Responses for eight ANS variables are reported for 30 male college students under four stimulus conditions in an investigation of individual response specificity and stereotypy. It was found that neither direction nor magnitude of reaction was significantly related to prestimulus level for any of the untransformed variables. Reactivity in this study was based, therefore, upon raw score changes, except for the customary transformation--"log palmar conductance change." Eight (27 per cent) of the subjects showed complete response specificity by response-level scoring (i.e., maximal activation occurred in the same variable under the four stimulus conditions). Only 2 (7 per cent) showed complete specificity by reactivity scoring, a result not significantly different from chance. A tendency toward response stereotypy (i.e., a stable hierarchy in autonomic response patterns to different stimuli) was found to occur under both methods of scoring, but to be much greater in responselevel scoring, whereby 22 subjects (73 per cent) demonstrated coefficients of concordance at or beyond the.01 level of confidence. For reactivity scoring only 4 (13 per cent) satisfied this criterion. Response specificity and stereotypy, when they occur, are interpreted as caused in part by method of measurement and in part by significant individual differences in resting ANS functions. A word of caution is included against overgeneralizing their significance.