412Observing behavior can be defined as behavior that produces stimuli that signal whether reinforcement is available without directly affecting the overall probability of reinforcement. For example, Wyckoff (1952) reported that pigeons would step on a treadle, the sole function of which was to produce a discriminative stimulus that indicated what schedule was currently in effect on a response key. In other words, observing responses changed the schedule of reinforcement from a mixed to a multiple schedule. Similarly, pigeons prefer an alternative that produces discriminative stimuli (e.g., a red stimulus associated with 100% reinforcement on some trials or a green stimulus associated with extinction on other trials) over an alternative associated with an equal probability of reinforcement that produces stimuli (e.g., yellow on some trials, blue on others), both of which are associated with 50% reinforcement (Roper & Zentall, 1999; see also Dinsmoor, 1983).One could conclude that the pigeons were showing a preference for the information provided by the discriminative stimuli. According to information theory, the amount of information transmitted is a function of the degree to which uncertainty is reduced. When, prior to the initial choice, the outcome is totally uncertain (50% reinforcement) and, following the choice, the probability of reinforcement either increases to 100% or decreases to 0%, uncertainty reduction should be maximal. Any increase or decrease in the overall probability of reinforcement should reduce the amount of information transmitted, because it would result in a decrease in uncertainty reduction. Consistent with information theory, Roper and Zentall (1999) showed that increasing the overall probability of reinforcement associated with the two alternatives decreases the preference for the alternative associated with the discriminative stimuli. On the other hand, contrary to information theory, they found that lowering the overall probability of reinforcement associated with both of the alternatives, which should also decrease the preference for the alternative that is followed by the discriminative stimuli, actually increases it.According to the law of least effort, given equal probabilities of reinforcement, organisms should prefer alternatives that require less effort over those that require more effort. However, Roper and Zentall (1999) found that under conditions of equal reinforcement, pigeons were willing to expend considerably greater effort to obtain the discriminative stimuli (up to 16 pecks) than the nondiscriminative stimuli, for which only 1 peck was required. Given the strong preference for discriminative stimuli found by Roper and Zentall, we were prompted to ask whether pigeons would prefer an alternative that provides discriminative stimuli, even if there was a substantial cost involved in the form of loss of reinforcement.According to the law of effect, given appropriate experience, organisms are presumed to favor alternatives that provide higher probabilities of reinforceme...