University of Southern California Dissertations and Theses

Differential effects of epinephrine and propranolol on shuttle box avoidance learning in rats of different ages

(USC Thesis Other)

Differential effects of epinephrine and propranolol on shuttle box avoidance learning in rats of different ages

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DIFFERENTIAL EFFECTS O F EPINEPHRINE A N D P R O P R A N O LO L O N SHUTTLE B O X AVOIDANCE LEARNING IN R A TS O F DIFFERENT A G E S b y John Jackson Herr A Dissertation Presented to the FACULTY O F THE G R A D U A TE S C H O O L UNIVERSITY O F S O U TH E R N CALIFORNIA In Partial Fulfillm ent of the Requirements fo r the Degree D O C T O R O F PHILOSOPHY (Psychology) September 1973 I! 74-921 HERR, John Jackson, 1946- DIFFERENTIAL EFFECTS OF EPINEPHRINE AND PROPRANOLOL ON SHUTTLE BOX AVOIDANCE LEARNING IN RATS OF DIFFERENT AGES. University of Southern California, Ph.D,, 1973 Psychology, experimental I University Microfilms, A XERO X Company, Ann Arbor, Michigan { THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. UNIVERSITY O F S O U TH E R N C A LIFO R NIA T H E G R A DU A TE SCHO OL U N IV E R S IT Y PARK LOS A NG ELES. C A L IF O R N IA 9 0 0 0 7 This dissertation, written by ................J oh n.. J ac ksq n.. H» r r under the direction of h \ 3 . Dissertation Com mittee, and approved by a ll its members, has been presented to and accepted by The Graduate School, in p artial fulfillm ent of requirements of the degree of D O C T O R O F P H IL O S O P H Y ( / Dean DISSERTATION COMMITTEE Chairman Acknowledgements F irs t, I would lik e to thank m y parents, Mr. and Mrs. John S. Herr, for th eir constant and dependable emotional (not to mention fin an cial) support from the f ir s t to the tw enty-first grade. Without them, m y aspirations for graduate study would s t ill be frozen out somewhere near the Vermillion River in South Dakota. Another couple w ho kept m e from throwing myself in the Vermillion River were Chris and Ray Drake. Without those evenings of hard boiled eggs and beer, the Midwest (and a certain department of psychology) would have surely consumed me. Next I would lik e to thank Mr. Garst Reese fo r his help in everything from designing the apparatus to gnashing out "exactly what question was I trying to ask." Turning to m y dissertation committee members, I would fir s t like to thank Dr. Herbert deVries whose research really led m e to the Andrus Gerontology Center. In fa c t, i t was m y interest in m ood and exercise (kindled by Dr. deVries) that eventually led to the topic of this dissertation. Next I would lik e to thank Dr. Jim Kahan, w ho taught m e that asking a question in s tatis tics did not invariably have a fa tal consequence. Jim's suggestions, beyond those of purely s ta tis tic a l nature, have always been very useful and have contributed in large measure to improvements in this dissertation. I would lik e to especially thank Dr. Jim Walker for his almost in fin ite supply of patience and helpful suggestions. Having Jim Walker there to turn to when som e methodological or theoretical crisis developed (as they regularly did) was not only of great practical value, but of great "psychological" comfort as w ell. Fin ally, I would lik e to thank Dr. James E. Birren. I s t ill have a copy of the fir s t paper I submitted to Dr. Birren in the aging seminar. Reading i t now, i t is a constant source of amazement to m e that he didn't throw m e out of the program on m y ear. Instead, Jim Birren took that jaded, cynical refugee from the Great Plains (whose only stated ambition was to get a "union card" in the shortest possible period of time) and molded him into a graduate student w ho genuinely enjoyed reading journal a rtic le s , attending seminars, and doing research (even shocking ra ts ). Thank you for your fa ith and your patience and for opening new doors. TABLE O F CONTENTS Page A C K N O W L E D G E M E N T S ................................................................................ i i PROBLEM........................... 1 Evolution of the Eisdorfer Hypothesis ........................ 2 Criticism of the Eisdorfer-Duke Experiments . . . . 6 Indirect Evidence Supporting the Eisdorfer Hypothesis............................................................................ 14 Directly Testing the Eisdorfer Hypothesis ................ 23 DESIGN A N D M E T H O D O L O G Y O F THE FIRST EXPERIMENT .................... 30 S ubjects.................................................................................... 30 A p p a ra tu s ................................................................................ 31 P ro ce d u re................................................................................ 32 Prehoc Inferences ................................................................ 35 RESU LTS O F THE FIRST EXPERIM ENT ................................................ 38 DESIGN A N D M E T H O D O L O G Y O F THE S E C O N D EXPERIM ENT ................ 40 R ESU LTS O F THE S E C O N D EXPERIMENT............................................... 43 DISCUSSION............................................................................................ 46 Consideration of the First Experiment ........................ 46 Consideration of the Second Experiment in Relation to the F i r s t ................................................ 50 Implications for Future Research .................................... 51 S U M M A R Y A N D CONCLUSIONS................................................... 54 R EFER EN C ES ........................................................................................ 56 L is t o f Tables Table Page 1. Analysis of Variance for the F irst Experiment . . . 38 2. Means, Standard Deviations, and Medians for the F irst Experiment ............................................................. 39 3. Analysis of Variance for Pooled Data from the First and Second Experiments............................................. 43 4. Means and Standard Deviations for the Pooled Data from the F irs t and Second Experiments . . . 44 5. Means, Standard Deviations, and Medians for the Second Experiment ......................................................... 45 v Problem The purpose of this dissertation was to test Eisdorfer's hypothesis that "a state of heightened rather than depressed autonomic end-organ arousal is responsible for the decrement of learning performance found in older age groups [Eisdorfer, Nowlin, & W ilkie, 1970, p. 1329]." From a review of the traditional psychophysiological lite ra tu re j on aging, Thompson and Marsh (1973) concluded: "The elderly have less autonomic reactivity than young subjects during states of focused attention and increased activation [p. 1 3 4 ] Just the opposite theory has been expressed by Eisdorfer (1968) based on a series of experiments with his associates at Duke University over the past several years. Unfortunately, the experimental work of the Eisdorfer-Duke group suffered from several design and methodological flaws. Since the consequences of elevated autonomic level in the aged have important implications for the direction of gerontological research from such diverse areas as cardiovascular disease to mem ory (Birren and Walker, 1972), this dissertation sought to conduct an experiment which overcame the previous design and methodological d iffic u ltie s of the Eisdorfer-Duke experiments. Evolution of the Eisdorfer Hypothesis While the original experiments and th e ir logical in te r pretation which culminated in Eisdorfer's hypothesis were reviewed extensively through 1968 elsewhere (Eisdorfer, 1968), they remain of sufficient importance to be reviewed here in somewhat less d e ta il. In the e a rlie s t study, Eisdorfer, Axelrod, and Wilkie (1963) found age decrements on the performance of a serial-ro te verbal learning task. B rie fly , the task consisted of memorizing a lis t of eight words which were presented one at a time on a projection screen. An asterisk appeared at the head of each li s t and served as a cue to recall the f ir s t word; the f ir s t word in turn served as a cue to I ” ~ " T l i recite the next word. The procedure continued until the entire lis t j had been presented several times. Eisdorfer, Axelrod, and Wilkie not only discovered that younger subjects performed the task better than older subjects; but also in a q u alitatively d ifferen t fashion. Older subjects tended to make the sam e number of errors of commission as younger subjects, but made considerably more errors of omission. They also found that older subjects showed disproportionately improved performance with increased exposure times to the stimulus words and that these performance increments were due to fewer errors of omission. Follow up experiments (Eisdorfer, 1965), however, indicated that the duration of the exposure time was not the determining factor for performance in the older group. Instead, i t was the duration of the interstimulus interval (IS I) that was c r itic a l. That is , subjects with a 4 second stimulus presentation and a 7 second ISI (4-on, 7-o ff) performed as well as subjects given a 10 second stimulus presentation with a 1 second ISI (10-on, 1 -o ff), although both groups outperformed those learning under 4-on, 1-off and 6-on, 1-o ff conditions. What made these results particularly interesting was that the average response latency was found to be 5.01 seconds. Since this time fe ll well within the 6-on, 1-off time duration and approximated the 4-on, 1 -o ff pacing schedule, Eisdorfer concluded i t was not simply that the ISI be of su fficien t duration for the older person to form a response, but som e generalized effect of pacing that w as c r itic a l. Further, since i t was errors of omission which varied inversly with pacing speed, i t was speculated that too rapid pacing led to response inhibition. The question that remained to be answered concerned the physiological mechanism for this age-related phenomenon. At this point, the Eisdorfer-Duke group's next set of experiments were considered most reasonably by viewing the chrono logical development of th eir experimental design and methodology to re fle c t an increasing amount of direct control exerted over their conceptual independent variable: peripheral autonomic and organ arousal. The f ir s t of these studies, which was correlational in nature (Powell, Eisdorfer, and Bogdonoff, 1964), replicated the fam iliar serial-ro te verbal learing task. Although younger subjects again were superior in performance to older subjects, surprisingly, the older subjects showed higher levels of autonomic arousal than the younger. In this case, arousal was operationally defined by the level of Free Fatty Acids (FFAs) in a series of blood samples. Such a result was not consistent with Malmo's (1959) activation theory (the fam iliar inverted "U" relationship between arousal and performance) unless the older subjects were over-aroused. As with a ll studies of a correlational nature, no legitim ate inferences of cause-effect relationships should have been drawn. However, the Eisdorfer-Duke group asked i f "increased autonomic a c tiv ity 'jamming' the perceptual and cognitive functions . . . [p. 195]" was the result of the increase in neurohumoral a c tiv ity which led to the increase of FFA level. In a later experiment, using a sim ilar behavioral task, Troyer, Eisdorfer, W ilkie, and Bogdonoff (1966) attempted to manipulate indirectly the arousal variable by varying both the task d iffic u lty (to increase arousal) and fam iliarization with the laboratory situation (to reduce anxiety). At f ir s t glance, the data seemed to refute the hypothesis that decrements in the performance of older subjects were due to over-arousal since the older lower FFA group performed more poorly than the older high FFA group. However, the study must really be considered inconclusive since other measures (p o s t-trial FFA levels) indicated the low FFA levels in the in ferio r group might have been due to a withdrawal response by the subjects under the rapidly paced situation. Based largely on the results of these experiments, Eisdorfer (1968) concluded: The relationship between learning and performance on the one hand and arousal on the other suggests that a curvi linear (inverted U-shaped) relationship between performance and arousal is a reasonable model. . . . I t has been im p lic itly assumed, however, that the aged are at a resting state of low internal arousal and sustain a low drive state. Our contention is that the aged m ay not be at a low state of arousal. Once aroused autonomically, perhaps because of a faulty a b ility to suppress an organ response or because of an altered feedback system, aged Ss appear to function as i f in states of high levels of autonomic a c tiv ity . . . . I t would be predicted, then, that where arousal or anxiety is diminished by experimental manipulation, older persons should improve th eir performance. [p. 215] In order to test this prediction, Eisdorfer, Nowlin, and Wilkie (1970) used the fam iliar serial-ro te verbal learning task with two groups of older subjects, one of which was treated with Propranolol HCI, a beta-adrenergic blocking agent. The results of this experiment were most impressive: the physiological effect of Propranolol should have been to lower Sympathetic Nervous System (SNS) a c tiv ity and this expected change was reflected in the observed decrease of the experimental subjects' circulating FFA level (as well as decreases in heart ra te ), while the behavioral effect of Propranolol resulted in the superior performance of the drug treated group relative to the control group. As supportive as these results appeared to be, Eisdorfer's hypothesis remained extremely tentative due to several design and methodological problems plaguing the f ir s t of the Eisdorfer-Duke group's experiments to the la st. Criticism of the Eisdorfer-Duke Experiments One of the keystones of the logic leading to Eisdorfer's hypothesis was that in older subjects response inhibition resulted from rapid pacing. Recent