Stimulus delay and the reduction of errors in the transfer of stimulus control

Abstract

Pigeons initially trained on a simultaneous discrimination of line orientation (81) were subsequently transferred to a wavelength discrimination (82). Three transfer procedures were employed The abrupt-tra~fer 8s were "abruptly" switched from 81 to the S2 dimension. The stimulus-compo~nding Ss were tram~d on a compound st~ulus consisting of Sl and 82 displayed in superimposition prior to the prese~tatlOn of S2 alone. The stimulus-delay Ss were trained on a compound stimulus in which the presentation of the 81 component was delayed for successively longer intervals as a result of a correct response . to the preceding trial. Stimulus-delay Ss transferred by responding to S2 prior to the pr.esentatIon of~1 and the resulting formation of the compound. Ss transferred by the stlmulus~com~oundmgan:d the abrupt-transfer procedures displayed 5 and 10 times as many errors to the 82 dimension, respectively, as Ss receiving the stimulus-delay procedure.An S forms a discrimination by learning to emit responses in the presence of one stimulus and by withholding responses in the presence of a second stimulus. Terrace (1963b) has shown that Ss may transfer from one discrimination to another without making errors. The absence of errors resulted from training pigeons to a compound stimulus consisting of an established wavelength discrimination displayed in superimposition with the to-be-transferred-to linetilt discrimination and gradually "fading out" the wavelength cue by reducing the intensity of that dimension.Reynolds (1961) has demonstrated that either dimension of a compound-stimulus display may come to exert stimulus control. Thus, when more than one stimulus dimension is correlated with reinforcement it is unclear which element of the stimulus display is, at any point in time, exerting stimulus control. Terrace's (1963b) transfer procedure employs a compound stimulus in which the relative intensity of the stimulus elements is systematically varied over several sessions. Considerable economy in the transfer program could be realized if a mechanism were available which would indicate that point in the transfer program by which the second dimension in the stimulus display has acquired stimulus control.Several procedures have been generated which minimize that occurrence of errors in the transfer of stimulus control from one dimension to another. These

Cross-modal transfer of stimulus control in the albino rat: A stimulus delay procedure

“…The results of the present experiment extend those of Brown and Rilling (1975) and Touchette (1971) to a new species and to stimuli in different modalities. In addition, the present results demonstrate that a gradually increasing stimulus delay is not essential to the effectiveness of the procedure.…”

“…A betweensubjects design was employed. Secondly, the present study extended the results of Brown and Rilling (1975) and Touchette (1971) by demonstrating that the stimulus delay procedure was effective in trans-341 ':erring stimulus control of a rat's leverpress across stimulus modalities.…”

“…If an incorrect response occurred on any trial, the duration of the delay was decreased by Yz sec on the subsequent trial. The use of such a gradually increasing delay not only resulted in the errorless transfer of stimulus control, but also allowed the specification of the exact point in time at which transfer occurred.The results of Touchette have been extended to animal subjects by Brown and Rilling (1975). Using pigeons and a gradually increasing delay procedure, Brown and Rilling were able to transfer stimulus control from a horizontal-vertical discrimination to a red-green discrimination.…”

“…The results of Touchette have been extended to animal subjects by Brown and Rilling (1975). Using pigeons and a gradually increasing delay procedure, Brown and Rilling were able to transfer stimulus control from a horizontal-vertical discrimination to a red-green discrimination.…”

“…It is important to note that, even in situations where stimulus control is well established, the disruptive effects of introducing a novel stimulus are considerable (Brown & Rilling, 1975). To eliminate such effects, the delay between the onset of the new and original discriminative stimulus was introduced only after several sessions during which the stimuli were simultaneously presented.…”