I INTRODUCTION
A QUICK glance at the chapter headings in these volumes suggests a greater emphasis on âneuro-â than upon â-psychology.â There is no doubt that the prospective experimenter will welcome the chapters of detailed instructions in the manipulation of the independent variable, involving as it does a relatively inaccessible and somewhat mysterious mass of tissue and drawing upon several disciplines for its specialized techniques. By contrast, the reliable recording of the consequent behavior changes usually seems pretty straightforward and routine, demanding little more than a modicum of common sense, an equipment catalogue, and a modest grant. At the gross level this may be adequate. It is sometimes obvious that the brain-operated animal eats nothing, that his bar-pressing for water is insufficient to keep him alive, that his learning is retarded, or that he is incapable of coordinated locomotion on a flat surface.
But as extirpation-by-teaspoon is replaced by functional ablation, stereotaxic localization, and microdissection, it is also appropriate to advance to a more fine-grained analysis of the dependent variable. This chapter, as well as Chapter 10 in Volume I of this series and Chapters 2, 3 and 4 in this volume, is predicated on the belief that electro/neuro/physiological sophistication may be wasted unless the subtleties on the measurement end of the enterprise are to a degree recognized.
Long a favorite behavioral target of the psychopharmacologist, general activity has in recent years been increasingly adopted as an assay tool by neuropsychologists. A change in activity level can indeed be a sensitive indicant of an altered internal state (motivation?), but its maximal usefulness requires attention to a number of potentially distorting factors. Failure to take these into account has led in many instances to unnecessary variability of the data, and on occasion, to conclusions that seemingly contradict each other. The identification of some of these pitfalls will be the major goal of the pages that follow.
In terms of its operational definition, and I question the present utility of any other, âgeneral activityâ must be regarded as multiple rather than unitary, and the first lesson to be heeded is that the differently measured general activities may diverge under identical biological conditions. Thus, the amount of the ratâs wheel running during the third day of continuous water deprivation is usually several times the ad lib. base-line (Finger and Reid, 1952), but measurement in a stabilimeter may yield no significant change, or even a decreased count (Campbell, 1964). The 4- or 5-day estrous cycle of activity is much more clearly revealed by wheel recording than by photocell recording or direct observation of the female ratâs movements about the home cage (Finger, 1961, 1969). Telencephalic lesions differentially affect activity in wheel and stabilimeter (Campbell and Lynch, 1969), and there is evidence that the behaviors reflected in the contrasting scores are mediated by different pathways (Lynch, 1970).
Clearly it is inappropriate, in enumerating the consequences of a neurological manipulation, to refer simply to a change or lack of change in general activity, without specifying the method of measurement. The specification of the situation must extend to the physiological and environmental conditions, for complex interactions seem to be the rule rather than the exception. To illustrate: hunger greatly accentuates the difference between frontal rats and controls in the stabilimeter, but not in the wheel (Campbell and Lynch, 1969), and deep frontal lesions elevate stabilimeter scores much more in the light than in the dark, but only during the first few postoperative days (Harrell and Isaac, 1969). As one examines in detail the major types of activity measurement, the suspicion is born that any neurological intervention can be shown to produce a change under some combination of circumstances. A dubious dividend is the limitation which such rich diversity imposes upon interpretation.
II METHODS OF MEASUREMENT
Any listing of methods that attempts to be both inclusive and restrictive would be an exercise in arbitrariness. It might be argued that the common label should be applied only to those procedures that yield similar results under comparable conditions, but at this point, the data required for such a classification are still fragmentary. In deciding which methods to describe, I have simply accepted the investigatorâs statement that what he is measuring can legitimately be termed general activity.
As a guide for the experimenterâs minimal coverage of the possibilities, it may tentatively be hypothesized that the revolving wheel or drum as normally used is in a category by itself, with all the other techniques in a second group. But even this prescription, it must be acknowledged, cannot yet be generalized beyond the rat. Further, there i...