PHYSIOLOGICAL NON-ESCAPE
RESPONSES
FROM
FIELD DEPENDENT
STRESS
DURING
ESCAPE
AND
IN FIELD INDEPENDENT
AND
SUBJECTS
CURT A. SANDMAN Department of Psychology, Ohio 43210, U.S.A.
Ohiv State
Uniwrdy,
1945 N. High Street,
Columhrts,
Accepted for publication 8 May 1974
Psychophysiological responses (GSP, GSR, heart rate and respiration) were monitored while the subjects viewed stressful stimuli. In balanced order the subjects were exposed to a condition in which they viewed the stimuli for 15 set and a condition in which they could escape the stimulus. The subjects were divided into four groups according to level of trait anxiety and rod and frame scores. The results suggested that escape from stress produced different patterns of physiological activity than non-escape conditions. The changes were most noticeable for tonic GSP, non-specific GSR and heart rate. Field independent subjects differentiated the conditions physiologically and behaviorally to a greater extent than field independent subjects. It was speculated that field independent subjects are more dependent upon physiological cues to evaluate their perceptions and emotions than field dependent subjects.
1. Introduction It is a curious observation that animals which defend themselves with movement, such as the turtle, show little evidence of fear (Solomon, 1927). Movement, escape or avoidance, therefore, may be considered a coping response to stress. For instance, during the typical active avoidance conditioning procedure, an animal initially becomes fearful of the conditioned stimulus, but as conditioning proceeds and the animal makes a series of correct responses, the fear subsides (Kamin, Brimer and Black, 1963). The avoidance response then, acts to reduce or cope with the fear. Physiological data have been generated in support of this contention since animals able to avoid (Weiss, 1968) or escape (Miller, 1969) shock suffer less severe gastric lesions than animals without a coping response. Even animals given the opportunity to fight when administered a shock are less fearful, as evidenced by decreased pituitary adrenocorticotropic hormone (ACTH), than animals without the opportunity to fight (Conner, Levine and Vernikos-Danellis, 1970). It has been further demonstrated that byexogenously inducing the putative physiological manifestation of the fear response (ACTH and related polypeptides) the coping 20.5
behavior is maintained (DeWied, 1966; Sandman, Kastin and Schally, 1971). The literature suggests, at least in subhuman animals, that any form of organized behavior may be considered a coping response to stress and thus short-circuit the physiological response to stress. Lazarus (1966, 1967, 1968) has convincingly argued that in human subjects the role of appraisal and reappraisal of a stimulus may lead to ego response mechanisms which constitute the avoidance, escape or coping response. Lazarus has shown that decreased physiological reactivity resulted when subjects were provided with defensive cognitive sets (Lazarus, Speisman, Mordkoff and Davison. 1962: Lazarus, Opton, Nomikus and Rankin, 1965). However, in the model proposed by Lazarus it is unclear what the effect of action such as avoidance or escape is on emotional and physiological responding (Lazarus, 1968; Averill. Opton and Lazarus, 1969). The present study was designed to determine the effects of direct action of a behavioral response upon the patterns of physiological responses in humans. Two conditions were examined, escape and non-escape from stress. Further, since it has been hypothesized that cognitive (Lazarus, 1967; Lacey and Lacey, 1970; Miller and Sandman, 1973) and affective (Dykman, Reese, Galbrecht, Ackerman and Sundermann, 1968: Miller and Sandman, 1973) variables affect physiological responses to stress, the influence of cognitive style (as measured by the rod and frame test) and trait anxiety on responses to escape and non-escape conditions were also examined. These two dimensions were included since they are orthogonal (Miller and Sandman, 1973) and each has been related to physiological responsivity. For instance field independence has been related to ‘accurate’ autonomic responding (Hein, Cohen and Shmavonian, 1964: Courter, Wattenmaker and Ax, 1965) and field dependence to diffuse nonspecific physiological responding. High trait anxious subjects have been reported to respond to stress with increased sympathetic-like physiological activity (Malmo, 1957: Smith and Wenger, 1965: Koepke and Pribram. 1966; Jackson and Barry, 1967; McDaniel and Sexton, 1970). In the present study it was hypothesized that: (a) The conditions of escape and non-escape will result in different patterns of physiological responses. (b) Defensive maneuvers consistent with the subjects’ cognitive style would be effective in short-circuiting the physiological responses to stress. Since escape mechanisms such as denial characterize the field dependent (FD) subjects’ reaction to stress (Witkin, l965), and stimulus consumption strategies such as intellectualization describe field independent (Fl) subjects’ response to stress, FD subjects should employ behavioral escape more readily than FI subjects and consequently be less physiologically labile than FI subjects. (c) Self-report measures of anxiety will be lower in the escape condition than the non-escape. Again, since escape is compatible with FD subjects’ cognitive
Physiological responses to stress
207
style, FD subjects should make best defensive use of escape and be less anxious during escape conditions. 2. Method 2. I. General design The statistical model was repeated measures 2 x 2 x 2 factorial design. The between-subject factors were field dependence-independence and high and low trait anxiety. The within-subject factor was the type of stress condition the subjects encountered, either non-escape or escape. The subjects were tested in the stress conditions as determined by a latin square design. 2.2. Subjects Twenty-four male subjects from introductory psychology classes with extreme scores on field independence-dependence as measured by the rod and frame test (Witkin, Lewis, Hertzman, Machover, Meissner and Warpner, 1954) and trait anxiety (Spielberger, Gorsuch and Lushene, 1969) were drawn from a larger subject population and divided into four equal groups. The means for the independent variables were: FD = 7.0; FI = I .7; HA = 48.3; and LA = 33.6. 2.3. Procedure All subjects chosen to participate in the experiment were administered the trait-state anxiety questionnaire (TSAI) at least a week prior to coming to the laboratory. Upon arrival at the laboratory they were again administered the state anxiety scale. As the electrodes were attached, the subjects were made familiar with a nine-point ‘distress scale’ which ranged from I, extremely pleasurable, to 9, extremely distressed, and told that they would be shown the scale during the experiment. They were then taken into the chamber, placed in a reclining chair and told to relax. They were informed that physiological responses to pictures were being investigated and that pictures would be projected on the screen in front of them. A IO min period allowing the subject to adapt to the laboratory followed. After the adaptation period the subject was told that he would see either slides of pictures he might find pleasurable or distressing. The slides, representing a continuum of affective values, consisted of pictures of dead bodies, nude and clothed males and females, landscape scenery, animals, etc. All of the slides had been previously rated with the distress scale by an independent sample of 37 male subjects. Three slides fell into each of the categories ‘extremely distressing’, ‘moderately distressing’, ‘moderately pleasurable’ and ‘extremely pleasurable’. In the non-escape condition the slide remained on for I5 sec. In the escape condition, the slides were on for a maximum of I5 set, but the subject was instructed that he could terminate the slide by pushing a key with his right index finger. The subject was
208
c.
