Relationship between Physiological and Self-Reported Stressful Response and Psychosomatic Disorders 1

Method

The original questionnaire was administered to 557 introductory psychology students. The revised one was administered to an additional 359 introductory psychology students.

The original test, modeled after the one used by Mandler, Mandler, and Uviller (1958), contained 26 items which the participant related on a seven-point scale to the degree anxiety affected his cardiovascular activity, respiration, muscle tension, perspiration, gastrointestinal activity, and general somatic conditions.

Results

Responses were coded into subtests containing only questions concerning one physiological system.

In the original questionnaire, those subtests on which internal consistency fell below .60 were dropped from this test form. Questions concerning somatic conditions during relaxation were eliminated.

The revised form contained questions regarding cardiovascular activity, respiration, muscle tension, perspiration, and gastrointestinal activity. Test-retest reliability over 3 mo. for this version was .72. Internal consistency for each subtest was above .60.

Method

From the same group of participants administered the revised test, the highest and lowest scoring 12.5% for each sex were selected for the experiment. Attrition reduced this group to 41 males (high 20; low 21) and 42 females (high 21; low 21).

The experiment was conducted in two adjoining soundproof, electrically shielded chambers. All physiological recordings were made on a Grass Model 7 polygraph. Standard preamplifiers and amplifiers were used except as otherwise described. Heart rate was recorded using silver-plated electrodes attached to the left wrist and leg; respiration was recorded by a thermistor placed slightly inside the nostril permitting the most air flow; integrated muscle tension was recorded from the frontalis muscle using standard placement with 1-cm. silver-silver chloride electrodes. A Consol BSR/GSR preamplifier providing a constant current of 16 A was used to record skin resistance. Electrodes similar to those used to record muscle tension were attached to the volar surface of the left palm and the upper left quadrant of the forearm. The electrolyte was a 50% saline solution in Unibase. Each participant was seated in a comfortable recliner chair.

Slide presentation via a Kodak Carousel projector was controlled by Grason-Stadler electromechanical programming equipment interfaced with the polygraph so as to provide a signal mark upon onset and offset of each slide.

Participants were individually tested by an experimenter of the same sex who was blind to the experimental condition. After attaching electrodes, a 15-min. adaptation period followed, with the last minute serving as the preassessment of tonic physiological levels. Participants were then instructed to make no unnecessary movements and to watch each slide closely. Each slide was presented for 15 sec. with a variable interslide interval (M = 45 sec.; range of 35 to 60 sec.). Four slides of geometric wooden block designs were followed by four stressful slides of bodily injury or burn treatment. The stressful slides had been subjectively rated previously by students scoring in the middle two quartiles on the revised test as those most disturbing from a larger pool of such stimuli.

Quantification of all dependent variables was performed by paid undergraduate assistants unfamiliar with the purpose of the investigation. Interscorer reliability exceeded .99 on all measures. Timing periods for scoring the physiological variables were the 15 sec. prior to the onset of the slide (for baseline) and the 15 sec. during slide presentation (for response). For each physiological variable, response to each slide was defined as a difference score (change was response minus baseline).

Electrodermal response to each slide was quantified by determining the least basal skin resistance during both timing periods. Each of these values in ohms resistance was then converted by a reciprocal transformation to micromhos conductance. The change in skin conductance level was determined by the difference score computation. Muscle tension was determined by measuring the maximum amplitude of the integrated response in millimeter deflection from chart baseline and converting to microvolts for each timing period. Respiration rate was determined by counting the frequency of completed respiratory cycles for each timing period and converting to cycles per minute. Heart rate was determined by counting R-waves from the electrocardiogram recording and converting to beats per minute. Data analyses were performed on an IBM 360 computer.

Results

A sex X condition X trials analysis of variance was performed on responses to both neutral and stress slides. No significant main effects or interactions were recorded for skin conductance level, muscle tension, respiration rate, or heart rate. However, while no significant main effects or interaction were observed for skin conductance responses to the neutral slides, a significant main effect of condition (F_1,73 = 5.54, p < .02) was found for the stressful slides.

OPEN IN VIEWER

Table 1

Mean Skin Conductance Responses for Stressful Slides

Slide No.	Baseline		Response
	Low	High	Low	High
1	4.67	7.56	9.22	11.56
2	6.22	7.33	7.11	9.11
3	5.22	6.78	7.00	9.00
4	5.55	9.00	7.55	7.67
At	5.42	7.67	7.72	9.34
SD	5.56	6.63	5.75	5.16

Method

The same individuals who participated in the previous experiment were tested again with the same equipment as that described in the first experiment. Also, perception of autonomic reactivity was recorded on a seven-button scale connected to an Esterline Angus recorder. The procedure was similar to that in the first experiment except for the rating procedure. After each slide was presented, another slide followed which named one of the somatic systems found on the revised questionnaire. The participant was instructed to estimate how much of a reaction the previous scene caused in the named system and indicate his estimate on the rating scale (1, extremely small reaction to 7, extremely large reaction).

Approximately one week after this experiment, the participant was mailed a medical history questionnaire with a self-addressed envelope enclosed. Part 1 of the history dealt with the experience of 30 major physical disorders. The rating scale was anchored at both ends by “1— not a problem at all” to “7— a very severe, intense, or frequent problem.” Part 2 investigated the extent to which the participant had sought medical treatment for the problems listed in Part 1. The scale extended the range from “0— never sought medical advise” to “4— hospitalized for the problem.”

Data reduction was accomplished in the same manner as in the first experiment. Quantification procedures of the physiological variables are specified there. Accuracy scores were derived by normalizing the means of self-report scores for each mode divided by the normalized mean of the mode scores.

Results

Sex X condition X trial analyses of variance for self-report and accuracy scores for all modalities were performed. No significant main effects or interactions for any of these analyses were recorded.

Total scores on the medical history did not differ when a t test was performed. The only item which differed significantly (p < .05) between high and low scorers was allergy.