The Cognitive And Social Functions Of Social Categorization

Categorization has been an area of interest arguably since Aristotle, whose classical view on classification by reference to mental definitions, that are comprised of rigid, clearly defined features and attributes. However, social categorization has been more extensively explored since the 1960’s with Tajfel’s theory of social identity. At the outset, it is important to establish a universal definition for both categories, categorization, and stereotypes. Categories are the mental representations of similar groups that we hold in memory that are based on essential features (Moskowitz, 2005).

Categorization is the cognitive process we use to organise and understand the world around us by classifying things into distinct groups (McGarty, Social Categorization, 2018). Stereotypes contain knowledge of what members of groups are like, what qualities they have and how we expect them to behave (Knippenberg & Dijksterhuis, 2000). This essay will examine the cognitive and social functions that categories serve for perceivers. It will then, critically discuss the research investigating how a perceiver’s inaccurate assumptions can be shaped by activated social categories that are incorrect. It will consider the discourse on whether categorization is a helpful efficient process or ultimately problematic for society and its members. This essay will conclude whether social perceivers can avoid relying on categorical stereotypes.

The functions of categorization

The functional explanation for the process of categorization has been consistently agreed upon across the decades. Allport stated that if we don’t categorize, then the amount of information required to process the stimuli would overwhelm our capacity (Allport, 1954). This is echoed 45 years later in McGarty’s book that explains that categorization attains cognitive economy because it translates the innumerable stimuli in the physical environment into usable and simplified entities (McGarty, Categorization in Social Psychology, 1999). Many psychologists would contend to the fact that categorization is a fundamental process because it serves the basic function of organizing knowledge we have about our environment, thus bringing coherence to the array of novel entities we interact with (Bodenhausen, Kang, & Peery, 2012). Categorization may be accused of dividing people by placing them in categories and creating an us versus them attitude. However without categorization we would not be able to understand or predict why others may be similar to us or different from others, moreover, when we categorize ourselves, it can create unity, shared goals, and an ability to learn whom we can cooperate with (McGarty, Social Categorization, 2018). This has been shown to be an almost innate process by Liberman, Woodward, Kinzler (2017) who used violation of expectation studies to investigate whether infants could form abstract social categories. Interestingly, they found that 11-month-old infants could use information about group membership to infer whether people would share properties and how they would interact (Liberman, Woodward, & Kinzler, 2017).

Suggesting that infants expect social relationships to be governed by what social categories the entities belonged to. When we come across a novel object or entity we look for and attend to recognisable features that associate with categories that we have in our memory, once we have placed said object or entity into that category we attribute them other features or stereotypes that they may have. But why do we do this? Largely, it is thought that schematic filter models may account for how and why stereotypes guide attention. These models propose that because humans are motivated to minimise their cognitive load, their attention relies on information that is easiest to comprehend, known as expectancy-consistent information (Allen, Sherman, Conrey, & Stroessner, 2009).

It could be said that when we categorize or assign stereotypes inaccurately it can improperly affect our future perception and cause unwarranted behaviour. However, abstract categorization may actually improve our behaviour. That after we categorize entities and apply stereotypes we can then compare those qualities to our self and this can subconsciously have an effect on our behaviour.

For example, Dijksterhuis and van Knippenberg (1998) reported that participants primed with the ‘professor’ category performed 13% better on a trivia test than participants primed ‘soccer hooligans’ (Dijksterhuis & van Knippenberg, 1998). This is because professors are associated with intelligence, unlike soccer hooligans, thus the idea is that trait activation leads to behaviour assimilation in abstract categories. Since this study, newer research has been conducted to see how far this idea can extend, in both spontaneous behaviours but also future behaviours. Nelson and Norton (2005) found that participants primed with the category ‘superhero’ volunteered twice as much as controls did and showed an increase in a prosocial behaviour (Nelson & Norton, 2005). In their third study, they found that ninety days after the initial exposure to the prime, participants primed with ‘superhero’ were four times more likely to show up to participate in the volunteering meeting than were people that had been primed with Superman (Nelson & Norton, 2005). This finding may be applicable beyond the laboratory setting, by demonstrating that primes can influence engagement in future behaviour it could be used to help people perform better, feel more confident or increase helpfulness. Therefore, this function can support the use of categorization, because without having an abstract category you cannot achieve behaviour assimilation through trait activation. However, prominent findings of priming have come under increased scrutiny and their efficacy and reliability debated.

In 9 experiments with 475 participants, Shanks as his colleagues employed the same intelligence priming procedures as Dijksterhuis and van Knippenberg, but found that none of the experiments obtained the effect. In contrast to this Lakens’ 2017 paper puts forward that 8 of those 9 studies were underpowered thus has low informational power (Lakens, 2017). Overall, there are many vital functions of social categorization; it helps us to make predictions and inferences, which can guide our decision making and how we interact with entities. Furthermore, due to our inability to cope with a high cognitive load, we are not able to process everything we encounter as we cannot treat every object or entity as unique. This is why our minds develop these mental representations of categories of objects. However, these categories are only informative if the stereotypes are accurate. Inaccurate interpretations. Categorization can sometimes have negative consequences when we categorize incorrectly or judge individuals by grouping them with similar entities and assigning them inaccurate characteristics.

