ABSTRACT
The aim of this chapter is to propose connectionist models as explanations for consumer behaviour—focusing on the feedforward artificial neural-network model—and to explore a consequent expansion of the theoretical framework of the Behavioural Perspective Model using connectionist constructs. This work originated in a research project where over 2,000 neural-network models of varying complexity (altering the number of hidden layer nodes) were examined for their ability to predict consumer behaviour. A comparison was made with the more traditional logistic regression (which itself turns out to be equivalent to a neural network with no hidden layer).
The findings show that neural networks are particularly useful in studying consumer behaviour and consistently demonstrate a performance superior to that of the traditionally employed consumer behaviour analysis method. Utilitarian and informational reinforcement variables are seen to make a noticeable contribution to explaining consumer choice. The main conclusion to be drawn from this work is that connectionist models show potential for predicting and explaining consumer behaviour. This should encourage us to explore the integration of connectionism into consumer behaviour analysis within the theoretical framework of the Behavioural Perspective Model using the neural-network model as an exemplar.
