Therapy and immunization of long-term analgesia in rats

Therapy and immunization of long-term analgesia in rats

LEARNING AND Therapy THOMAS MOTIVATION 12, 133-148 and Immunization B. MOYE, (1981) of Long-Term Analgesia DEBORAH J. COON, JAMES STEVEN F. M...

1MB Sizes 0 Downloads 2 Views

LEARNING

AND

Therapy THOMAS

MOTIVATION

12, 133-148

and Immunization B. MOYE,

(1981)

of Long-Term

Analgesia

DEBORAH J. COON, JAMES STEVEN F. MAIER University

in Rats

W. GFUU, AND

of Colorado

Three experiments are reported which examine the effects of experience with escapable shock either subsequent to (Experiment 1) or prior to (Experiments 2 and 3) a session of inescapable shock on the subsequently produced long-term analgesic reaction in rats. Experment 1 demonstrated that experience with escapable shock 4 hr after a session of inescapable tail shock completely reverses the analgesic response that is normally observed 24 hr later upon reexposure to shock. The escapability of the shock was shown to be the important factor in reversing the analgesic reaction, since subjects given inescapable shock in amounts equivalent to escape subjects exhibited no reduction in analgesia. Experiment 2 showed that experience with escapable shock 4 hr prior to a session of inescapable tail shock could also completely eliminate the long-term analgesic reaction. Experiment 3 replicated the results of Experiment 2, but employed a different escape task and temporal parameters in order to extend the generality of the findings, and to more closely match the procedures employed in behavioral experiments reported by J. L. Williams and S. F. Maier (Journal of Experimental Psychology: Animal Behavior Processes, 1977, 3, 240-253). The implications of these results for the areas of pain control and learned helplessness were discussed.

Organisms exposed to stressful stimulation often exhibit reductions in their reactivity or sensitivity to painful stimuli. For example, exposure to stressors such as footshock (Akil, Madden, Patrick, & Barchas, 1976), cold water swims (Bodnar, Kelly, & Glusman, 1978), severe immobilization (Amir & Amit, 1978), rotation (Hayes, Bennett, Newlon, & Mayer, 1978), and food deprivation (Bodnar, Kelley, Spiaggia, & Glusman, 1978), have each been shown to produce decreased responsiveness on a number of different analgesimetric tests. This reduction in responsiveness to painful stimulation, or stress-induced analgesia, generally persists for at This research was supported by NSF Grant BNS 7800508 and RSDA MH 00314 to S. F. Maier. The authors would like to thank James P. Rodgers for assistance in data analysis and R. L. Jackson for assistance in running subjects in Experiment 3. Experiment 3 was submitted in partial fulfillment of the requirements for the Master of Arts degree by the second author. Requests for reprints should be sent to Thomas B. Moye, Department of Psychology, University of Colorado, Boulder, CO 80309. 133 0023-%90/81/020133-16$02.00/O Copyright 0 1981 by Academic Press, Inc. All rights of reproduction in any form reserved.

134

MOYE

ET AL.

