In lieu of an abstract, here is a brief excerpt of the content:

EVOKED BRAIN POTENTIALS: SIGNS OR CODES? WILLIAM R. UTTAL* Introduction In the last few years a number ofreports relating evoked cortical activity and various psychological response parameters have appeared. Because of the especially significant nature ofthese findings, it is particularly important that we consider carefully some ofthe philosophic and technical inferences which might be drawn on the basis of these initial results. The purpose of this paper is to look at some of these recent reports and to discuss some of the problems and difficulties which arise when electrophysiological recordings from the brain are compared to subjective results. The evoked-potential techniques provide a route of attack on one of the most profound questions ever asked by philosopher-scientists: that of the relation between discriminative behavior ("mind" in the olderjargon) and information state ofthe neuronal matrix ("brain" in the olderjargon). These techniques, refined by means of the averaging process for use on the normal, healthy, and intact human subject, are an almost unique approach to the problem ofthe physiological basis ofbehavior. It is particularly important, then, that we avoid certain misconceptions, some of which are less than obvious, as we begin to accumulate an empirical foundation for the notion of physiological reductionism, which heretofore has been more properly described as an article offaith rather than as a theorem. Progress toward such a physiological analysis is hindered by our current inadequate definitions of a "correlate" or a "basis" of behavior. Never- * Department ofPsychology and Mental Health Research Institute, University ofMichigan, Ann Arbor. Some of my research reported here was supported by National Institutes of Health grant MH-8786-02 and by National Science Foundation grant GB-2000. This paper was written while I was an NIH special postdoctoral fellow at the Kyoto (Japan) Prefectoral University of Medicine. I would like to express my appreciation for the cordial hospitality ofProfessor Hisato Yoshimura, head of the First Department of Physiology. I am grateful to my colleague, Dr. David Krantz, for his careful reading ofand provocative commentary on a manuscript ofthis paper. 627 theless, the current philosophical orientation in our consideration ofthese electrical potentials and their relation to behavior is predominantly mechanistic . Behavior, particularly those components which represent states ofawareness (the processes we usually call "sensations" or "perceptions"), is considered to be a direct reflection ofthe state ofthe underlying neurophysiological mechanism. Workers in this field are trying to equate concepts from an older psychological vocabulary with specific patterns of electrical activity within and among neurons in the highest parts of the central nervous system. Some recent suggestions have been made concerning the functional characteristics ofboth glial and macromolecular components ofthe same tissues, but there is still an enormous gulfbetween these speculations and data on behavior, particularly in the dimensions of the temporal scales describing each of the processes. But the time scale of many behavioral phenomena is roughly ofthe same order ofmagnitude as that ofevoked brain potentials, and evoked potentials most probably are neurogenic in origin. Therefore, there is a tendency to draw conclusions a little more freely than we might if the correlated activity came from the liver, for example. At the present stage in the development of this new psychobiology, certain precautions are appropriate. In this paper, we emphasize one such precaution dealing with the nature of our interpretations ofthe relationship of the electrophysiological signal to the correlated behavior. We distinguish between those neuroelectric fluctuations that are true information carrying codes and those that are merely signs of stimulus states, but are not actually used by the nervous system in defining behavioral states. Codes and Signs In general, a code may be defined as a set ofsymbols and transformation rules which allows an economical representation ofa corpus of information , in a way amenable to decoding by some interpretive mechanism. For the special problem we are discussing here, it is also necessary that the code actually be interpreted by some subsequent mechanism. A representation which is not so interpreted, but is lost at some more central level ofinformation processing, is, in this context, a sign. A sign may be useful to an external observer who may decode it or measure its prop628 William R. Uttal ยท Evoked Brain Potentials Perspectives in Biology and Medicine...

pdf

Share