Category Learning in Psychology

Updated April 17, 2017

Human beings automatically and effortlessly categorise their environment, their world and their experience to a degree not even remotely shared by other creatures. According to "Perspectives on Thinking, Learning and Cognitive Styles" by Robert Sternberg and LI-fang Zhang of Yale University, we categorise everything: When we decide whether or not to take an umbrella in the morning, we do so based on categorising the weather into "umbrella needed" or "umbrella not needed." A physician uses symptoms and medical tests to diagnose patients into known categories of injury or disease. We use "stereotypes" to categorise the people around us.

Why We Categorize

The reason we use categories is to promote "cognitive economy," according to the authors of "The Embodied Mind: Cognitive Science and Human Experience," Varela, Thompson and Rosch, researchers from Yale University. We encounter a large amount of objects in the world everyday. Our brains are designed to note the ways these objects are different or similar from each other. This is the basis of forming concepts and categories. Without this ability, we would be forced to remember each object we encountered in detail, storing it individually in memory, along with all that we know about it. While our brains are big, they're not big enough to store that much information, let alone think and reason about all that information. Categorisation breaks down and organises the world into meaningful units so we can think about them.

Human Learning

So powerful is our talent to learn and so fundamental to that talent, is our ability to categorise the world that we're now trying to teach computers how to categorise the world. Research suggest that we have multiple neuro- and cognitive-based learning systems to tackle different aspects of the problem of categorisation. The two main systems are: explicit reasoning--or hypothesis testing-based system--and an implicit procedural system--a system that learns by doing, or by experience.

Machine Learning

Given a set of explicit categorisation rules, computers excel with this type of categorisation method. However, this method alone fails to capture the subtleties and nuances of distinguishing even everyday objects. Humans employ more powerful techniques, not yet well understood, and questions about this technique are fuelling the drive to create machines that are ever more intelligent.

Visual Categorization

One form of implicit procedural learning is our ability to categorise objects visually. Often, we can't easily describe the differences between the objects we see, but we "know" based on their visual features--that they belong to different classes of objects. While later we can attach labels to those categories, for example "cats," "dogs;" the initial computation is preverbal and somewhat automatically unconscious. In fact, pigeons and other animals can discriminate between complex classes of objects as described in an article in the Journal of Experimental Psychology.

Motor System-Based Categorization

Another form of implicit procedural learning is carried out by our motor system. In short, our bodies seemed designed to interact with the world in a way that supports automatic categorisation of the world. Simply interacting with objects causes the brain, somehow, to create automatic categorised representations of those objects, and these representations are used to help us think about and understand our world. The incredible conclusion is that if we couldn't touch and interact with the external world, we would not understand it as well as we do--just seeing the world and thinking about it isn't enough.

bibliography-icon icon for annotation tool Cite this Article

About the Author

Comte de Dirac is a self-taught writer, professional business blogger and SEO professional who began writing professionally in 1995. He has technical and research articles in several print and online publications including the "Asian Journal of Engineering Research" and the "Journal of Applied Psychology."