studies, however, by Kinsbourne and Berryhill (1972) indicated that by controlling total time available for learning, the effect of pacing on performance drops out. If the pacing effect was simply a rtifa c t, then the evidence for response inhibition due to pacing was also a rtifa c t. Therefore, i f Kinsbourne and Berryhill are correct, no hypothesis of altered autonomic arousal needed to be invoked to explain the Eisdorfer- Duke group's early behavioral data. I t must be pointed out, however, that Kinsbourne and B erryhill's experiment w as based on a paired- associates learning task (as opposed to the Eisdorfer-Duke group's serial-ro te verbal learning task). Consequently, the interaction ; I of pacing and performance m ay be task specific. Another problem of the Eisdorfer hypothesis relating to response inhibition was the inconsistency with which inferences were drawn from experimental data. To be specific, when the results of Troyer et al_. (1966) were opposite of those that were expected (the rapidly paced group had lower FFA values and performed worse than ; the slowly paced group), the results were interpreted to mean that the rapidly paced group "withdrew" from a d iffic u lt situation and thus did not show a characteristic FFA response indicating involvement. However, in an e a rlie r study (Powell et, al_., 1964), i performance differences between rapidly and slowly paced groups were attributed to differences in arousal: the rapidly paced group having been considered over-aroused! In other words, i f the Eisdorfer-Duke group postulated that the reasons performance on the rapidly paced task was in ferio r for older Ss was that they were over aroused, then how on the sam e task could they explain the sam e poor performance at a la te r date by saying the poorly performing S s withdrew? B y th e ir ow n d e fin itio n , what made the task d iffic u lt (thus causing the Ss to withdraw ) had to have been that they were over-aroused. The contradiction is obvious. Yet another problem of the Eisdorfer-Duke studies revolved around the use of FFAs to measure Autonomic Nervous System (ANS) arousal. Elevation of FFA levels merely accompany A N S arousal, they do not in itia te i t . Even the references cited by the Eisdorfer- Duke group to substantiate the v a lid ity of using FFA as a measure of A N A arousal freely admitted this fact (Bogdonoff, Estes, Harlan, Trout, and Kirschner, 1960; Bogdonoff, Estes, Friedberg, and Kline, 1961). :ersuing this point, Bogdonoff e t al_. (1960) in a study to determine metabolic and cardiovascular effects of mental stress concluded that the elevated levels of FFA a fte r stress were due to "secretion of adrenaline by the adrenal medulla and secretion of noradrenalin at the nerve endings of the sympathetic nervous system [p. 1338]. . . ." However, in a review paper by Bogdonoff et a l. (1961), the authors concluded that many other factors besides A N S arousal could contribute to elevating or lowering FFA levels. Since Eisdorfer himself was su fficien tly curious about the role of the major circulating cathecholamines (Epinephrine and Norepinephrine) in FFA mobilization with age to have been the f ir s t author of an a rtic le about the subject (Eisdorfer, Powell, Silverman, and Bogdonoff, 1965), i t was surprising that he failed to choose a more direct measure of A N S arousal. This problem of indirect inference was particularly apparent in the latest study (1970) where inferences of decreased S N S a c tiv ity , after Propranolol injection, were drawn from observed drops in circulating FFA levels. Unfor tunately, the situation was not nearly so simple, since the action of adrenergic blocking agents is not necessarily related to th eir A N S blocking function (Innes and Nickerson, 1965). Since i t was concluded that FFA level was an inappropriate measure of A N S arousal under any circumstances, Eisdorfer's terms 9 "peripheral end organ" and "autonomic end organ arousal" were examined more closely to determine what might make a good measure. Since Eisdorfer used these two terms interchangeably, for the purposes of this dissertation they were lumped together into the more inclusive term "Peripheral Autonomic End Organ Arousal (PAEOA)." Unfortunately, naming something was m uch easier than knowing what i t was, and neither Eisdorfer nor the Eisdorfer-Duke group was ever clear on defining a peripheral autonomic end organ. In the most recent experiment (1970), however, the Eisdorfer- Duke group did at least im p lic itly operationally define P A E O A since they used Propranolol, a beta-adrenergic blocking agent, to experimentally manipulate P A E O A level. Since the beta-adrenergic receptors are stimulated by the catecholamines (being more sensitive to Epinephrine than Norepinephrine), and since FFAs are also mobilized by Epinephrine (E) and Norepinephrine (NE), i t was concluded that despite what the Eisdorfer-Duke group have said, they must mean that P A E O A is determined by either the total amount of physiologically active E and N E or changes in the responsiveness (toward hypersensitivity) of those S N S receptors to E and NE. W hy the beta- adrenergic receptors were chosen for study over alpha-adrenergic receptors (which are more sensitive to N E than E) was not discussed. Another d iffic u lty remaining for the concept of P A E O A was the lim iting term "Peripheral." I t is well accepted that A N S arousal involves both Central Nervous System (CNS) as well as peripheral structures (Grossman, 1967), yet the Eisdorfer-Duke group tenaciously held to ascribing performance decrements with age due to over-PAEOA to peripheral "jamming" of the C N S (Powell, et al_., 1964). Later, the sam e concept was more eloquently stated, "feedback from the peripheral manifestations of this arousal might, in its e lf, actively contribute to performance decrements [Eisdorfer et al_., 1970, p. 1327] In a ll probability the issue arouse because the Eisdorfer-Duke group was reluctant to suggest a heightened arousal of the C N S with age. Recent research by the Russians, in particular Frolkis and his associates (Frolkis, Bezrukov, Bogatskaya, Verkhratsky, Zamostian Shevtchuk, and Shtchegoleva, 1970) supported the idea that this cautiousness may be legitim ate. I t was noted, however, that amphetamines, which have the a b ility to cross the blood brain barrier (Davis, 1970) as well as exert a peripheral sympathomimetic e ffe c t, elevate both C N S arousal and P A E O A and have been shown to improve the performance of both old rats (Doty and Doty, 1966) and old m en (Jeffrey, 1971, 1972). While by virtue of his lim ited definition of arousal, Eisdorfer's hypothesis is not refuted, i t certainly is not supported by these data. Obviously, the Eisdorfer-Duke group was free to lim it th eir discussion of peripheral effects providing they were able to manipulate the arousal of those isolated peripheral structures. In other words, Propranolol was to have no direct central effects. Although the Eisdorfer-Duke group freely admitted that Propranolol crosses the blood brain barrier (BBB), they claimed that "there is no j evidence of resulting central nervous system activation or deactivation [p. 1237]." This statement certainly was and continues to be equivocal. In fa c t, Eisdorfer, Nowlin, and Wilkie (1970) fa ile d to support th e ir assumption with even one citation from the i lite ra tu re . Actually, Propranolol crosses the B B B with as m uch ease as i t crosses in the heart (Black, Duncan, and Shanks, 1965) and tends to remain in those areas of the brain most generally ascribed to emotional behavior, the limbic areas (Masuoka, 1970). Several d ifferen t behavioral effects have been attributed to Propranolol. W aal (1967) as well as Axford and G ilchrist (1970) have reported mental depression to accompany treatment with Propranolol for cardiovascular disorders. Propranolol has also been used experimentallyasa psychiatric drug. Granvi11 e-Grossman and Turner (1966) reported i t exerted a beneficial effect on anxiety, although i t must be stated that som e theory holds that anxiety is re ally based on a peripheral phenomenon (P itts and McClure, 1967) and the r e lie f produced by Propranolol in anxiety is of a peripheral nature (P itts , 1969). Atmonson and his associates (Atmonson, Blum, Wijsenbeek, Maoz, Steiner, and Ziegelman, 1971) found Propranolol was an effective agent against both manic and hebephrenic symptoms, and suggested direct C N S j involvement of the drug. In animal studies, Margules (1971) found the presence of alpha- and beta-adrenergic receptors in the amygdala. H e concluded that the a c tiv ity of beta-receptors was to increase the suppression of operant behavior by punishment. Studies by Richardson and his associates (Richardson, Stacey, Cerauskis, and Musty, 1971) reported that treatment with Propranolol had a sim ilar effect on the learning of a DRL-20 task as ablation of the baso-lateral amygdala in rats. They concluded that blocking of the beta-adrenergic receptors in the C N S results in the organism's in a b ility to withold a response. While the in a b ility to withold a response would obviously lead to decrements in som e forms of behavior (fo r example, a DRL-20 task where the a b ility to withold a response is cru c ia l), i t might have a fa c ilito ry effect on other forms of behavior. One such type of behavior might be learning an active avoidance response in which the greater the number of responses made by the S _ , the higher the probability that S ^ has of successfully avoiding the aversive stimulus. In one such study, a p o st-trial injection of Propranolol was found to sign ificantly improve the performance of rats learning of C A R when the learning tra ils were distributed over several days (Merlo and Izquierdo, 1971). There is at least one study, either overlooked or ignored by Eisdorfer and his associates, which upholds th e ir assumptions about the direct C N S effects of Propranolol. Laverty and Taylor (1968) reported that in rats, large dosages (both oral and intraparatoneally) caused no alteration of biogenic amine metabolism in the brain. In addition, they found p retrial injections (both of higher and lower 13 dosages than used by Merlo and Izquierdo) led to no changes in per formance over a variety of behavioral tasks, including the C A R . While the biochemical studies were reported in som e detail and offered good support to the Eisdorfer-Duke group's assumptions (since the biogenic amines are intim ately involved with C N S arousal), the behavioral studies were reported very poorly (perhaps because the journal was in medical pharmacology and therefore behavioral data was of secondary in terest). Since the specific tasks, apparatus, s ta tis tic a l analysis and even the sample size v/ere omitted from the a rtic le , i t was impossible to compare Laverty and Taylor's results with those of Margules (1971), Richardson et al_. (1971) or Merlo and Izquierdo (1971). Obviously, the question of the direct central effects of Propranolol remains unresolved. As i f to demonstrate this assertion, i t was recalled that the Ss of the older groups in the serial-ro te verbal learning task did poorly because they appeared to withold responses. O n the basis of the theoretical background fo r the central effects of beta- adrenergic blockers (that is , they impair the organism's a b ility to withold a response), i t would have been predicted that Propranol would fa c ilita te the behavior of old people in the Eisdorfer-Duke task. Peculiarly, i t did fa c ilita te behavior, but not by decreasing significantly the number of errors of omission. As a resu lt, response inhibition was not a factor in the experiment. This result was, of course, inconsistent with not only the entire chain of logic leading to the Eisdorfer hypothesis, but with the theories 141 of direct Propranolol C N S effects as w ell. Since the matter remains unresolved, the possibility that Propranolol might have had som e general fa c ilito ry effect of the performance of the serial-ro te verbal lea.'ning task (independent of any peripheral effect) s t ill exists. I t was at this point where the Eisdorfer-Duke group's most serious design weakness was apparent. Since they used no younger Ss in the 1970 experiment, i t was impossible to attribute the positive effect of Propranolol to lowering P A E O A when i t might well have been that young people would have benefited ju st as greatly from Propranolol as old because of som e C N S effect (although Eisdorfer's hypothesis would predict th eir performance to be impaired, or at least remain the same). I t is obvious, after reviewing the lite ra tu re this fa r, that while the Eisdorfer hypothesis is provocative and potentially testable, i t has failed to be conclusively (or even convincingly) supported by direct experimental evidence. Indirect Evidence Supporting the Eisdorfer Hypothesis I f som e form of over-PAEOA with age is indeed a fa c t, then i t was reasonable to look at other organ systems effected by P A E O A to determine i f they reflected age changes consistent with Eisdorfer's hypothesis. Such evidence would bolster considerably the rather weak arguement submitted thus fa r in terms of direct experimental evidence supporting the hypothesis. In Western societies, i t is well established that blood pressure appears to rise as a function of age (Guyton, 1971). Since blood 15 pressure is obviously effected by PAEO A, i t was legitim ate to ask i f the age-related changes in blood pressure could have been accounted for by age related changes in PA EO A . Since som e authorities suggested that blood pressure is more a function of environment than age (Walker, 1966; Henry and Cassel, 1969), the relationship between P A E O A and environmental factors leading to hypertension was also included for consideration. C lin ic a lly , the level of blood pressure at which one is diagnosed as being hypertensive varies from authority to authority. In general, systolic blood pressure over 160 m m H g (diastolic over 100) is considered hypertensive while systolic blood pressure between 140 and 