Psychological Laws

Psychology is a science. Sciences are supposed to feature laws, that is, laws of nature, generalizations that are strongly projectible past the actual data that confirm them. Yet in general, psychologists are reluctant to dub their generalizations "laws," even when they have great confidence in those generalizations; some are even uncomfortable in talking of psychological laws at all.

Over the decades, a few generalizations or regularities have explicitly been called laws, in GESTALT PSYCHOLOGY (the laws of organization), in the theory of conditioning (the law of effect, the law of exercise, Jost's law, the law of generalization), in neuropsychology (Bowditch's laws, Hebb's law), and of course in psychophysics (Weber, Fechner, Thurstone, Steven, Bloch, Ricco, Bunsen-Roscoe, Ferry-Porter, Grassmann, Yerkes-Dodson, Schachter). For the most part -- the law of effect being arguably an exception -- these are empirical generalizations.

Most such laws are equations relating one measurable magnitude to one or more other, independently measurable, magnitudes. Laws of this particular type are of course found throughout the natural and social sciences. But there are psychological generalizations of other kinds that are laws or lawlike as well, even if they are not commonly called by that name. For example, some of the empirical laws are thought to be explained by higher-level theoretical principles or hypotheses, which would themselves have to be considered lawlike in order to play that explanatory role. And surely there are qualitative rather than quantitative generalizations, truths of the form "Whenever organism S is in state A and X occurs, S goes into state B," that are lawlike.

Such qualitative laws would be derived from a standard explanatory pattern in psychology: the function-analytical explanation in the sense of Cummins (1983), as found in much of current cognitive theory, perceptual psychology, and PSYCHOLINGUISTICS. (The pattern is ubiquitous in biology, in computer science, and in electronics as well.) Some psychological questions take the form, "How does S ?," where " " designates some accomplishment carried out by the organism S (e.g., "How do speakers of English understand novel sentences?" or "How does an experimental subject estimate the distance in miles from her present location to the place she was born?" or "How do dogs recognize individual human smells?"). In answer to such a question, the theorist appeals to a functional or componential analysis of S. S's performance is explained as being the joint product of several constituent operations, individually less demanding, by components or subagencies of the subject acting in concert. The components' individual functions are specified first, and then the explanation details the ways in which they cooperate in producing the more sophisticated corporate explanandum activity.

For example, to explain language understanding, a psycholinguist posits a phonological segmenter, a parser, a SYNTAX, a LEXICON, and (notoriously) a store of real-world knowledge, and starts to tell a story of how those components interact. Of any functionary figuring in that story, we might want to ask in turn how it performs its own particular job. This is another psychological question of just the same form as the first, only this time about the functional organization of one of the subagencies. It is important to see that we can go on asking our function-analytical questions at considerable length. (What neural structures ultimately realize the lexicon?)

Thus we can see human beings and other psychological subjects as being simply functionally organized systems, corporate entities that have myriad behavioral capacities by virtue of their internal bureaucratic organizations. An organism's complete psychological description would consist of a flow chart depicting the organism's immediate subagencies and their routes of cooperative access to each other, followed by a set of lower-level flow charts -- "detail" maps or "blowups" -- for each of the main components, and so on down. At any given level, the flow charts show how the components depicted at that level cooperate to realize the capacities of the single agency whose functional analysis they corporately constitute.

Function-analytical laws of two kinds could be read off such diagrams. First there would be qualitative laws of the sort mentioned above; a given flow chart would show what a creature would do, given that it is in such-and-such a state and that (say) it received a stimulus of a certain sort. These "horizontal" laws would be of direct use in the prediction and control of behavior. Second, there would be "vertical" laws, relating lower-level states to the higher-level states that they constitute at a time. Of course, all such laws would be qualified by normalcy or "ceteris paribus" clauses because exceptions to them can be created by hardware breakdown or by perturbation of the system by some external agent.

