Abstract
Two studies assessed construct and concurrent validity of two scales designed to measure reinforcement sensitivity. Two non-probabilistic samples of college students from Mexico City were used. The first study showed that while the BIS/BAS (behavior inhibition and behavior activation scales) scales may possess construct validity, they lack in concurrent one; conversely, the SPSRQ (sensitivity to punishment and sensitivity to reinforcement questionnaire) did not show construct validity but showed evidence of concurrent one. Convergent validity was assessed using self-report questionnaires. The second study showed that the BAS scales may predict response distribution, in the IOWA-GT, in a way that is consistent with Reinforcement Sensitivity Theory (RST); this was not the case with the BIS scale. The results suggest that either the scales developed to assess RST, or the theory itself (or both) require a thorough revision.
