BLOOD PRESSURE REACTIONS TO EXPRESSED AND UNEXPRESSED ANGER IN LOW GUILT AND HIGH GUILT SUSIECTS Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY Salvamre Gambaro I966 III/II HI ”2 rm“ 7&6”! ”0’30'7 LIBRA£12I Michigan St a t‘? Uni:> 5:353)‘ ' fl This is to certify that the thesis entitled BLOOD PRESSURE PEACTIONS TO EXPRESS ED AND UNEXPRESSED ANGER IN LOW GUILT AI‘ID HIGH GUILT SUBJECTS presented by Salvatore Gambaro has been accepted towards fulfillment . .I - ' t. of the requirements for WW Maior nrofessor Date /% /§:1 ///[’ 0-169 "' I ABSTRACT BLOOD PRESSURE REACTIONS TO EXPRESSED AND UNEXPRESSED ANGER IN LOW WILT AND HIE-l GUILT SUBJECTS w Salvatore Ganbaro The psychosomatic theory of essential hypertension involves the following major postulates: Inhibition of anger instigated aggression raises diastolic blood pressure and. by implication. the expression of aggression under these conditions lowers diastolic blood pressure. In an effort to experimentally investigate these postulates. basically two questions were considered in this research: "that effect do psychological stimuli, designed to arouse anger, have on blood pressure; and what is the effect on blood pressure of directly and indirectly expressing anger following these stimuli?" Also considered in this research was the variable of ''aggression guilt" as it relates to blood pressure reactions in the expression and non-expression of anger follow- ing the experimental manipulations. Subjects were randomly divided into four groups with twenty subjects in each group. Group I subjects went through a I'frustration sequence” and were allowed "direct expression" ,of anger by shocking the frustrator (the experimenter's assistant). Group II subjects went through the ”frustration sequence” and were allowed ”indirect expression" by shocking the experimenter. Group III subjects went through the "frustration sequence" but were allowed "no expression" through the 1 Salvatore Gambaro shock apparatus. Group IV subjects did not go through the ”frustration sequence." They were allowed to complete the presented task and in the "expression" phase they were engaged in "fill-in" activities. Blood pressure was measured indirectly in the subject's left arm by means of an E and M Physiograph. utilizing an mectrosphygnograph transducer. Resting blood pressures were recorded for all subjects. Post ”frustration" and post "expression" recordings were also obtained. Subjective ratings of anger were also made at the end of the "frustra- tion sequence." At the end of the experimental procedure. the Masher Incomplete Sentence Test was administered and scored for aggression- guilt. Subjects were categorized as High Chilt or Low (hilt on the basis of the median guilt score for all 80 subjects. Subjects who underwent the ”frustration sequence" rated themselves as significantly more angry than control subjects. Diastolic blood pressure was significantly elevated as a result of our experimental procedure. In the post "expression" phase for Groups I and II and post ”fill-in" phase for Ms II; and IV. a comparison of the mean diastolic decreases of subjects in the High (hilt and Low Gailt categories was made. Across all conditions of "expression" and ”non- expression," the "low guilt" subjects exhibited a significantly greater decrease of diastolic blood pressure. "Direct expression," shocking the frustrator, resulted in a significantly greater diastolic decrease than was measured in the ”no expression" group. "Indirect expression" did not result in a simificantly greater diastolic decrease than was measured in the "no expression“ group. However. when the variable of aggression-guilt was taken into account, a different picture emerged. Salvatore Gambaro "High guilt“ and "low guilt" subjects did not differ in their diastolic and systolic elevations in anger; almost all subjects showed blood pressure cheases in this phase. In the "expression phase" both the High (hilt and Low (hilt sub-groups in the ”direct expression" group exhibited significant mean diastolic decreases. In the "indirect expression" group, we found that the group as a whole did not differ significantly from the ”no expression" group in mean diastolic decrease. But. when the High (hilt and Low (hilt sub-groups were compared in the "indirect expression" phase. we found that the Low Guilt group had a diastolic decrease sigxificantly greater than the High Guilt group had and the Low (hilt group's decrease was as great as that shown by the total “direct expression" group. The "high guilt," ”indirect expression" subjects were actually showing a slight mean increase after "expression"! The results were seen as supporting those aspects of the psycho, somatic theory of hypertension which were put to an experimental test and also as introducing the importance of the aggression-guilt variable into the theory. mm... MWA—a Committee Chairman Date Jig/Q BLOOD PRESSURE REACTIONS TO EXPRESED AND UNEXPRESED ANGER IN LOW GUILT AND 310'! GUILT SUBJECTS By Salvatore Gsmbaro A THBIS Submitted to Michigan State University in partial fulfillment of. the requirements for the degree of IDCTOR OF PHILOSOPHY Department of Psychology 1966 ACKNOWLEDCHENTS The completion of graduate studies and of a dissertation involves many more people than the writer. Some contribute actively with sugges- tions and encouragement. some contribute by example, and others contribute by their publications. I wish to acknowledge gratefully the assistance of the following peeple who have contributed to my studies and this dissertation in various ways: Professor Paul Bakan. Dr. Glenn Hatton and Dr. Bertram Karon served on the dissertation committee and made helpful comments and crit- icisms of the initial thesis proposal as well as the finished manuscript. Professor Albert Rabin served as aw cmittee chairman for both the thesis committee and the graduate guidance committee. Hr. Stiefel of the Physiology Department was most generous of his time in explaining the operation of the physiograph he gave me on loan. Dr. Alfredson. Chairman of the Physiology Department, assisted by providing an experimental room to which I had access during all hours of the day and night. W thanks also go to Mrs. Leroy Quance for her careful typing of the final manuscript. . Of these I wish to express particular gratitude to Professor Rabin. His numerous comments and suggestions have helped shape this thesis from a ”rough idea" to a finished product which I take pride in. He has also contributed in ways unknown to him. Perhaps it is largely because of his interest in students and his example as a psychologist that I am writing this today. by mother and father. Mr. and Mrs. Antonio Gambaro, have always contributed in ways that they could. They have given encouragement to looming and have always assisted financially when needed, even when it was difficult to do so. Finally. I wish to say that without the love and companionship of my wife Marcelle. graduate studies would never have been begun. She has made the difficult times lighter and the good times happier because they were shared. ii TABLE OF CONTENTS Chapter I. III. IV. V. VI. MRO mo T ION C O O I O O O O O O O O O O O O O O AimsofThisResearch ............ Ib'pertension: ARelatedProblem. . . . . .. RelevantResearch ............. Prospectus.................. Hemcdynamics Briefly Considered . . . . . . . Anger .................... Motion and Transient Rises of Blood Pressure RelevantResearch ............. The Problem of Specificity . . . . . . . . . . The Constructs of (hilt and Defense Preference (hilt ................... DefensePreference............. mom AND mmm O O O O O O O O O O O O O mnon O O O O O O O O O O O O O O O O O O O O O A. Subjects . . . . . . . . . . . . . . . . . B. InstrumentsandApparatus ........ C.Procedure................ D. TreahnentoftheData .......... WTS O O O O O O I O O O O O O O O C O O O 0 Additional Findings and Data From Dmloratory Aspects Of tha S tudy O O O O O O O O O O O O O O O O DIS cuss 1m 0 O O O O O O O O O O O O O O O O O O SUMMARYANDCONCLUSIONS ............ mom 0 O O O O O O O O O O O O O O O O O O O O 0 AWDICE O O O O O O O O O O O O O O O O O O O O O 0 iii 0 O O O O O O O O O O O O O O O O O O O 0 O O O 0 C O O O O O O O O O O O O O O O O O O O O I O O O O O O O O O O Page \lmtt-‘H H 12 15 15 SEE-{'2' 25 26 26 26 32 37 41 $5 66 70 71+ LIST OF TABLES AND FIGURES Table Page 1. Summaryofoperationaldesign............... 39 2. Diastolic blood pressure increases in anger . . . . . . . . 1+2 3. High (hilt and Low (hilt sub-group diastolic increases in thethreeangerstimulatedgroups . . . . . . . . . . . . 1+3 1}. Diastolic blood pressure changes after the expression ofangerorfill-inactivity . . . . . . . . . . . . . . 45 5. Sumary of diastolic decreases of the several groups and Iowmilt,Highmiltsub-groups ............ 148 6. Systolic blood pressure changes after the expression ofangerorfill-inactivity . . . . . . . . . . . . . . 50 7. Correlation between defense preference and degree of diastolic increase after frustration to anger sequence . 52 8. Correlation between defense preference and degree of diastolic decrease after expression of anger . . . . . . 53 9. Correlation between defense preference and degree of systolic increase after frustration to anger sequence . . 