Philosophers have raised several deeper, skeptical questions about putative laws that are more distinctive to psychology than to the natural sciences, though none of these questions is directed at the acknowledged empirical laws with which we began. The questions stem from the widely shared assumption that many psychological states are representational (see MENTAL REPRESENTATION). One might suggest that what distinguishes psychological laws from those of other natural sciences is that they concern representational states of organisms, but that would seem to rule out laws of conditioning.

Contemporary cognitive and perceptual psychology do traffic in representations, information-carrying states produced and computationally manipulated by psychological mechanisms. By its nature, a representation represents something. That is, it has a content or, as the medieval philosophers called it, an "intentional object." And, remarkably, that content or object need not exist in reality; for instance, through deceptive environment or malfunction, a visual edge detector may signal an edge that is not really there (see INTENTIONALITY).

A first skeptical question is this: Representation of a possibly inexistent object is a property not found in physics or chemistry -- and it is a relational property, ostensibly a relation that its subject bears to an external thing, determined in part by factors external to the organism. Yet natural laws are causal, and some philosophers (notably Fodor 1987 and Searle 1983) have argued that because an entity's causal powers are intrinsic to the entity itself, either representational properties cannot properly figure in psychological laws or there must be a kind of "narrow" representational content that is intrinsic to the subject and independent of environmental factors. (See INDIVIDUALISM. Against that view, see Burge 1986; McClamrock 1995; and Wilson 1995.)

A second set of questions concerns commonsense mental notions, such as those of believing, desiring, seeing, feeling pain, and the like (see PROPOSITIONAL ATTITUDES). Jerry Fodor (1975, 1981) argues that only slightly cleaned-up versions of those concepts have a home in scientific psychology and indeed will figure explicitly in laws; certainly some current psychological experiments make unabashed reference to subjects' beliefs, desires, memories, etc.

However, Donald Davidson (1970, 1974) and Daniel Dennett (1978, 1987) contend that such commonsense concepts correspond to no NATURAL KINDS, not even within psychology. This is in part because they are ascribed to subjects on grounds that are in large part normative rather than empirical, and in part because (Davidson alleges) the "ceteris paribus" clauses needed as qualifying such commonsense generalizations as "If a man wants to eat an acorn omelette, then he generally will if the opportunity exists and no other desire overrides" (1974: 45) are open-ended and unverifiable in a way that such clauses are not when they occur in biology or chemistry. For somewhat different reasons, Paul Churchland (1989) agrees that commonsense concepts correspond to nothing real in psychology or biology and that they will very probably be dropped from science altogether (see ELIMINATIVE MATERIALISM).

Additional links

-- William Lycan

References

Burge, T. (1986). Individualism and psychology. Philosophical Review 95:3-45.

Churchland, P. M. (1989). A Neurocomputational Perspective. Cambridge, MA: Bradford Books/MIT Press.

Cummins, R. (1983). The Nature of Psychological Explanation. Cambridge, MA: MIT Press/Bradford Books.

Davidson, D. (1970). Mental events. In L. Foster and J. W. Swanson, Eds., Experience and Theory. Amherst, MA: University of Massachusetts Press, pp. 79-101.

Davidson, D. (1974). Psychology as philosophy. In S. C. Brown, Ed., Philosophy of Psychology. London: Macmillan, pp. 41-52.

Dennett, D. C. (1978). Brainstorms. Montgomery, VT: Bradford Books.

Dennett, D. C. (1987). The Intentional Stance. Cambridge, MA: MIT Press.

Fodor, J. A. (1975). The Language of Thought. Hassocks, England: Harvester Press.

Fodor, J. A. (1981). RePresentations. Cambridge, MA: MIT Press.

Fodor, J. A. (1987). Psychosemantics. Cambridge, MA: MIT Press.

McClamrock, R. (1995). Existential Cognition. Chicago: University of Chicago Press.

Searle, J. (1983). Intentionality. Cambridge: Cambridge University Press.

Wilson, R. A. (1995). Cartesian Psychology and Physical Minds. Cambridge: Cambridge University Press.

Psychological Laws

See also

Additional links

References

Further Readings