Pausing and escape on fixed-interval schedules: Effects of stimuli correlated with different reinforcer magnitudes and different reinforcement rates.

Research has shown that both pausing and escape are controlled jointly by past and upcoming conditions of reinforcement. The present research was designed to investigate the contextual variables that influence pausing and escape on FI schedules and to assess further the functional similarities between pausing and escape. It was of particular interest that the effects of past events have been observed long after they had been removed from the environment of the organism. To shed light on the nature of control by such past events on pausing and escape, three experiments were arranged with fixed-interval schedules that manipulated stimuli correlated with past and upcoming reinforcer magnitude (Experiments 1& 2) and past and upcoming reinforcement rate (Experiment 3). The results were consistent with other research in that pausing was jointly controlled by past and upcoming schedule conditions, but escape behavior was not produced reliably. Although presenting stimuli correlated with past conditions had little effect on pausing or escape, when reinforcement rate was manipulated, local response rates were influenced by the past reinforcement rate and that effect increased when stimuli signaled past conditions. The effects of within-session manipulations of reinforcer magnitude and reinforcement rate on pausing and escape are discussed in terms of Gibbon\u27s (1977) scalar expectancy theory that predicts the ratio between local and overall reinforcement density must reach a threshold value before responding begins after reinforcement