54 10. Correlation between defense preference and degree of systolic decrease after expression of anger . . . . . . . 55 Figure Page 1. Schema to demonstrate the main features of the circ‘llatow ”at” O O O O O O O O O O O O O O O O O O O 8 2. Curve of blood pressure throughout the circulation from leftventricletorightauricle............. 9 3. Curve of velocity throughout the circulation - from left mtr1CI° to ng‘t auriCIB O O O O O O O O O O O O O O 0 9 14. Diagram to illustrate the storing of energy of the heart's contraction in the elastic arterial walls (at root of aorta) during systole and its release during diastole . . 10 iv LIST OF APPENDICES Appendix A. B. J. ORIENTING INSTRUCTIONS GIVEN TO SUBJEXITS . . . . . . . INSTRUCTIONS GIVEN TO THE SUBJMTS mo WERE STIMULATED TOANERBYTHEASSISTANT ............. INSTRUCTIONS FORTHECOU'NTING TASK . . . . . . . . . . ASSISTANT'S ROLE HIRING THE MANIPULATION TO ANCER mm 0 O O O O O O O O O O O O O O C C O O O 0 INSTRUCTIONS FOR THE MWONAL GUESSING GAME: mOUPs I AND H O C O O O O O O O O I O O O O O O 0 INSTRUCTIONS FOR THE INTERPERSONAL GUESSING GAME: mOUPS III MD Iv . O O O O O O C O O O O O O O O O SUBJECTIVERATDIGOFANCERSCALE. . . . . . . . . . . AGGIESSION-GUET ITEMS OF THE MOSHER INCOMPLETE WMCE mT O O O O O O O O O O O O O O O O O O 0 INSTRUCTIONS FOR THE BLACKY TEST AND THE DESCRIPTIVE STATEMENTS TO BE RANKED FOR EACH CARD. (Only those cartoons Elum identified as hostility situations were used.) 0 O O O O O O O O O O O O O O O O O O 0 DATA PERTAINING TO DIASTOLIC PRESSURE UNDER VARYING CONDITIONS, SUBJETIVE RATINCB OF ANCER AND GUILT POSITION O O O O O O O O O O O I O O O O O O O O O O Page 75 76 77 78 79 8O 81 82 83 86 CHAPTER I INTROWCTION ADIS OF THIS RESEARCH The primary purpose of this investigation was to study the inter- relationship of a psychological state (anger) and a concomitant physiological change (blood pressure). with the intention of contributing to the understanding of the clinical syndrome of essential hypertension. Further, in a more exploratory fashion. the variables of guilt and defense preference. as they relate to blood pressure changes under certain conditions. were investigated. HYPERTENSION: A RELATED PROBLEM There is a clinical syndrome which is knovm'as “essential hyper- tension.“ The use of the term ”essential" is to distinguish between two major types of arterial hypertension. In one type. called primary. or essential hypertension. the condition occurs unpreceded by known pathological changes in the kidneys. the vascular system or other organs, and in which emotional factors are believed to play an important role. Essential hypertension. therefore, may cause but is not itself the result of kidney disease. In the other form, i.e. renal hypertension, the Impertension is a consequence of chronic kidney disease (Best and Taylor, 1958). The clinical syndrome of hypertension appears in the form of chronically elevated diastolic blood pressure. Usually the systolic pressure is also raised. but it need not be. Also, systolic hypertension 1 may occur without a diastolic pressure elevation. Systolic hypertension alone is not believed to be clinically important unless very high levels threaten the integrity of the blood vessels. In the more important diastolic lvpertension. the chronic, abnormally high pressure is due to spasms of the arterioles. This condition leads to serious consequences; it throws extra work upon the heart and subjects the arterial system to undue strain (Best and Taylor. 1961). Most of the damage caused by the persistence of high blood pressure can be summed up in one word-- sclerosis. This scarring of the walls of the arterioles and arteries ultimately leads to closure of some of these vessels (Page. 1951). The organic tradition in medicine has generally favored a purely physical view of the etiology. pathogenesis and treatment of essential hypertension. There is a psychosomatic approach. on the other hand. which does not ignore the physical prohlus involved but includes a consideration of the role of experience and emotions. It emphasizes the multiple factors involved in etiology and pathogenesis. Briefly stated. the psychosomatic theory is as follows: During the early phase of hyper- tension marked lability of the blood pressure is common. mis lability makes it probable that no permanent. structural vasoconstriction or sclerosis exists. The transient blood pressure elevations in the pre- mertensive are easily induced by psychic factors. then the emotions causing the transient mortension are prolonged over long periods. permanent vasospasm and hypertension occur. The theory supposes that long. continued excitation of the mediating mechanisms causes sclerosis and permanent mertension (Alexander. 1939). One of the most energetic "schools'I of research on hypertension is that which promotes a "repressed-hostility" hypothesis. Beginning with Alexander's (1939) theoretical discussion. this hypothesis asserts that hypertension develops as a result of the chronic repression of feelings of anger. This repressed anger or tension is somatically «pressed in increased blood pressure. which. over time. leads to the permanent vascular changes associated with twpertension. Alexander's speculations were based on the psychoanalysis of hypertensive patients. A number of other clinical observers have also worked along the line of Alexander's repressed hostility hypothesis (Singer, Ackerwan. Cohn. Schroeder and Steel. 19%; Milton. 1961; Saul. 1939; Schwarts. 191w; Weiss. 19142). Their general conclusions seem to be that although all varieties of character and neurotic disturbances occur in hypertensive individuals. a common problem seems to be the presence of ”emotional tension" due to I'chronically repressed hostility.“ These investigators feel that their findings are consistent with this sort of causal sequence: 1) g inhibits impulses to action (insituations where others would aggress). 2) thereby raising his .blood pressure. 3) Such frequent inhibitions result in heightened blood pressure outlasting the inhibition period; and 1+) eventually. S's blood pressure remains at a high level. He is always under stress. Their method is that of interpreting the data obtained during psycho- therapy. Disregarding for a moment the question of the method itself, these observers could not avoid errors due to preselection of the material which they studied. It will remain an open question as to how representa- tive of the general papulation of hypertensives are those in psychotherapy; it is quite likely that they are a small and select part of the pepula- tion. However. the observations of such a large group of keen clinical observers cannot easily be dismissed. regardless of their methodolog. and in this case they led to a series of‘primarily correlational studies. Relemt Research One of the most creative tests of the hostility hypothesis was done by Saul. Sheppard. Selby. Sachs and Master (l95h) . They asked 17 chronic hypertensives and 16 non-hypertensive patients to report their dreams. The gs did not know the experimental hypothesis. Over a two month period 78 dreams were collected. These were rated for hostility by two judges who did not know the hypothesis being investigated and did not know the S's blood pressures. Dreams of hypertensives had content rated sig- nificantly higher for hostility than did dreams of the non-hypertensive patients. In a similar study. Kaplan, Gottschalk. Magliocco. Rohobit and Ross (1961) compared a group of 10 hypertensive patients with 10 non- hypertensive patients on measurements of hostile content in samples of five-minute verbal productions and hypnotic dreams. Their results showed a striking difference. The Importensive group's productions showed a greater mean amount of hostility (p < .002) as measured by this method. Milton (1961) investigated the relationships over time among blood pressure. personality dynamics discussed during therapy. and overt behaviors for a single hypertensive patient. A 38 year old. white. male hypertensive served as the S. Over a nine month period. interrelations among 15 variables were factor analysed and an approximate simple structure was achieved. Five factors were isolated. Two of these factors were related to blood pressure and they involve hostility and identification conflict. In all three of the above experiments, however. the possibility that the hostile verbal content might be due to eleva- tion of blood pressure, rather than vice versa, or that both might be dependent variables of some other critical factor, cannot be discounted. A study which attempted to get at this question (Gottschalk. Gleser. D'hnura and Hanenson. 