Pioneering work in the 1940s and 1950s suggested that the concept of 'chunk
ing' might be important in many processes of perception, learning and cogni
tion in humans and animals. We summarize here the major sources of evidence
for chunking mechanisms, and consider how such mechanisms have been implem
ented in computational models of the learning process. We distinguish two f
orms of chunking: the first deliberate, under strategic control, and goal-o
riented; the second automatic, continuous, and linked to perceptual process
es. Recent work with discrimination-network computational models of long- a
nd short-term memory (EPAM/CHREST) has produced a diverse range of applicat
ions of perceptual chunking. We focus on recent successes in verbal teaming
, expert memory, language acquisition and learning multiple representations
, to illustrate the implementation and use of chunking mechanisms within co
ntemporary models of human learning.