.‘I.
sNtltiri7nr7
informed that the slide would remain on several seconds after he pressed the key. This was done to ensure that each slide remained on for at least 5 sec. Immediately following each slide, the distress rating scale was projected for 3 set and the subject was requested to rate his reaction to the slide on the ninepoint scale. There was a I5 set intertrial interval ( ITl) between the distress scale and the subsequent stimulus. Following both the escape and non-escape conditions a IO min ‘rest’ period ensued during which the subjects took the state anxiety questionnaire.
Subjects were tested in an 80 db sound attenuated and electrically shielded chamber. The stimuli were projected on a screen in the chamber by a Kodak Carousal slide projector programmed for automatic advance and stimulus presentation with BRS Foringer Digibit logic units. A key was placed on the arm of the subject’s chair to abort the stimulus in the escape condition. Physiological recording was done with a Grass, model 7 polygraph, equipped with appropriate bridges, pre-amplifiers and driver amplifiers. The voltage or resistance measures were written out on Grass oscillographs providing graphic representation of the physiological activity. 2.4.1. Respiration. The respirometer consisted of a sliding piston mounted on an elastic band and placed around the subject’s chest. A photocrystal in one end of the piston served as an arm of the resistance bridge. The resistance of the photocrystal was modulated by a small light in the opposite end of the piston and was determined by the distance between light and the photocrystal. The distance between the light and the photocrystal was determined by the excursion of the subject’s chest which relates most directly to respiration. The varying voltage of the resistance bridge was written out on Grass oscillographs providing graphic representations of the subject’s respiratory activities. Heart rate. Grass El05 cup electrodes filled with EKG solution were placed on the lower left rib cage and the right collar bone. The signal was amplified by a Grass wideband a.c. pre-amplifier and averaged by a Grass cardiotachometer which provided a beat-by-beat record of the heart rate. 2.4.2.
2.4.3. Exosomatic GSR. As described by Shmavonian, Miller and Cohen (1968), a curved Ag-AgCl electrode of 3 cm” placed on the volar surface of the second phalange of the right hand served as the active electrode, and a curved armband of 58 cm’ placed on the upper arm served as the inactive electrode of a monopolar placement. A constant current of 20 /LA supplied by a Biophysics model 201 GSR pre-amplifier was impressed at the active site. A bandage with a 2 cm’ hole in it restricted the current density to IO /IA/cm’ at the active electrode, whereas the current density at the reference electrode was
0.314 /lA/cm’. A 0.05N NaCl solution suspended in an inert plastic medium served as electrode paste. Surface body oils were removed by swabbing the electrode sites with acetone. 2.4.4. Endosomatic GSP. Grass El05 cup electrodes were converted into lowstanding potential (less than 50 pV), low d.c. drift, Ag-AgCI electrodes (Venables and Martin, 1967). The volar surface of the second phalange of the second finger of the left hand served as the active electrode site and the inner surface of the left forearm served as the reference site. The electrode sites were swabbed with acetone to remove oils, with activity at the reference being abolished by brisk rubbing with a gauze pad. The electrode medium was 0.05N NaCl in an inert plastic medium. The difference in the standing potential at the two sites was amplified by a I MQ input impedance Grass model 7 PI pre-amplifier and written out by the Grass oscillograph. A bucking voltage of opposite polarity supplied by the pre-amplifier balanced the d.c. potential and kept the oscillograph on channel. 2.5. Physiological data reduction To ensure comparable time intervals in the escape and non-escape conditions, the physiological data were reduced only during the first 5 set of the stimulus presentation. 2.5.1. Exosomatic GSR. The number of non-specific responses of 100 Q magnitude occurring within the ITI was tallied. Basal conductance changes as reflected in pre-stimulus levels and defined as the reciprocal of resistance were calculated and presented as micromhos. Skin conductance change AC was calculated as the reciprocal of the sum of all responses during the presentation of the stimulus, I set after onset, by the formula (I/R, - 1/RI) x I 06, where the pre-stimulus resistance level = R, and R, = the resistance level following the stimulus. 2.5.2. Endosomaric GSP. Total negative baseline shifts (tonic response) 1 set after stimulus onset and during the first seconds of the stimulus wereexpressed as millivolt change AP. Frequency of a-wave (Forbes, 1964) or phasic responses during the stimulus was tallied. 2.5.3. Heart rate. The fastest and the slowest heartbeats for the 5 set, prestimulus period and the first 5 set stimulus period were averaged. Change scores were obtained by subtracting the 5 set stimulus period from the prestimulus baseline period. 2.5.4. Respiration. Rate of respiration was determined by peak-to-peak for each 5 set period and reduced like the heart rate data.
count
Means for physiological Heart
responses rate change
Non-escape
Escape
Table I. during exnpc Number
and non-esc:tpc
non-specific
Non-escape
GSR
from stress Delta potential
(mVf
Escape
Non-escape
Escape
FILA
-2.6
+1.5
I.1
0.3
0.14
0.14
FIHA FDLA FDHA Total
-3.1 -0.7 -0.6 -1.8
+1.3 +0.2 -2.1 +0.2
1.3 1.3 I .4 1.3
0.3 0.9 1.o 0.6
0.44 0.40 0.14 0.30
0.22 0.20 0.1 I 0.17
It was hypothesized that conditions in which subjects could escape stressful stimuli would result in physiological patterns that would differ from forced exposure to stressful conditions. This hypothesis was partially confirmed. The findings were not significant for the entire system of physiological variables as determined by a multivariate analysis of variance (asymptotic chi square = 28.2, df = 32). Flowever, separate 23 univariate analysis for each dependent measure indicated a significant change in tonic GSP (F( I ,20) = 27.04, p < O-01), number of non-specific GSR response; (F( 1,20) = 14.28, p < 0.01) and heart rate (F( I ,20) = 18.76, p < 0.01). From table I, it is apparent that relative to non-escape conditions, escape from stress resulted in a decrease in both electrodermal measures (tonic GSP and non-specific GSR) but yielded a slight increase in heart rate. The relationships indicated in table I were consistent for each of the stimuli individually except in one case, tonic GSP during one stimulus in which there was no difference between conditions. Further analysis indicated that this one exceptional case was due to the inconsistent responses of field dependent subjects to that particular stimulus. There were no differences between escape and non-escape conditions for the non-stressful stimuli. It was expected that self-reported measures of state anxiety would be lower during conditions of escape than during forced exposure to the stimuli. This hypothesis was not directly confirmed (F(3.60) = I .45). However, the experimental conditions interacted significantly with trait anxiety (F(3,160) = 4.46, p < 0.01) and produced a reliable change in state anxiety. From fig. I, it is apparent that changes in state anxiety were most dramatic for all groups of subjects in response to exposure to the laboratory and were relatively unaffected by the experimental conditions. 3.2. Inditlidual diflerenws The second hypothesis, which suggested that FD subjects would view the stimulus for less time and display different physioio~ical response patterns than
PhJsiologieal
211
responses to stress
50 I
30-
C----o 25
FOHA FDLA FIHA FILA
Prk Lbb Non I
-
Escape
EXPERIMENTAL Fig. I.
