SUMMARY Three adolescent male baboons were trained to press a bar in response to a bell or a light to receive food, and not to press in response to a tone. This simple conditioning paradigm was repeated more than 1,000 times until the subjects responded correctly no matter what time of day or night, despite the presence of strange people in the test chamber, and after interruptions in the testing schedule lasting as long as 30 days, thus illustrating the stability of the acquired responses. During this phase of the experiment the subjects were also tested for baseline levels and diurnal variations in motor activity, body temperature, blood pressure (BP), and electrocardiogram (EKG). Once performance was stable and the baseline data had been established, the animal were subjected to a stressor condition in which the normal 12-hour light/ 12-hour dark cycle (12L-12D) was replaced by a succession of cycles, such as 6L-6D or 24D-24L, and feeding times were irregular. Each time the diurnal records indicated that the subject's activity rhythm was becoming entrained to a new cycle, the cycle was altered. The subjects were studied for 10 weeks under the stressor condition and for 6 to 16 weeks after cessation of the stressor condition. All subjects showed disruption of the CRs within 12 days after onset of the stressor condition. They did not respond to the CSs when tested at night, and one subject gave paradoxical responses in the morning sessions. After 52 days, performance was disrupted in all subjects in all sessions. One subject responded to all signal stimuli in almost all sessions (excitatory response to stressor); one seldom responded at all (inhibitory response to stressor); and the third gave a mixed response. All failed to take the food reward at times. General behavior paralleled the changes in the performance of conditional responses. The first subject was extremely agitated and aggressive towards all humans in the test situation; the second was hypoactive and sat head-down during the test sessions; the third alternated between agitated behavior (including self-rnutilative biting) and hypoactivi ty. During the stressor condition all subjects showed unusually wide fluctuations in BP from day to day. Two showed EKG changes characteristic of coronary insufficiency. One showed a flattening of the normal biphasic diurnal fluctuation in body temperature, and one showed increased frequency of defecation during the test sessions. In the poststressor period the diurnal activity patterns returned to normal within 4 to 13 days, with only periodic aberrations thereafter. Although general behavior in the testing sessions reverted to approximately the patterns seen before the stressor condition was imposed, CR performance remained very poor in all three subjects. Two subjects became severely ill within two to three months after the stressor phase ended. One died of myocardial infarction, without signs of coronary atherosclerosis. The other suffered a two-day illness of undetermined origin characterized by hypothermia and hypotension. One month later, the animal was found to have hypertension. METHODS Subjects: Three male baboons (Papio hamadryas), 26 to 34 months old, weighing 5 to 7.9 kg: "Nosorog" (subject N), "Dog" (subject D), and "Bronenosetsn (subject B). None had a previous history of illness of any kind. Settinfil Manipulation of light/dark cycles and feeding times was done while the animal were in their home cages; most testing procedures were carried out in a special smooth-walled testing chamber (2.0 m x 2.5 m 2.0 m). The testing chamber rested on balloons, which were connected by a closed air system of tubes to a kymograph allowing diurnal recording of activity. A response bar in the cage was attached to the kymograph to allow recording of the force applied to the bar.. The kymograph also recorded stimulus markers corresponding to the signal stimuli used in the experiment. Procedures: Operant discrimination conditioning; stressor = disruption of diurnal stereotype. The experiment consisted of a prestressor phase (12 months), a stressor phase (10 weeks), and a post-stressor phase (2 to 4 months). Strong-CS = bell; Weak-CS = light; OS = tone; CR = bar press; UCS = a bit of candy, sugar, grape, or other preferred food. Stimulus duration was 10 sec with UCS presented immediately after the CR. The stimuli were always presented in the same order (Strong-CS, Weak-CS, DS), repeated three or four times to constitute one session. Intertrial interval was l min. Routine sessions occurred between 10:00 h and 12:00 h. These sessions were run daily in the prestressor period; during the stressor periods, they were held on days when 10:00-12:00 fell in "daylight" for the animal; in the poststressor period they were run once a week. In addition, each subject was tested on one to three occasions at 4-hour intervals around the clock before the stressor period. During the stressor period, such diurnal testing was carried out once after 12 days and again after 52 days. Independent Variables: Light-dark cycles and the timing of feedings relative to the light-dark cycle. During the prestressor periods, all subjects were on l 2L-l 2D beginning at 08:00 h. During the stressor phase, the subjects were shifted from one cycle to another every few days with l to 2 days of the normal cycle (12L-12D) after days 12 and 52. On the latter occasions, conditional performance, physiological status, and spontaneous behavior were studied at 4-hour intervals for one 24-hour period. To a certain extent the interval between shifts in cycle was tailored to the individual subject. When the subject seemed to be adapting to a new cycle {as judged by entrainment of gross activity to the current light-dark rhythm), the cycle was changed; this required changes every l to 2 days toward the end of the experiments. A sample sequence is presented in Table 116-1. This sequence of changes was followed by several weeks in which the normal (prestressor) light-dark schedule was reestablished, but feeding occurred during the day or night on a succession of different schedules. Table 116-1 Light (L) - Dark (D) Schedule Feeding Duration (calendar days) 12D-12L during L 5 6D-6L-6D-6L during L 9 6L-6D-6L-6D L 7 6D-6L-6D-6L during L 9 6L-6D-6L-6D during L 9 24L