A set of concepts practices findings and beliefs that dominates a discipline and affects its activity structure and progress

Paradeigma (Greek for pattern) was used by Plato in various meanings of 'example', including an ideal instance of a concept, a model and a standard (e.g. Statesman 278bsq.; Guthrie 1975). In the scientific literature, 'paradigm' is occasionally used to denote a standard concept, method, or procedure (definition 3, e.g. 'the "classical conditioning paradigm', 'the "conditioned taste aversion paradigm', etc.). It was, however, Kuhn (1962) who endowed 'paradigm' with its broader meaning and central position in the philosophy, sociology, and history of science (definition 4).

Kuhn treated the culture of science from a historical perspective. He differentiated five stages in the ontogeny of a scientific discipline: immaturity, normal science, crisis, revolution, and resolution. 'Immaturity' is characterized by fact gathering in the absence of an accepted conceptual and practical framework. 'Normal science' is 'research firmly based upon one or more past scientific achievements... that some particular scientific community acknowledges for a time as supplying the foundation for its further practice' (Kuhn 1962).

These achievements form 'paradigms' (definition 4). Paradigms are constructs of thought (shared concepts and beliefs), methodology (standard methods and practices), and sociology (they underlie the behavioural code and esprit des corpes of the discipline, as reflected in professional societies, journals, and meetings; "culture). Achievements that give rise to paradigms delineate the workings of the discipline, yet are sufficiently open ended to leave problems for resolution in the future. Crisis arises when anomalies pop-out in the worldview that is advocated by the paradigm, and success in solving problems and advancing the field slows down. The crisis ultimately leads to a revolution, in which a new paradigm starts to emerge. The new and old paradigms, according to Kuhn, are incommensurable, and the revolution is characterized by a struggle between the disciples of each. Finally, the struggle is resolved with the new paradigm prevailing—till the next cycle.

Kuhn's analysis was very influential, itself approaching the status of a paradigm in the philosophy and history of science. Typical of Kuhnian paradigms, it encountered crisis and gave way to new accounts of scientific practice (Hacking 1981; Bechtel 1988), in which 'paradigms' were replaced by other concepts, such as 'research programmes' (Lakatos 1978) or 'traditions' (Laudan 1977). For the purpose of the present discussion, 'programmes' and 'traditions' could be described as families of theories with some shared characteristics, which, in contrast to Kuhnian paradigms, coexist to various extent throughout the ontogenesis of the discipline. Hence, in the philosophy of science, Kuhnian paradigms and revolutions are not so trendy anymore, but still, they retain some usefulness in contemplating the state of the art and the progress of a given discipline.

It is only fair to note that even if Kuhn's account of the ontogeny of scientific disciplines were valid, the question would still remain whether the science of learning and memory has attained the status of'normal science'. The answer is not simple. Kuhn, similarly to many philosophers and historians of science, drew mostly from the analysis of physics, in which theories and paradigms are easier to delineate. Research on learning and memory combines multiple disciplines, ranging from psychology via neurobiology to computational science. In each of these disciplines, the level of 'normality' (in the Kuhnian sense) is different. Nevertheless, for the sake of argument, analysis of the field of learning and memory with Kuhnian tools is of substantial interest. It sharpens attitudes, illuminates the source of current trends, and provides perspectives on popular scientific as well as social practices.

So which are the paradigms of memory research? The discipline at large draws on some paradigms that are shared by other disciplines in the natural sciences. Because these paradigms are at a higher levels of "generalization, we can consider them as 'metaparadigms'. Here are examples:

1. *Reductionism. Well, this is surely a successful paradigm. The secret of using it smartly is to focus on those "levels of analysis that fit the discipline. This is not at all trivial in memory research ("internal representation, "learning; Dudai 1992).

2. Modularity. This paradigm considers biological systems as composed of modules, or elementary systems, either cross or within levels. Modularity serves reductionism. At the same time, it could lure us away from considering the system as an integrated whole. For the fingerprints of the modularity paradigm in the cognitive and the brain sciences, see "engram, "functional neuroimaging, "phrenology; also the critique in Fuster (2000a).

3. The Panglossian paradigm.1 This paradigm posits that natural selection is an optimising agent, and therefore biological systems are neatly adapted to perform particular tasks (Gould and Lewontin 1979; Dennett 1983). The Panglossian paradigm guides us to search for an adaptationist explanation whenever we encounter an incomprehensible system. The truth might be that the system is as is because of built-in, accumulating internal constraints, irrespective of the assumed adaptation; or that the ongoing evolution of the system is still far from the optimum.

In addition to the above 'metaparadigms', other paradigms are more specific to memory research. In the history of psychology, several schools have approached the status of the dominant paradigm at their time (e.g. Weimer and Palermo 1973; Gholson and Barker 1985). Most notable were introspective psychology, that has characterized the emergence of the 'new psychology' in Germany a century ago (Boring 1950a), and the subsequent reaction to it, *behaviourism. In current neuroscience, a few paradigms fuel the "zeitgeist. A major example is provided by the prevailing concept that long-term memory is embodied in synaptic remodelling and growth. This conceptual framework has definitely reached the status of a Kuhnian paradigm, influencing so much of the current research in molecular and cellular neuro-biology (see "consolidation, "CREB, "development, "immediate early genes, "late response genes, "protein synthesis). The classification of memory systems into "declarative and nondeclarative ("taxonomy) is another contemporary paradigm. And the assumption that Hong-term potentiation (LTP) is memory is yet another example, still a mini-paradigm, but ambitiously aiming higher.

Given the current paradigms of memory research, and adhering for the sake of argument to the Kuhnian terminology—will revolutions ensue? As noted above, when facts are difficult to explain in terms of the existing conceptual framework, be it a paradigm or a research programme or a tradition, they herald new concepts and ultimately change worldviews (e.g. Lightman and Gingerich 1991). It would be naive to assume that statements such as memory = growth, or LTP = memory, will not succumb to anomalies. Behaviourism did, to the complexity of language (Chomsky 1959) and of learning (Dickinson 1980). The exclusive-ness of instantaneous associations in "associative learning, if it was ever substantiated in the data, was diminished by the documentation of effective long-delay associations in nature (Garcia 1981). Even the hegemony of laboratory experiments, whose practitioners came to worship reaction times, avoidance boxes, and "mazes, was shattered by the complexity of "real-life memory (Neisser 1978). Interestingly, in all the above cases, the prevailing paradigm did not disappear, but rather lost some of its inflated status, while retaining its usefulness side by side with newer paradigms. The piecemeal evolutionary accounts of Lakatos (1978) and Laudan (1977) seem thus to be more realistic than the Kuhnian sharp survival of the fittest. This means that on the one hand, even paradigms that currently seem amazingly robust are not immortal, but on the other, their spirit will survive in their progeny.

Selected associations: Development, Bias, Culture, Reduction, Zeitgeist

1Dr Pangloss ('entirely-language'), the teacher of Candide (Voltaire 1759), was the ultimate incurable optimist. He justified every disaster in the world by the *a priori assumption that everything under the sun is for good cause. Pangloss was hanged, which, judged by his own philosophy, couldn't be but good.

0 0

Post a comment