1961+) reported the relationship between blood pressure and hostility and also the changes that would occur in this relationship following continuous administration of a diuretic agent, such as hydrochlorothiazide. which is capable of significantly reducing the blood pressure of hypertensive patients. (It was hoped that this experimental aspect of the study might provide some clues as to whether the hostility might be a result. in part, of pathOphysio- logical factors rather than simply a causal factor in the elevated blood pressure). Twelve hypertensive female patients with an absence of known renal or hormonal pathology were observed. Taped five minute speeches in response to stmdardized instructions from the same examiner were rated for levels of "hostility inward” and ”hostility outward." Their results report that statistically significant positive correlations (p < .05) were found between hostility inward levels and average systolic and diastolic blood pressures. Sigiificant negative correlations (p < .05) were found between hostility outward levels and blood pressure. No significant changes in hostility levels occurred while women were taking hydrochlorothiaside even though there were significant decreases in blood pressure during this time. Not all studies present results which support the repressed hostility hypothesis. A study which presents ambiguous results. at best, is that by Hatarazzo (195“). He had gs. well matched fer age, sex.and socio-economic status. but with either hypertensive or normal blood pressure, tell stories to TAT pictures. As they were telling the stories. theirjperfbrmance was criticized by the investigator. Pro-test indices of aggressiveness were obtained from the Rosenzweig Picture-Frustration Study and the Allport Ascendance-Submission Scale. TAT stories were coded for aggressive content befOre and after the criticism. which began after several stories had been told. There were no differences between hypertensive and normal blood pressure gs on either'pre or post-test aggressiveness using these variables. However. 12 out of 40 of the non-hypertensive So refused to continue telling TAT stories when criticism began while only 1 out of the #0 hypertensive gs quit. Matarazzo interprets quitting as aggressivity toward §,and notes that the hypertensives, who are at least theorized to be less Openly aggressive, generally did not quit. Neiberg (1957) gave a multiple choice form of the Rosenzweig P-F Study to a group of hypertensive Se and non-hypertensive So before and after a criticism experience. Subjects were criticized while going through the wisconsin Card Sort. Verbalized aggression by §_fbllowing stress was recorded. He found no differences in pre orIpost-stress aggressivity between the hypertensive and the non-hypertensive gs. PROSPECTUS In an effort to experimentally investigate those aspects of the psychosomatic theory of hypertension which state that the inhibition of aggression will raise blood.pressure and which imply that the expression of aggression will lower blood pressure, basically two questions are considered in this research: ”What effect do psychological stimuli, designed to arouse anger, have on blood pressure; and what is the effect on blood pressure of'directly and indirectly expressing anger following these stimuli‘!’I Also considered in this research, but in a more exploratory manner, are the variables of "defense preference" and "aggression-guilt“ as they relate to blood pressure reactions in the expression and non-expression of anger’following the experimental manipulations. HEHODYNAMICS BRIEFLY CONSIDERED Arterial blood pressure is the result of'the discharge by the heart into the arterial system of a volume of blood which cannot all escape through the peripheral vessels into the venous system much befOre the next heart beat occurs. The arterial system during life is at all times overfilled. That is, the elastic arterial walls are always more or less stretched. The height of the blood pressure is therefore dependent upon two main factors: the output of the heart. and the caliber of the peripheral resistance (Best and Taylor. 1958). The small vessels at the periphery-the arterioles and the capillaries--but especially the arterioles, offer resistance to the passage of blood from the arterial system into the veins. Within the closed circulatory system. illustrated in Figure l, we can see that the dilation or contraction of the arterioles can control circulation and pressure in the whole system. '0 No.3 8 I" (0?; “QV‘IQS {In-I9 rIw/{é'Jn— _—,\"'~_" g,— ‘ ..... l Vet'ns A... a. A Fig. 1. Schema to demonstrate the main features of the circulatory system The arterioles in the body represent stapcocks through which. by alterations in their caliber, the volume of blood flow entering the capillaries is metered. The arterioles are supplied with nerves which, by bringing about contraction or relaxation of the rings of smooth muscle. control their calibers. Constriction of the peripheral vessels reduces. temporarily, the outflow from the arterial system and a rise in arterial pressure occurs. provided the minute volume of the heart is not at the same time reduced. The arterial system must. therefore. contain more blood than formerlyt and the arterial walls are more fercibly stretched. The pressure then rises to the point at which the quantity of blood leaving the arterial system again equals that which is pumped into the heart; no further rise in pressure then occurs (Best and Taylor, 1958). we can see in the illustration of Figure 2 that within the vascular system a large part of the ventricular contraction is expended in overcoming the resistance offered by the small calibered arterioles. If the output of the heart remains unchanged. any constric- tion of the arterioles would further increase blood pressure. ' .3 In I: | ‘2 ' I m. I s. I a. I30~ .1 A .2 l ~21 . I :E | .9 : ‘IO‘ : I '5. I I ° 1 l b ‘ 60" ' I l 30‘r | I o .1 Fig. 2. Curve of blood pressure throughout the circulation from left ventricle to right auricle (Best and Taylor. 1958. p. 130) The velocity of the blood also varies in different parts of the vascular system but it does not, like blood pressure. show a continuously pro- greesive diminution from the aorta to the right ventricle (see Figure 3). Ies M. m. PM See . So '0 ‘Ioo... ”a ‘Ie Hes --—’— I I I I I I I I I I I I I I ‘ I I Fig. 3. Curve of velocity througiout the circulation - from left ventricle to right auricle (Best and Taylor. 1958. p. 136) 10 Each ventricular systole forces blood into an already filled and stretched aorta. partly by further distention of the aorta and its branches. During diastole the gradual recoil of the elastic walls causes the flow of blood to continue out of the arterial system through arterioles into capillaries. The steady flow through capillaries is thus achieved by the compression chamber function of the aorta. and the volume of capillary flow is determined by vasomotor changes in arterioles which act as stopcocks of the circulatory system (Wiggers. 1952). In Figure it we can see how the elastic recoil of the arterial wall acts as a pump to keep the blood moving forward. SysTour , II: 3 ll DooS'I’OIC- —- ’TI‘ K’A- A----7+ ~\ N ~l—Lii’l" / \J\\.’I\ TJ-"/’ Fig. A. Diagram to illustrate the storing of energy of the heart's contraction. in the elastic arterial walls (at root of aorta) during systole and its release during diastole. Small arrows indicate the direc- tion of movement of the arterial walls and of the force exerted during the two phases. Large arrows indicate direction of the blood flow (Best and Taylor, 1958. P- 150). 11 The whole pressure curve is designated the "pressure pulse." The maximal pressure reached during systole is the systolic pressure; the minimal pressure realized at the end of diastole is the diastolic pressure, and the numerical difference between these represents the pulse pressure. The expansion of the vessels which the pressure pulse produces in any part of the arterial system is the pulse (Wiggers. 1952). Although blood pressure has been measured for more than half a century, there is still little agreement on what constitutes "nomal" blood pressure (mm, 1962). This is due, in good part, to the fact that arterial blood pressure is influenced by a number of physiological and psychological conditions-age, muscular exercise, digestion, emotion, etc. In the newborn baby. the systolic pressure averages 40 mm. Hg. but reaches a value of around 80 mm. by the end of the first month. It rises more gradually throughout childhood, attaining a level of about 100 mm. by the twelfth year and 120 or so by the seventeenth. It ranains around this level for the next five to eight years. After about the twenty-fifth year, the renewed increase in systolic pressure averages about 0.5 mm. Hg per year (Best and Taylor, 1958). From "blood pressure by age norms" (Fishberg. 1939), general guidelines have been established for what to consider high or "abnormal" blood pressure. As a practical criterion, the physician assumes that normal systolic blood pressure. taken after the subject has had a short rest, 13 in the range of 100 plus his age in years. The rule for diastolic is a little more than one-third less than the systolic pressure. 