160 m m H g is considered borderline hypertensive. Since hypertension is considered one of the major risk factors for cardiovascular disease, and since surveys have reported as many as 40% of som e populations to demonstrate at least borderline hyper tension (Mulrow, 1970), hypertension has been intensly investigated. Because the level of blood pressure appears to vary d irectly with age, i t wasn't surprising that both the incidence and prevalence of borderline hypertension and hypertension has been observed to increase with age. According to Gutmann and Benson (1971), "Approximately 90 per cent of patients with hypertension are classified as 'e s se n tial', which by d efin itio n , means no etiology is known." [p. 543] While the etiology of essential hypertension w ill undoubtedly prove to be m ultiple in nature, over-PAEOA was suspected to account for m uch of 16 the variance since "the most effective means of treating these [hyper tense] patients s t ill remains the use of drugs which in one way or another block Sympathetic Nervous System a c tiv ity (Engelman, Portnoy, and Sjoerdsma, 1970)." [p. 1141] Even diuretics have been shown to exert an influence on the SNS, perhaps at a metabolic level (Neuvonen, 1971). The complexities of the circulatory control by the A N S were reviewed by Edis and Shepherd (1970). While many factors are responsible fo r maintaining blood pressure, probably the two most important are cardiac output and total peripheral resistance. In normotensive individuals at rest, changes in ore direction of either of these parameters results in a s h ift to the opposite direction by the other, thus maintaining the blood pressure within normal range. These homeostatic changes are effected both at the tissue level as well as through reflex arcs. According to Julius and Schork (1971), in borderline hypertension, cardiac output appeared to be increased over normal while peripheral resistance fa ile d to be reduced. O n the other hand, in essential hypertension, cardiac output appeared to be normal while peripheral resistance was increased. Edis and Shepherd (1970) suggested one of the two possible explanations for ANS-related essential hypertension was increased re activity of peripheral arterio les. That is , by being more reactive they were more easily constricted and as a resu lt, increased peripheral resistance. Relating this idea to age-related hyper tension, i t was interesting to note that Goldman (1970) suggested: 17 Although masked by the reduced cardiac output and the loss of elastic blunting of the systolic thrust, the hypertension of the elderly is dependent upon a rte rio la r control, ju st as is the 'essential' hypertension of younger persons, [p. 775] In a study relating environmentally induced changes in blood pressure with circulatory changes, i t was found that primates engaged in a 72 hour C A R f ir s t showed elevated blood pressure due to increased cardiac output, but with time the cardiac output returned to its baseline value while the blood pressure remained elevated due to increased peripheral resistance (Forsyth, 1971). Even more interesting was the fact that peripheral resistance was centered in the skeletal muscles, kidney and gastrointestinal tra c t, precisely where the greatest resistance is in essential hypertension. I t was noted that the cardiac output-peripheral resistance relationship changed in this experiment over time, m uch as does the sam e rela tionship change in the progression of borderline hypertension to essential hypertension. A series of related experiments with humans, conducted by Brod and his associates (Brod, Fencl, H e jl, and Jirka, 1959; Brod, Fencl, H e jl, Jirka, and Ulrych, 1962; Brod, 1964) found that sim ilar circulatory changes occurred with mental stress (solving arithmetic problems) as with essential hypertension. W hen they compared the circulatory responses to mental stress of normotensive and hypertensive Ss, they found that both groups showed the sam e degree of vasoconstriction in the skin and viscera but the hypertensive Ss d idn't demonstrate vasodilation in the muscles. In addition, they found the elevations in blood pressure resulting from the stressor to be of longer durations for the hypertensive Ss. At this point, several d ifferen t lines of experimental evidence began to converge. F irs t, blood pressure, a widely accepted psychophysiological variable thought to re fle ct arousal, was known to increase with ace. Second, the circulatory progress of environmentally induced elevations in blood pressure seemed to parallel closely age-related elevations in blood pressure. Third, the Eisdorfer hypothesis at least implied that either the le v e l, or sen sitivity to, primary modulating agents of S N S arousal (that is , E and NE) was altered with age. Fourth, environmental stressors are generally accepted to increase the secretion of E and N E (Levi, 1967; Mason, 1968; Brady, 1970; Frankenhaueser, 19/0). Since a ll these lines of evidence were related to the role of catecholamine (CA) concentration cr sen sitivity, i t seemed reasonable to examine further the relationship between CAs, blood pressure, and aging. The concept that the lin k between environment (or personality) and hypertension may be the C A s is not novel (Frankel, 1969). In fa c t, i t was extended by Raab (1966) to cover a ll forms of ischemic heart disease. However, from an extensive review of the environment- blood pressure interaction lite ra tu re , Gutman and Benson (1971) concluded that the etiological role of the C A s "remains unestablished, [p. 550] Even leaving environmental factors aside, those researchers most sympathetic to the idea that C A s are in som e manner involved with hypertension agree that the evidence is "slim [Nestel] 19 and Esler, 1970) [p. 1175]." However, there is a reasonable amount of compelling evidence, including even that of a metabolic (Brunjes, 1964; deChamplain, Krakoff, and Axelrod, 1969) and morphological (Matsumoto, 1969) nature. Examining more general studies, Engelman et al_. (1970) found plasma C A levels in hypertensive Ss were almost twice as high as in normal Ss. They suggested that other experimenters had been less successful in finding these differences because of less sensitive measuring techniques (p articularly of NE) and suggested that the enzymatic double-isotope derivative assay which they used superior to previously used methodologies. In urine studies, Serrano, Figueroa, Torres, and Del Angel (1964) found significantly higher excretion of C A s in hypertensive Ss than normotensive Ss, although they claimed this difference disappeared i f the nine highest C A excretors were eliminated from the sample. Since those nine subjects represented almost 10% of the total sample, such a modification of the data was really inappropriate. Although Serrano et al_. indicated that there was no significant relationship between blood pressure and C A excretion in normotensive Ss, Nestel and Doyle (1968) discovered a low correlation (r=.38) for (combined) groups of both hypertensive and normotensive Ss. However, they found no difference on urinary free C A s between the two groups. In observing th e ir data, however, i t appeared they over looked an important relationship. Blood pressure and C A level appeared to be independent of one another for the normotensive individuals, while for the hypertensive Ss, they appeared to be rather highly correlated. A reasonable interpretation of the data thus fa r was that som e individuals suffering from essential hypertension have a re la tiv e ly high response to those that are present (although not necessarily elevated). A reasonable expectation was that hypertensive Ss would show greater response in either C A levels or blood pressure increments to environmental stressors. Experimental reports (Nestel, 1969; Nestel and Esler, 1970) bore out this expectation. Further, Ikoma (1965) found that by dividing patients with essential hypertension into groups based on high or low S N S a c tiv ity (in response to mental or physical excitement), he was able to predict the effect of various hypotensive agents. In the case of Ss responding with high S N S a c tiv ity , C A levels seemed intim ately involved with the hypertension and those drugs resulting in lowered C A excretion also lowered blood pressure. In the case of the Ss responding with low S N S a c tiv ity , the relation between blood pressure and C A level was weak and drugs influencing C A level were ineffectual These studies taken together suggested that the most serious problem faced by advocates of a C A theory of essential hypertension w as that they generally faile d to recognize that essential hypertension was of m ultiple-etiology, and while C A level and/or sensitivity is of considerable importance to many individuals with borderline or essential hypertension, i t is of insignificant importance to many others. This problem alone has contributed greatly to the confusion resulting from conflicting experimental reports. In this regard, i t was particularly impressive to find that hypertensive Ss are generally considered to show greater blood pressure elevations to exogenously administered C A s than normotensive Ss in both the animal (Okamoto, 1966) and hum an lite ra tu re (Goldenberg, Pines, Baldwin, Greene, and Roh, 1948; Kaplan and Shilah, 1964; Trendelenburg, 1966; Mendlowitz, 1967. Examining the data relating sensitivity to exogenously administered C A s and age, i t was found that lite ra tu re was consistent with the Eisdorfer hypothesis since i t would have predicted either greater a v a ila b ility of C A s with age or a greater response to them. In a review of the lite ra tu re , Bender (1970) concluded: "The available evidence strongly indicates that there is an increase with age in the sen sitivity to exogenously administered amines [p. 220]." More recent studies by Frolkis et al_. (1970) suggested that in old animals the threshold to C A s for a vascular response drops markedly with age. Shevchuk (1972) reported, in rabbits, the age differences in vascular re a c tiv ity steem from increased sen sitivity of the beta- adrenergic receptors. While reports that confirmed an increased sen sitivity to exogenous C A s were consistent with Eisdorfer's hypothesis, they were of l i t t l e importance i f the a v a ila b ility of endogenous C A s decreases with age. The case does not appear to be the fa c t. Reviewing most of the lite ra tu re , and based on the results of his ow n study, Fisher (1971) concluded sympathetico-adrenal function "appears to be well preserved in the elderly [p. 259]." Shamovian, M ille r, and Cohen (1970) reported that the resting urinary C A excretion of th eir older Ss was higher than for th e ir younger S s (although the difference w as not s ta tis tic a lly sig n ifican t). However, elevated urinary excretion levels with age have been reported recently in both the Ita lia n (Giorgino, Scardapane, and N ardelli, 1969) and the Russian (Eidelman, Maximov, and Tyagileva, 1972) lite ra tu re . While the lite ra tu re reviewed to this point does tend to lend support to the Eisdorfer hypothesis, i t was noted e a rlie r in the discussion that Edis and Shepherd (1970 referred to two possible autonomic explanations fo r essential hypertension. To this point, only the explanation related to increased reactivity of the peripheral arterioles has been considered. The alternative explanation involved an alteration in baroreceptor reflex function. Put another way, the changes in blood pressure that occur with age and hypertension m ay not be due to any increases in S N S activation or sen sitivity but instead could be considered due to decreases in the sen sitivity or reactivity of the A N S negative feedback sensors and effectors (Dilman, 1971; Gribben, Pickering, and Sleight, 1969). In fa c t, based on reports by Nelson and Gellhorn (1958), i t could be argued that S N S activation or sensitivity actually decreases with age; i t is only the fact that the Parasympathetic Nervous System decreases in efficiency that leads to the erroneous conclusions about S N S changes. However, this alternative explanation for the data presented thus far did not invalidate its use in the support of the Eisdorfer hypothesis. The logic of the Eisdorfer hypothesis freely admits that P A E O A exerts an influence on behavior through the sensory overload of som e C N S mechanism. I t is apparent that the overload need not be absolute, only re la tiv e . As a consequence, the actual source of the overload is a moot point, providing an overload exists. In the case of blood pressure changes with age and essential hypertension, i t is abundantly clear that an S N S overload exists (otherwise, homeostasis would occur and blood pressure would not remain elevated), despite what the absolute level of a c tiv ity or sen sitivity might be. O n the basis of a ll the material presented, i t was concluded that Thomspon and Marsh (1973) were correct in th e ir assessment that: Continued investigation and comparisons of pharacologic agents which modify either central or peripheral autonomic nervous system a c tiv ity should prove to be extremely f r u it ful in the study of brain-behavior relationships in the elderly [p. 139]. D irectly Testing the Eisdorfer Hypothesis Probably the best test of the Eisdorfer hypothesis would have been to use the sam e se rial-ro te verbal learning task with both young and old groups treated under Propranolol, placebo, and som e type of s tric tly peripheral sympathomimetic drug conditions. This experiment was impossible to perform, however, for practical reasons. To replicate and improve the experiment of Eisdorfer, Nowlin, and W ilkie (1970) would have required, in addition to healthy elderly volunteers, the fa c ilitie s of a medical laborabory as well as of 24 I I j a physician. Such fa c ilitie s were not available to this experimenter. As a resu lt, i t was decided to use rats as Ss. Aside from the fact that a large body of behavioral and biochemical lite ra tu re exists for the ra t, and the obvious advantages in cost and management j over larger animals, rats provide good Ss for experiments with an age variable since th eir lif e span is rather short and animals of varying ages can be obtained without insurmountable d iffic u lty from commercial breeding laboratories. Since verbal learning tasks are obviously inappropriate for nonhuman Ss, the best test of Eisdorfer's hypothesis was seen to be an experiment in which arousal was acknowledged to play such a su ffic ie n tly large role that performance differences with age might be most parsimoniously attributed to arousal differences. Thus drugs affecting the P A E O A level might be expected to affect the performance of younger and older Ss d iffe re n tia lly . The task that was chosen to f u lf il these c rite ria was the acquisition of an active conditioned avoidance response (CAR). In sexually mature animals, probably the best predictor of performance of a C A R is strain (Barrett and Ray, 1970). • However, other experimenters have interpreted strain differences to represent differences in adrenergic (and cholinergic) system in both rats (Leith and B arrett, 1971) and mice (Yen, Katz, and Krop, 1971). In these studies, however, i t was not clear whether i t was central or peripheral arousal that was of importance. 