The purpose of the present paper is to state what seem to be the important determinants of behavior in a testing situation and to examine their implicatione for the problem of prediction. It should be made very clear at the outset that there is nothing new in the statements of these determinants; nevertheless, they require listing and brief description beciuse the implications to be drawn from them seem to call for a revised approach to the problem of prediction. Attention will be given below to those determinants which are present regardless of the sex, age, and social class of the individual. In addition, there is no intention in the following discussion to convey the impression that the determinants are independent of each other; on the contrary, it is implicitly assumed that they are interdependent variables.
A. The nature of the stimulus materials. The content of the individual's responses, the accompanying expressive behavior, the needs and attitudind factors engendered are in part a function of the nature of the stimulus materials. How an individual will react to a thematic card cannot be determined from know- ing how he reacted to a comprehension item on the Binet.' Regardless of the continua (familiarity, physical attributes, shape, etc.) along which one may order stimulus materials, the behavior of the individual will in part be a function of thae point on a particular continuum that the stimulus material falls.
B. The nature of the instructions. An individual who is told that he may respond in whatever way he wants, or that he must complete a task within a certain time limit, or that he must reproduce a 1In the subsequent discussion words like bekvior, reactions, or responses refer not only to what an individual says or does but also to the ~eompmying needs and attitudes which may be overtly manifested or lnferred from other behavioral data.