In order to understand the differences in individual behavior across different contexts, this study introduces legitimate behavior and its deviation into a utility function. We hypothesize that people have preferences for adhering to the legitimate behavior that is required by the behavioral norm embedded in a particular context; furthermore, deviating from this legitimate behavior may generate a utility loss for them. We apply our model in the context of conditional contributions in a public goods experiment; moreover, we verify that the behavioral pattern of this conditional cooperation is derived from subjects' preferences for complying with the legitimate behavior required by the norm of the conditional cooperation activated in the experimental context. Furthermore, we attempt to measure the individual-level degrees of respect for the legitimate behavior in the given context using observable experimental data. The measurement results reveal that the subjects' relative sensitivities to deviations are highly centrally distributed; additionally, most subjects have a relatively high degree of respect for the legitimate behavior required by the conditional cooperation norm. Accordingly, this paper will help to improve our understanding of the micro mechanism underlying individual behavior.
Keywords: conditional cooperation; deviation; legitimate behavior; measurement; utility function.