Associationism and cognition: Human contingency learning at 25

Abstract

A major topic within human learning, the field of contingency judgement, began to emerge about 25 years ago following publication of an article on depressive realism by Alloy and Abramson (1979). Subsequently, associationism has been the dominant theoretical framework for understanding contingency learning but this has been challenged in recent years by an alternative cognitive or inferential approach. This article outlines the key conceptual differences between these approaches and summarizes some of the main methods that have been employed to distinguish between them.

Acknowledgments

I am very grateful to the United Kingdom Economic and Social Research Council and Biotechnology and Biological Sciences Research Council for their longstanding financial support. I thank Tom Beckers, Jan De Houwer, and Helena Matute for very helpful comments on this article. It is a particular pleasure to acknowledge colleagues with whom I have collaborated over many years, especially Tony Dickinson, Francisco López, Klaus Melchers, and José Perales.

Notes

¹ There had been some prior work on human contingency learning (Jenkins & Ward, 1965; Smedslund, 1963) but it was the connection with associative theory that made Alloy and Abramson's (1979) paper so significant (and hence the reference to 25 years in the title of this article).

² Many of the advocates of the associative approach to human contingency learning have been well known as animal learning researchers. Their concern to generalize their animal work to humans has been supplemented, perhaps, by an understandable desire to widen their research in a climate in which funding for purely behavioural work with animals has become harder and harder to obtain.

³ Of course, in casual language we might refer to a symbolic proposition such as “the light predicts shock” as an association, but I am explicitly avoiding such usage. In common with the Humean tradition, I am assuming that associations have the defining properties of automatically carrying the mind from one idea to another and of being semantically transparent or contentless (Fodor, 2003). Neither of these is true of propositions.

⁴ Associative processes, of course, play a substantial role even in information-processing theories of higher cognitive functions such as memory. To describe something as a “retrieval cue” for a “memory representation”, for instance, is to refer to an associative mechanism, and this is quite explicit in many theories of retrieval.

⁵ This dual-process view of the mind has gained many influential followers in recent years (e.g., Kahneman, 2003). The evidence for two systems is indeed very persuasive (Osman, 2004; Shanks & St. John, 1994). Where the evidence is, in my view, much weaker is in the claim that one of these systems is procedural, implicit, or unconscious.

⁶ There may appear to be a contradiction here. Earlier I argued that the use of verbal measures is important to ensure that we remain within the realm of explicit processes, whereas I am now arguing that we should focus on intuitive judgement as associative theory will not be able to challenge cognitive approaches in the domain of “rational” or “reflective” judgements. But intuitive judgements are invariably explicit rather than implicit (Tunney & Shanks, 2003). Intuitive judgements may be made without inference but they are still conscious and can usually be justified.