25 N o such ambiguity existed in a study by Latane and Schachter (1972). Latane and Schachter found that moderate doses of E (.125 mg/kg) led to improved acquisition of a shuttle box CAR, while higher doses (2.5-5.0 mg/kg) did not. They interpreted th eir results to support the hypothesis that "there is a nonmonotonic, inverted U relationship between adrenalin concentration and avoidance learning [p. 371]." Since adrenalin (or E) doesn't cross the B B B (Glowinski and Axelrod, 1966; Axelrod, 1971), the differences in performance can be d irectly attributed to the effects of increased PA EO A . The lite ra tu re concerning the acquisition of a C A R with age was considerably more confusing. Denenberg and Smith (1963) reported avoidance learning behavior to improve with age in Purdue- Wistar rats (50, 100, 150, and 200 days old). However, in an e a rlie r study (Denenberg and Kline, 1958) using hooded rats, Denenberg w as unable to find significant age difference in perform ance. In other studies using Long-Evans (hooded) rats, no significant age differences were found fo r either simple (Doty and Doty, 1966) or discriminated (Doty, 1966) C A R tasks, even though the age range of the animals was over 600 days. Kirby (1963), using Wistar's from 25 to 100 days old, also reported no learning differences with age. Other experimenters found age decrements in C A R learning. In (DBA/2J) mice, Oliverio and Bovet (1966) found 60 day old Ss to be superior to 180 and 360 day old Ss. Freund and Walker (1971), using a d ifferen t strain (C57BL/6J), found age decrements began to appear at 6 months. In albino rats, Battig and Grandjean (1959) reported C A R acquisition best in 3 month old Ss and worst in 25 month old Ss (Ss tested were 1, 3, 5, 7, 9, 13, and 25 months old). Battig and Grandjean suggested the less-than-6 month old rats performed better than the over-12 month old S _ s . In probably the most complete study of a ll, (although limited in age range) Barrett and Ray (1970) tested four d ifferen t strains (Simonson hooded, Holtzman albino, a Charles River derived albino from Z iv ic -M ille r, and an inbred albino from Charles River) at two different ages (60 days and 9-11 months) and of both sexes. The Ss were tested over a variety of tasks, including a shuttle box C A R . Barrett and Ray found a significant main effect for age and strain, but not for sex. Their results indicated a decrement in performance with age. However, they were careful to point out that strain accounted fo r the largest proportion of the variance. One possible explanation for the.conflicting data reported for C A R performance and age might rest in a relationship overlooked by a ll the investigators. Since strains vary in baseline levels of arousal, i f Eisdorfer's hypothesis is correct that P A E O A increases with age, conflicting reports of C A R performance differences with age could best be understood by an Age X Strain interaction. Under this model, each strain would show performance peaks (representing th e ir position at the "top" of the inverted "U" curve) at d iffe re n t, but predictable, ages. Therefore, one strain might show improved performance from 3 to 9 months while another would show increments from 3 to 6 months but decrements from 3 to 9 months. Unfortunately, Barrett and Ray (1970) failed to report any results of the Strain X Age interaction for the shuttle box C A R in th e ir experiment. Until the results (either positive or negative in nature) of an Age X Strain interaction are reported, i t was concluded that within any given strain, age differences in the performance could be most parsimoniously explained by differences of PA EO A . The selection of drugs to a lte r P A E O A was re la tiv e ly straight forward. To increase P A E O A level, i t was decided that E was the best choice. This decision was reached prim arily for the reason that Latane and Schachter (1962) demonstrated its effect on performance to be related to inverted "U" function. In addition, they demonstrated which dosages effectively improved the performance of 4 month old rats. Since E is the primary (naturally occurring) stimulus for the beta-adrenergic receptors (Innes and Nickerson, 1965) and since sim ilar quantities of E as those injected by Latane and Schachter have been demonstrated to have d iffe re n tia l effects with age on the bioelectric a c tiv ity of the brain of rats (Frolkis, Bezrukov, and Sinitsky, 1972), E was the best drug for increasing PA EO A . A s was noted above, Axelrod, as recently as 1971, has stated clearly that the B B B exists fo r both E and N E (in both directions). Cohen and Shmavonian (1967) suggested that the B B B breaks dow n with age, based on th e ir study which showed E E G to be correlated with excreted levels of E in older S s but not in young Ss. This notion of age-related B B B changes to E has been rejected more recently by Mankovsky (1972). Despite the major drawbacks to using Propranolol, i t was chosen in an attempt to generalize the Eisdorfer, Nowlin, and Wilkie (1970) experimental results to rats. I t was f e lt that the findings of Merlo and Izquierdo (1971) were very damaging to the 1970 experiment because Merlo and Izquierdo found that p o st-trial injections of Propranolol actually improved performance of a C A R in rats. Since Eisdorfer et al_. (1970) failed to use young groups, i t was d iffic u lt to know whether or not the effect of Propranolol might not simply be a general aid to learning, independent of its peripheral beta-blocking effects. In fa c t, Merlo and Izquierdo suggested the reason fo r the improved performance of the Propranolol treated .S s in th e ir experiment "could be central and independent of th e ir adrenergic blocking [p. 185]. . . ." I t was decided to use the sam e dosage and method of administration as Merlo and Izquierdo except to administer the Propranolol p re tria l, as in the Eisdorfer et alL (1970) experiment. Since the peripheral effects of Propranolol are presumably to lower PAEO A, i t was expected that i f the Eisdorfer et al_. (1970) assumption about Propranolol modifying performance through peripheral influence was true, the younger Ss should show decrements in performance with Propranolol. I f they did better under the Propranolol condition, then clearly the Eisdorfer et aK conclusions based on the 1970 experiment were invalid. The inferences that were to be drawn from the performance of the other experimental groups are more complicated and w ill be discussed in detail at the conclusion of the next chapter. Design and Methodology of the F irst Experiment The basic design of the experiment followed a 3 x 2 A NO VA. Conceptually, the independent variables were Age and P A E O A level the dependent variable was the acquisition of a shuttle box CAR. Operationally, (the) Age (variable) was defined by chronological age; P A E O A was defined by drug treatment. The Ss treated with E were considered to have elevated PAEO A, while the Ss treated with Propranolol were considered to have lowered PAEO A. Control Ss (receiving placebo injections) were included to establish a baseline of performance for both age groups. Consequently, within each age group, the Ss were randomly assigned to a drug treatment group. Therefore, there were a total of six experimental groups: Younger Propranolol (YP), Younger Control (YC), Younger Epinephrine (YE), Older Propranolol (OP), Older Control (OC), and Older Epinephrine (OE). The acquisition of the shuttle box C A R was defined as the total number of avoidance (as opposed to escape or no escape) responses made during 80 tria ls of a shuttle box avoidance task. P ilo t data suggested that performance of the task was asymptotic for both age groups a fte r this many tr ia ls . Subjects The Ss were 102 Simonsen albino rats (a Sprague-Dawley d e riv itiv e ) obtained from the Simonsen Laboratories in G ilroy, C alifornia. All Ss were female, although this precaution was probably 30 31 unnecessary since other investigators have not reported performance differences on the shuttle box avoidance to be a function of sex. At purchase, one-half the Ss were approximately 3 months old and one-half were 8 months old. Since the experiment was not completed for almost one month a fte r th e ir purchase, the average age at testing of the younger groups was 3.5 months and for the mature adult groups was 8.5 months old. The mean weight at testing for the younger adults was approximately 268 g (S.D. = 17 g); fo r the mature adults, 374 g (S.D. = 44 g). The Ss were housed in the University of Southern California (main campus) V ivaria, 5 Ss per cage, and maintained with food (Purina Rat Chow) and water ad 1ibitum. Apparatus The shuttle box consisted of a 24 inch long x 8 inch wide x 12 inch deep compartment resting on a grid of % inch diameter stainless steel rods spaced % inch apart (center to center). The compartment was divided into two 12 x 8 x 12 inch deep subcompartments by a stain less steel g u illo tin e door (powered by a 28 volt electrical motor) which could be lowered to h inch of the grid to separate the two sides of the compartment (but not the ra t from her t a il) or raised 4 inches from the grid to provide a more than adequate passage from one subcompartment to the other. The entire shuttle box (with the exception of a hinged clear plexiglass front side) was constructed of wood; the three remaining wooden sides as well as the top were covered with fiberglass (to discourage chewing) and painted black. 32 | Symmetrically located on either side of the door were four 5-watt bulbs (2 per side) which served as the conditioned stimulus for the learning task. Microswitches were placed under movable sections of the grid to detect i f the had crossed sides. Shock was delivered to the grid by a constant current device designed by Bintz (1970). The intensity of the shock was approximately 110 microamperes. From data presented by Pare (1969) this intensity of shock appeared (by extrapolation) to be sufficient to cause escape or passive avoidance 100% of the time in animals of several d ifferen t strains, weights and ages. The entire shuttle box apparatus was connected by cable to a remote automated logic and counting system. Because the logic and counting system was electromagnetic in nature, i t was isolated in the next room. The shuttle box and automated logic and counting system were designed and constructed by J E . Procedure Three days before a was to be tested that rat was transferred from the University Vivaria to the basement of the Ethel Percy Andrus Gerontology Center (a distance of approximately h mile) where was housed individually and continued to be maintained on an ad libitum food and water schedule. Since the move resulted in shaking and loud noises, three days were allowed for recovery. Unfortunately, since University zoning restrictions would not allow the entire colony to be housed at the Andrus Gerontology Center, the move was unavoidable. r 33 : Each was deprived of food approximately 18 hours before the testing was to begin since the interactions between a recent feeding and drug effect were considered to be a potential source of experimental error. Generally, six Ss were tested per day. O n the day of testing, Ss were moved from the basement of the building to E _'s third floor laboratory under cover to minimize trauma. The Ss were then randomly assigned a testing order (having been previously assigned a drug condition). The experimental protocol for the f ir s t S ^ then began. ^ was weighed, and on the basis of this determination the amount of drug to be injected was loaded into the syringe. I f the injection was Propranolol (Inderal, manufactured by Ayerst Laboratories), the injection was given I. P. at a dosage of 2 mg/kg (the identical route of administration and dosage used by Merlo and Izquierdo). I f the injection was Epinephrine (Adrenalian in Oil Ampoules, 1:500; manufac tured by Parke-Davis), i t was administered subcutaneously in the nape of the neck at a dosage of .125 mg/kg (the identical route of administration and dosage used by Latane and Schachter). One-half the control Ss received I. P. injections of h cc o il; one-half subcutaneous injections of h cc peanut o il. All injections were delivered through a 27 guage needle (disposable). After the injection, ! S was returned to its cage for 5 minutes during which £ tested the logic and counting system and cleaned the shuttle box chamber and grid with a disinfectant solution (Lysol). After the grid and box were wiped dry and the 5 minute recovery .............. ' .' ... 