The specificity or patterning of psychophysiological responses to behavioral stimuli is of importance for basic and applied psychophysiological research. However, the statistical tools used to describe specificity are heterogenious and the attempts to describe temporal stability or reliability of such response profiles are even more complicated. The aim of this paper is to present an approach designed to clarify the relationships between different measures of temporal stability of specificity. This is done with a compilation of statistical measures based on variance components. This approach incorporates features of response specificity and reliability. Individual, situational and motivational response specificity are defined in a uniform way with respect to temporal stability. The use of the formulae is demonstrated with an empirical example. As is found in the research on single parameters, the profile reliability of level scores is higher than that of change scores. Furthermore, the stability of individual response patterns is dependent upon the question of whether the data is aggregated across situations or not. Situational response patterns are stable when the data is aggregated across subjects, whereas the stability is low without such aggregation. Issues of inconsistencies of variance estimates are addressed, and the relationships between temporal stability, consistency and covariation of physiological variables are discussed.