are teams where you have got players who, from a distance,

look almost identical. And, of course, with more black players
coming into the game, they would not mind me saying that it can
be very confusing… if there were five or six black players in the

team and several of them are going for the ball it can be difficult.

It’s difficult to know from this comment alone, if John Motson is

prejudiced or if he is a victim of the Cross Racial Identification Bias.

Lack of familiarity with outgroup members can lead to the feeling

that “They” all look alike, but there’s more to it than that. Research

using brain imaging has found that merely categorizing people as

ingroup or outgroup members influences how we process information

about them even if we are familiar with the outgroup. In one study,

participants were exposed to unfamiliar faces of people who were the

same race as the participants. Some of these faces were ingroup

members (they were from the same university as the participants) or

outgroup members (they were from a rival university). Participants

processed the faces more completely, less simplistically and less

superficially (this was verified by more activity in the orbitofrontal

5

cortex) when they had been categorized as being from their ingroup

than they did when they had been categorized as members of the

outgroup. Deciding that someone is like me or not, makes a difference

even at the level of the brain.

We may not only process outgroup members’ faces more

superficially – we may sometimes process them more like objects than

like fellow human beings. In a study by Harris and Fiske, when

participants saw pictures of people from a variety of groups, fMRI

showed activation in their medial prefrontal cortex, which is the part of

the brain responsible for social thought. However, this activation was

not evident in response to images of either nonhuman objects or people

from particularly extreme outgroups, such as drug addicts or the

homeless. We just don’t think about them at all.

When we do think about them though, we may not see them as

fully human.

Cuddy and her associates conducted a study two weeks after a

major hurricane devastated the lives of thousands of people.

Participants read about a mother whose young son had died during the

hurricane. Some participants read about an African American mother

while others read the same story but in this case, the woman was

European American. Participants felt that the woman, regardless of her

race, was distressed, panicked, fearful and angry about her son’s

death. These are emotions that people believe animals as well as

humans experience.

We have other emotions though, secondary emotions, that are

believed to be uniquely human. These emotions require human

capabilities such as the ability to think complexly and abstractly and the

ability to think ethically. These emotions include: anguish, mourning,

grief, remorse, resentment, and guilt. Participants were less likely to

6

believe that the woman had experienced these uniquely human and

humanizing emotions if she was from a different racial group. People of

color (Black and Latinx participants) were less likely to indicate that the

White mother had experienced these secondary, human emotions.

Similarly, White participants indicated that the Black mother had not

felt the same complex human emotions that a White mother in the

same situation would have felt. In other words, people felt that the

tragedy of losing a child had less of an effect on someone who was not

a member of their group. Although they acknowledged that the

outgroup mother would be distressed by her son’s death, they did not

feel that she experienced complex and uniquely human emotions that

someone from their group would experience. Here’s an example: Since

they did not believe that the non-racially similar mother would feel guilt

(since guilt is a complex secondary emotion), they would not consider

or believe that she might stay awake night after night thinking about

what she could have done to save her son. Similarly, they would not

think that her life would be destroyed by her son’s death (since anguish

is a secondary emotion) but they would imagine that someone from

their group would experience incredible grief and anguish in this

situation.

This type of dehumanization has played a role in atrocities that have

occurred throughout history.

Additional Consequences of Ingroup/Outgroup Categorization

There are several other consequences that occur as a result of

seeing someone as part of the ingroup or outgroup.

Generalizing from the behavior of a single outgroup member

People are more likely to generalize from the behavior of one

member of the outgroup to other members of the outgroup. This is

7

particularly likely when the behavior is negative, stereotypical and

socially or personally threatening. So, for example, if a member of

another group harms you in some way, you are more likely to infer that

most people in that group behave this way than if a member of your

group harmed you.

This has been demonstrated experimentally. White participants who

overheard a fake phone conversation about an assault by a Black

person later rated Blacks as more violent than did participants who

heard the same conversation about a White assailant. Observing a

single instance of negative behavior involving a member of an outgroup

led people to evaluate all members of that group negatively.

The Ultimate Attribution Error: But what if a member of your group

has done something negative? People are more generous and forgiving

when it comes to their own group. According to the ultimate attribution

error, we see our own group’s negative behavior as something

temporary – it’s due to the situation – but the same behavior by a

member from a different group is likely to be attributed to that group’s

permanent horrible tendencies. This effect is particularly strong when

the groups involved have a history of intense conflict or when the

people involved are deeply prejudiced. It also occurs if the situation is

one that provokes strong emotion.

Can you think of real life examples of these two effects?