34 period had elasped, the hinged plexiglass door was opened and was placed in the le ft side of the chamber. A note was made of the time and was given 10 minutes of exploration with the door up (open). The room lights were turned o ff and J E went to the adjoining room where the logic and counting equipment was located. At the conclusion of the 10 minutes of exploration, the automated learning tria ls began when the lig h t in the occupied side of the box went on. W hen S _ fa ile d to cross over to the opposite side of the box and depress the grid (with her weight) within approximately 6 seconds (a ll Ss did fa il on the f ir s t t r i a l ) , the shock was delivered to the grid of the lighted side of the chamber. W hen fin a lly crossed to the dark side, the lig h t was extinguished and the door was lowered (closed); simultaneously, the shock was turned o ff and the tr a il was counted as an escape. Approximately 30 seconds after the onset of the f ir s t t r ia l, the second tr a il began. Again, the lig h t (CS) cam e on in the occupied side of the box. This time, at the onset of the lig h t, the door was raised. I f crossed into the dark (safe) side, the tr ia l was counted as a successful avoidance and no shock was applied; instead, the lig h t went out and the door was lowered. I f fa ile d to cross within 6 seconds, the sam e procedure occurred as described in the conditions leading to an escape. This program was repeated every 30 seconds until 80 tria ls were completed (approximately 40 minutes). At the conclusion of 80 tr ia ls , the following data was recorded from the counting system: the number of successful avoidances (by blocks of ten tr ia ls ) and the number of tria ls required to complete six successful avoidance in a row. The Ss were generally tested between 10:00 a.m. and 5:00 p.m. While the total number of rats purchased provided for 17 Ss in each group (c e ll), only the data from 15 Ss were actually entered into the s ta tis tic a l analysis. Elimination of Ss stemmed from Experimenter error (£ dropped _ S , £ was bitten by ,S , £ injured _ S while holding for injection) as well as equipment fa ilu re (when the learning tria ls were interrupted for more than 1 minute) and was without regard for Ss' performance. Tw o Ss had to be eliminated (randomly) at the conclusion of the experiment to insure equal sample size. Prehoc Inferences From the work of Latane and Schachter (1962), i t is clear that performance differences on shuttle box C A R can be interpreted to represent differences of PA0EA on the inverted "U" curve. Since Eisdorfer's hypothesis explained age differences on the sam e curve, differences in performance between the Y C and 0C groups were assumed to re fle c t d ifferen t positions with respect to arousal on the inverted "U" curve. However, Eisdorfer's assumption that older S s would always be to the righ t (or higher level of PAEO A) than the younger Ss was abandoned. As a resu lt, the inferences about P A E O A which would be drawn from knowing only the performance of the Y C and 0C groups was rather lim ited. However, i f this information was to be combined with the results of the other groups, specific inferences supporting or refuting the Eisdorfer hypothesis could be drawn. To be more specific, i f Y C was superior to OC, than this difference would have been interpreted by the Eisdorfer hypothesis to m ean that the O C group was further to the right on the inverted "U" than the YC. Regardless of this assumption, i f E improved the performance of the older Ss (0E>0C), then obviously Eisdorfer's hypothesis would have been refuted. Unfortunately, i f the opposite case occurred (0E<0C), Eisdorfer's hypothesis would not necessarily have been supported since the in itia l differences (YC> OC) might not have been due to the older Ss being over-aroused (as the Eisdorfer hypothesis would have assumed) but really under-aroused, and a smaller quantity of Epinephrine might have actually improved th e ir performance. Therefore, i t was decided that i f the Older Control Ss showed in fe rio r performance to Y C Ss, and i f the O E group was in ferio r to the O C group, more Ss would need to be tested (at d ifferen t lower dosages of E) before any conclusions could be drawn. O n the other hand, i f the O C group proved superior to the Y C group (indicating higher P A E O A am ong older Ss under both Eisdorfer's and Latane and Schachter's assumptions): and the O E group in ferio r to the O C group, then providing E d idn't also impair the performance of the younger Ss (YE<YC), these results could be taken d irectly to support fu lly the Eisdorfer hypothesis since i t has never been suggested that less than four month old rats are "over the top" of the inverted "U." The results of the Y P and O P groups were of secondary interest since this part of the experiment was merely a replication of 37 Eisdorfer, Nowlin, and W ilkie's (1970) experiment generalized to a d ifferen t species and task, with a complete (young-old) design. I f Propranolol had sign ificantly improved the performance of the YPs compared to the YCs, then the effect of Propranolol on behavior could no longer be considered to be a simple function of PA EO A . Since i t was demonstrated that there is a considerable amount of other evidence in directly supporting the Eisdorfer hypothesis, ju st because Eisdorfer might have been righ t for the wrong reasons, doesn't mean his hypo thesis is necessarily incorrect. O n the other hand, significant decrements in the performance of the shuttle box avoidance in the younger Ss (YP<YC) would indicate that p retrial injection of Propranolol does not appear to improve performance on this task, and therefore, while not proving the Eisdorfer hypothesis, i t would at least support i t . Unfortunately, any support offered in favor of the hypothesis by the Propranolol treated groups would remain suspect because of the possible direct C N S effects the drug might have on behavior. Therefore, the re ally crucial test of Eisdorfer's hypothesis in this experiment was the interaction effect between Age and Epinephrine. N o matter what the results of the Propranolol groups indicated, i f Epinephrine sign ificantly improved (or had l i t t l e or no effect on) the performance of younger Ss (YE> YC) and sig n ifican tly impaired the performance of older Ss (0E<.0C), then the Eisdorfer hypothesis was borne out (conclusively, providing 0C>YE; ten tatively, i f 0C<YE). 38 Results of the F irst Experiment The data fo r the total number of successful avoidances were analyzed using the m ultivariate analysis of variance package (developed at the Psychometric Laboratory of the University of North Carolina) available at the University of Southern California Computer Center. The results of that analysis are summarized in Table 1. Table 1 Analysis of Variance fo r the F irst Experiment Source df M S £ P< Error 84 188.349 Age 1 1480.295 7.859 0.006 Drugs (B) 2 1000.420 5.312 0.007 A X B 2 487.792 2.590 0.081 I t was found that both the Age and Drug main effects were significant beyond an alpha level of .05 (p ^ .0 5 was the prehoc level of significance considered to be acceptable). The Age X Drug interaction approached significance, but fa ile d to attain i t . Upon examination of the means and variances (summarized in Table 2 ), i t was apparent that on the whole, the older Ss performed better than the younger Ss. In the case of the Drug main e ffe c t, analysis using Dunnett's £ (Winer, 1971) demonstrated the significance was due to poorer performance exhibited by the Propranolol groups (jt = 3.18, df = 84, p<,.005) rather than the Epinephrine groups. Table 2 Means, Standard Deviations, and Medians for the Second Experiment Measure Group Mean S.D. Medi 3.5 Month Old S Epinephrine 59.53 13.16 64 Control 58.33 10.28 61 Propranolol 43.67 21.44 42 8.5 Month Old S Epinephrine 59.20 13.41 66 Control 67.27 9.59 71 Propranolol 59.40 10.93 63 Nonparametric analysis of each Epinephrine group compared to its respective control, however, yielded a different result. Two- tailed Mann-Whitney tests (Siegel, 1956) disclosed that the Older Epinephrine group performed significantly worse than the Older Control group (U ^ = 64; ni,n2 = 15; p< .002) while the Younger Epinephrine group faile d to d iffe r significantly from the Younger Control group (jJ = 89; n j, n2 = 15). Design and Methodology of the Second Experiment At the conclusion of the fir s t experiment i t was decided to replicate (with som e modification) the experiment for the Epinephrine and Control groups for both ages. This decision was based on several factors. The foremost of these factors was to gain additional confidence that the s ta tis tic a l inferences drawn from the data of the f ir s t experiment could be borne out by the results of a (p a rtia l) replication of that experiment. This is , i f a phenomenon is legitim ate (as opposed to a s ta tis tic a l "mirage"), there is a high probability i t w ill occur a second time, and in a sense the second occurrence validates the inferences drawn from the analysis of the f ir s t (or original) data. The second factor in the decision to replicate the experiment involved the time interval between the injection of Epinephrine and the start of the avoidance tr ia ls . Latane and Schachter (1962) used a period of approximately 50 minutes for this time interval. In the f ir s t experiment of this dissertation, i t was decided that this length of time was too long i f both the Epinephrine and Propranolol groups were to be subjected to the sam e exact experimental conditions since the physiological h a lf-life of Propranolol is between 30 and 60 minutes (Black, Duncan, and Shanks, 1965). Therefore, in the fir s t experiment, a shorter duration was chosen since a ll available evidence indicated that the physiological effects of Epinephrine (even in o il) were almost immediate. 40 Since the rather large effects of Epinephrine on performance reported by Latane and Schachter were not observed in the data from the f ir s t experiment, the question naturally arose as to how c ritic a l the injection-avoidance tr ia l interval actually was. Postexperimental data, based on three remaining Ss (shipping "extras"- one older, two younger) confirmed this possibility might be the case. Therefore, 34 more Ss were ordered from the Simonsen Laboratories to be used in a second experiment. These Ss were of the sam e strain , sex, and ages (h alf younger, half older) as the Ss of the f ir s t experiment. Three S ^ s from each age group were treated with the sam e dosage of Epinephrine and subjected to the exact sam e experimental conditions as in the f ir s t experiment except that the injection- avoidance tr ia l interval was 60 minutes rather than 15 minutes (as in the f ir s t experiment). I t w as apparent from these animals that older Ss were severely impaired while the younger Ss appeared to be slig h tly impaired. Therefore, i t w as decided to decrease the dosage by a factor of one-half. I t was hoped that this dosage would not significantly impair the younger Ss (and even possibly improve th eir performance). The remaining 28 S s (14 of each age) were randomly assigned to either the Control or Epinephrine treated condition and except for the modification stated above, the methodology and procedure for the f ir s t experiment was followed exactly. The data from one Older Control had to be eliminated because of an equipment fa ilu re which interrupted the learning tria ls for over an hour fo r that particular S _ . A ll other Ss were included (despite any trauma they might have suffered which would have excluded them from the f ir s t experiment). Fin ally, the younger groups were of unequal size (8 Epinephrine treated, 6 Control) due to a clerical error by J E when assigning S s to conditions. Results of the Second Experiment The data from the second experiment were analyzed using both parametric and nonparametric techniques. Under parametric analysis, the data from both the f ir s t and second experiments were analyzed together using a 3x2 A N O V A where the three Drug conditions were Control (pooled over both experiments), Epinephrine 1 (from the f ir s t experiment), and Epinephrine 2 (from the second experiment). The Age conditions remained the same. The results of this analysis are summarized in Table 3. Table 3 Analysis of Variance fo r Pooled Data from the F irst and Second Experiments Source df M S £ P< Error 81 180.743 Age (A) 1 153.044 0.847 0.360 Drug (B) 2 2181.955 12.072 0.001 A X B 2 656.629 3.633 0.031 I t was found that both the Drug and Age X Drug interaction were significant beyond an alpha level of .05. The main effect for Age was not s ta tis tic a lly significant. Upon examination of the means (summarized in Table 4) i t was apparent that the Age X Drug interaction was due to the d iffe re n tia l effect of Epinephrine on the performance of the C A R based on age. I t was observed that while Epinephrine had l i t t l e or no effect on younger Ss, Older Ss seemed to be substantially impaired by the drug. 43 44 Table 4 Means and Standard Deviations for the Pooled Data from the F irs t and Second Experiment Measure Group Mean S.D. 3.5 Month Old Ss Epinephrine 1 59.53 13.17 Epinephrine 2 47.50 17.35 Control 56.86 12.77 8.5 Month Old S s Epinephrine 1 59.20 13.41 Epinephrine 2 35.57 21.39 Control 66.09 8.89 W hen the data from the second experiment alone were analyzed using nonparametric techniques, essentially the sam e pattern emerged: Older Ss were sign ificantly impaired by Epinephrine (U ^ = 3; nj = 7, r\2 - 8; p< .572) when compared to th eir respective controls. I t was noted that in the second experiment, unlike the f ir s t , the performance of the O C group was not s ta tis tic a lly sign ificantly superior to the Y C group (J J = 11.5; n^, r\2 = 6; p<.350). However, neither the Y C nor the O C groups of the second experiment differed sign ificantly from th e ir respective Y C or O C group in the f ir s t experiment. Since both the means and the medians for the Y C and O C groups in the second experiment (summarized in Table 5) indicated the sam e direction of effect as in the f ir s t experiment (0(T>YC), and 45 since in the second experiment the cell sample size was quite small (n = 6 ), i t was concluded that this lack of significance was easily explained by sampling error. Table 5 Means, Standard Deviations, and