A bstracr In a reconsideration of the achievement-motivation theory it is assumed that moderate stimulation releases positive affects and very low or high stimulation negative affects. It is further assumed that as fat as the achievement-motivationsystem is concerned the most stimulating situation is that where the probability of success (P,) is about S O , the least stimulating that where P, is near I . OO given impulsion to vigorous research. According to this theory the total motivation or tendency to undertake an activity is a summation of motivation components. The theory focuses upon the individual's motive to achieve success (M,) and motive to avoid failure (M,), which are assumed to be aroused in achievement situations. The motivation for an achievement task is then conceived as made up of motivation to achieve success, motivation to avoid failure and various extrinsic Resnick, 1965) to positive (e.g., Spence, 1956). As to the latter results, it should also be noted that the anxiety questionnaires are developed within research traditions which do not hypothesize a simple negative relationship between anxiety strength and performance level but rather that the relationship may vary from negative to positive? depending on the characteristics of the task or situation (cf. Mandler and Sarason, 1952; Taylor, 1956).Inconsistencies between empirical results or between theory and results have often forced investigators to re-examine their notions. This has been the case within the achievement-motivation tradition, too. Thus, in the face of unexpected results such as a higher performance level among subjects with a low M, than among those with a high M, one has, for example, questioned the assumption that individuals who score high and those who score low on the motive tests are equal in all performance-relevant respects except achievement motivation (Smith, 1966). Implicitly it has been suggested that superiority in 'other' personality characteristics or motives has compensated for the inferiority in achievement motivation and even resulted in higher performance among low M, subjects than among high M , subjects. However, while it is easy to approve the questioning of the assumption of all other things equal, the compensation viewpoint is harder to accept. On the whole, correlation rather than compensation seems to be the rule (e.g., Anastasi, 1958; Vernon, 1961). Therefore, contrary to what has been suggested within achievement-motivation research, the author finds it more likely that individuals who are inferior in achievement motivation are also inferior with respect to other performance-related characteristics, such as, for example, abilities. And even if the compensation viewpoint should turn out to be valid, it is not too obvious how it can explain the fact that sometimes those with a weak motive and sometimes those with a strong motive have performed at the highest level. This diversity in A reconsideration of the achievement-motivation theory 63 results can only be explained by assuming that the effects...