[Page 1]

Autonomic Correlates of Inter-Action Process [*]

Jerome L. Berlin and Eugene T. Gendlin, University of Wisconsin

In psychotherapy, one of the most frequently discussed variables is the quality of the inter-personal relationship between client and therapist. This therapeutic quality of the relationship has often been viewed as a variable separate from the many topics and special events which occur only in psychotherapy. Rogers [6], especially, applies his basic theory of inter-personal relationships not only to psychotherapy, but equally to family, friendships, business, education, and posits the same therapeutic effects in all these contexts. He posits four "conditions" of an "improving" relationship: genuineness, empathic understanding, positive regard, and unconditionality of regard. When one person offers these at high levels, and is perceived as such by the other, the relationship will "improve" and the individuals will change therapeutically, becoming more open to their experiences, less defensive, more expressive, more fully themselves.

This theory of perception of general relationship "conditions" and resulting therapeutic change is carried further by the theory of experiencing [2] which concerns the moment to moment process occurring in the individual. In terms of this theory, the experiencing process in an individual is an aspect of a unitary psychophysiological organism. Whatever one person's perception of the other's general attitudes is, the quality of the relationship in terms of experiencing would be viewed as the differential qualities of concretely occurring process in each individual during inter-action.

From these theoretical considerations, the prediction arises that if, in an inter-personal relationship, one person's general perception of the other's

[Page 2]

attitudes shows a high level of Rogers' conditions, then the moment to moment psychophysiological process in the individual will be more eventful, more involving and tense at times, and more relaxing at times, than in inter-actions occurring in relationships perceived as low in conditions. Also, for the overall comfort level, we predicted overall lower tension levels during inter-action in a high "condition" relationship as compared with one of low condition levels.

Hypotheses:

IA. There will be higher peaks of tension during the inter-action in the high level relationship, than in the low level one.

1B. There will be higher degrees of relaxation during the inter-action in the high level relationship, than in the low level one.

2. Overall tension levels will be lower during the inter-action in the high level relationship, than in the low level one.

Procedure:

The subjects were 13 sorority girls, asked to choose one girl in the sorority with whom she had a "good and close" relationship and one with whom she had a "cold and distant" relationship. For each of these relationships, the subjects were asked to complete a 72 item questionnaire, the "Relationship Inventory," developed by Barrett-Lennard. Table 1 shows the differences between the two types of relationships chosen. Table 2 compares the condition levels in the positive relationships with Barrett-Lennard's previous condition level findings in successful and unsuccessful psychotherapy.

The 13 subjects (A person) were then (within one week) recorded on the polygraph during 20 minutes of inter-action with the girl chosen as the good relationship (B person), and separately, 20 minutes of inter-action with the girl [Page 3] chosen as the poor relationship (C person).

The procedure employs the same individual subject during inter-action in a high and low condition level relationship. Since the subject is her own standard for the comparison, the well-known stereotypic differences between individuals are controlled for.

The physiological indices for this experiment consisted of three variables: galvanic skin response, skin temperature, and heart rate. Constant measures of galvanic skin resistance were obtained through the use of a Fels Dermohmeter. The subject current was maintained at 70 microamperes and the Dermohmeter output was fed through a Grass Model 5A driver amplifier with the resulting readout being recorded on a Grass Inc. oscillograph. Measurement of skin temperature was accomplished through the use of a Yellow Springs tele-thermometer the range of which was set at 23 to 41° C. The output was fed into a Grass Model 5E1 chopper preamplifier and the final amplifier and readout stage was the same as for the galvanic skin resistance system. A Grass Model 5P4 cardiotachometer was employed to obtain a direct measure of heart rate.

The subject (A) was taken into the experimental room 15 minutes prior to the commencement of the experiment. In order to control for preceding ambient temperatures the right hand (site of the thermister) was placed in warm water (+33° C.) for five minutes. After patting the hand dry, the subject was seated in a cushioned chair and the electrodes were attached.

The experimental room was controlled for temperature and humidity at all times. The temperature was held at 73° F. ± 1 and the relative humidity range was 40% to 45%. Facing the subject was an identical chair for the B and C persons, who were treated and recorded identically as A subjects.

After the electrodes were properly placed the experimenter placed the instruments into operation. At the appointed time the B or C subject (with sensors already attached) was brought into the experimental room and seated in [Page 4] the appropriate chair. After 30 seconds the following instructions were given to the two persons:

"We would like you to engage in the kind of conversation that helps people to get to know each other as deeply as possible."

The instructions were repeated a second time whereupon the experimenter left the room.

Operationally, increase in tension was defined as a decrease in galvanic skin resistance, a lowering in skin temperature and an increase in heart rate. Conversely, relaxation was defined as being an increase in skin resistance, an increase in temperature and a decrease in heart rate.

In order to test hypothesis 1A the amplitude of the three periods of greatest continuous increase in tension for each of the three measures was ascertained using as a base measure the period immediately preceding the measure. The data for hypothesis 1B was obtained in the same manner except that the greatest periods of relaxation were used. Overall tension levels (hypothesis 2) were measured by obtaining the means of each variable measured at one minute intervals, as well as beginning to end differences in GSR and skin temperature.

Findings:

The operational forms of the two hypotheses and the findings concerning them are shown in Table 3, Table 4, Table 5 and Table 6.

Significantly higher tension peaks were found on all three physiological indices for the inter-actions occurring in the high condition level relationship. Significantly higher relaxation peaks were found only on one physiological index of the three. Overall tension levels were significantly different, but in the reverse direction from our prediction, i.e., the high condition level relationships showed higher overall tension level on GSR during the inter-actions, than the low level ones, and more tension drops over the whole period for both GSR and temperature.

[Page 5]

Discussion:

The findings indicate that inter-personal factors have autonomic correlates or conversely, that the autonomic aspects of present body life have inter-personal dimensions. The findings support the general theory that an inter-personal inter-action process has organismic dimensions.

Specifically, the predicted increased eventfulness (more open to experience, more expressive, more fully themselves) appears to occur autonomically, in a relationship characterized by high condition levels. (Higher peaks of both tension and relaxation.)

On the other hand, the hostility, mistrust, and suspicion manifest in the subject's questionnaire responses did not result in high autonomic tension during inter-action. Our interpretation of this finding is that under low conditions the individual withdraws, keeps from being involved, and that he experiences not the hostility or mistrust as such, but instead a limited, narrowed and minimized process of experiencing.

The findings must, of course, be considered with caution, until many replications are completed. However, the implication is that physiologically, a more fruitful way of considering psychological contents is in terms of their maximizing or minimizing the actually ongoing process of experiencing, not as contents of experience that are (or are not) experienced.

Summary:

Each of thirteen subjects spent 20 minutes in inter-action with an individual previously chosen, with whom the relationship was perceived as "good and close" and another individual with whom the relationship was perceived as "cold and distant." The 72 item "Relationship Inventory" separately established these differences in the subject's perception of the two relationships. The Inventory measures the level at which the subject perceives Rogers' four basic "conditions" of a relationship. The autonomic correlates recorded during inter-action show greater peaks of tension (on three autonomic indices) [Page 6] and greater peaks of relaxation for the high condition subjects; however there occurred greater tension decrease during inter-action in those relationships in which the levels of "conditions" were perceived as low. The findings are interpreted to mean that the psychophysiological process of experiencing which occurs in the individual during inter-action is more eventful, involved, at times more tense, at other times more relaxed, than in poor relationships.

References

[1] Barrett-Lennard, Godfrey. Dimensions of perceived therapist response related to therapeutic change. Unpublished doctoral dissertation, University of Chicago, 1959.

[2] Gendlin, E. T. Experiencing: A variable in therapeutic change. American Journal of Psychotherapy (in press).

[3] Gendlin, E. T. and Berlin, J. I. Galvanic skin response correlates of different modes of experiencing. Journal of Clinical Psychology, 17, 73-77, 1961. (Also presented at the American Psychological Association convention, 1959)

[4] Lacey, J. I., Psychophysiological approaches to the evaluation of psychotherapeutic process and outcome. In Research in Psychotherapy. Proceedings of a conference, Washington, D.C.: American Psychological Association, 1959, P. 160.

[5] Lacey, J. I. and Lacey, Beatrice C. Chapter V. The Brain and Human Behavior. Williams and Williams, Baltimore: 1958.

[6] Rogers, C. R. A theory of therapy, personality, and interpersonal relationships as developed in the client-centered framework. In S. Koch (Ed.) Psychology: A Study of a Science, Vol. Ill. Formulations of the Person and the Social Context. New York: McGraw-Hill, 1959, 184-256.

[Page 7]

Table 1--Differences between B forms and C forms on the four conditions of a relationship

	AB	AC
			t	P<
R	43.23	2.00	9.840	.01
U	15.62	-15.54	7.018	.01
E	21.15	-10.08	10.074	.01
C	32.31	-6.62	8.196	.01
TOTAL	112.31	-30.24	15.46	.01

R = Level of Regard

U = Unconditionality of Regard

E = Empathy

C = Congruence

Table 2. Comparison of the relationship measures for "more changed" and "less changed" and AB group

	"More Changed" GLB Group	AB Group	"Less Changed" GLB Group
	Mean	Mean	Mean
1. Level of Regard (R)	40.7	43.23	27.9
2. Unconditionality (U)	30.1	15.62	21.7
3. Empathic Understanding (E)	28.8	21.15	17.7
4. Congruence	36.8	32.31	22.9

[Page 8}

Table 3.--Mean of greater amplitude of tension responses for G.S.R., skin temperature, and heart rate (Hypothesis 1A)

	of AB	of AC	t	p 2 tail
G.S.R. Low, Resistance ohms x 1,000	29.28	21.77	2.378	.05
Temp. Low, Centigrade	6.56	4.16	2.365	.05
Heart Rate - High	48.85	36.91	3.50	.01

Table 4.--Mean of greater amplitude of relaxation responses for G.S.R., skin temperature, and heart rate (Hypothesis 1B)

	of AB	of AC	t	p 2 tail
G.S.R. High, Resistance ohms x 1,000	24.08	25.39	.3777	NS
Temp. High, Centigrade	7.00	10.03	1.14	NS
Heart Rate - Low	44.76	35.36	3.89	.01

[Page 9]

Table 5.--Difference of means over interviews for G.S.R., skin temperature, and heart rate (Hypothesis 2)

	of AB	of AC	t	p 2 tail
G.S.R. Mean, ohms x 1,000	89.08	106.62	3.52	.01
Temperature - Mean, Centigrade	27.91	28.59	.882	NS
Heart Rate - Mean, HBM	86.28	84.36	1.55	NS

Table 6.--Beginning to end difference in G.S.R. and skin temperature

	of AB	of AC	t	p 2 tail
G.S.R. E-B	6.3	17.78	3.10	.01
Temp. E-B	-.977	.79	2.13	.05

G.S.R. - ohms x 1,000

Temp. - degrees Centigrade

Eugene T. Gendlin