Changes
CONDITIONS
in state anxiety as a function of experimental settings.
FI subjects was also partially confirmed. Field dependent subjects displayed significantly greater electrodermal activity than FI subjects during both escape and non-escape conditions with respect to number of non-specific GSRs (F(1,20) = 5.87, p < 0.02) skin conductance (F( 1,20) = 5.77, p < 0.02) and phasic GSP responses (F( 1,20) = 5.13, p < 0.03). Significant interactions between cognitive style and escape conditions were evident for heart rate (F(1,20) = 10.21, p < 0.01) and number of non-specific GSR responses (F(1,20) = 5.38, p < 0.03). The nature of the interactions are apparent from table 1. During the non-escape conditions the FI groups evidenced a large deceleration in heart rate, whereas the FD groups displayed only a small heart rate decrease. The pattern was dramatically reverse3 during the escape conditions, however, since heart rate acceleration characterizes the FI groups and heart deceleration occurs for the FD groups. In addition, the FI groups evidenced four times as much non-specific GSR activity during the non-escape condition when compared with the escape condition. The FD subjects did not differentiate the experimental conditions to the extent the FI subjects did. The patterns of all the physiological responses to the stressful stimuli are presented in standard score form in fig. 2. As hypothesized, the FD subjects viewed the stimulus for less time than FI subjects. The trend, apparent in table 2, failed to reach a level of statistical significance. The relationship between the physiological variables and the escape behavior for each of the stimuli is presented in table 3. The data strongly
c NON
ESCAPC
ESCA,R
h
4
I!
\
! \ ! i i
/
‘\ ‘1 ./.i
‘\
j!
i--J
!
!
i
!
\
i I !
\ ’ \j
i! !j’
‘\./
i
I i
i
\ b
I
RC
AC
NS
PI
AP
Pta
HP
DEFENDENT “ARIARIES
FDNA
?.o-
Fig. 2. Patterns of response for: (a) FILA, (b), FIHA, (c) FDLA, and (d) FDHA groups during escape and non-escape from stress. The values for the physiological variables are transformed into standard scores: R = respiration; BC= base conductance; dC= skin conductance; NS= number of non-specific GSRs: Pi= base potential: dP= tonic GSP: Ph = phasic GSP; and HR = heart rate.
Mean viewing
time during
Table 2. impersonal
stress-escape
Time (set)
Groups Field Field Field Field
independent-low independent-high dependent--low dependent-high
anxious anxious anxious anxious
suggest that the behavior
(escape)
physiological
than
correlations
responses are significant
conditions.
IO.52 9.34 8.50 9.17
2 5.17 + 4.36 i_ 3.93 i 4.21
of FI sub.iects is more closely
is the
behavior
li,r the FI subjects
of
FD
than
related
to their
hub_jects. More
of the
thr the FD
sub.jects. and
0.14 -0.01
-0.16 -0.09
-0.27* 0.04
-0.08 0.58t
*p cc0.05, tp < 0.01.
FILA FIHA FDLA FDHA
Respiration curve
Respiration base
0.25* 0.37t
0.341 0.02
Base conductance
0.11 0.22
0.05 -0.36t
Delta conductance
0.29* -0.15
-0.17 -0.4Ot
No. NS CSR
0.16 0.25*
0.24* -0.42t
Basal potential
Table 3. Correlations of physiological variables with viewing time.
-0.02 -0.12
-0.36t -0.01
Delta potential
0.23” 0.24*
0.21* -0.23*
No. phasics
-0mt 0.62t
4.35t -0.50t
Basal heart rate
-+.I7 0.03
-0.33t -0.33t -0.43t
Changes in heart rate
2 D
$ * 3
G
Ler g_. g 2. I) z 2
214
C. A. Sandman
more than twice as many are significant at the 0.01 level of significance. This finding suggests that FI subjects may differentiate the content of the stimuli to a finer degree than do FD subjects. From fig. 1 it appears that both high and low trait anxious FD subjects become more anxious in the escape conditions when compared with the nonescape conditions. The reverse pattern was evident for the FI subjects. Even though this pattern failed to reach a level of statistical significance it is interesting to note that FD subjects responded on the self-report dimension as well as the physiological and behavioral dimensions in an almost opposite way from FI subjects. 4. Discussion The effect of escape on the physiological responses to stress was a change in responding for three variables : tonic GSP, number of non-specific GSRs and heart rate. The results for the electrodermal responses may be interpreted as being consistent with the data reported for lower organisms in which avoidance (Weiss, 1968), escape (Miller, 1969) or organized behavior (Conner et al., 1970) led to a reduction in physiological activity. It has also been reported in human subjects that some form of organized behavior (Cannon, 1929) passive avoidance (Goldstein and Adams, 1967) or control of the stimulus contingencies (Haggard, 1943; Wachtel, 1968; Stotland and Blumenthal, 1968) leads to a decrease in sympathetic-like nervous system activity. The general heart rate deceleration observed during forced exposure to stressful stimuli is consistent with earlier findings (Libby, Lacey and Lacey, 1973; Hare, 1973). The escape conditions acted to short-circuit this cardiac response to stress. Therefore, the present study provides additional support for the notion that direct action may act to change the physiological response to stress. However, the preponderance of research with human subjects has demonstrated that cognitive factors are of primary importance for short-circuiting the response to stress (Lazarus, 1968; Weinstein, Averill, Opton and Lazarus, 1968; Averill et al., 1969). It was apparent from the present study that cognitive factors can reliably predict physiological and behavioral responses to stress. It was expected that FD subjects would be affected by escape to a greater degree than FI subjects, but that was not the case. It was apparent that FI subjects demonstrated dramatically different response profiles during escape and non-escape conditions, especially for heart rate and non-specific GSRs. The FI subjects also had higher correspondence between physiological variables and escape behavior than did the FD subjects. It might be speculated that the FI subjects differentiate situations and utilize physiological cues to determine perceptual or affective states to a greater extent than do the FD subjects. Surprisingly, the affective variables, neither alone nor interacting with cognitive variables, exerted the expected effect on the physiological responses
Physiological
responses to stress
215
to stress. Additionally, the self-reported changes in anxiety during escape conditions were not even related to trait anxiety as much as to cognitive style. Therefore it does not appear that affective variables relate consistently to vicarious stress conditions. Acknowledgements This investigation was based in part on a doctoral dissertation completed at Louisiana State University. The support and assistance of Lyle H. Miller, Temple University, is gratefully acknowledged. References Averill, J. R., Opton, E. M. and Lazarus, R. S. (1969). Cross-cultural studies of psychophysiological responses during stress and emotion. International Journal OfPsychology, 4, 83-102.
