Cognitive Mechanisms of Learning

9Symbolic Processing System Models in Cognitive Psychology

9.1. Why formalize?

Formalization is a widely accepted practice in the field of psychology. However, not everyone saw the introduction of symbolic processing system (SPS) models in the 1980s as a promising development in furthering understanding of psychological phenomena. In a collective work published in 1985 on modeling in psychology, we presented the following arguments:

“Experimental psychology is based on the premise of a certain intra-individual coherence…A processing model¹ is an interpretation of observed behaviors…the model explains observed behaviors using non-observable mechanisms [which produce them]. The key characteristics of this type of model are:

1. 1) Unlike structural models, which are essentially timeless, they take account of the dynamic character of mental activities;
2. 2) Unlike stochastic models, the psychological content of a processing model is in the model itself;
3. 3) Unlike data analysis models², based on aggregates of observed data…the mechanisms assumed in a processing model are inherently intra-individual…the operation [of the model] reproduces individual patterns of observable behaviors, on a qualitative level”.

In this chapter, we present four different works which address the question of formalization in the context of the SPS metaphor (sections 9.1, 9.2, 9.3 and 9.4). The first of these, by Niels Taatgen and Frank Lee, models the three stages of learning to solve a complex problem ...