Medians for the Second Experiment Group Epinephrine Control Epinephrine Control Measure Mean S.D. 3.5 Month Old Ss 47.50 17.35 53.17 18.26 8.5 Month Old Ss 35.57 21.39 63.17 6.76 Median 52 59 36 60 Discussion Consideration of the F irst Experiment While there were prehoc inferences which could be stated about the ultimate interpretation of the data, there were no prehoc expectations as to what results might be expected. Therefore, in the interests of economy and because there was no a priori reason to suspect that the data wouldn't be amenable to parametric s ta tis tic s , i t was decided to analyze the data using a 3x2 A N O VA . Since this experiment used the Simonson Albino strain , i t is of at least anecdotal interest to note that this strain appears to be superior to a ll but one other used in the studies reviewed above. A note of caution is in order, however, since i t is very d iffic u lt to equate experimental conditions from one study to the next. While the main e ffect for Age was sign ificant, the fact that the older Ss were superior to the younger Ss neither proves nor disproves Eisdorfer's hypothesis. However, taking into consideration the assumptions made regarding the importance of P A E O A to the task, i t was concluded that these results support the concept that the baseline P A E O A (during the shuttle box task) was higher for the older Ss than fo r the younger. Regardless of this interpretation, i t was interesting that of the nine previously mentioned studies, only one (Denenberg and Smith, 1963) reported age changes in the sam e direction as this investigation. Since none of the previous experiments used the Simonsen Albino strain, i t seemed again a reminder of the potential amount of variance that 46 an Age X Strain interaction might account fo r i f a ll studies showing an Age effect on C A R learning could be pooled. The fact that the significance of the Drug main effect was due to the effects of Propranolol (as opposed to Epinephrine) demonstrated several things. F irst i t showed that p retrial injections of Propranolol lead to decrements in performance of the shuttle box C A R by both 3.5 and 8.5 month old animals, the opposite of the results reported by Merlo and Izquierdo (1971) w ho used p o st-trial injections. Since the sam e dosage was used in both experiments, i t was concluded that Propranolol, when given before the learning tr ia ls , does not have a fa c ilito ry effect on performance across a ll learning situations. While these results do not prove the Eisdorfer-Duke group were ju s tifie d in using Propranolol to lower P A E O A (in their specific learning situ atio n ), they at least do not refute the assumption that the central effects of Propranolol on behavior are re la tiv e ly secondary to its peripheral effects. Since the main reason the Propranolol groups were included was to test the notion that Propranolol might have been far more psychoactive than had Eisdorfer, Nowlin, and Wilkie had anticipated, these results are of at least som e importance. The fact that both the younger and older Ss were impaired by Propranolol is consistent with Eisdorfer's hypothesis and the assumptions underlying the shuttle box avoidance task set forth by Latane and Schachter (1962). That is , since the Y C group acquired the C A R less well than the O C s (indicating the O C s were higher on the inverted "U" as well as more highly aroused), i t was reasonable that Propranolol resulted in decrements to both age groups since the baseline P A E O A for the older group was not yet sufficient to result in d eficits in performance of the task. Unfortunately, to this point, the best data offered were only not inconsistent with the Eisdorfer hypothesis or previous experiments. They did not re ally prove that Eisdorfer was correct. I t was at this point that the results of the Age X Drug interaction was of the greatest disappointment since i t approached, but feel short of significance. Even more disappointing was the fact that the Epinephrine treated groups failed to be significantly d ifferen t from the control groups, particularly since they were the only groups in which P A E O A was manipulated d irectly and without direct C N S effect. I f , however, the cell means (in Table 2) were examined by rank order only, within ages, the interaction effect was apparent. That is , am ong the younger Ss, the best performance was YE> YC>YP; am ong the older, 0C> O E and OP. Examining the medians (also in Table 2), i t was apparent that the sam e qualitative relationship existed, but was quantitatively more d istin ct. The fact that the medians for a ll the groups, in particular the Epinephrine groups, showed rather large departures from their respective means, suggested the underlying distribution of scores might not be normally distributed, this making the A N O V A unsuitable for delineating significant differences in central tendency between the treatment groups. The fact that the Simonsen Albino strain acquires the avoidance so quickly (the : S with the median score of the O C group missed only 9 of 80 possible avoidances) suggested that while there was ample opportunity fo r in ferio r performance to be expressed, there was insufficient opportunity fo r the expression of superior performance. A s a consequence of this skewed distrib utio n, the best discrimination of differences in central tendency m ay no longer have been in the parametric approach. At this point, a new problem presented its e lf. N o non- parametric methods exist fo r examining a ll the effects and in te r actions analysed by the ANO VA. Further, the use of multiple nonindependent nonparametric tests suffers from the sam e problems (spuriously significant results) of multiple nonindependent jt tests. Put another way, one could have the results of the A N O V A i f he were w illing to stipulate that the departures from the normal curve of the underlying distribution of scores were insubstantial. A s a resu lt, he could make no statements about the interaction effects of Epinephrine treatment and Age. O n the other hand, i f he would have been w illing to technically sacrifice the results of the ANO VA, he would have been able to make (at least) two independent non parametric tests to determine the interaction between Age and Epinephrine. Since the results of the A N O V A were interesting, but inconclusive, and since the purpose of the Propranolol groups w as really secondary in nature; based on the evidence of non-normality, the option for nonparametric analysis was exercised. 50 The nonparametric analysis indicated that Epinephrine significantly impaired the performance of the older Ss while i t fa ile d to sign ificantly effect the performance of the younger Ss. Consideration of the Second Experiment in Relation to the F irst O n the whole, the results of the modified replication of the Epinephrine and Control groups supported the inferences drawn from the f ir s t experiment. That is , Epinephrine impaired the performance of older Ss while having no s ta tis tic a lly significant effect on the performance of the younger Ss. This case held true in both the parametric and nonparametric analysis. Since the interaction effect was replicable from one experiment to the next, the results of the experiments were interpreted to consistently and fu lly support the Eisdorfer hypo thesis. The second experiment also demonstrated the c ritic a l importance of the time course of Epinephrine action on behavior. Even with one-half the dosage, the effects of Epinephrine on the older S s was fa r more profound with the linger injection-avoidance tria ls in te rv a l. Taking into consideration the results of both experiments, speculation was made as to the relative (and absolute) positions on the inverted "U" of each of the groups. Under this model, the O C s were near the "top (suggesting they were at an optimal level of arousal)," being driven dow n by either increments (OE) or decrements (OP) in PA EO A . The YC s were somewhat to the " le ft" and "down" from the O C s (suggesting they were at a suboptimal level of arousal--on the s lig h tly under-aroused side), and although they showed l i t t l e or no chance in performance with increased P A E O A (YE), th e ir performance was impaired with decreased P A E O A (YP). The results of this dissertation are exciting and provocative fo r several reasons. F irs t, i t has been demonstrated that Eisdorfer's hypothesis can be successfully generalized to a d ifferen t task and species to predict the effect of age and P A E O A on performance. Second, this dissertation demonstrated that P A E O A can be operationally defined and experimentally manipulated independently of the need for Propranolol or reliance upon the determination of circulating FFAs, thus giving greater credence to the concept of PA EO A . Fin ally, this dissertation demonstrated that the effect on behavior of age-related changes in P A E O A is a process that is not always decremental in nature. Implications for Future Research The implications of Eisdorfer's hypothesis are enormous for research in many d ifferen t areas. For example, while i t has been demonstrated that operant conditioning can be successful in decreasing systolic blood pressure in hypertensives (Benson, Shapiro, Tursky, and Schwartz, 1971), what might be the effects of the sam e training on the performance of tasks previously demonstrated to show decrements with age and cardiovascular disease (Simonson, 1965)? Eisdorfer's hypothesis, of course, would predict that per- formance would improve as P A E O A declined (measured in this case by decreases in blood pressure). In fa c t, the entire area of autonomic operant conditioning lends its e lf to testing and applying Eisdorfer's hypothesis. Using this technique, one could condition a decrease in peripheral resistance by plythesmograph, or even skin temperature feedback. Presumably these changes would re fle c t decreased P A E O A and should result in improved performance on those tasks previously associated with age decrements. Other areas of both acute and chronic manipulation of P A E O A should be investigated. While something is known about the physiology of meditation (Wallace and Benson, 1972), how might any chronic change in P A E O A (as is claimed to occur with mediation) effect the blood pressure or behavior of older individuals? Further a fie ld , is the result of emotional catharsis to lower P A E O A (Hokanson and Burgess, 1962) or to raise i t (Kahn, 1966)? Are there age-related differences in behavior after catharsis? Exercise (chronic physical training) has been demonstrated to lower blood pressure in older subjects (deVries, 1970). What might one expect the effect of chronic physical training to be on the performance of behavioral tasks? What might be the effect of massage and hot baths? H ow might these effects change with the nature of the task and the age of the subject? I f , as the Eisdorfer hypothesis suggests, the older organism is predisposed to over-PAEOA, then i t would be of particular practical importance to identify those potentially hazardous environments which lead to hyperarousal. In addition, coping strategies should be developed to detoxify these environments when they are unavoidable. Sum m ary and Conclusions The purpose of this dissertation was to test the Eisdorfer hypothesis that performance decrements with age can be explained by the inverted "U" relationship between arousal and performance by assuming older S s are over-aroused. Using young rats, Latane and Schachter demonstrated that the acquisition of a shuttle box avoidance task is a function of the sam e inverted "U" curve. Using the shuttle box avoidance task to test Eisdorfer's hypothesis, female Simonsen Albino rats of 3.5 and 8.5 months of age (N = 102) were injected with Epinephrine, Propranolol HC1, or peanut oil (Control). Older Control Ss performed significantly better than Younger Control Ss (suggesting older S s were at a higher in itia l level of PAEOA). Propranolol (which lowers PA EO A ) significantly impaired the performance of Ss of both ages. Epinephrine (which elevates P A E O A ) sign ificantly impaired the performance of older Ss but fa ile d to have a s ta tis tic a lly significant effect on younger Ss. A subsequent experiment which replicated (with slight modification) the Epinephrine and Control groups for both ages of Ss (N = 28) resulted in sim ilar findings as those of the fir s t experiment. From the results of the litera tu re review and the two experiments of this dissertation, the following conclusions were drawn: 54 1. The design and methodology of the Eisdorfer-Duke group's experiments preceding and subsequent to the exposition of the Eisdorfer hypothesis were so severely flawed as to call into question the v a lid ity of the hypothesis its e lf. 2. A large body of indirect evidence exists which conveniently links A N S status with age-related changes in blood pressure and can be considered to support the v a lid ity of the Eisdorfer hypothesis. 