Cannon, W. B. (1929). Bodily Changes in Pain, Hunger, Fear and Rage: An account of Recent Researches into the Function of Emotiona ! Excitement. Appleton-Century-Crofts: New York. Conner, R. L., Levine, S. and Vernikos-Danellis, J. (1970). Shock-induced fighting and pituitary adrenal activity. Proceedings of the 78th Annual APA Convention, 201-202. Courter, R. J., Wattenmaker, R. A. and Ax, A. F. (1965). Physiological concomitants of psychological differentiation. Psychophysiology, 1,282-290. _ _ DeWied. D. (1966). Inhibitorv effect of ACTH and related oeotides on extinction of conditionkd avoidance behavior in rats. Proceedings of the Socieyy for Experimental Biology and Medicine,
122,28-32.
Dykman, R. A., Reese, W. G., Galbrecht, C. R.. Ackerman, P. T. and Sundermann, R. S. (1968). Autonomic responses in psychiatric patients. Annals sf the New York Academy of Sciences, 147, 237-303. Forbes, T. W. (1964). Problems in measurement of electrodermal phenomena+hoice of method and phenomena potential, impedance, resistance. Psychophysiology, 1,26-30. Goldstein, M. J. and Adams, J. N. (1967). Coping style and behavioral response to stress. Journal of Experimental
Research in Personality,
2,239-251.
Haggard, E. (1943). Some conditions determining adjustment during and readjustment following experimentally induced stress. In: Tomkins, S. (Ed.), Contemporary Psychology. Harvard University Press: Cambridge, Mass., 529-544. Hare, R. D. (1973). Orienting and defensive responses to visual stimuli. Psychophysiology, 10, 453464.
Hein, P., Cohen, S. and Shmavonian, B. M. (1964). Perceptual mode and Pavlonian typology. In: Waitis, J. (Ed.), Recent Advances in Biological Psychiatry. Plenum Press, New York. 71-78. Jackson, B. T. and Barry, W. F. (I 967). The vasomotor component of the orientation reaction as a correlative of anxiety. Perceptual and Motor Skills, 25, 5 14-5 16. Kamin, L. J., Brimer, C. J. and Black, A. H. (1963). Conditioned suppression as a monitor of the CS in the course of avoidance training. Journal of Comparative and Physiological Psychology,
56,497-501.
Lacey, J. I. and Lacey, B. C. (1970). Some autonomic-central nervous system interrelations, In: Black, P. (Ed.), Physiological Correlates of Emotion. Academic Press: New York, 205-227. Lazarus, R. S. (1966). PsychologicalStressand the Coping Process. McGraw-Hill: New York. Lazarus. R. S. (1967). Stress theory and .psychophysiological research. Forsvarsmedicin, 3. . __ 152-177. Lazarus, R. S. (I 968). Emotions and adaptation conceptual and empirical relations. Nebraska Symposium
on Motivation,
175-270.
Lazarus, R. S., Opton, E. M., Nomikus, M. J. and Rankin, N. 0. (1965). The principle of short-circuiting of threat: further evidence. Journal of Personality, 33,622-635.
216
C. A. Sandman
Lazarus, R. S., Speisman, J. C., Mordkoff, A. M. and Davison, L. A. (1962). A laboratory study of psychological stress produced by a motion picture film. PwcholonicalMonoaraahs. -_ 76 (whole no. 34): Libby, W. L., Lacey, B. C. and Lacey, J. I. (1973). Pupillary and cardiac activity during visual attention. Psychophysiology, 10,270-294. Malmo, R. B. (1957). Anxiety and behavioral stress. Psychological Review, 64,276-287. McDaniel. J. W. and Sexton. A. W. (1970). Psvchoendocrine studies of natients with soinal cord lesions. JournalofAb~ormal ksych&og;, 76, 117-122. Miller, L. H. and Sandman, C. A. (1973). Psychophysiological response patterns as a function of cognitive style and state-trait anxiety. Paper presented at the Southeastern Psychological Association, New Orleans. Miller, N. E. (1969). Psychosomatic effects of specific types of training. Anna/s of the Net+ York Academy
of Sciences,
159, 1025-1040.
Sandman, C. A., Kastin, A. J. and Schally, A. V. (1971). Behavioral inhibition as modified by melanocyte-stimulating hormone (MSH) and light-dark conditions. Physiology and Behavior,
6,4548.
Shmavonian, B. M., Miller, L. H. and Cohen, S. I. (1968). Differences among age and sex groups in electrodermal conditioning. Psychophysiology, 5, 119-l 3 1. Smith, D. B. D. and Wenger, M. A. (1965). Changes in autonomic balance during phasic anxiety. Psychophysiology, 1,267-271. Solomon, M. (1927). The mechanism of the emotions. British Journal ofMedicalPsycho/ogy, 7,301-314.
Spielberger, C. D., Gorsuch, R. L. and Lushene, R. E. (1969). The State-Traif Anxiefy Inventory. Consulting Psychologists Press: Palo Alto. Stotland, E. and Blumenthal, A. (1966). The reduction of anxiety as a result of the expectation of making a choice. Canadian JournalofPsychology, 18,138-145. Venables, P. H. and Martin, 1. (1967). Manual of Psychophysiological Merhods. American Elsevier Publishing Co., Inc. : New York. Wachtel, P. L. (1968). Anxiety attention and coping with threat. Journal of Abnormal Psychology, 73, 137-143. Weiss, J. M. (1968). Effects of coping responses on stress. JournalofComparatiae andPhysiological Psychology,
65,25 I-260.
Weinstein, J., Averill, J. R., Opton, E. M. and Lazarus, R. S. (1968). Defensive style and discrepancy between self-report and physiological indexes of stress. Journal of Personality and Social Psychology, 10,406413. Witkin, H. A. (1965). Psychological differentiation and forms of pathology. Journal of’ AbnormalPsychology,
70,317-336.
Witkin, H. A., Lewis, H. B., Hertzman, M., Machover, K., Meissner, P. B. and Wapner, S. (1954). Personality Through Perception. Harper & Row: New York.