most 2 hr following the stressful event (Bodnar er al., 1978; Akil et al., 1976). In a recent study reported by Jackson, Maier, and Coon (1979) it was demonstrated that an analgesic reaction could be elicited in rats 24 hr following exposure to inescapable tail shock provided that the subjects were briefly reexposed to shock immediately prior to analgesia testing. The design of the experiment was as follows: rats in one group were given inescapable shock via fixed tail electrodes while restrained in Plexiglas tubes. Rats in another group were merely restrained in the tubes without shock. Twenty-four hours later, half of the subjects from the preshocked and restrained groups received five escape trials in a shuttlebox in which one crossing (FR-1) terminated shock (for a rationale for this reexposure procedure, see Maier & Jackson (1979)). The other half did not receive shock. The subjects then immediately received either a hot plate or a tail-flick analgesia test. As expected, inescapably shocked rats not reexposed to shock 24 hr later were not analgesic. However, rats that were reexposed to shock (only a total of lo-15 set of shock), displayed a pronounced decrease in responsiveness on both hot plate and tail-flick tests. The reexposure procedure was not in itself sufficient to induce an analgesic response in subjects which had not been inescapably shocked 24 hr earlier. This study raises a number of interesting possibilities. Perhaps some of the behavioral consequences of exposure to inescapable shock occur because the organism has become analgesic (Maier, Coon, McDaniel, Jackson, & Grau, 1979). For example, poor escape performance following inescapable shock in learned helplessness experiments might occur partly because the aversive stimulus has been rendered less painful, causing the animal to be less motivated to escape the stimulus. Other sequelae of inescapable shock exposure such as decreased shock-elicited aggression (Maier, Anderson, & Lieberman, 1972), might also be mediated in this fashion. From the opposite perspective, some aspects of changes in pain responsivity might be explicable in terms of learned helplessness effects and might be influenced by variables known to be involved in the modulation of learned helplessness effects. An evaluation of these possibilities requires an assessment of the behavioral characteristics of the analgesic reaction which is aroused by brief exposure to shock 24 hr after exposure to inescapable shock, and a comparison of these characteristics with those of the learned helplessness effect. Unfortunately, very little is known about the behavioral nature of either this long-term antinociceptive reaction or the short-term reaction which rapidly dissipates without reexposure to the stressor. Experiments investigating the short-term effect have been almost exclusively restricted to pharmacological (e.g., Bodnar, Kelley, Spiaggia, Ehrenberg, & Glusman, 1978), biochemical (e.g., Madden, Akil, Patrick, & Barchas, 1977), and surgical (Amir & Amit, 1979) manipulations.

THERAPY

AND IMMUNIZATION

OF ANALGESIA

135

Two experiments have examined some behavioral characteristics of the long-term effect, but by themselves are not sufficient to draw strong conclusions about possible parallels with learned helplessness. In the first experiment, Jackson ef al. (1979) demonstrated that the long-term analgesic reaction, like the escape-learning deficit, was specific to the inescapability of the initial shocks. Some rats were first given escapable shocks, others received yoked inescapable shocks, and others no shock. Twenty-four hours later all subjects were given five shuttlebox escape trials followed by a pain reactivity test. Only the subjects that had initially received inescapable shocks became analgesic. In the other study, Maier et al. (1979) found a similar time course for both the shuttlebox escape deficit and the long-term analgesic reaction which follows inescapable shock. These two findings encourage the notion that there is a parallel between the changes in pain responsivity, inescapable shock and the learned helplessness effect. However, the learned helplessness effect has many other known characteristics and more study will be needed before a strong conclusion is warranted. Two of the most intriguing findings in the learned helplessness literature are that (1) prior experience with escapable shock will prevent inescapable shock from interfering with later escape acquisition (the so-called “immunization” effect (Seligman & Maier, 1967; Williams & Maier, 1977)), and that (2) subsequent experience with escapable shock, either by means of forcible exposure to the escape response, or the use of an escape task that the inescapably shocked subjects can perform, eliminates the interference with subsequent escape acquisition (the so-called “therapy” effect (Seligman, Maier, & Geer, 1968; Williams & Maier, 1977)). The purpose of the experiments to be reported was to determine whether immunization and therapy would occur with regard to the analgesic reaction which occurs 24 hr after exposure to inescapable shock. If prior experience with escapable shock prevents, and subsequent experience with escapable shock eliminates the nociceptive change produced by inescapable shock, a strong parallel between the learned helplessness effect and changes in pain responsivity would be suggested. In addition, little is known concerning the mechanism(s) which produce the nociceptive change, and the present results would constrain the possibilities. If immunization and therapy do occur, the nociceptive mechanism(s) must be readily altered by the experience of escape learning. EXPERIMENT

1

This experiment investigated the effect of experience with escapable shock following exposure to inescapable shock. Three groups of rats received a session of inescapable tail shock in restraining tubes. An escape-yoked procedure was not necessary since escapable shock does not produce an analgesic reaction (Jackson et al., 1979). Four hours later,

136

MOYE

ET AL.

