This post is our first that is wholly dedicated to a tenet of the social identity approach. That tenet is ‘cognitive categorization’.
Why cognitive categorization?
We need to have some familiarity with cognitive categorization in order to understand self-categorization; self-categorization being at the heart of the social identity story. Without self-categorization, for instance, there can be no such thing as an ingroup. But we will get to that later.
Cognitive categorization is “the process of understanding what something is by knowing what other things it is equivalent to and what other things it is different from”.[1]
What is this thing? What are its boundaries? What qualities does it have? Is it a table or an aardvark? These questions can only answered by determining whether ‘the thing’, whatever it is, is different from, or the same as, ‘other things’.
Let’s look at an example. In the below image is a collection of knives, forks, spoons. But forget for a moment that you know what cutlery is.
I knew my eclectic tableware collection would come in handy.
Without knowledge of common eating implements what we have in front of us is a wholly disorganized mess of items; each unique and different from the others, yet in many ways all the same. Can you see it? The buzzing confusion of features and shapes?
In order to get a conceptual grip on these things we need a way of appreciating patterns in the data. How do we establish that three round items can be classed together and distinguished from the other items? What do we need before we can give them a label (“spoons” perhaps) and focus on the potentially useful characteristics they share (e.g. can hold and transport liquids)?
The answer is cognitive categorization, and cognitive categorization does this via a phenomenon called accentuation.[2] More specially, via the “accentuation of intraclass similarities and interclass differences”.[3]
What this means is that, when we categorize, we perceive more differences between categories and more similarities within categories. Thus, categorization is not just passively observing intrinsic similarities and differences (really there are no such intrinsic similarities or differences; only that buzzing confusion). Categorization is instead the active psychological process of imbuing our sensory experience with meaning by adding similarities and differences.
The empirical work helps make this clear.
The classic demonstration is the line length experiments conducted by Tajfel and Wilkes.[4] In that series of experiments the researchers asked research participants to estimate line lengths. Eight lines of different lengths were sequentially presented in random order. As an experimental manipulation, the eight lines were given different labels for different participants. An illustration is given below.
In the first set the labels correspond to the length of the lines, while in the second the labels are random.
What Tajfel and Wilkes found was that the addition of a systematic categorization scheme (i.e. shorter lines labelled ‘A’ and longer lines labelled ‘B’) made participants reliably exaggerate differences at the group boundary. In other words, despite the same lines being presented to all participants, the difference between lines 4 and 5 was perceived as bigger when the organized ‘A’ and ‘B’ categories were present. They also found some indication that the differences within each category (e.g. within all the ‘A’ lines) were perceived as smaller. In sum, the participants found a way of ‘seeing’ the short lines vs. long lines.
Since those early studies accentuation effects have been replicated countless times in a variety of contexts. My personal favourite is the accentuation of interclass differences and intraclass similarities in computer generated fish.[5]
Looking back at my cutlery, the take home message is that our understanding of ‘stuff’ is not an inherent property of that stuff, but rather is a product of cognitive categorization. This means that something as simple as perceiving spoons requires some serious cognitive work. In particular, the detection of specific patterns in the stimuli and the subsequent suppression of the potential differences among the spoons (e.g. all the spoons are different sizes) and the exaggeration of the potential differences between the spoons and the items (e.g. only the spoons are round). Categorization, through accentuation, allows us to convert data into information.
That is cognitive categorization in a nutshell: an omnipresent perceptual tool for understanding our world. And while the process introduced above is extremely general, some comprehension of the above is a prerequisite for understanding self-categorization, and subsequently the social identity approach.
Although I think you can predict where this is going… Spoiler alert! Self-categorization is just the cognitive categorization by a perceiver of themselves. That is, creating understanding of the self via the active perception of who ‘I’ is equivalent to and who ‘I’ is different from.
Yet despite the simplicity of the idea, the implications are enormous. Self-categorization is foundational to social stereotyping, group cohesion, ethnocentrism, cooperation, altruism, empathy, collective action, normative behaviour, social influence, and more.[6] Pretty big stuff.
And maybe you already have an inkling of how these ideas connect. But maybe not. Maybe the connection between perceiving spoons and organizational behaviour is pretty damn opaque. Either way, stick around to find out what the social identity approach has to say.
[1] McGarty, C. (1999). Categorization in social psychology: Sage Publications Ltd. p. 1
[2] Tajfel, H. (1959). Quantitative judgement in social perception. British Journal of Psychology, 50(1), 16-29.
[3] Turner, J. C. (1982). Toward a cognitive definition of the group. Social identity and intergroup relations, p. 28.
[4] Tajfel, H., & Wilkes, A. L. (1963). Classification and quantitative judgement. British Journal of Psychology, 54(2), 101-114.
[5] Corneille, O., & Judd, C. M. (1999). Accentuation and sensitization effects in the categorization of multifaceted stimuli. Journal of Personality and Social Psychology, 77(5), 927-941.
[6] Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the Social Group: A Self-Categorization Theory. Blackwell: Oxford.
Social Identity Resources 
Good post on our tendency to categorise. Given the results of the lines test, do we only assign categories once we aware of the existance of categories a and b?
Therefore are we only capable of recognising social caegories that are already made known to us?
If this is the case, are there times when we expend extra effort to recognise a new category that was previously unrecognised?
Hi Miles, good question.
It is actually sort of the opposite. That is, we are aware of the categories A and B only after we assign them. Said another way, perceiving a category is actually the same act as creating a category.
That probably doesn’t sound very intuitive, but it might become clearer when you think of the subjective experience of participants in the Tajfel and Wilkes study. Although they might have been familiar with the labels, they had no prior experience with the line task, and certainly no prior experience with the relationship between these particular labels and line lengths. Really, what the participants were exposed to was a set of novel stimuli with various labels and various lengths. The participants then derived from those stimuli a categorization scheme that made sense to them; short lines vs. long lines. This might have been categorization scheme hypothesised by the researchers, but the participants still had to do that cognitive work.
In some ways this is the answer to your second question. Do we at times spend the extra effort to recognize a new category? No. It is actually the case that we always spend the extra effort.
Does that make sense? It is pretty tricky to talk about. Because we always categorize (and don’t just have a tendency to categorise), there is no intuitive way to talk about non-categorized stimuli. That is, we don’t have the right language for it. I used the term ‘data’ in the blog, but that isn’t perfect either.
I should also say that the above does not mean that one person’s subjectively experienced categories cannot be communicated to a second person. It is just that the subsequent use of those categories will still require the internal and effortful creation of those categories by the second person.
If the above is all sounding very constructivist then you are right. Indeed, there is a strong social constructivist theme in much of the social identity literature. This will become more and more apparent as the blog continues. It will also become apparent that there is some dissensus around these views among social identity theorists. For example, some see more of a role for the use of categories that are stored in memory and can be applied to data ‘ready made’. Another time though.
Cheers
Andrew
Pingback: A tale of three Trumps: Social identity and cognitive categorisation (Part 1) | ayoubbouguettaya·