12 The systolic level of blood pressure is mainly determined by the activity of the heart. whereas the diastolic level is a tonic mnction of the peripheral arteries (Page, 1951). The diastolic level is the more important of the two in hypertension (Best and Taylor, 1961) and the one we will be most concerned with in this experiment. It is the "run off” pressure from the great vessels into the tissues. and is the pressure which must be developed by the heart before it can even begin to eject blood into the vessels (Page, 1951). ANGER Anger is considered to be an emotional reaction with prominent skeletal-facial components. It may be conceptualized as a drive state; like anxiety in this respect (Buss. 1961). Because the expressions and characteristics of anger undergo modification. during socialization. the skeletal and facial aspects of anger are best seen in children. Children can be taught to modulate the expression of anger; a scowl and a threatening position of the arms can be punished, and with continued punishment these expressions can be inhibited. However. while the postural features of anger may be suppressed, physical aspects are not changed by socialization (files. 1961). Society is concerned mainly with responses that have interpersonal consequences, and both overt anger and attacking responses are shaped and inhibited by society. However. the heightened level of physiological activity is less public, and there are no taboos concerning such physiological arousal. Furthennore. it is difficult to control autonomic responses; though they may be conditioned in the laboratory (Razran, 1961). it is extremely difficult to modulate or inhibit them in everyday situations. 13 Thus. the autonomic aspects of anger remain intact while the postural and facial aspects are modulated and inhibited. The pattern of autonomic arousal that occurs in anger constitutes a physiological tension state (Buss. 1961). The individual is aware of this tension state because interoceptors are sensitive to these internal stimuli. but it is difficult to verbalize precisely what the tension state is like. The subject typically reports being stirred up, aroused, tense, excited, tight, etc. (Buss, 1961). The tension state is sufficiently different from the resting state to be discriminated but too diffuse to be described in detail. Folklore has frequently associ- ated this tension state with blood pressure changes occurring with increases or decreases of anger. Note how often we have heard the expression to an angered person, “Now watch your blood pressure!" Folk- lore also holds that an angered person often feels better after striking at the source of his annoyance, either directly through physical actions or indirectly through verbal responses. These thoughts. essentially, contain the notions of Bollard. Doob. Miller, Mower and Sears (1939) on the ”catharsis hypothesis." Specifi- cally, the catharsis hypothesis reads: "The occurrence of any act of aggression is assumed to reduce the instigation to aggression" (p. 50). Extrapolating from their notions conceming the joint operation of displacement and catharsis. we are led to believe that displaced aggression. even though forced. might also lead to some degree of arousal reduction. Thus we might expect a reduction in tension level. as measured by blood pressure. in a person who not only aggresses against his frustrator but also in one who aggresses against a 1'4 'non-frustrator." The response is a displaced one. however, and in relation to anger at least. we would not expect as great a reduction in “arousal level“ as we would if the response was direct. The bodily and chemical changes that occur in anger are by no means clear. In recent years adrenaline and noradrenaline have been assigned central roles in attempts to account for physiological changes in both fear and anger (Cbldenberg, Pines. Baldwin, Green and Rob. 19148). These investigators found that injections of adrenaline caused a sharp drOp in peripheral blood vessel resistance and "striking increases" in cardiac output. Injections of noradrenaline caused "striking increases” in peripheral resistance with little or no change in cardiac output. In a review of this, and other studies which com- prise only a paucity of knowledge about bodily changes in human anger, Buss (1961) summarizes the following account of human anger: . . . An anger stimulus impinges on a receptor. which initiates an afferent neural impulse to the cerebral cortex. The cortex in turn stimulates the hypothalamus. the amygdala. and related structures in the rhinencephalon (which are evidently necessary for an integrated rage pattem). The hypothalamus relays impulses in two directions. First as the “head ganglion" of an autonomic nervous system. it initiates or relays impulses to the sympathetic nervous system. The action of the sympa- thetic nervous system is only moderate in anger. there being only slight increases in heart rate, galvanic skin response, etc. The hypothalamus also stimulates the adrenal medulla to secrete both noradrenaline and adrenaline, with the major com- ponent being noradrenaline. Noradrenaline in overflow quantities may also be produced at sympathetic nerve endings. and its effect is a large increase in peripheral vascular resistance. a rise in both systolic and diastolic blood pressure, and a diminution in the activity of the stomach and intestines. The changes are different from these in fear. but there is an over- lap in the physiolog of fear and‘anger (p. 105). It may be that the actual carrying out of an aggressive response is accompanied by the release of adrenaline. Adrenaline elevates 15 cardiac output so that more blood is delivered; and it acts as a vasodilator, allowing the blood to reach skeletal muscles and decreas- ing peripheral resistance. In decreasing the peripheral resistance the diastolic blood pressure level would also be decreased. EMOTION AND TRANSIENT RISES OF BLOOD PRESSURE It seems necessary, before getting into our’problem area, that we should supply a sampling of evidence that the vascular system is responsive to emotional stimuli and conditioning, resulting in at least a transient rise in blood pressure. That transitory blood pressure changes might fellow changes in mood or stimulation was systematically noted as early as 1773 by Stephen Hales (Hamilton, 19h2). However, although it has long been well known that changes in circulatory dynamics may occur as part of the total response to physical and psychological stress, many details of the interrelationships between the emotional and circulatory phenomena so evoked are still unclear. In fact, the problem that plagues the cardiologist even today (Ackner, 1956) is that he sees evidence daily of the influence of emotion on the circulation, but has at hand no extensive body of'data that might illuminate his clinical observations. Relevant Research Wilhelmj, Homire, McDonough, Waldmann and McCarthy (1953) took continuous blood pressure readings from trained dogs and demonstrated that many trivial factors such as strangers in the laboratory, noise and confusion in the kennels, changes in the daily experimental routine, 16 an unexpected change of experimenters or the presence of a female dog in estrus may cause pronounced and sudden elevations of blood pressure. A Russian experimenter, Andreev (1952), demonstrated. with dogs as subjects, that it was possible to develop conditioned-reflex heightening of blood pressure to sound. He reports that '. . . the conditioned increase in blood pressure is maintained for months without reinforcement and is accompanied by changes in the physiochemical preperties of the blood serum as well as by an increase in the amount of vase-constrictive substances in the blood." In an earlier study by Menzies (1937). very interesting for its implications, vasoconstriction was elicited by imersing one hand in ice water; warm water was used to produce vasodilation. The vaso- constriction was suocessfully conditioned to: (a) the sound of a bell: (b) to the sound of a buzzer and to the subject 's whispered repetition of a nonsense word; (c) to movements and postures of various parts of the subject's body; (d) to a pattern of light. The frequency and the extent of both the original and the vasomotor changes were detennined by taking continuous records of skin temperature from one hand. A themopile and high sensitivity galvanometer provided a means of measuring slight changes of temperature. This experiment is highly interesting for its demonstration that changes in the blood vessels can be controlled by verbal processes. It indicates that autonomic re- sponses may become linked to symbolic activities such as bodily attitudes, gestures, implicit speech and imageryy The experimenter himself concludes: It has been found to be relatively easy to establish a sym- pathetic conditioned reflex in man. Using the vasoconstriction 17 due to dipping the hand in water as the unconditioned response, and appropriately coupling it with a conditioned stimulus (buzzer, light, or verbal stimulus), the conditioned stimulus soon becomes effective in evoking vasoconstriction. It is of particular interest that the intensity of the conditioned response was greater 30-60 days after the conditioning than it was immediately after the conditioning period, although no reinforcement was employed during the interval. In this work even recalling or thinking of the conditioned visual stimulus sufficed to evoke the vasoconstrictor effect in some subjects. This suggests that any factor associated with an emotional situation may evoke sympathetic discharges long after the crucial event (p. 115). There seems to be little or no disagreement as to the importance of emotional factors in causing circulatory