one group was given training on a shock escape task in a two-way shuttlebox. This task consisted of 24 trials in which one crossing of the shuttlebox was required to terminate shock, followed by 24 trials in which two crossings were required for shock termination. Rats in a yoked group received equivalent amounts of inescapable shock in a shuttlebox, while the other inescapably shocked group was merely restrained in a shuttlebox for an equivalent period of time. Two other groups of rats received no inescapable shock in the first phase, but were merely restrained in the tubes. Four hours later, one group received escape training in the shuttlebox, while the other group was simply restrained in the box. Twenty hours later all subjects were given five single crossing shuttlebox escape trials followed by a tail-flick pain reactivity test. Method Subjects. The subjects were 43 male albino rats obtained from the Holtzman suppliers of Madison, Wisconsin. The animals were from 90 to 100 days old at the start of the experiment. They were maintained on a 12-hr-light/lZhr-dark cycle, with food and water continuously available in the home cages. Apparatus. Inescapable shock or restraint was administered in four Plexiglas tubes which measured 23.4 cm in length, and 7.0 cm in diameter. The rat’s tail was attached by adhesive tape to a Plexiglas rod which extended from the rear of the tube. Unscrambled shocks were delivered by four separate shock sources through electrodes coated with electrode paste and taped to the rat’s tail. The tubes were located in separate sound-attenuating chambers. Subsequent escape training, yoked inescapable shock, or restraint was carried out in four identical two-way shuttleboxes. The shuttleboxes measured 34.9 x 20.5 X 19.5 cm. Each chamber was divided into two equal-sized compartments by a metal wall which spanned the width of the box from floor to ceiling. A rectangular opening 5.4 cm high and 5.5 cm wide was cut in the bottom of the metal wall to allow rats to cross back and forth between compartments. The floors consisted of stainless-steel grids 0.35 cm in diameter and spaced 1 cm apart. Scrambled 0.6-mA shocks were delivered across the grids by four separate constant-current shock sources. Analgesia testing was conducted using a tail-flick device, which consisted of a 43.0 x 17.7 x &O-cm metal box which supported a 7.4 X 3.0-cm aluminum plate. A shallow slot was cut in this plate, and the rat’s tail was placed in this slot during a trial. A photocell receiver was mounted in the bottom of the slot. A General Electric 150-W spotlight was mounted above the slot which held the rat’s tail. A condenser lens was located 6.5 cm above the slot, and served to focus the light on the rat’s tail. A lateral deflection of the tail of at least 5 mm actuated the photocell receiver and automatically terminated the trial.

THERAPY

AND

IMMUNIZATION

OF ANALGESIA

137

Procedure. The subjects were randomly assigned to one of five groups (n = Wgroup). On the first day of the experiment, the subjects were placed in the restraining tubes. Three of the groups (PE, PY, and PR) were given 80 inescapable I.O-mA shocks through electrodes fixed to the tail. Each shock lasted for 5 sec. The shocks were delivered on a variable time 60-set schedule (range of 5-200 set). Subjects in the remaining two groups (RE and RR) were restrained in the tubes for an equivalent amount of time, but received no shocks. Four hours after the end of the inescapable shock session, subjects from Groups PE and RE were given shuttlebox escape/avoidance training. Trials were presented on a variable time 60-set schedule (range 5-150 set). The beginning of each trial was signaled by a lOOO-Hz tone which raised the background noise level from approximately 70 to 75 db (re 0.0002 dyn/cn-&‘). If no response occurred within 5 set of tone onset, a 0.6-mA shock was delivered, and terminated whenever a response occurred. If no response occurred within 35 set of tone onset, the trial was automatically terminated. During the first 24 trials of escape/avoidance training, one crossing (FR-1) of the shuttlebox after tone onset either prevented or terminated the shock. During the next 24 trials, the subject was required to cross the shuttlebox twice (FR-2) in order to avoid or escape shock. Each subject in Group PY was randomly assigned a partner in Group PE. During escape/avoidance training, subjects from Group PY were placed in the shuttlebox and received amounts of uncontrollable shock equal to that which their partners in Group PE received. Each member of a PE-PY pair received tone and shock at the same time, and the events terminated whenever the PE subject responded appropriately. Subjects in Groups PR and RR were simply restrained in the shuttlebox for periods of time equal to subjects in Groups PE, RE, and PY. Subjects in all groups were tested 20 hr later. The test procedure consisted of 5 FR-1 escape/avoidance trials in the shuttlebox (as described above) followed immediately by tail-flick testing. Each subject received 3 tail-flick test trials. After each trial, the subject was returned to the shuttlebox for approximately 3 min. On each test trial, the experimenter (who was unaware of group membership) held the subject in his/her hand and placed the subject’s tail in the grooved metal plate. A switch activated the lamp and started a timer. The light beam was focused on a spot about halfway between the base and tip of the tail. The heat lamp was initially adjusted to produce tail-flick latencies of about 6-8 set in naive subjects from the same shipment as used in the study. A trial was terminated if a tail-flick had not occurred in 20 set, and a 20-set latency was recorded. This was necessary in order to prevent tissue damage to the tail. Parametric procedures were used for all statistical analyses and a .05 rejection region was adopted.

