Chunking Theory

Chunking it Together
(link to source)

Let's Define Chunking...

Several definitions have been proposed for the definition of chunking. Based off of De Groot's chess experiment, Chase and Simon (1973) began to clear up what is called the chunking theory, which has pretty much been the basis of expertise. A pretty, straightforward definition centers the idea of a chunk around long-term memory (LTM). In this respect, the chunk is defined as the grouping of information in a significant way as a single entity. This entity is encoded then as a perceptual unit of information.

As will be seen with comparison of chunking theory to template theory, both theories offer strong predictions about both short-term memory (STM) capacity as well as chunk size. The use of the chunk is not seen in the other two theories-long-term working memory theory and SEEK theory-as strongly as it is seen in these two theories. Chunking can be sub-divided into goal-oriented chunking and perceptual chunking. Goal-oriented chunking refers to the chunking that occurs under strategic control and is goal-oriented. Perceptual chunking is automatic and occurs during perception. This is the type of chunking that will be the focus of discussion and explanation regarding the chunking theory in relation to expertise.

A Little About the Computational Model

The chunking theory is closely related to the computational model known as EPAM (Elementary Perceiver and Memorizer) derived by Edward Feigenbaum and Herbert Simon. This model suggests that in order to become an expert, information is obtained by learning a sizeable database of chunks cataloged by a discrimination net. In the discrimination net, tests about the features of perceptual stimuli are carried out. Further information about this computational model is available by clicking on the above link.

Why Are Chunks Important?

Chunks are essential in the development of conditions of productions. The ability of experts to offer solutions so quickly, or base their ability as a basis of intuition is an example of the role of productions. Evidence for the use of productions is implied by the fact that experts of various disciplines use a forward search approach as opposed to a backward search approach-characterized by novices-when solving problems. The forward search approach suggests that experts use productions that are based on pattern recognition. Additionally, regardless of the capacity of STM, chunks explain why experts are able to remember larger amounts of information in such a short amount of time. The reason for this phenomena is that experts, unlike novices, do not store the information as separate units in their STM, but store the chunks that have been established in their LTM.

The final component of the chunking theory suggests something of the obvious: it takes a long time to learn the large number of chunks to become an expert. It is not a fast process; taking usually about 10 years to incorporate all those chunks, which is something like incorporating from 10,000 to 100,000 chunks!

Read on about the other theories to learn about what they postulate with regard to expertise in memory...

(Gobet & Clarkson, 2004),(Gobet et al., 2001), (Gobet, 1998)