Jon Hung

This week I take a look at some of the work of UC Berkeley professor Eleanor Rosch.  Rosch created a new paradigm in the field of cognitive psychology, known as prototype theory. Her investigation into the way humans employ categorical reasoning is important to understanding how humans are able simplify the complexity of the world into simple, yet meaningful and distinct concepts.  Anyone working with information architecture, product cataloging, or marketing might want to have heard of her work.

More after the jump.

(read more)….

Some examples of questions which employ some degree of categorical reasoning:

Should I get cable or DSL?

Is this piece of code better defined as a sub-class or a class on its own?

Should this piece of content be tagged with this label or that label?

What should I wear to this business casual event?

When humans act in the world, we classify stimuli in the world as belonging to meaningful groups and deduct appropriate behaviors from what knowledge we have about those groups. This is categorical reasoning. If we never reason that the orange and black striped animal is a tiger and figure out we need to get the hell away from it, we’d be a person lacking categorical structure in addition to being tiger-food.

Scholars have long theorized about how categories are represented in the mind and guide human behavior.

• Classical theory: a list of necessary and sufficient conditions
  Problem: no such hard boundaries, too many exceptions and “fuzzy” scenarios.
• Examplar theory: a category is a list of examples.
  Problem: your brain cannot hold that many examples, and we don’t seem to reason by matching new instances to prior ones, which would be resource demanding.

Rosch is responsible for an explanation of categorical reasoning known as prototype theory.  She makes the claim that categories are a graded structures, rather than a category with boundary defined by a definite set of features. Instead, members of the category are structured around a few central members, called prototypes (or best examples). Other non-prototypical members of the category are around the fringe and are ‘a-typical’ to various degrees. Chair, sofa, table are better examples of the concept furniture than are stove, refrigerator or telephone.

Rosch’s experiments show us that people classify an everyday object or experience on an example that they regard as the best representation of the appropriate category.

In addition to developing prototype theory, Rosch notes in her research that many category members are in-themselves categories. Within the animal category, there are many types of dogs, such as. To explain this phenomena of greater or less specificity, she develops a hierarchy of categories, describing them as basic, superordinate, or subordinate.

The middle of the hierarchy is called the basic level, and contains the categories most commonly used in everyday speech. This is because at the basic level, categories are maximally distinct, or minimizing perceived similarities across contrasting categories.  If I were to ask you what you are you sitting on,  you would respond “I’m sitting on a chair.”  The expected response is a category at the basic level, because you only need to specify that you are sitting on a chair and not a bed, or on a pencil, or floor.  My question does not imply a superordinate (I’m sitting on a piece of furniture) or subordinate (I’m sitting on an Ikea NOCHT chair) response, so a description like this would be inappropriate.  Descriptions at this basic level provide the maximum information while requiring the least cognitive ability.

Some points to take home from this discussion:

1) When you’re marketing items or selling a product, it’s important to keep the idea of the prototypical in mind. How similar is your product to the prototypical member of the same category? Companies are often in battle about being the most salient, and prototypical member of their product category. Coke versus Pepsi, Nike versus Reebok,  I’m not trying to tell you to be the same. But make sure you’re managing the space between your product and the main prototype. Are you marketing that difference enough? Often times, people don’t have the cognitive resources to develop such distinctions. You’ve got to know what your audience will consider a distinct item.

2) Know your audience’s basic level of categorization and account for their ability to make distinctions.  Every user population or target market has a level of knowledge they can apply to make distinctions between data items. This applies to you information architecture people. If you’re developing a site for cancer doctors, you can get into fine-grained distinctions. If you’re cataloging all types of diseases, on the other hand, your audience may not require every oncology term.

3) A follow up to point 2 is that cultural trends and movements in the marketplace often affect a psychological category. New products that enter the market are consistently shifting people’s ideas of everyday items. What is a computer versus a mobile device? The divide between them grows thinner with every new SmartPhone. What defines a green product? Standards are always changing. As you position yourself into the market, one needs to be painfully aware of how the majority of people view the prototypical example.

4) There are concepts that have a tricky time being categorized using any of the above theories. A commonly brought up objection is the concept of a “game”. Or what is an experience? Is there a basic level? I mean, we never talk about going outside to “experience the mall”. We say we’re going to meet up friends, or going shopping, or watching a movie. These are specific subordinate categories of expierneces that are all grouped within the “going to the mall” experience. This gives us license to be creative in category formation, to be inclusive of what range of products or services can be considered ‘your category’.

These points are just a small sampling of what can be taken away from Rosch’s ideas as well as the field of categorical researching in general. If you haven’t heard of Eleanor Rosch or her work, you can skim her early work to see how the field began, or read up on categorical reasoning research. I really enjoy the paper because it emphasizes the fact that we are cognitively limited in absorbing all the stimulus out there in the world, and that categorical reasoning is one of many mental apparati which helps us act in the world without needing extensive resources. Her theory confirms our intuitions about how we label objects in the world, create conceptual categories, and navigate our symbolic landscape, and reminds us of the power a prototype can have in definining a niche.