Academic journal article The Psychological Record

Varying Temporal Placement of a Stimulus Correlated with Non-Reinforcement in a Temporally Defined Schedule

Academic journal article The Psychological Record

Varying Temporal Placement of a Stimulus Correlated with Non-Reinforcement in a Temporally Defined Schedule

Article excerpt

The study of the functional effects of so-called neutral stimuli on operant behavior has dealt mainly with the analysis of steady-state performance under stimuli continuously present in each component of multiple schedules of reinforcement (Rilling, 1977; Terrace, 1966). However, some experimental studies have explored the varied effects that may be produced by discrete non-continuous stimulus presentation in interval and temporally defined schedules.

Morse and Skinner (1957) described what they called a "second type of superstition" as an example of "sensory" control of behavior. Pigeons were exposed to a non-contingent incidental stimulus presented during selected periods of a variable-interval schedule of food (VI 30 min), with daily sessions varying between 6 and 20 h in length. Inter-reinforcement intervals varied between 1 and 59 min. Performance under the VI schedule consisted of low response rates, with some local irregularities. An incidental blue light was projected on the response key instead of the original orange light during a 4-min period every hour in different temporal locations. This incidental stimulus was non-contingent both on the occurrence of responding and the occurrence of reinforcement. Two kinds of "sensory superstition" were observed: positive, when the rate of responding increased in the presence of the stimulus accidentally paired or close to reinforcement, and negative when a decrease in rate occurred in its presence associated with non-reinforcement.

Farmer and Schoenfeld (1966a) studied the effects of an added neutral stimulus presented in different positions of a fixed-interval schedule of reinforcement (FI 1 min). The added stimulus consisted of a change in the key illumination, independent of the pigeon's responding. In the first study, the change of light lasted 6 s and was presented during ten different temporal placements of the FI schedule. Depending upon the temporal placement of the stimulus change, response rates preceding, during, and following it were similar to those observed during discriminative cueing, conditioned reinforcement or chaining. In the second experiment, the stimulus change occurred twice during the FI interval, one fixed during the last 6 s of the interval, and another at varied placement along the interval, as in the first study. In this experiment, the absolute amount of responding during the stimulus change at varied temporal placements decreased systematically as its separation from the fixed placement increased. Similar results were obtained when a response-contingent 10-s stimulus was added at five different placements of an FI 1-min schedule of reinforcement (Farmer & Schoenfeld, 1966b).

Dews (1962) interrupted a bright house light during alternate succeeding 50-s periods in a FI 500-s schedule with pigeons. The interval ended with the house light absent as the [S.sup.D] for reinforcement delivery. Although the key light was on during the whole interval, responding was interrupted when the house light was present, which interrupted the sequence of responding typical of FI schedules. Nevertheless, the overall pattern was scalloped despite the exchange of patterns of no responding and responding during the presence and absence of the house light. However, the possible function of the absence of the house light as an [S.sup.D] might have been confounded with the salience of the key light during the periods of chamber darkness, although this was discarded by Dews (1965) by using the house light period as an [S.sup.D]. In a subsequent study (Dews, 1966), when the absence of the house light occurred only during two 50-s periods of the 500-s interval, the rate of responding during a segment of an uninterrupted fixed interval was dependent upon the temporal position of the segment in the interval. Responding was lower at the interval onset, and responding increased progressively as the probe was presented in successive locations of the interval closer to reinforcement, an effect that was called the "Cheshire phenomenon". …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.