RESPONSES
FROM
FIELD DEPENDENT
STRESS
DURING
ESCAPE
AND
IN FIELD INDEPENDENT
AND
SUBJECTS
CURT A. SANDMAN Department of Psychology, Ohio 43210, U.S.A.
Ohiv State
Uniwrdy,
1945 N. High Street,
Columhrts,
Accepted for publication 8 May 1974
Psychophysiological responses (GSP, GSR, heart rate and respiration) were monitored while the subjects viewed stressful stimuli. In balanced order the subjects were exposed to a condition in which they viewed the stimuli for 15 set and a condition in which they could escape the stimulus. The subjects were divided into four groups according to level of trait anxiety and rod and frame scores. The results suggested that escape from stress produced different patterns of physiological activity than non-escape conditions. The changes were most noticeable for tonic GSP, non-specific GSR and heart rate. Field independent subjects differentiated the conditions physiologically and behaviorally to a greater extent than field independent subjects. It was speculated that field independent subjects are more dependent upon physiological cues to evaluate their perceptions and emotions than field dependent subjects.
1. Introduction It is a curious observation that animals which defend themselves with movement, such as the turtle, show little evidence of fear (Solomon, 1927). Movement, escape or avoidance, therefore, may be considered a coping response to stress. For instance, during the typical active avoidance conditioning procedure, an animal initially becomes fearful of the conditioned stimulus, but as conditioning proceeds and the animal makes a series of correct responses, the fear subsides (Kamin, Brimer and Black, 1963). The avoidance response then, acts to reduce or cope with the fear. Physiological data have been generated in support of this contention since animals able to avoid (Weiss, 1968) or escape (Miller, 1969) shock suffer less severe gastric lesions than animals without a coping response. Even animals given the opportunity to fight when administered a shock are less fearful, as evidenced by decreased pituitary adrenocorticotropic hormone (ACTH), than animals without the opportunity to fight (Conner, Levine and Vernikos-Danellis, 1970). It has been further demonstrated that byexogenously inducing the putative physiological manifestation of the fear response (ACTH and related polypeptides) the coping 20.5
behavior is maintained (DeWied, 1966; Sandman, Kastin and Schally, 1971). The literature suggests, at least in subhuman animals, that any form of organized behavior may be considered a coping response to stress and thus short-circuit the physiological response to stress. Lazarus (1966, 1967, 1968) has convincingly argued that in human subjects the role of appraisal and reappraisal of a stimulus may lead to ego response mechanisms which constitute the avoidance, escape or coping response. Lazarus has shown that decreased physiological reactivity resulted when subjects were provided with defensive cognitive sets (Lazarus, Speisman, Mordkoff and Davison. 1962: Lazarus, Opton, Nomikus and Rankin, 1965). However, in the model proposed by Lazarus it is unclear what the effect of action such as avoidance or escape is on emotional and physiological responding (Lazarus, 1968; Averill. Opton and Lazarus, 1969). The present study was designed to determine the effects of direct action of a behavioral response upon the patterns of physiological responses in humans. Two conditions were examined, escape and non-escape from stress. Further, since it has been hypothesized that cognitive (Lazarus, 1967; Lacey and Lacey, 1970; Miller and Sandman, 1973) and affective (Dykman, Reese, Galbrecht, Ackerman and Sundermann, 1968: Miller and Sandman, 1973) variables affect physiological responses to stress, the influence of cognitive style (as measured by the rod and frame test) and trait anxiety on responses to escape and non-escape conditions were also examined. These two dimensions were included since they are orthogonal (Miller and Sandman, 1973) and each has been related to physiological responsivity. For instance field independence has been related to ‘accurate’ autonomic responding (Hein, Cohen and Shmavonian, 1964: Courter, Wattenmaker and Ax, 1965) and field dependence to diffuse nonspecific physiological responding. High trait anxious subjects have been reported to respond to stress with increased sympathetic-like physiological activity (Malmo, 1957: Smith and Wenger, 1965: Koepke and Pribram. 1966; Jackson and Barry, 1967; McDaniel and Sexton, 1970). In the present study it was hypothesized that: (a) The conditions of escape and non-escape will result in different patterns of physiological responses. (b) Defensive maneuvers consistent with the subjects’ cognitive style would be effective in short-circuiting the physiological responses to stress. Since escape mechanisms such as denial characterize the field dependent (FD) subjects’ reaction to stress (Witkin, l965), and stimulus consumption strategies such as intellectualization describe field independent (Fl) subjects’ response to stress, FD subjects should employ behavioral escape more readily than FI subjects and consequently be less physiologically labile than FI subjects. (c) Self-report measures of anxiety will be lower in the escape condition than the non-escape. Again, since escape is compatible with FD subjects’ cognitive
Physiological responses to stress
207
style, FD subjects should make best defensive use of escape and be less anxious during escape conditions. 2. Method 2. I. General design The statistical model was repeated measures 2 x 2 x 2 factorial design. The between-subject factors were field dependence-independence and high and low trait anxiety. The within-subject factor was the type of stress condition the subjects encountered, either non-escape or escape. The subjects were tested in the stress conditions as determined by a latin square design. 2.2. Subjects Twenty-four male subjects from introductory psychology classes with extreme scores on field independence-dependence as measured by the rod and frame test (Witkin, Lewis, Hertzman, Machover, Meissner and Warpner, 1954) and trait anxiety (Spielberger, Gorsuch and Lushene, 1969) were drawn from a larger subject population and divided into four equal groups. The means for the independent variables were: FD = 7.0; FI = I .7; HA = 48.3; and LA = 33.6. 2.3. Procedure All subjects chosen to participate in the experiment were administered the trait-state anxiety questionnaire (TSAI) at least a week prior to coming to the laboratory. Upon arrival at the laboratory they were again administered the state anxiety scale. As the electrodes were attached, the subjects were made familiar with a nine-point ‘distress scale’ which ranged from I, extremely pleasurable, to 9, extremely distressed, and told that they would be shown the scale during the experiment. They were then taken into the chamber, placed in a reclining chair and told to relax. They were informed that physiological responses to pictures were being investigated and that pictures would be projected on the screen in front of them. A IO min period allowing the subject to adapt to the laboratory followed. After the adaptation period the subject was told that he would see either slides of pictures he might find pleasurable or distressing. The slides, representing a continuum of affective values, consisted of pictures of dead bodies, nude and clothed males and females, landscape scenery, animals, etc. All of the slides had been previously rated with the distress scale by an independent sample of 37 male subjects. Three slides fell into each of the categories ‘extremely distressing’, ‘moderately distressing’, ‘moderately pleasurable’ and ‘extremely pleasurable’. In the non-escape condition the slide remained on for I5 sec. In the escape condition, the slides were on for a maximum of I5 set, but the subject was instructed that he could terminate the slide by pushing a key with his right index finger. The subject was
208
c.
