ABSTRACT
Over the past decade two additional techniques have become increasingly popular and important in the assessment of conditioned reinforcement: the study of observing responses, which has evaluated the reinforcing strength of stimuli that signal the availability of, or provide information about, primary reinforcement; and the study of choice for stimuli associated with schedules of primary reinforcement. This chapter discusses research with these techniques and the important implications of this research for the theory of conditioned reinforcement. It introduces the theories and distinguishing among them on conceptual grounds and in terms of empirical predictions. The uncertainty reduction hypothesis states that stimuli associated with negative outcomes should also be reinforcing; the conditioned reinforcement hypothesis requires that these stimuli be aversive—or at least not positively reinforcing. The pairing hypothesis, which is more parsimonious than either the delay reduction or uncertainty reduction hypotheses, even when formulated in terms of reinforcement density, has a rich history.
