The objective study of behavior, which is the stated aim of our society, does not impose restrictions on the levels of explanatory constructs that are used. The only restriction is that the evidence concerning those constructs be stated in an objective or scientifically communicable way. Thus the concepts that we employ to explain behavior range from the sociologic to the biochemical. This article's underlying thesis is that behavior needs to be investigated at various levels, and that these levels should be clearly differentiated in order to bring these investigations into a scientifically meaningful relationship. The thesis is here illustrated by examining evidence and arguments concerning the utility and status of a noninvasive index of myocardial performance: T-wave amplitude (TWA). The examination begins at a psychophysiological-index level, wherein TWA is considered in terms of how well this noninvasive physiological index differentiates psychological processes. Secondly, at a lower physiological-index level, we consider the assumption that TWA reflects a relatively unitary physiological process, myocardial beta-adrenergic sympathetic influence. Both the grounds for and implications of this assumption are discussed. Finally, at the physiological-index level of discourse, brief reference is made to the mechanism by which changes in beta-adrenergic sympathetic innervation may produce correlated changes in TWA. The overall aim of the article is to differentiate these three levels of investigation, and yet also to consider the interrelationship among these three levels in order to provide a fuller scientific understanding of the phenomena involved.