The contingent learning approach outlined in this paper is the first step in a larger research program. It aims to develop a theory that enriches the standard economic approach in a methodologically compatible way by introducing theoretical elements and evidence from other fields with respect to learning behavior in economically relevant situations.
The approach therefore includes determinants in the form of situational constraints that human learning can be assumed to be contingent upon. Hence, in addition and in analogy to the economic restrictions in traditional economic theory – and to the cognitive restrictions in approaches of bounded rationality – the CLA introduces learning restrictions that account for the interplay between the individual and the situation.
In other words, whereas standard economics focusses on a specific type of restrictions (i.e., on endowments, budget constraints, relative prices), and the bounded rationality approach focusses on cognitive restrictions, the CLA focusses on the interplay between situational and cognitive factors, thereby bridging part of the gap between traditional economics and psychology.
This additional perspective may be especially important if economics is understood as a behavioral social science whose strength lies – due to the prevailing methodological paradigm – in the explanation and prediction of changes in individual behavior as a reaction to changes in restrictions and relative prices. The development of such a theory is the first step. It is followed by testing the hypotheses that are deduced from theory.
Since the CLA hypotheses presented above are quite general and –some may argue– straightforward, it seems to be a good research strategy to follow Popper’s advice in trying to find examples where the hypotheses can be falsified, instead of finding supporting evidence, as many researches do in this field today. This enterprise is on its way and a first series of experiments has been conducted (SLEMBECK, 1998a, 1998b). The results are encouraging in that it appears to be difficult to falsify at least some of the hypotheses. It seems that some cases that appear to present counter-examples to the CLA hypotheses at first sight, show to be no such examples once overlapping effects of learning and other variables are untangled.
But it has to be admitted that it is difficult to create experiments that provide tests that are truly ceteris paribus. The most important and most challenging task, however, is the application of the theory to a wider range of economically relevant phenomena. Hence, the real test for the power of an economic theory enriched by contingent learning lies within its application, and may emerge in competition with other approaches that aim to explain the same phenomena.[73]
73 For a general comparison of the CLA to current theoretical and experimental approaches see the appendix.
Prof. Tilman Slembeck
Next: APPENDIX: A Comparison of Learning Approaches
Summary: Index