138

MOYE ET AL.

Results and Discussion One subject from Group PE and one subject from Group RE failed to respond throughout the escape/avoidance training following inescapable shock or restraint in the tubes. The PE subject and his PY partner were dropped from the study and replaced with two more subjects. The RE subject was also excluded from the study and replaced. All other subjects in Groups PE and RE performed the FR-1 and FR-2 shuttlebox tasks. The mean latency to escape during the 48 total trials was 8.2 set for Group RE, whereas Group PE responded with a mean latency of 8.9 sec. Analysis of variance indicated that these latencies did not differ reliably [F( 1, 14) = 2.81. However, there was a significant main effect of group membership on the escape latencies during the 5 FR-1 escape trials that immediately preceded analgesia testing [F(4, 35) = 6.871. The mean latencies to escape were 5.8,6.1,7.5,8.2, and 12.5 set for Groups PE, RE, PR, RR, and PY, respectively. Thus some groups received more or less shock than other groups during the analgesia reinstatement procedure. The tail-flick results can be seen in Fig. 1, which shows mean tail-flick latencies for each group along with the standard errors of the mean. As can be seen, the usual analgesic effect of inescapable shock was observed (Group PR). Subjects given inescapable shock and tested 24 hr later (Group PR) were slow to flick in response to radiant heat. However, exposure to escapable shock between inescapable shock and testing prevented this analgesic effect from occurring (Group PE). Moreover, the escapability of the second shock exposure was critical in producing this reversal-a second experience in which the shock was inescapable (Group PY) had no reversing effect. Finally, the escape training procedure per se had no effect on tail-hick latencies (Group RE as compared to RR).

PEPR PI

GROUP

FIG. 1. Mean tail-flick latencies for subjects given escape training (PE), yokedinescapable shock (PY), or restraint (PR) 4 hr following a session of inescapable tail shock. The other groups received either no inescapable tail shock, then escape training 4 hr later (RE), or were not shocked at all (RR).

THERAPY

AND

IMMUNIZATION

OF ANALGESIA

139

These conclusions were confirmed statistically. Since the amount of shock received during the reexposure before testing differed between groups, an analysis of covariance was performed on the mean latency to flick. The first- through fourth-order terms of FR-I escape latency immediately prior to analgesia testing were treated as covariates, thus removing their contribution to the tail-flick latencies. This analysis revealed a substantial Groups effect [F(4, 31) = 13.851. Planned orthogonal comparisons following the analysis of covariance revealed that Groups PR and PY had reliably longer latencies to tail-flick than Groups RR, RE, and PE [F( I,3 1) = 72.991. Furthermore, Groups PR and PY did not differ from one another [F( 1,31) < 1.01, nor did Groups RR, RE, and PE differ from each other [F(2, 31) < 1.01. Thus the data clearly show that exposure to escapable shock 4 hr after a session of inescapable shock can reverse the analgesic response that is normally observed 24 hr later. This result is particularly striking since the treatment that reverses the long-term analgesic response is exposure to more shock. The critical factor, however, is that the shock must be escapable, since exposure to inescapable shock did not reduce the analgesic response whatsoever. EXPERIMENT

