No abstract

Although the use of timeout from positive reinforcement is widespread and has been shown to be an effective punisher in clinical settings, the factors responsible for its effectiveness are not fully understood. The present experiment was designed to evaluate one of these factors, the reinforcement rate underway during periods of time-in. Rats' lever pressing was maintained on variable-interval schedules of food reinforcement in a multiple schedule with two components. In one component, no timeouts were delivered. In the other component, when a variable-ratio schedule was met, lever presses were followed by a 30-s timeout during which a tone sounded, the component light was extinguished, and the food schedule was suspended. Across conditions, the reinforcement rates in both components ranged from 0.5 to 6 pellets per min and the degree of response suppression was evaluated. In the leanest condition, 0.5 pellets per min, the timeouts failed to suppress responding. Generally, as the reinforcement rate was increased, the degree of suppression was enhanced systematically, showing that the punitive function of timeout is directly related to time-in reinforcement rate. The results are discussed in relation to laboratory research on timeout avoidance and shock punishment.

Disruptions in operant behavior in the form of extended pausing reliably occur when there is a discriminable transition from a relatively rich schedule of reinforcement to a relatively lean schedule (i.e., a rich-lean transition). The present experiments assessed the reinforcing and aversive functions of the stimuli correlated with the transitions between rich and lean schedules using observing and escape procedures. In both experiments, pigeons' key pecks were reinforced with food on a compound schedule with two FR components. Completing the FR in the rich component produced a large reinforcer; completing the FR in the lean component produced a small reinforcer. The components alternated semi-randomly to produce four types of transitions: lean-lean, lean-rich, rich-lean, and rich-rich. In the observing experiment (Experiment 1), pigeons responded on a mixed schedule, in which the food key color was not correlated with the transitions. Pecks on a second key, the observing key, converted the mixed schedule to a multiple schedule by changing the food key color to a color correlated with the current transition. Across conditions, the consequences of the observing responses were manipulated by limiting the stimuli that could be produced. The stimuli correlated with the lean-rich and rich-rich transitions maintained observing, indicating that these stimuli were reinforcing. In contrast, the stimuli correlated with lean-lean and rich-lean transitions neither maintained nor punished observing, indicating that these stimuli were neither reinforcing nor aversive. In the escape experiment (Experiment 2), pigeons responded on a multiple schedule. Pecks on a second key, the escape key, darkened the food key, suspended the FR schedule, and changed the stimuli in the chamber. Across conditions, the magnitude of the stimulus changes that occurred following responses on the escape key was manipulated. The magnitude of the stimulus changes had no systematic effect for most of the pigeons; however, escape was more likely to occur during the rich-lean and leanlean transitions than during the lean-rich and rich-rich transitions, indicating that the lean stimuli were aversive. Overall, results from the present experiments indicate that stimuli correlated with lean schedules of reinforcement can acquire aversive functions, but the detection of these functions may depend on the experimental procedure.