3. The Eisdorfer hypothesis can be generalized to a d ifferen t species, task, and manipulation of P A E O A to successfully predict performance differences based on age. References Atmonson, A ., B. Blum, I . , Wijsenbeek, H., Maoz, B ., Steiner, M., and Ziegelman, G. The short term effects of adrenergic- blocking agents in a small group of psychotic patients. Psychiatr. Neurol. Neurochir. , 1971, 74, 251-258. Axelrod, J. Noradrenaline: fate and control of its biosynthesis. Science, 1971, 173, 598-606. Axford, A. T ., and G ilc h ris t, L. Propranolol in the treatment of hypertension. British Journal of Clin. P ract., 1971, 25, 326-328. “ B arrett, R. J ., and Ray, O.S. Behavior in the open fie ld , Lashley I I I maze, shuttle box and Sidman avoidance as a function of strain , sex and age. Developmental Psychology, 1970, 3 ^ , 73-77. B attig, K., and Grandjean, E. Beziehung zwischen a lte r und erlernen einer bedingten fluchtreaktion bei der weissen ra tte . Gerontologia, 1959, 266-276 (English summary). Bender, A.D. The influence of age on the a c tiv ity of catecholamines and related theraputic agents. Journal of the American Ger. Soc. , 1970, 18, 220-232. Benson, H., Shapiro, D., Tursky, B., and Schwartz, G.E. Decreased systolic blood pressure through operant conditioning techniques in patients with essential hypertension. Science, 1971, 173, 740-742. Bintz, J. A. well regulated constant-current shocker. Behavioral Res. Method, and Instrumentation, 1970, 2, 258-259. Birren, J. E., and Walker, J. P. Hypertension, A N S arousal, memory and aging. In Research Program: Physiology of Memory and Aging, Los Angeles: University of Southern C alifornia, 1972. Bogdonoff, M.D., Estes, E. H., Friedberg, S. J ., and Klien, R. F. Fat mobilization in man. Ann. In t. Med., 1961 , 55, 328-338. Bogdonoff, M. D., Estes, E. H., Harlan, W.R., Trout, D. L ., and Kirschner, N. Metabolic and cardiovascular changes during a state of acute central nervous system arousal. Journal C lin . Endocrin. Metab ., 1960, 20, 1333-1340. Brady, J. V. Endocrine and autonomic correlates of emotional behavior. In P. Black (Ed.) Physiological Correlates of Emotion, N ew York: Academic Press, 1970. 56 57 Brod, J. Circulation in muscle during acute pressor responses to emotional stress and during chronic sustained elevation of blood pressure. American Heart Journal, 1964, 68, 424-426. Brod, J ., Fencl, V ., H ejl, A ., and Jirka, J. Circulatory changes underlying blood pressure elevation during acute emotional stress (mental arithm etic) in normotensive and hypertensive subjects. Clin. Sci., 1959, 1 J 3 , 269-279. Brunjes, S. Catechol amine metabolism in essential hypertension. N ew England Journal of Medicine, 1964, 271, 120-124. Cohen, S. I . , and Shmavonian, B. M. Catechol amines, vasomotor conditioning and aging. In L. Gitman (Ed.) Endocrines and Aging, Springfield, 111.: Charles C. Thomas, 1967. Davis, J. M . Theories of biological etiology of affective disorders. In C. C. P fe iffe r and J. R. Symthies (Eds.) International Review of Neurobiology. Vol. 12, N ew York: Academic Press, 1970. de Champlain, J ., Krakoff, L. R ., and Axelrod, J. In ter relationships of sodium intake, hypertension, and norepinephrine storage in the ra t. Circ. Res., 1969, 24 (Suppl. 1), 175-192. deVries, H. A. Physiological effects of an exercise training regimen upon m en aged 52-88. Journal of Gerontology, 1970, 25, 325-336. Deneberg, V ., and Kline, N. The relationship between age and avoidance learning in the hooded ra t. Journal of Comparative and Physiological Psychology, 1958, 51_, 488-492. Deneberg, V. and Smith, S. Effects of in fa n tile stimulation and age upon behavior. Journal of Comparative and Physiological Psychology, 1963, 56, 307-312. Dilman, V. M. Age-associated elevations of hypothalamic threshold to feedback control, and its role in development, aging, and disease. Lancet, 1971, 1, 1211-1219. Doty, B. A. Age and avoidance conditioning in rats. Journal of Gerontology, 1966, 2]_, 287-290. Doty, B., and Doty, L. F a c ilita tiv e effects of amphetamine on avoidance conditioning in relation to age and problem d iffic u lity . Psychopharmacologia, 1966, 9, 234-241. Edis, A. J ., and Shepherd, J. T. Autonomic control of the peripheral vascular system. Archives of Internal Medicine, 1970, 125, 716-724. 58! Eidelman, M . M., Maximov, S. V ., and Tyagileva, V. P. The influence of vitamin Bi on age peculiarities of the sympatho-adrenal system ! functional state. In 9th International Congress of Gerontology. Vol. 3. Section Sessions. Kiev, USSR, 1972. Eisdorfer, C. Verbal learning and response time in the aged. Journal Genet. Psychology, 1965, 107, 15-22. Eisdorfer, C. Arousal and performance: experiments in verbal learning and a tentative theory. In G. A. Tall and (E d.), H um an Aging and Behavior. N ew York: Academic Press, 1968. Eisdorfer, C., Axelrod, S ., and W ilkie, F. L. Stimulus exposure time as a factor in serial learning in an aged sample. Journal of Abnormal and Social Psychology, 1963, 67, 594-600. Eisdorfer, C., Nowlin, J ., and W ilkie, F. L. Improvement of learning in the aged by modification of autonomic nervous system a c tiv ity . Science, 1970, 170, 1327-1329. Eisdorfer, C., Powell, A ., Silverman, G ., and Bogdonoff, M. The Characteristics of lip id mobilization and peripheral disposition in aged individuals. Journal of Gerontology, 1965, 20, 511-514. Engelman, P., Portnoy, B., and Sjoerdsma, A. Plasma catecholamine concentrations in patients with hypertension. Circ. Res., 1970, 26-27 (Suppl. I ) , 1-141. Fisher, R. H. The urinary excretion of 4-hydroxy-3-methoxy mandelic acid in the elderly. Gerontol. C lin ., 1971, 13, 257-260. Forsyth, R. P. Regional blood-flow changes during 72-hour avoidance schedules in the monkey. Science, 1971, 173, 546-548. Frankel, E. Coronary disease and personality. B rit. Med. Journal, 1 , 382-383. Freund, G., and Walker, D. W . The effect of Aging on acquisition and retension of shuttle box avoidance in mice. Life S c i., 1971, 10. (Pt. 1), 1343-1349. Frankenhaueser, M. Behavior and catecholamine release during stress. Sartryck ur Forsvarsmedicin, 1970, 6., 17-24. Frolkis, V. V ., Bezrukov, V. V ., Bogatskaya, L. N., Verkhratsky, N. S ., Zamostian, V. P ., Shevtchuck, V. G., and Shtchegoleva, I. V. Catecholamines in the metabolism and functions regulation in aging. Gerontologia, 1970, 16, 129-140. 59 Giorgino, R., Scardapane, R ., and N ardelli, G. M. (Adrenal glands and aging; aspects of medullary a c tiv ity ). Folia Endocr., Roma, 1969, 22, 215-224 (Biological Abstracts, 1970, No. 101099). Glowinski. J ., and Axelrod, J. Effects of drugs on the diposition of H 3-norepinephrine in the ra t brain. Pharmacol. Rev., 1966 18, 775-785. Goldenberg, M., Pines, I. L ., Baldwin, E. F ., Greene, D. G., and Roh, C. E. Hemodynamic response of m an to norepinephrine and epinephrine and its relation to the problem of hypertension. American Journal Med., 1948, 5, 792-806. Goldman, R. Speculations on vascular changes with age. Journal of the American Geriatrics Society, 1970, 18, 765-799. Granvi11 e-Grossman, K. L ., and Turner, P. The effect of propranolol on anxiety. Lancet, 1966, 788-790. Gribbin, B., Pickering, T. G., and Sleight, P. Decrease in baroflex sen sitivity with increasing a rte ria l pressure and with increasing age. B ritish Heart Journal, 1969, 31_, 792. (abstract) Grossman, S. P. A textbook of physiological psychology. New York: Wiley and Sons, 1967. Gutmann, M., and Benson, H. Interaction of environmental factors and systemic a rte ria l blood pressure: a review. Medicine, 1971, 50, 543-553. Guyton, A. C. Textbook of medical physiology. (4th ed.) Philadelphia: W . B. Saunders, 1971. Henry, J. P ., and Cassel, J. C. Psychosocial factors in essential hypertension: recent epidemiologic and animal experimental evidence. American Journal of Epidemiology, 1969, 9 C U 171-199. Hokanson, J ., and Burgess, M. The effects of three types of agression on vascular process. Journal of Abnormal and Social Psychology, 1962, 64, 446-449. Ikoma, T. Studies on catechols with reference to hypertension. Jap. Circ. Journal, 1279-1286, 29, 1965. Innes, I . R ., and Nickerson, M. Drugs acting on postganglionic adrenergic nerve endings and structures innvervated by them (sympathomimetic drugs), in L. S. Goodm an and A. Gilman (Eds.), The pharmacological basis of therapeutics. (3rd ed.) N ew York! Macmillan, 1965. Jeffrey, D. W . The effect of dextroamphetamine on arousal and on performance in young and old men. Paper presented at the meeting of the Gerontological Society, Houston, Texas, October, 1971. Jeffrey, D. W . Arousal and speed of responding as affected by dextroamphetamine in healthy young and old men. In 9th in te r- national congress of gerontoloqy. Vol. 3. Section session. Kiev, USSR, 1972. ~ Julius, S ., and Schork, M. A. Borderline hypertension—a c ritic a l review. Journal of Chronic Disease, 1971, 23, 723-754. Kahn, M . The physiology of catharsis. Journal of Pers. and Social Psychology, 1966, 3 _ , 278-286. Kaplan, N. M., and Shilah, J. G. Effect of angiotensin I I on the blood pressure in hum ans with hypertensive diseases. Journal of Clinical Investigation, 1964, 43, 659-669. Kinsbourne, M., and B erryh ill, J. L. The nature of the interaction between pacing and the age decrement in learning. Journal of Gerontology, 1972, 27, 471-477. Kirby, R. Acquisition, extinction and retention of an avoidance response in rats as a function of age. Journal of Comparative and Physiological Psychology, 1963, 56, 158-162. Latane, B., and Schachter, S. Adrenalin and avoidance learning. Journal of Comparative and Physiological Psychology, 1962, 55, 369-372. Laverty, R., and Tyalor, L. M. Propranolol uptake into the central nervous system and the effect on rat behavior and amine metabolism. Journal of Pharmacy and Pharmacology, 1968, 20, 605-609. Leith, N. N., and B arrett, R. J. Relationship of adrenergic and cholinergic systems to strain differences in avoidance performance. Pharmacologist, 1971, 13^ 232. (abstract) Levi, L. Stress. N ew York: Liveright, 1967. Margules, D. L. Alpha and beta adrenergic receptors in amygdala: reciprocal inhibitors and fa c ilita to rs of punished operant behavior. European Journal of Pharmacology, 1971, 16, 21-26. Mason, J. W . A review of psychoendocrine research on the sympathetic- adrenal medullary system. Psychosomatic Medicine, 1968, 30, 631-653. — Matsuoka, E. Propranolol. A C TA Pharmacol. , 1970, 2 5 ^ , 447-449. Matsumoto, M. Morphological studies on the A N S of hypertensive rats. Jap. Circ. Journal, 1969, 33^ , 411-416. Malmo, R. B. Activation: a neuropsychological dimension. Psychological Review, 1959, 66^ , 367-386. Mankovsky, N. B. The aging of m an and the function of nervous system. In 9th international congress of gerontology, Vol. 1. Plenary and sections sessions. Kiev, USSR, 1972. Mendlowitz, M . Vascular re activity in essential and renal hyper tension in man. American Heart Journal, 1967, 73, 121-128. Merlo, A. B. and Izquierdo, J. A. Effect of p o st-trial injection of beta-adrenergic blocking agents on a conditioned reflex in rats. Psychopharmacol ogia, 1971 , 2 2 ^ , 181-136. Mulrow, P. J. (Ed.) Hypertension. Vol. 19. S alt, hormones and hypertension. N ew York: American Heart Association, 1971. Nelson, R. and Gellhorn, E. The influence of age and functional neuropsychiatric disorders on sympathetic and parasympathetic functions. Journal of Psychosomatic Research, 1958, 3 ^, 12-26. Nestel, P. J. Blood pressure and catecholamine excretion after mental stress in lab ile hypertension. Lancet, 1969, 1 _ , 692-694. Nestel, P. J. and Doyle, A. E. Excretion of free noradrenaline and adrenaline by healthy young subjects and by patients with essential hypertension. Aust. Ann. Medicine, 1968, T7, 295-299. Nestel, P. J. and Esler, M . D. Patterns of catecholamine excretion in urine in hypertension. Circ. Res., 1970, 26-27 (Suppl. I I ) , 1175-1181. Neuvonen, P. J. Influence of diuretics and diazoxide on ions and vascular re activity in normotensive and spontaneously hyper tensive rats. Ann. Med. Exp. Biol. Fenn., 1971, 4£, 109-119. 62 Okamoto, K. Pharmacodynamic studies on the cardiovascular system of spontaneously hypertensive rats. Jap. Circ. Journal, 1966, 30, 987-1007. ~ O liverio, A. and Bovet, D. Effects of age on maze learning and avoidance conditioning of mice. Life Sci. , 1966, 5, 1317-1324. Pare, W . P. Age, sex and strain differences in the aversive threshold to grid shock in the ra t. Journal of Comparative and Physiological Psychology, 1969, 69, 214-218. P itts , F. N., Jr. The Biochemistry of anxiety. S cientific American, 1969, 220 (2 ), 69-75. P itts , F. N. and McClure, J. N. Lactate metabolism in anxiety neurosis. N ew England Journal of Medicine, 1967, 277, 1329-1326. Powell, A ., Eisdorfer, C., and Bogdonoff, M. Physiologic response patterns observed in a learning task. Arch. Gen. Psychia., 1964, K), 192-195. Raab, W . Emotional and sensory stress factors in myocardial pathology. American Heart Journal, 1966, 72, 538-564. Richardson, J. S ., Stacey, P. D., Cerauskis, P. W., and Musty, R. E. Propranolol interferes with inhibitory behavior in rats. Journal of Pharmacy and Pharmacology, 1971, 23, 459-460. Serrano, P. A ., Figueroa, G., Torres, M ., and Del Angel, A. R. Adrenaline, noradrenaline, and dopamine excretion in patients with essential hypertension. American Journal of Cardiology, 1964, 13, 484-488. Shevchuk, V. G. Age induced peculiarities of the adrenomimetic and adrenolytic substances effect on the hemodynamics. In 9th international congress of gerontology. Vol. 3. Section sessions. Kiev, USSR, 1972. Siegel, S. Nonparametric s ta tis tic s , N ew York: McGraw-Hill, 1956. Simonson, E. Performance as a function of age and cardiovascular diesease. In A. T. Welford and J. E. Birren (Eds.) Behavior, aging and the nervous system, Springfield, 111: Charles C. Thomas, 1965. Thompson, L. W., and Marsh, G. R. Psychophysiological studies on aging. In C. Eisdorfer and M. P. Lawton (Eds.) Psychology of adult development and aging, Washington, D. C.: American Psychological Association, 1973. 63 Trendelenburg, U. Mechanisms of supersensitivity and subsensitivity to sympathomimetic amines. Pharmacol. Rev. , 1966, T8, 629-640. Troyer, W . G., Eisdorfer, C ., W ilkie, F ., and Bogdonoff, M. D. Free fa tty acid responses in the aged individual during performance of a learning task. Journal of Gerontology, 1966, 21, 415-419. Von Euler, U. S. Adrenergic neurotransmitter functions. Science, 1971,173,202-206. Waal, H. J. Propranolol-induced depression. B rit. Med. Journal, 1967, 2, 50. Walker, A. R. P. Minimum vascular sequelae of aging. G eriatrics, 1966, 21(2), 161-165. Wallace, R. K ., and Benson, H. The physiology of meditation. S cientific American, 1972, 226 (2), 85-90. Winer, B. J. S tatistical principle in experimental design. 2nd ed., San Francisco: McGraw-Hill, 1971. Yen, H. C. Y ., Katz, M. H., and Krop, S. Effects of som e psychoactive drugs on conditioned avoidance response in aggressive mice. Arch. In t. Pharmacod.yn. Ther. , 1971 , 190, 103-109.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Age Differences In Primary And Secondary Memory Processes