.‘I.
sNtltiri7nr7
informed that the slide would remain on several seconds after he pressed the key. This was done to ensure that each slide remained on for at least 5 sec. Immediately following each slide, the distress rating scale was projected for 3 set and the subject was requested to rate his reaction to the slide on the ninepoint scale. There was a I5 set intertrial interval ( ITl) between the distress scale and the subsequent stimulus. Following both the escape and non-escape conditions a IO min ‘rest’ period ensued during which the subjects took the state anxiety questionnaire.
Subjects were tested in an 80 db sound attenuated and electrically shielded chamber. The stimuli were projected on a screen in the chamber by a Kodak Carousal slide projector programmed for automatic advance and stimulus presentation with BRS Foringer Digibit logic units. A key was placed on the arm of the subject’s chair to abort the stimulus in the escape condition. Physiological recording was done with a Grass, model 7 polygraph, equipped with appropriate bridges, pre-amplifiers and driver amplifiers. The voltage or resistance measures were written out on Grass oscillographs providing graphic representation of the physiological activity. 2.4.1. Respiration. The respirometer consisted of a sliding piston mounted on an elastic band and placed around the subject’s chest. A photocrystal in one end of the piston served as an arm of the resistance bridge. The resistance of the photocrystal was modulated by a small light in the opposite end of the piston and was determined by the distance between light and the photocrystal. The distance between the light and the photocrystal was determined by the excursion of the subject’s chest which relates most directly to respiration. The varying voltage of the resistance bridge was written out on Grass oscillographs providing graphic representations of the subject’s respiratory activities. Heart rate. Grass El05 cup electrodes filled with EKG solution were placed on the lower left rib cage and the right collar bone. The signal was amplified by a Grass wideband a.c. pre-amplifier and averaged by a Grass cardiotachometer which provided a beat-by-beat record of the heart rate. 2.4.2.
2.4.3. Exosomatic GSR. As described by Shmavonian, Miller and Cohen (1968), a curved Ag-AgCl electrode of 3 cm” placed on the volar surface of the second phalange of the right hand served as the active electrode, and a curved armband of 58 cm’ placed on the upper arm served as the inactive electrode of a monopolar placement. A constant current of 20 /LA supplied by a Biophysics model 201 GSR pre-amplifier was impressed at the active site. A bandage with a 2 cm’ hole in it restricted the current density to IO /IA/cm’ at the active electrode, whereas the current density at the reference electrode was
0.314 /lA/cm’. A 0.05N NaCl solution suspended in an inert plastic medium served as electrode paste. Surface body oils were removed by swabbing the electrode sites with acetone. 2.4.4. Endosomatic GSP. Grass El05 cup electrodes were converted into lowstanding potential (less than 50 pV), low d.c. drift, Ag-AgCI electrodes (Venables and Martin, 1967). The volar surface of the second phalange of the second finger of the left hand served as the active electrode site and the inner surface of the left forearm served as the reference site. The electrode sites were swabbed with acetone to remove oils, with activity at the reference being abolished by brisk rubbing with a gauze pad. The electrode medium was 0.05N NaCl in an inert plastic medium. The difference in the standing potential at the two sites was amplified by a I MQ input impedance Grass model 7 PI pre-amplifier and written out by the Grass oscillograph. A bucking voltage of opposite polarity supplied by the pre-amplifier balanced the d.c. potential and kept the oscillograph on channel. 2.5. Physiological data reduction To ensure comparable time intervals in the escape and non-escape conditions, the physiological data were reduced only during the first 5 set of the stimulus presentation. 2.5.1. Exosomatic GSR. The number of non-specific responses of 100 Q magnitude occurring within the ITI was tallied. Basal conductance changes as reflected in pre-stimulus levels and defined as the reciprocal of resistance were calculated and presented as micromhos. Skin conductance change AC was calculated as the reciprocal of the sum of all responses during the presentation of the stimulus, I set after onset, by the formula (I/R, - 1/RI) x I 06, where the pre-stimulus resistance level = R, and R, = the resistance level following the stimulus. 2.5.2. Endosomaric GSP. Total negative baseline shifts (tonic response) 1 set after stimulus onset and during the first seconds of the stimulus wereexpressed as millivolt change AP. Frequency of a-wave (Forbes, 1964) or phasic responses during the stimulus was tallied. 2.5.3. Heart rate. The fastest and the slowest heartbeats for the 5 set, prestimulus period and the first 5 set stimulus period were averaged. Change scores were obtained by subtracting the 5 set stimulus period from the prestimulus baseline period. 2.5.4. Respiration. Rate of respiration was determined by peak-to-peak for each 5 set period and reduced like the heart rate data.
count
Means for physiological Heart
responses rate change
Non-escape
Escape
Table I. during exnpc Number
and non-esc:tpc
non-specific
Non-escape
GSR
from stress Delta potential
(mVf
Escape
Non-escape
Escape
FILA
-2.6
+1.5
I.1
0.3
0.14
0.14
FIHA FDLA FDHA Total
-3.1 -0.7 -0.6 -1.8
+1.3 +0.2 -2.1 +0.2
1.3 1.3 I .4 1.3
0.3 0.9 1.o 0.6
0.44 0.40 0.14 0.30
0.22 0.20 0.1 I 0.17
It was hypothesized that conditions in which subjects could escape stressful stimuli would result in physiological patterns that would differ from forced exposure to stressful conditions. This hypothesis was partially confirmed. The findings were not significant for the entire system of physiological variables as determined by a multivariate analysis of variance (asymptotic chi square = 28.2, df = 32). Flowever, separate 23 univariate analysis for each dependent measure indicated a significant change in tonic GSP (F( I ,20) = 27.04, p < O-01), number of non-specific GSR response; (F( 1,20) = 14.28, p < 0.01) and heart rate (F( I ,20) = 18.76, p < 0.01). From table I, it is apparent that relative to non-escape conditions, escape from stress resulted in a decrease in both electrodermal measures (tonic GSP and non-specific GSR) but yielded a slight increase in heart rate. The relationships indicated in table I were consistent for each of the stimuli individually except in one case, tonic GSP during one stimulus in which there was no difference between conditions. Further analysis indicated that this one exceptional case was due to the inconsistent responses of field dependent subjects to that particular stimulus. There were no differences between escape and non-escape conditions for the non-stressful stimuli. It was expected that self-reported measures of state anxiety would be lower during conditions of escape than during forced exposure to the stimuli. This hypothesis was not directly confirmed (F(3.60) = I .45). However, the experimental conditions interacted significantly with trait anxiety (F(3,160) = 4.46, p < 0.01) and produced a reliable change in state anxiety. From fig. I, it is apparent that changes in state anxiety were most dramatic for all groups of subjects in response to exposure to the laboratory and were relatively unaffected by the experimental conditions. 3.2. Inditlidual diflerenws The second hypothesis, which suggested that FD subjects would view the stimulus for less time and display different physioio~ical response patterns than
PhJsiologieal
211
responses to stress
50 I
30-
C----o 25
FOHA FDLA FIHA FILA
Prk Lbb Non I
-
Escape
EXPERIMENTAL Fig. I.
Changes
CONDITIONS
in state anxiety as a function of experimental settings.
FI subjects was also partially confirmed. Field dependent subjects displayed significantly greater electrodermal activity than FI subjects during both escape and non-escape conditions with respect to number of non-specific GSRs (F(1,20) = 5.87, p < 0.02) skin conductance (F( 1,20) = 5.77, p < 0.02) and phasic GSP responses (F( 1,20) = 5.13, p < 0.03). Significant interactions between cognitive style and escape conditions were evident for heart rate (F(1,20) = 10.21, p < 0.01) and number of non-specific GSR responses (F(1,20) = 5.38, p < 0.03). The nature of the interactions are apparent from table 1. During the non-escape conditions the FI groups evidenced a large deceleration in heart rate, whereas the FD groups displayed only a small heart rate decrease. The pattern was dramatically reverse3 during the escape conditions, however, since heart rate acceleration characterizes the FI groups and heart deceleration occurs for the FD groups. In addition, the FI groups evidenced four times as much non-specific GSR activity during the non-escape condition when compared with the escape condition. The FD subjects did not differentiate the experimental conditions to the extent the FI subjects did. The patterns of all the physiological responses to the stressful stimuli are presented in standard score form in fig. 2. As hypothesized, the FD subjects viewed the stimulus for less time than FI subjects. The trend, apparent in table 2, failed to reach a level of statistical significance. The relationship between the physiological variables and the escape behavior for each of the stimuli is presented in table 3. The data strongly
c NON
ESCAPC
ESCA,R
h
4
I!
\
! \ ! i i
/
‘\ ‘1 ./.i
‘\
j!
i--J
!
!
i
!
\
i I !
\ ’ \j
i! !j’
‘\./
i
I i
i
\ b
I
RC
AC
NS
PI
AP
Pta
HP
DEFENDENT “ARIARIES
FDNA
?.o-
Fig. 2. Patterns of response for: (a) FILA, (b), FIHA, (c) FDLA, and (d) FDHA groups during escape and non-escape from stress. The values for the physiological variables are transformed into standard scores: R = respiration; BC= base conductance; dC= skin conductance; NS= number of non-specific GSRs: Pi= base potential: dP= tonic GSP: Ph = phasic GSP; and HR = heart rate.
Mean viewing
time during
Table 2. impersonal
stress-escape
Time (set)
Groups Field Field Field Field
independent-low independent-high dependent--low dependent-high
anxious anxious anxious anxious
suggest that the behavior
(escape)
physiological
than
correlations
responses are significant
conditions.
IO.52 9.34 8.50 9.17
2 5.17 + 4.36 i_ 3.93 i 4.21
of FI sub.iects is more closely
is the
behavior
li,r the FI subjects
of
FD
than
related
to their
hub_jects. More
of the
thr the FD
sub.jects. and
0.14 -0.01
-0.16 -0.09
-0.27* 0.04
-0.08 0.58t
*p cc0.05, tp < 0.01.
FILA FIHA FDLA FDHA
Respiration curve
Respiration base
0.25* 0.37t
0.341 0.02
Base conductance
0.11 0.22
0.05 -0.36t
Delta conductance
0.29* -0.15
-0.17 -0.4Ot
No. NS CSR
0.16 0.25*
0.24* -0.42t
Basal potential
Table 3. Correlations of physiological variables with viewing time.
-0.02 -0.12
-0.36t -0.01
Delta potential
0.23” 0.24*
0.21* -0.23*
No. phasics
-0mt 0.62t
4.35t -0.50t
Basal heart rate
-+.I7 0.03
-0.33t -0.33t -0.43t
Changes in heart rate
2 D
$ * 3
G
Ler g_. g 2. I) z 2
214
C. A. Sandman
more than twice as many are significant at the 0.01 level of significance. This finding suggests that FI subjects may differentiate the content of the stimuli to a finer degree than do FD subjects. From fig. 1 it appears that both high and low trait anxious FD subjects become more anxious in the escape conditions when compared with the nonescape conditions. The reverse pattern was evident for the FI subjects. Even though this pattern failed to reach a level of statistical significance it is interesting to note that FD subjects responded on the self-report dimension as well as the physiological and behavioral dimensions in an almost opposite way from FI subjects. 4. Discussion The effect of escape on the physiological responses to stress was a change in responding for three variables : tonic GSP, number of non-specific GSRs and heart rate. The results for the electrodermal responses may be interpreted as being consistent with the data reported for lower organisms in which avoidance (Weiss, 1968), escape (Miller, 1969) or organized behavior (Conner et al., 1970) led to a reduction in physiological activity. It has also been reported in human subjects that some form of organized behavior (Cannon, 1929) passive avoidance (Goldstein and Adams, 1967) or control of the stimulus contingencies (Haggard, 1943; Wachtel, 1968; Stotland and Blumenthal, 1968) leads to a decrease in sympathetic-like nervous system activity. The general heart rate deceleration observed during forced exposure to stressful stimuli is consistent with earlier findings (Libby, Lacey and Lacey, 1973; Hare, 1973). The escape conditions acted to short-circuit this cardiac response to stress. Therefore, the present study provides additional support for the notion that direct action may act to change the physiological response to stress. However, the preponderance of research with human subjects has demonstrated that cognitive factors are of primary importance for short-circuiting the response to stress (Lazarus, 1968; Weinstein, Averill, Opton and Lazarus, 1968; Averill et al., 1969). It was apparent from the present study that cognitive factors can reliably predict physiological and behavioral responses to stress. It was expected that FD subjects would be affected by escape to a greater degree than FI subjects, but that was not the case. It was apparent that FI subjects demonstrated dramatically different response profiles during escape and non-escape conditions, especially for heart rate and non-specific GSRs. The FI subjects also had higher correspondence between physiological variables and escape behavior than did the FD subjects. It might be speculated that the FI subjects differentiate situations and utilize physiological cues to determine perceptual or affective states to a greater extent than do the FD subjects. Surprisingly, the affective variables, neither alone nor interacting with cognitive variables, exerted the expected effect on the physiological responses
Physiological
responses to stress
215
to stress. Additionally, the self-reported changes in anxiety during escape conditions were not even related to trait anxiety as much as to cognitive style. Therefore it does not appear that affective variables relate consistently to vicarious stress conditions. Acknowledgements This investigation was based in part on a doctoral dissertation completed at Louisiana State University. The support and assistance of Lyle H. Miller, Temple University, is gratefully acknowledged. References Averill, J. R., Opton, E. M. and Lazarus, R. S. (1969). Cross-cultural studies of psychophysiological responses during stress and emotion. International Journal OfPsychology, 4, 83-102.
Cannon, W. B. (1929). Bodily Changes in Pain, Hunger, Fear and Rage: An account of Recent Researches into the Function of Emotiona ! Excitement. Appleton-Century-Crofts: New York. Conner, R. L., Levine, S. and Vernikos-Danellis, J. (1970). Shock-induced fighting and pituitary adrenal activity. Proceedings of the 78th Annual APA Convention, 201-202. Courter, R. J., Wattenmaker, R. A. and Ax, A. F. (1965). Physiological concomitants of psychological differentiation. Psychophysiology, 1,282-290. _ _ DeWied. D. (1966). Inhibitorv effect of ACTH and related oeotides on extinction of conditionkd avoidance behavior in rats. Proceedings of the Socieyy for Experimental Biology and Medicine,
122,28-32.
Dykman, R. A., Reese, W. G., Galbrecht, C. R.. Ackerman, P. T. and Sundermann, R. S. (1968). Autonomic responses in psychiatric patients. Annals sf the New York Academy of Sciences, 147, 237-303. Forbes, T. W. (1964). Problems in measurement of electrodermal phenomena+hoice of method and phenomena potential, impedance, resistance. Psychophysiology, 1,26-30. Goldstein, M. J. and Adams, J. N. (1967). Coping style and behavioral response to stress. Journal of Experimental
Research in Personality,
2,239-251.
Haggard, E. (1943). Some conditions determining adjustment during and readjustment following experimentally induced stress. In: Tomkins, S. (Ed.), Contemporary Psychology. Harvard University Press: Cambridge, Mass., 529-544. Hare, R. D. (1973). Orienting and defensive responses to visual stimuli. Psychophysiology, 10, 453464.
Hein, P., Cohen, S. and Shmavonian, B. M. (1964). Perceptual mode and Pavlonian typology. In: Waitis, J. (Ed.), Recent Advances in Biological Psychiatry. Plenum Press, New York. 71-78. Jackson, B. T. and Barry, W. F. (I 967). The vasomotor component of the orientation reaction as a correlative of anxiety. Perceptual and Motor Skills, 25, 5 14-5 16. Kamin, L. J., Brimer, C. J. and Black, A. H. (1963). Conditioned suppression as a monitor of the CS in the course of avoidance training. Journal of Comparative and Physiological Psychology,
56,497-501.
Lacey, J. I. and Lacey, B. C. (1970). Some autonomic-central nervous system interrelations, In: Black, P. (Ed.), Physiological Correlates of Emotion. Academic Press: New York, 205-227. Lazarus, R. S. (1966). PsychologicalStressand the Coping Process. McGraw-Hill: New York. Lazarus. R. S. (1967). Stress theory and .psychophysiological research. Forsvarsmedicin, 3. . __ 152-177. Lazarus, R. S. (I 968). Emotions and adaptation conceptual and empirical relations. Nebraska Symposium
on Motivation,
175-270.
Lazarus, R. S., Opton, E. M., Nomikus, M. J. and Rankin, N. 0. (1965). The principle of short-circuiting of threat: further evidence. Journal of Personality, 33,622-635.
216
C. A. Sandman
Lazarus, R. S., Speisman, J. C., Mordkoff, A. M. and Davison, L. A. (1962). A laboratory study of psychological stress produced by a motion picture film. PwcholonicalMonoaraahs. -_ 76 (whole no. 34): Libby, W. L., Lacey, B. C. and Lacey, J. I. (1973). Pupillary and cardiac activity during visual attention. Psychophysiology, 10,270-294. Malmo, R. B. (1957). Anxiety and behavioral stress. Psychological Review, 64,276-287. McDaniel. J. W. and Sexton. A. W. (1970). Psvchoendocrine studies of natients with soinal cord lesions. JournalofAb~ormal ksych&og;, 76, 117-122. Miller, L. H. and Sandman, C. A. (1973). Psychophysiological response patterns as a function of cognitive style and state-trait anxiety. Paper presented at the Southeastern Psychological Association, New Orleans. Miller, N. E. (1969). Psychosomatic effects of specific types of training. Anna/s of the Net+ York Academy
of Sciences,
159, 1025-1040.
Sandman, C. A., Kastin, A. J. and Schally, A. V. (1971). Behavioral inhibition as modified by melanocyte-stimulating hormone (MSH) and light-dark conditions. Physiology and Behavior,
6,4548.
Shmavonian, B. M., Miller, L. H. and Cohen, S. I. (1968). Differences among age and sex groups in electrodermal conditioning. Psychophysiology, 5, 119-l 3 1. Smith, D. B. D. and Wenger, M. A. (1965). Changes in autonomic balance during phasic anxiety. Psychophysiology, 1,267-271. Solomon, M. (1927). The mechanism of the emotions. British Journal ofMedicalPsycho/ogy, 7,301-314.
Spielberger, C. D., Gorsuch, R. L. and Lushene, R. E. (1969). The State-Traif Anxiefy Inventory. Consulting Psychologists Press: Palo Alto. Stotland, E. and Blumenthal, A. (1966). The reduction of anxiety as a result of the expectation of making a choice. Canadian JournalofPsychology, 18,138-145. Venables, P. H. and Martin, 1. (1967). Manual of Psychophysiological Merhods. American Elsevier Publishing Co., Inc. : New York. Wachtel, P. L. (1968). Anxiety attention and coping with threat. Journal of Abnormal Psychology, 73, 137-143. Weiss, J. M. (1968). Effects of coping responses on stress. JournalofComparatiae andPhysiological Psychology,
65,25 I-260.
Weinstein, J., Averill, J. R., Opton, E. M. and Lazarus, R. S. (1968). Defensive style and discrepancy between self-report and physiological indexes of stress. Journal of Personality and Social Psychology, 10,406413. Witkin, H. A. (1965). Psychological differentiation and forms of pathology. Journal of’ AbnormalPsychology,
70,317-336.
Witkin, H. A., Lewis, H. B., Hertzman, M., Machover, K., Meissner, P. B. and Wapner, S. (1954). Personality Through Perception. Harper & Row: New York.