Previously, there has been a lot of research on the perception of African American models and how their stereotypic associated traits can change people’s perception of their behaviour. For example, a well-known study showed that a mildly aggressive shove was perceived as more threatening when performed by an African American actor in comparison to a white actor. (Duncan, 1976). Similarly, Sagar and Schofield (1980) presented 11-year-old boys with drawings and verbal accounts of an ambiguous interaction between two people, for example, bumping into each other in a corridor. The ethnicity was manipulated by the shading used in the drawings. They found that the boys perceived the actions of the African American stick drawing to be more threatening than the depicted white figure (Sagar & Schofield, 1980). Curiously, they found that this racial bias was similar for both the white and the African American boys; that the tendency to see African American behaviour as more hostile did not depend on their ethnicity. This suggests that the threatening bias reflects the application of a known stereotype and is not internalised prejudice.

But why is this the case, why are some categories more accessible when viewing particular entities? Research has shown that when we categorize individual entities, we assign them the characteristics and traits associated with their racial group. For example, in America, hostility is linked to African Americans as shown by work done by Correll. In his 2002 study, he asked 40 participants to play a simple video game that presented either a white target or an African American target, and a held object, either a gun or a mundane commodity. The objective of the game was to shoot a target holding a gun and not shoot an innocent target. The ideology was that a participant would not need to process the ethnicity of the target to determine whether they were armed. The findings revealed that participants were more likely to mistakenly shoot an African American target that a white target (Correll, Park, Judd, & Wittenbrink, 2002). In a similar vein, recent research by Lundberg, Neel, Lassetterand Todd (2018) has confirmed this categorization bias through a weapon identification task, where participants had to classify objects as guns or tools after briefly being presented a black or white male face. They found that dangerous objects and words were identified more easily after seeing faces that we black than white. These faster response times and lower error rates indicate that there is a racial bias of black men being viewed as more hostile, aggressive, and threatening in automatic processing (Lundberg, 2018). This could lead to a social problem as associating black ethnicity with guns is an inaccurate interpretation, thus supports the idea that categorization is accusable of leading us to perceive something that isn’t reality.

With these findings, it is important to state that this association is not due to prejudice or racism, but is an automatic process that involves people on some level associating two concepts, like race and level of hostility. However, there is conflicting research regarding this statement. For example, Devine (1989) believes that stereotypes are automatically activated, unlike personal beliefs which require more conscious attention (Devine, 1989). Basing her second study off Bargh and Pietromonaco (1982) and using procedures that produce attentionless processing of primes, Devine found that priming participants tachistoscopically with pejorative labels and stereotypes of black people automatically activated the cultural stereotype of hostility regardless of whether they were high or low in prejudice (Devine, 1989). The explanation given was that all types of people are equally knowledgeable about cultural stereotypes so regardless of being a high- and low-prejudice person the stereotype-congruent responses are equally as strong and inescapable. Whereas Lepore and Brown (1997) found that only more prejudiced participants form a more negative and less positive impression of a target person after being primed with the social category of African American (Lepore & Brown, 1997).

However, this argument that only more prejudice people apply stereotypes, can be refuted by looking back at the Sagar and Schofield (1980) study, where both white and black children rated the African American drawing as more hostile. It is unreasonable to suspect that the black children are prejudiced against their own race, but that they are knowledgeable about the stereotypical associations with the category. This is illustrated by Devine who comments that culturally defined stereotypes are a part of people’s social heritage rather than personal beliefs (Devine, 1989). So, if racial stereotyping is culturally inherited, is gender also a cultural and social problem? A recent study by Weisman, Johnson and Shutts (2015) looked into whether we have evolved cognitively to track gender but not race.

The research investigated automatic social encoding of race and gender in three- to six-year-olds. The children were asked to remember facts about unfamiliar target children which varied in race or gender. They found that the children were sensitive to gender, but not race, in their attempts to remember facts about individuals (Weisman, Johnson, & Shutts, 2015). This could mean that gender is a more fundamental category for children and salient when encountering and encoding an individual. This is supported by earlier research by Frazier (2012) and Shutts, Banaji, & Spelke (2010) where they respectively demonstrated that young children are more likely to accept food and novel objects offered by same-gender individuals than other-gender individuals, but, have no preference on accepting objects connected with same- and other-race individuals (Frazier, 2012) (Shutts, Banaji, & Spelke, 2010). Conclusion In conclusion social categorization emerges as a process that is generally adaptive, it is a quick automatic first step that helps us think about others’ thoughts, beliefs and actions, thus helping us to navigate social interactions.

On the other hand, social categorization can sometimes be problematic. Prejudice may be an inevitable outcome, as stereotypes are automatically applied to members of a categorized group. Yet this is a consequence of having limited capacity to fully process entities. Understanding what causes social categories to become automatically encoded could be useful in ameliorate category-based judgments and social bias.