PDF

Audience Effect And Age Differences On Tasks Of Increasing Complexity

PDF

Biofeedback Control Of The Eeg Alpha Rhythm And Its Effect On Reaction Time In The Young And Old

PDF

Changes In Memory As A Function Of Age

PDF

Influences Of Interoperative Experience And Age On Recovery Of Visual Function Following Two-Stage Lesions Of The Striate Cortex

PDF

The Effect Of Exercise Training On Systolic Time Intervals In Elderly Men

PDF

Protein synthesis in the hippocampus of rats during learning assessed by radioautography

PDF

Psychomotor Performance And Change In Cardiac Rate In Subjects Behaviorally Predisposed To Coronary Heart Disease

PDF

Signal Processing Time And Aging: Age Differences In Backward Dichoptic Masking

PDF

An Experimental Investigation Of The Relationship Of Age To Various Parameters Of Muscle Strength

PDF

Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

PDF

Effects Of Exercise Upon Joint Mobility Of Young And Old Men

PDF

Death Anxiety In Leukemic Children

PDF

Electrophysiological Properties Of Reaction Time And Aging

PDF

The Retroactive Effect Of Strenuous And Exhaustive Exercise On Maze Task Learning

PDF

The Effect Of Three Different Pace Plans On The Cardiac Cost Of 1320-Yardruns

PDF

Verbal And Motor Mediation In Normal Children Of Two Ages And Retarded Individuals

PDF

The Relationship Between Work Capacity And Motor Learning

PDF

Conflicting Motives In The Prisoner'S Dilemma Game

PDF

The Effect Of Participation In A Girls Inter-University Athletic Program Upon Selected Physiological Variables

Asset Metadata

Creator Herr, John Jackson (author)

Core Title Differential effects of epinephrine and propranolol on shuttle box avoidance learning in rats of different ages

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Birren, James E. (committee chair), deVries, Herbert A. (committee member), Kahan, James Paul (committee member), Walker, James Paul (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-760618

Unique identifier UC11364413

Identifier 7400921.pdf (filename),usctheses-c18-760618 (legacy record id)

Legacy Identifier 7400921

Dmrecord 760618

Document Type Dissertation

Rights Herr, John Jackson

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA