Psychology ETDs

Publication Date

9-10-1970

Abstract

The purpose of the present simple reaction time (RT) study was to evaluate the application of the variable criterion model to the effects of false feedback and stimulus intensity. The variable criterion model is a decision-theory approach relating stimulus intensity and response evocation. According to the model, stimulus intensity determines the rate of neural events or "impulses" originating from stimulus onset, whereas contextual and motivational variables determine the S's decision criterion, or the number of impulses that must accumulate before a response is made. The time required to reach the criterion value is the reaction time. Reaction time is thus an increasing function of the criterion level and a decreasin1g function of stimulus intensity. Any experimental manipulation that raises or lowers the criterion should have a corresponding effect on RT.

A new class of experimental manipulation, false feedback, was investigated. The basic procedure was to supply Ss, by means of post-trial feedback, with false but credible information about their RT performance which was presumed to alter the location of their decision criterion. Three separate groups, 20 Ss each, responded to irregular presentations of white noise of 40-, 70-, and 100-db. (SPL) loudness level for 127 trials. All trials consisted of a visual ready signal (0.5 sec, duration), irregular foreperiods of 1.0, 1.5, and 2.0 sec., an auditory response signal, and an ITI of 10,0 sec. All Ss were undergraduate women. One feedback group was led to believe it was performing fast in comparison with a previously established norm, the other group slow. The third group was the control receiving no feedback. The main finding was that presentation of either fast or slow feedback resulted in decreased mean RTs and attenuated within-S stimulus intensity effects that persisted through a second day as well, The variable criterion model, when applied to a plot of the data, rendered a very good description and was congruent with the assumption that Ss seemingly lower their response criterion as a result of feedback.

To investigate the more immediate effects of feedback on RT performance, two additional groups, 15 Ss each, were run: the first group received slow feedback only during the first half of the RT series whereas the second group received slow feedback only during the second half of the series. An unusual amount of between-group variability, unfortunately, obscured any attempt to examine changes in RT performance as a result of the feedback shift.

A final factor of concern was the investigation of stimulus sequence effects. To test the "neural noise" hypothesis that effects of Trial n-1 stimulus intensity should decrease as the length of the ITI is increased, two other ITI groups (in addition to the 10.0 sec. ITI control group), 15 Is each, of 6.0 sec. and 15.0 sec. were included. Reaction time was found to be an increasing function of the intensity of the preceding stimulus. However, the occurrence of sequence effects was not dependent upon ITI length, as suggested by the "neural noise" hypothesis. Stimulus sequence effects were most readily attributed to trial-to-trial compensatory variations in the S's response criterion.

Degree Name

Psychology

Level of Degree

Doctoral

Department Name

Psychology

First Committee Member (Chair)

G. Robert Grice

Second Committee Member

Henry Carleton Ellis

Third Committee Member

Frank Anderson Logan

Fourth Committee Member

Karl Peter Koenig

Fifth Committee Member

Douglas Peter Ferraro

Language

English

Document Type

Dissertation

Comments

PHS Grant MH 16400 from National Institute of Mental Health

Download

Included in

Psychology Commons

Share

COinS