A theory of sensory interaction: an experimental investigation of the relationship between autonomic activity and visual sensitivity
MetadataShow full item record
The chief objectives of this study were as follows: 1. To construct a theory of sensory interaction which will generate testable hypotheses concerning the relationship between autonomic activity and visual sensitivity. 2. To investigate experimentally the relationship between autonomic activity and visual sensitivity. 3 To demonstrate the predictive power of the theory. A review of the literature on intersensory sensitivity effects clearly indicates that the work in this area is not nearly so extensive as that in the investigation of single sensory effects. Furthermore, analysis of the work that has been done is complicated by a lack of standardization of experimental procedure. It seems, however, on the basis of what has been reported, fairly safe to assume that stimulation of one sensory system can affect performance in another sensory system; that the result of this interaction may be either facilitating or inhibiting; that the intensity of an auxiliary stimulus probably determines whether that stimulus will be facilitating or inhibiting in its effects. Also, it appears that interoceptive stimuli may operate in a similar fashion to exteroceptive stimuli in respect to intersensory phenomena. The theories proposed to explain intersensory effects have, in general, been based upon either neurological or gestalt models. None of the theories reviewed yields an explicit quantitative statement of change in sensory response as a function of an auxiliary stimulus. The theory presented in this study was intended to be applicable to all sensory systems. It was used in this study, however, to generate an experimental hypothesis relating specifically to vision and autonomic activity. The theory was presented in two designs. The first is a formal structural model. The second is a mathematical model in intervening variable form which was suggested by the structural model. The experimental hypothesis in this study was deduced from the set of mathematical postulates. In general terms, the theory states the relative magnitude and direction of change in sensitivity to a given stimulus(delta K1) as a function of the intensity of concurrent stimulation in another modality (S2). It is predicted that as S2 increases in intensity, the form of delta K1 will show an increase of sensitivity followed by a decrease of sensitivity. Two male college freshmen were used as subjects in this experiment. For each subject, GSR was recorded continuously and concurrently with a recording of visual threshold responses. GSR's were evoked by reading so-called neutral and complex words to the subject. Visual thresholds obtained during the thirty second period just before the reading of a word, and the thresholds obtained during the thirty seconds following the onset of the GSR were each averaged. The differences and the direction of the differences between these threshold averages represented the change in visual sensitivity associated with each given GSR. The obtained GSR values were converted into conductance units and in this form were considered to be measures of S2. When the obtained measures of change in visual sensitivity(D) were plotted against their associated GSR transformations in log conductance units(log GSR_cond), it was possible to test the following experimental hypothesis: When D is plotted against log GSR_cond it will be found that the form of D is fitted by the development of Ps1. In this statement, Ps1 represents a theoretical function which is based upon the intensity of log S2 and certain empirical constants. Applying an analysis of variance test for goodness of fit, it was found that the experimental hypothesis was supported by the results obtained from each of the subjects. When the distributions obtained from each subject were tested for a straight line fit, it was found that the null hypothesis was rejected in the case of Subject A, but could not be rejected in the case of Subject B. This latter case may be due to the fact that Subject B's GSR responses were more limited in range than Subject A's and corresponded to the flatter part of the theoretical curve. Consideration of the range of GSR obtained from each of the subjects indicates that heteromodal experiments specifically designed to test the upper limits of GSR would be pertinent for further validation and development of the theory. Such experiments would give more information about the relationship between S2 and delta K1. They would also indicate whether the potential range of GSR is essentially the same in all individuals; whether there are significant differences in potential range; or whether there is a typology in this respect. Related to the matter of potential GSR range is the problem of the empirical constants to be used in the theoretical formulae. It is suggested that considerably more data are needed before it can be decided whether a set of constants applicable to all individuals can be used, or whether it would be necessary to compute constants separately for each individual. The results of this experiment are, in general, consistent with the body of heteromodal studies. This experiment goes beyond the other reported studies, however, in that it demonstrates facilitation and inhibition within a single subject as a systematic function of S2. The results obtained suggest that a reported relationship between poor dark vision and anxiety might be explained by postulating an intensity of autonomically generated S2 sufficient to inhibit visual sensitivity. It is further suggested that GSR's will normally occur together with the independent variable in almost any heteromodal experiment, so that effects related to such GSR's must be controlled or partialled out if the heteromodal effects specific to the independent variable are to be measured. In relation to the concept of perceptual defense, two suggestions are offered. First, a more parsimonious explanation of the process might be achieved if the autonomic events supposed to be initiated by the taboo stimulus were considered to be the avoidance process rather than serving as cues for another avoidance process. This might occur if, as S2 events, they were sufficiently intense to inhibit visual sensitivity. Secondly, it is suggested that much of the conflict about the validity of perceptual defense might be resolved if the "tabooness" of a word were judged not by the nature of the word as such but by the magnitude of autonomic response it evoked in any given situation. Conclusions: 1. It appears that autonomic activity, measured in GSR conductance units, is associated with changes in sensitivity in visual threshold performance, and that these changes may be in the direction of facilitation or inhibition depending upon the intensity of the concurrent autonomic activity. 2. The postulate set presented in this study may validly predict visual sensitivity changes in heteromodal situations similar to that used in this experiment.
Thesis (Ph.D.)--Boston University