changes. But the problem of how important such factors are is difficult to answer quantitatively and accurately. Subjects can be emosed to comparable stimuli or environmental factors. but the emotional significance of a given factor to the individual determines the occurrence and to a large extent the severity of the bodily response to it. Characterizing the feelings experienced by subjects is difficult, nor is there any certainty that such feelings as they describe retrospectively may have been present at the time of examination. In sumary (and in reference to the present proposal), it is recognized that psychological stress is not just something external which can be applied; for its occurrence depends on factors of emotional reaction inherent and conditioned within the individual. mly the external factors pro- ducing stress can be clearly defined. THE PROBLEM OF SPECIFICITY Cannon's classic work (1932) describing the physiological prepa- ration for "fight or flight" in animals has been taken as the basis for attractive hypotheses concerning the possible role of acute and 18 chronic emotional states in the genesis of various psychosomatic dis- orders in man. Yet Cannon's data do not supply clues as to any differentiation of physiologic responses in association with specife ically different emotions. In fact, his work suggested that emotion was a generic term to apply to an emergency pattern of physiologic changes regardless of the psychological components, i.e. anger or fear. In his description of the bodily changes in reaction to stress, con- siderable emphasis is laid upon the diffuse character of the reaction. This diffuseness was said to be related to the widespread effects of sympathetic stimulation and the release of adrenalin into the blood stream (Cannon, 1932). Efforts to clarify the mechanisms and specificity of emotionally induced circulatory changes have led to conflicting results. Some investigators have advanced data to support the notion that patterns of circulatory response are specific fer each individual and that a single person will react similarly regardless of the implications or nature of the stimulus for emotional experience (Lacey, 1952; Lacey, Bateman and Van Lehn, 1953; Halmo and Shagass. l9u9). Other workers submit evi- dence in support of the opposite thesis: that the pattern of circulatory activity is specifically related to the nature of the emotional response evoked in the subject and that a single individual will exhibit consistent and different circulatory responses to different emotional states (Ax, 1953; Funkenstein. 1953: Walf and Wolff, 19%). In line with this later general hypothesis, we would expect distinct patterns for anger and fear across individuals despite individual differences. Thus we would say that anger tends to increase blood pressure; in one 19 subject the increase might be small and in another subject it might be large, but both subjects should show an increase. According to this formulation the direction of blood pressure change is always up, regardless of individual differences in extent of change. However, the prevailing view until recently has been the hypothesis put forth by Cannon (1932) that emotions cannot be differentiated on the basis of physiological changes that accompany them. The physiological components of anger and fear were believed to be the same and were treated under the heading of "emotion." , The first definitive study to challenge this view was offered by Ax (1953). Forty-three subjects were stimulated in the laboratory to ”fear" and ”anger." The fear stimulus consisted of a gradually in- creasing intermittent shock stimulus to the little finger which never reached an intensity sufficient to cause pain. men the subject reported the sensation, the experimenter expressed surprise, checked the wiring, pressed a key which caused sparks to jump near the subject, and then exclaimed with alarm that this was a dangerous high-voltage short circuit. The anger stimulus was provided by an assistant who entered the room stating that he must check the wiring because some of the calibrations might be off. He then turned off a radio, criticized a nearby nurse and told the subject sarcastically that it would have helped if he had been on time. He checked the electrodes, roughly adjusted the subject and criticized him for moving, non-cooperation and other behavior. The stimulation periods of fear and anger were alter- nated so that 22 subjects received the fear stimulation first and 21 subjects received the anger stimulation first. During these states the 20 following physiological reactions were recorded: (1) heart rate, (2) face temperature. (3) ballistocardiogram, (h) respiration rate, (5) hand temperature, (6) skin conductance, (7) integrated muscle potential, (8) systolic and diastolic blood pressures. The scores used were the maximum rise and maximum fall from the preceding resting level and the number of responses of a critical value per unit of time. Of the scores thus obtained. 7 showed significant discrimination between anger and fear. gastolic blood pressure rises, heart rate falls. number of rises inskin conductance and average intmsity of muscle potential increases, are gLeater for anggr than for fear, whereas skin conductance increases, frequency of muscle potential in- creases and respiration rate increases were greater for fear than for anger. Similarly, Schadrter (1957) exposed subjects to situations intended to produce emotions of pain, fear and anger in that order. For the pain experience subjects kept their hands in a bucket of ice water (3 degrees centigrade for 1 minute). The anger and fear situations were similar to those used by Ax. Subjects responded to these stimulus conditions with a significant increase in diastolic blood pressure to the pain and anger conditions and a significant systolic increase to the fear condition. Another experiment along this line (Funkenstein, King and Drolette, 1957) used frustrating tasks to induce anger. The results were similar to Schachter's. Buss (1961) compared the results of the above three studies in terms of blood pressure changes. He asserts that in general these experiments support the notion that physiological emotion is not as 21 Cannon hypothesized. that is, all the same emergency pattern. as also concludes that for fear (or anxiety) stimuli there appears to be a consistent blood.pressure pattern of response: a rise in systolic blood pressure (and also heart rate) and little or*no increase in dumflolic pressure. For anger’there appears to be a consistent rise in diastolic pressure. Although these results could be taken as a refutation of Cannon's hypothesis of a unitary visceral excitement reaction, it could still be argued that there is universal visceral excitement but with different organs reacting in intensity according to the type of emotion experienced. THE CONSTRUCTS 0F GUILT AND DEFENSE PREFERENCE eat As stated, the exploratory aspect of this research is concerned with the variables of aggression-guilt and modes of defense. Do people who score high on a measure of guilt differ from people who score low on a measure of guilt in their blood pressure reactions to expressed or unexpressed anger? Is a person's defense preference a variable which is related to the blood pressure reactions elicited by an anger provoking situation? Historically, the guilt construct has played an important role in psychoanalytic theory. Guilt and the superego are seen as intimately related conceptualizations (Fenichel, l9h5). The superego-the incor- porated moral standards of society-is the structural component of personality concerned with coercing the ego, or reality oriented aspect 22 of the psyche, into considering its moral, ethical or religious standards of proper behavior. The superego, which is said to develop with the passing or*partial resolution of the oedipus complex (anichel. l9h5), influences the ego through the medium of gilt feelings. In psycho- analytic conceptions of guilt, several important considerations become apparent. Guilt refers to painful feelings of self blame, self criticism and self remorse which arise from internalized standards of proper behavior. The violation may be real or imagined. Chilt may result from failure to attain goals and to live in accordance with ideals, as well as by transgressions of internalized standards of’proper behavior. Some investigations have evaluated guilt as an inhibitor (Mackinnon. 1938; Clark, 1952; Radar, 1957). Collectively, the studies of Hackinnon, Clark and Rader’provide evidence supporting the hypothesis that guilt serves to inhibit morally tabooed behavior, such as cheating, the aggression of verbal agggession and the telling of sexual stories to TAT cards. Alexander (1938) also stresses the inhibitory effect of guilt: As a fbrm of anxiety, the fearful expectation of an inevit- able and deserved suffering. the sense of guilt is primarily an inhibitory phenomenon. Under its pressure the individual is apt to avoid the expression of’those impulses which have evoked and contributed to his guilt feelings (p. #3). From the psychoanalytic viewpoint, a possible transgression of taboos circumscribing aggressive impulses would arouse guilt. Thus we might expect the inhibitory phenomenon of guilt to reflect itself somatically in a situation where a person is angered or called upon to aggress. 