before noon 1 24L after noon 2 12D-12L during L l 24L before noon l 24L after noon 1 Dependent Variables: Performance in a conditioning paradigm, general behavior, and physiological functions. Conditioned performance was judged by the appropriateness of responses to the CSs and DS, the latencies to respond, and the force applied to the response bar. General behavior was evaluated by the level and distribution of general activity (represented by round-the-clock recordings of home cage activity), behavior of the animal toward the experimenter, etc. The physiological functions monitored included body temperature and BP (measured at each session), and the EKG (recorded at longer intervals). RESULTS Each subject underwent conditioning sessions for approximately 12 months (more than 190 sessions involving more than 1,000 trials with each CS and more than 600 with the DS). At the end of this time they performed at almost 100% accuracy with four strange persons in the room and after one month's pause in the testing procedure. The subjects pressed the response bar with greater force to Strong-CS than to Weak-CS (law of intensity). When testing sessions were run at 4-hour intervals around the clock, the only differences between performance during the day and at night were that the latencies to respond were increased and the force applied to the response bar was reduced at night. The first sign of disruption of the dynamic stereotype occurred approximately 12 days after onset of the stressor condition. When test sessions were run at 4-hour intervals around the clock, the subjects failed to respond to the CSs .in the 16:00 h session and in subsequent test sessions through the night. Subject B pressed more forcefully in response to the Strong-CS than to the Weak-CS; otherwise, performance of all subjects in the 08:00 h and 12:00 h sessions remained intact. When the same procedure was carried out after 52 days of the stressor condition, the performance of all subjects was disrupted in all sessions, but in different ways. Subject N was agitated, pressed repeatedly in response to all signal stimuli, and frequently pressed the bar between signals (excitatory response); 1t often failed to take the food reward. Subject D usually failed to respond to any of the signal stimuli. Occasionally, it responded once or twice late in the session but failed to take the food (inhibitory response). Subject B's performance varied. In the morning sessions it showed the same excitatory response as subject N; in the afternoon it often responded appropriately to the DS but with paradoxical responses to the CSs and frequent presses between trials; in the night sessions it did not respond to any of the stimuli. The subjects' general behavior also changed dramatically. With each change in the light-dark cycle the diurnal pattern of activity became irregular for a few days and then gradually began to follow the new light-dark rhythm. The rate at which activity entrained to the new rhythm increased until, by the end of the experiment, it was necessary to change the rhythm every day to prevent entrainment. As the experiment progressed subject N became more agitated, running about the cage, screeching in response to the signal stimuli, and attacking the experimenters and animal handlers. Subject D, by contrast, sat ln one corner of the test chamber with head down and vocalized occasionally. Subject B was agitated, aggressive, and self-mutilative (biting its own arm and leg) in the morning sessions. In later sessions it was unresponsive. There were also physiological changes in all three subjects. By the twelfth stressor day the biphasic temperature cycle disappeared in one animal (subject N); in the other two it remained synchronous with the prestressor day-night cycle. Great day-to-day fluctuations were seen in the BP of all three subjects, and two (subjects N and B) showed EKG signs of coronary insufficiency (wide QRS interval and negative T-waves in all leads). Finally, subject N showed increased frequency of defecation during its periods of extreme agitation. In the poststressor period, when the cycle returned to 12L-12D, the distribution of activity returned to normal in 4 to 13 days; it continued to show periodic aberrations in all subjects, however, even 2 months after the stressor period. Performance in the experimental chamber also remained disrupted for a long period of time. All three subjects were inconsistent in their responses to the CSs and frequently pressed the bar in response to the DS. Subjects N and D never responded in the midnight and 04:00 h sessions, while subject B pressed weakly but regularly in response to all signal stimuli around the clock. Two of the subjects became severely ill, and one died within 3 months after the stressor period. Subject N suffered a myocardial infarction. Two months after the stressor conditions ended, this animal showed a sudden onset of sluggishness and anorexia, followed in a few hours by total immobility. Auscultatory percussion yielded no diagnostic evidence, but an EKG the following day showed signs of infarction of the anterior wall of the left ventricle (negative T-wave in leads I and V4, absence of QRS complex in III, and marked elevation of the ST segment in V4). The animal died after 4 days and the initial diagnosis was confirmed at autopsy. The heart was normal in size, with the right atrium and ventricle slightly dilated and containing blood clots. The cardiac valves, pulmonary vessels, and intima of the aorta and coronaries were unremarkable. Microscopy revealed scattered areas in the left ventricle with severe dystrophic changes in the cardiac muscle fibers, including atrophy, loss of cells, and micronecrotic areas with peripheral areas of early fibroblastic proliferation. Other findings included mild erosions of the gastric mucosa, erythematous patches in the lungs, and nutmeg liver. Three months after the end of the stressor condition, subject B was ill for two days, displaying sluggishness, shivering and poor appetite. Its temperature was lower than normal (37.5°C) and it was hypotensive (BP= 100/60 mm Hg) relative to the prestressor period (BP= 130/80 mm Hg). The lungs were clear; EKG and stool cultures were normal. One month later this animal's BP had increased to 160/120 mm Hg.