23 Defense geference It was after psychoanalysis shifted the emphasis of psychiatric thinking to the irrational aspects of man's thinking that a dynamic view of human functioning was formed and the notion of defense mecha- nisms evolved. Thus the concept of defense mechanisms also originated from psychoanalytic theory (Freud, 1936). Freud (1936) used the term "defense'I to mean a protection of the ego against the instinctual demands of the id. Since that time the meaning of “defense" has been considerably enlarged by the Ego Psychologists and the origin of defense preferences is usually seen from one of four points of view. (Thelen, 1961}, gives an overview of the literature covering these four perspectives.) Since the present study is concerned only with the relationship between type of defense most used in aggression-defined situations and blood pressure reactions, and not with the origin of defenses, we will avoid this broader issue. In general the Ego Psychologists' position on defenses is that defenses involve a type of “perceptual maneuvering" to escape experi- encing anxiety. For enmple, in the book The $2 and the Mechanisms of Defense (193?) . Anna Freud expanded the meaning of "defense“ to include dealings with external reality and danger. From her initial expansion of the concept the broader definition was arrived at which generally accepted the meaning of defenses as being any behavior designed, usually unconsciously, to protect an individual from the discomfort aroused by an anxiety provoking situation. In relation to hostility, defenses can change one's perception of a hostile situation, thereby rendering it innocuous. For purposes of this study, the broader definition is accepted. 24 Our exploratory concern with defenses here is with their differen- tial relationship with one aspect of‘physiological functioning. For example, do blood pressure changes in an angereprovoking or anger expressing situation vary according to the typica1.mode of defense used in aggression situations? Is the degree of change related to the degree of use of, say, "reaction formation?" ”intellectualization?" "avoid- ance?" "regression?" “projection?" CHAPTER. II mom AND HIPOTHESES As stated. basically two questions were considemd in this research: "mat effect do psychological stimuli, designed to arouse anger, have on blood pressure; and what is the effect of directly and indirectly expressing anger following these stimuli?” In line with these questions the following hypotheses were tested: 1. 2. The stimulation to anger of experimental subjects will significantly raise mean diastolic blood pressure readings. The direct expression of anger following a period of frustration will significantly lower mean diastolic blood pressure readings below that of a group not allowed expression of anger following frustration. The indirect expression of anger (not at source of frustration). following a period of frustration. will significantly lower mean diastolic blood pressure readings below that of a group not allowed expression of anger following frustration but will not lower it as much as direct expression will. In addition to investigating these main hypotheses, this research explored the relation of aggression-guilt and psychological defenses to the degree of blood pressure changes in the expression and non-expression of anger. 25 CHAPTER III METHOD A. Subjects Subjects used in this study were 80 male undergraduates attending Michigan State University who were required to serve as research subjects as part of the requirements of enrollment in introductory psychology courses. The data on sex differences in anger are certainly not conclusive, but there is evidence that men tend to become angrier than women and tend to ventilate their anger'more (Buss, 1961). Use of’men and women may tend to obscure otherwise consistent physiological patterns. Men and women may even have different physiological patterns in anger. B. Instruments and Apparatus 1. Blood Pressure measurement Blood pressure was measured indirectly in the subject's left arm by means of an E and M Physiograph, utilizing an Electrosphygnograph transducer. The Electrosphygmograph is a transducerapreamplifier for the recording of indirect systolic and diastolic blood pressures of humans and many animals. The Electrosphygmograph combines a pressure transducer and an amplifier to produce single channel recordings of occluding cuff pressure and superimposed Korotkoff sounds.1 1These are the sounds heard over an artery with a mechanical listen- ing device. As the air'pressure in a cuff which circles the subject's arm is ually decreased and the blood resumes its flow through the once occ uded artery, faint tapping sounds are heard which are synchronous with the heart beats. 25 27 The method of blood pressure measurement with this instrument can be best illustrated by the description of another, more common technique. This method is the one used in clinical medical practice in which an instrument called a gphggomanometer is used. The method of measuring blood pressure with this instrument is called the auscultatory technique because the examiner employs a listening device, a stethoscgpe, in order to make his measurements. To take a blood pressure reading the eminer encircles the subject's upper arm (usually his left arm) with a rubber tube that can be inflated with air. The amount of air pressure in the cuff is measured by a column of mercury. Hence, blood pressure readings are usually stated in millimeters (mum) of mercury (Hg). The examiner inflates the cuff so as to completely step the flow in the main, or brachial, artery of the arm. He then releases pressure in the cuff until he can hear a beating sound through his stethoscOpe which is applied to the artery below the cuff. The height of the mercury column at the time when he first hears the sound is the sEtolic blood pressure. The examiner then releases more air from the cuff until he can no longer hear the beating sound. The height of the mercury at that somewhat indeterminate time when the beating sound can no longer be heard is the diastolic blood pressure. 7 The autanatic device used in this experiment works in a similar fashion, except that the sounds marking the systolic and diastolic read- ings are recorded automatically. 2. Aggression-Glut Measurement The Mosher Incgplete Sentences Test (MIST) was our instrument for the measurement of aggression-guilt. The test is a 50 stem sentence 28 completion blank which was develOped and validated as a measure of guilt by Mbsher (1961). The responses to the test can be quantitatively scored fer guilt. The scoring of the sentence completions incorporates the referents fer guilt which have been suggested by psychoanalytic theory. The MIST can be scored to obtain a total gpilt score and three subscale scores: hostile gpilt, sex gpilt and.moralitz conscious gpilt. For purposes of this study, only the hostile gpilt stems and score were used. The MIST was devised to measure manifest and relatively covert admission of moralistic standards and feelings of self blame, self remorse and self criticism for failing to live in accordance with internalized moral ideals. Responses are rated on a "guilt continuum,” from '5' to '0', with high guilt being given a high number. Numerous examples for each scoring category are provided. The responses were scored independently by two judges so that an interhrater reliability check could be made. Reliability The inter-rater reliability found in this study was .86. This figure is comparable to the inter-rater reliability of .90 Mbsher (1961) reports for the scoring of his hostile-guilt stems. Construction of the MIST To construct his sentence completion test, Mosher collected 200 sentence stems from three psychologists. The stems were those the psychologists thought would be related to guilt on the basis of notions contained in psychoanalytic theory. The stems were then divided into two comparable sentence completion ferns and administered to 38 hos- pitalized patients in a V. A. hospital. 29 The patients' responses to each stem were listed on a card and two clinical psychologists independently selected the stems which they felt best elicited information on the guilt dimension. Stems which yielded stereotyped responses or extremely brief responses were eliminated. 'lhose stone which elicited a wide range of "guilt-no guilt” responses and on which both raters agreed, were chosen. The psychologists then independently rated the remaining stens into three classes: hostge, 33:5, and morality conscious guilt. The stems they agreed upon for each category were retained as part of the sentence completion test. Finally, the test was administered to 51 males and 60 females in an introductory psychology course. Their responses formed the basis of the scoring emlss which make up the scoring manual. Separate scoring norms were made up for males and females. Construct vgidim of the MIST To investigate the construct validity of the MIST, Mosher used responses to it as the Operational definition of guilt. in an experimental test of a hypothesis generated by theoretical discussion of guilt. Grilt was defined as a generalized expectancy for self-mediated punishment for violating. anticipating the violation of, or failure to attain inter- nalized standards of preper behavior. As part of his dissertation, Mosher (1961) examined the relationship of guilt to verbal conditioning. He predicted that subjects high on hostile guilt would condition less readily to hostile content than subjects scoring low on hostile guilt. The rationale for this hypothesis stewed from the psychoanalytic notion that guilt over hostility should be aroused by the hostile content and 30 lead to a tendency to inhibit the use of the hostile content or to a disruption of the learning process due to the evocation of irrelevant responses related to the guilt motivated inhibitory response. The hypothesis was clearly supported. Crite r t For purposes of this study the median guilt score was calculated. Subjects scoring above this cutoff point were categorized as having high hostile flat. Subjects scoring below this cutoff point were categorized as having low hostile gpilt. 3. Defense Preference Measurement The measure of defenses utilized in this study was the Defense Preference mum (DPI). This is a test which uses the original M Test Cartoons (mum, 1956) to arouse feelings associated with psycho- sexual conflict. mly the cards (i.e., cartoons II, III, IV, VII and VIII) which Blum identified as hostility situations were used. In administration, the subject is given five statements with each cartoon and is asked to rank the five statements in terms of how well they fit the situation. (A copy of the instructions and test can be found in the Appendix.) Each statement represents an Operational defi- nition of one of the five defensive modes; namely, intellectualization. regression, projection, reaction formation and avoidance (the generic term for the repression-denial family). Sporing Objectivity of scoring is achieved by having the subject simply rank order a given set of statements. The total score for each defensive mode is arrived at by simply adding the ranks for each individual statement. 31 RE flit} Slum (1956). using the test-retest technique, obtained reliability coefficients of .145 and .136. Intervals of three to four weeks between test-retest were used. Congtruct vgidfiy of the DPI Blum (1956) feels that the successful employment of the DPI in a variety of research situations (i.e., the hypotheses were confirmed), lends support to the validity of the defense construct and his instru- ment for its measurement. Goldstein (1952). in the first use of this instrument, explored the consistency of defense preferences among a mixed group of male and female undergraduates. He found that subjects who tended to employ the same type of defense with ll'a number of cartoons" manifested more dis- turbance in their spontaneous stories to the test than those subjects with more flexible defense preference. Shire (1951+) also reported that these "general defenders” manifested significantly more maladjustment as measured by the Munro Inspection Technique (with the Rorschach) and on spontaneous stories of the Blacky Test. His measure of the "general defenders” was the DPI. More recently, Thelen (1961;), using the DPI. found that male adolescents were significantly more similar to their fathers in defense preferences than to non-related adult males. ’4. Subjective Rating of Anger Measurement Subjective ratings of anger were obtained from all subjects participating in the experiment. The rating scale consists of six items (see Appendix) to which the subject is instructed to respond by placing a check mark at an apprOpriate point along the line, i.e.. 32 ”Please check one. Knish most honestly expresses your feelings during the past part of the experiment": Very Pleased _; Pleased _; Indifferent _; Mildly Annoyed ____; Angry _; Very Angry _. The responses are weighted from one to six, with the highest rating indica- tive of the highest subjective perception of anger. C. Procedupg The procedure used was essentially that designed by Hokanson (1961). which not only submits the subjects to a frustrating condition but also gives them an Opportunity to overtly aggress by administering an electric shock. This overt aggression is what we will call "the expression of anger." The presence of anger is thus determined by Operational defini- tion--the presence Of an anger inducing situation. Earlier use of this technique (Hokanson, 1959). indicates that subjective feelings of anger are markedly increased in the process, as reported by the experi- mental subjects. The experiment was introduced to each subject as one involving blood pressure responses to routine intellectual tasks and a brief ESP guessing enrperiment. (See Appendix for the instructions given at each stage of the investigation.) Following this, the subject was given a ten minute adaptation period, during which he was sitting at rest and usually engaging in conversation with the experimenter. After ten minutes, the first and resting blood pressure recording was taken. 921.22 1 For purposes of the experiment the subjects were randomly divided into four groups, with 20 subjects in each group. Group I went through the frustration sequence and was allowed "direct expression” by shocking 33 the frustrator. Group I; went through the frustration sequence and was allowed "indirect expression" by shocking the experimenter. Group III went through the frustration sequence but was allowed “no expression" by not allowing them to administer shock. Group IV’did not go through the frustration sequence and did not administer shock. Eppstratlon §gguppce (Groups I. II and III) A crucial role in the ”anger’manipulation” is played by an accom- plice of the experimenter. If the experimenter himself is the frustrating agent he has the dual burden of trying to maintain his subject's interest in the experiment as well as provoking the desired affect. An accomplice can better’play the provocative role because he can more easily assume a peer role and would not have to interact with the subject on a friendly level. The technique involves putting the assistant in some disrepute. He is described to the subject as not the regular assistant but as one who had been fired fer incompetence and arrogance, but due to the illness of the regular assistant he had to be employed fer that day. Thus he is labeled as a suitable target for anger for the subject. This technique attempts to polve the artificiality problem by making the attack appear to be an incidental, unrelated aspect of the laboratory situation. After the first blood pressure reading, the subject was told that the experimenter wanted to see what effect counting would have on his blood pressure. He was then asked to count backwards from 99 to 1 by two's as quickly as possible. When each subject in the three "frustra- tion groups” reached a count in the sixties, the assistant entered the room and gruffly told the experimenter that he should not have begun befbre he (the assistant) arrived. He then turned to the subject and 34 said, ”Stop! I'll tell you when to start again!" In an annoyed tone the experimenter told the assistant to watch the counting carefully, and then occupied himself with another piece of apparatus. The assist- ant then tOld the subject to begin again. During the counting the assistant stepped the subject two more times and criticized the subject for counting too slowly. After the third interruption, the experimenter, again sounding annoyed, told the assistant that he was ”fouling things up again” and demanded that he go busy himself with another piece of apparatus. The experimenter then took another blood pressure recording and immediately administered the subjective rating of anger scale. No {Estratgon gpguence (meup IV) Subjects in the "no frustration condition, " Group IV, were allowed to count backward from 99 to 1 without any interference. After complet- ing the counting sequence blood pressure recordings were taken and the subjects were administered the subjective rating of anger scale. The gpression of AggLession Phase (Groups I and II) Immediately following the ratings, qup I subjects (direct expression) were given an Opportunity to aggress physically against the frustrator (the assistant) by administering electric shocks to him. Group I; subjects (indirect expression) were given an opportunity to aggress physically against the experimenter who took no direct part in the frustration. Group II; and Group IV subjects were not given the same opportunity to aggress. The situation was structured'as follows: Subjects were told that the next task involved a brief experiment in extrasensory perception in which they were to think of a number from 1 to 10, following which the 35 experimenter/assistant was going to try to guess the number. (The instructions and conditions were'identical for Group I and Group II except that the assistant read the instructions and received the shock from (soup I while the experimenter read the instructions and received the shock from m.) may were told that if the experimenter's/ assistant's guess was wrong, they were to signal this error by pressing a button On their panel. The pressing of this button, they were told, would turn on a light on the experimenter's/assistant's panel and give him a shock via the two leads attached to his fingers. The subject was told that the shock was being used so that the effect of pain on his subsequent guesses could be studied. For credibility, each subject wa given a sample of the shock he thought he would be administering (20-30 volts, .5 amps). However, after this the experimenter surrepti- tiously turned a potentiometer which reduced the voltage to a much lower level. The shock received by the experimenter and assistant was very small and felt like a tickle but they responded as though it hurt. The subjects were further told that if the experimenter's/assistant's guess was correct, they (the subjects) were to do nothing. Each subject was allowed to administer five shocks. Blood pressure was again recorded at the end Of this phase. Control Procedpre (Groups III and IV) Subjects in Group g1 and Group IV went through the same procedure except that the signaling of the experimenter's errors was done by simply flashing the light instead Of also administering shock. To con- trol for the condition in which subjects in Group I and Group II were given a shock for credibility, subjects in Group II; and Gropp IV were 36 asked to feel the same shock. They were told that we would also be looking for the effect of this shock on their blood pressure changes. After five incorrect guesses (five light flashes) this phase of the experiment was ended and blood pressure recordings were taken. en cs 0 etion Task for Guilt After the final (third) blood pressure recording was taken, all subjects were given a sheet of paper with the 15 sentence stems on it and told: ”Complete this as quickly as possible and let me know when you are through.” . Defggse aggregggce Task For the defense preference task, each subject was given the five Elacky Pictures along with the appropriate sentences fbr each and the instructions were first read verbatum (see Appendix) and then explained further if necessary. This was administered after the sentence comple- tion task for guilt. The total time spent with each subject was approximately #5 minutes. All 80 subjects were seen within a period of three weeks. Throughout the experiment, all but five of the 80 subjects made their appointments. These five were later replaced by others. This high rate of cooperation was. no doubt. due to the procedure involved in reminding subjects of their appointments. Several days befOre their appointments each subject was sent a letter of reminder and a campus map with the location of the experiment circled. This information was sent in enve10pes bearing the official heading of the quartment of Psychology. 37 D. Treatment of the Data 1. To test our hypothesis that stimulation to anger will signifi- cantly raise mean blood pressure readings. a test of mean differences was required. We wanted to know whether or not the mean blood pressure increase was greater for the "anger group” than for the ”non-anger" group. For this phase of the experiment, Grou s I II and III. which went through the frustration sequence but were not allowed expression of anger, comprised the experimental group. Group IV, the group that did not go through the frustration sequence. was the control. (The control group was similar to the experimental group except that the control group was allowed to finish the counting sequence.) At the beginning and end of the counting and "manipulation to anger" sequence, blood pressure readings were taken for both groups. It was hypothesized that the experimental group, Groups II II and _I_I__I_. would show a significantly higher mean change in blood pressure readings than the control group. Ch‘ogp IV. A I't" test of the mean differences of change was used here with the application of a one tailed test, since the direction of change was also predicted. 2. To determine the effects on blood pressure of direct, indirect and no emression of aggression after arousal. a two way analysis of variance was employed. The other variable involved here was aggression- guilt. The papulation sample was dichotomized into a High Guilt and a Low (hilt group on the basis of the median guilt score for the 80 subjects. In this manner we were also able to determine the interaction effects of guilt and type of expression. Graphically we had: 38 Direct Indirect No No Expression Expression Expression Arousal Mean Change High milt in B.P. Low (hilt We were then able to examine the effects of the type of expression and the interaction when "guilt" was considered to be a variable. The results were examined with and without "guilt" considemd as a variable. 3. The subjective ratings of anger were scored in such a way that the "highest anger" response was scored '6" and the item at the extreme other end of the continuum was scored '1.” Each item is separated by a score of '1." The mean "anger score" for the three groups subjected to the "anger manipulation" was compared with the mean ”anger score” of the group not stimulated to anger. We hypothesized that the "stimu- lated group"I would give a higher mean rating for subjective feelings of anger. A "t' test of mean differences was applicable here. The level of significance used was (p < .05) with the application of a one tailed test, since the direction of the difference was also pmdicted. h. Although we were primarily interested in diastolic pressure because of its greater importance in the clinical syndrome of hyper- tension, all of the above statistical tests involving blood pressure measures were repeated for the systolic measures. 5. To detemine the relationship between degree of blood pressure change in anger, expression and type of defense utilized, a product moment correlational analysis was mm. The degree of blood pressure change in each experimental condition was correlated with each of the five defen- sive modes. 39 source of frus- tration (the assistant). Shock was admin- istered as a result of the experimenter's inaccurate guess administer shock. Table l. Sumary of Operational design Steps in Groups (20 subjects per group) Procedure i III IV Subjects were told that they would be participants in an 1. experiment to Same Same Same determine blood pressure reactions to intellectual activities. After a ten minute rest. systolic and 2. diastolic blood Same Same Same pressure recordings were made. Subjects were given Subjects were a counting task. allowed to count They were inter- all the way back 3. rupted fbur times Same Same 'without the while counting. interference the treated gruffly other groups and not allowed to received. complete the task. Blood pressure h. recordings were Same Same Same taken. Subjective ratings 5. of anger were Same Same Same obtained. Subjects were This group This group went This group went allowed to 'ex- aggressed indi- through the same through the same press anger" rectly (shock procedure except procedure except through a shock to experimenter). that their that their apparatus. This efforts would effbrts would group aggressed only turn on a only turn on a 6. directly at the light and not light and not administer shock. (continued) Table l . continued W M Steps in Groups (20 subjects per group) Procedure I fiIlZL JILL IV during the "inter- personal guessing 6. game." Each sub- (cont'd.) ject was allowed to shock exec tly five times. Blood pressure recordings were 7. taken at the end Same Same Same of this phase of the experiment. The sentence com- pletion task to be scored for guilt 8. (MIST) and the Same Same Same Defense Preference Inquiry (DPI) were administered. CHAPTER IV RESULTS HYPOTHESIS I: The stimulation to anger of experimental subjects will signifi- cantly raise mean diastolic blood pressure readings. It was hypothesized’that if our experimental procedure was effec- tive in arousing anger, we would see this manifested in elevated diastolic blood pressure recordings for our three experimental groups involved in the frustration sequence. As well as predicting a diastolic blood pressure rise in these groups, we expected that the change would be greater than that seen in the control group which did not experience the frustration sequence. To make this initial comparison, mean dias- tolic recordings for our three experimental groups (Groups I. II and III) were taken befbre and after the frustration sequence. The post- frustration mean of 87.3# was compared with the pro-frustration mean of 79.38 (Table 2). Using a one tailed test of significance, the post- frustration mean was found to be significantly greater (p <:t02). This mean increase of 7.96 m. was then compared with the mean increase in blood pressure of our control group, the group which was allowed to complete the counting sequence. The mean blood pressure of the control group actually dropped slightly, resulting in an average change of -.2 mm. When compared to our average increase of +7.96 mm. in the experi- mental groups. we fbund the difference of these differences to be 41 142 significant beyond the .001 level. A one tailed test was used again, since the direction of the difference was predicted. Table 2. Diastolic blood pressure increases in anger Experimental Control Groups moup (N=®) H gomo "IWflffliflfljflmflflfliflfl'flffliflmfli“