2

The findings of Experiment 1 clearly demonstrated that experience with escapable shock subsequent to a session of inescapable shock could completely reverse the long-term analgesic response that normally occurs. The purpose of Experiment 2 was to investigate whether experience with escapable shock prior to exposure to inescapable shock blocks the changes in nociception that would normally be observed. The same treatment conditions were employed in this experiment as in Experiment 1, the only difference being that the treatments occurred prior to either inescapable shock or restraint in the tubes. Also, since in Experiment 1 there were significant group differences in the amount of shock received duirng the 5 FR-I shuttlebox trials immediately prior to analgesia testing, subjects in this experiment were simply given five brief inescapable foot shocks as the analgesia reinstatement procedure. Method Subjects. The subjects were 40 rats of the same sex, strain, and age as in Experiment 1. Apparatus. The apparatus used was the same as in Experiment 1 with the exception that the brief reinstating shocks administered immediately prior to analgesia testing were given in six chambers with the dimensions 34.5 x 24.0 x 22.5 cm (L x W x H). The front and rear walls of each chamber were constructed of aluminum, and the side walls and ceiling were Plexiglas. The floor of each chamber consisted of 26 stainless-steel

140

MOYE

ET AL.

rods 3 mm in diameter spaced 1 cm apart. Scrambled shock was administered through the grids of each chamber by six separate constant current shock generators. Each chamber was housed in a separate sound attenuating chamber with a ventilating fan which also provided masking noise of approximately 70 db. Illumination in each chamber was provided by a 7.5-W bulb mounted on the far wall of the sound-attenuating chamber. Procedure. The subjects were randomly assigned to one of five groups (n = 8/group). On the first day of the experiment, subjects in two groups (EP and ER) were given escape/avoidance training in two-way shuttleboxes as described in Experiment 1. Subjects in Group YP were each assigned a partner in Group EP. During escape/avoidance training, subjects in Group YP were placed in a shuttlebox and received equal amounts of uncontrollable shock as their partners in Group EP received. Tone and shock for each EP-YP pair began simultaneously and terminated whenever the EP subject responded appropriately. Subjects in Group RP and RR were simply restrained in the shuttlebox for periods of time equal to subjects in Groups EP, ER, and YP. Four hours after the end of the shuttlebox session, subjects in Groups EP, RP, and YP were given 80 inescapable tail shocks in restraining tubes as in Experiment 1. Subjects in Groups ER and RR were merely restrained in the tubes for an equivalent period of time. Subjects in all groups were tested 20 hr later. In order to equate the amount of shock received by each group in the reinstatement procedure, subjects in all groups were given five 0.6-mA inescapable foot shocks prior to analgesia testing. Each shock was 1.5 set in duration. Immediately following the five foot shocks, subjects were given tail-flick analgesia tests as in Experiment 1. Results

and Discussion

All subjects in Groups ER and EP responded on each trial during the 24 FR-I and 24 FR-2 shuttlebox escape/avoidance trials. The mean latency to escape during the 48 trials was 8.6 set for Group EP and 8.8 set for Group ER. These latencies did not differ reliably [ F( 1, 14) < 11. The data of primary interest are shown in Fig. 2. The mean tail-flick latencies for Groups RP and YP were longer than those for Groups RR, ER, or EP. Thus, the usual effect of inescapable shock on tail-flick responding was again demonstrated (Group RP). Further, the antinociceptive response was blocked by prior experience with escapable shock (Group EP). Again, the block only occurred if the shock was escapable and not if it was inescapable (Group YP). Also as before, escape training per se had no effect (Group ER). Analysis of variance performed on the mean latency to tail-flick demonstrated a reliable effect of treatment condition [F(4, 35) = 10.971. Newman-Keuls comparisons ((Y = .05) indicated that both Groups RP

THERAPY

AND IMMUNIZATION

OF ANALGESIA

141

GROUP

FIG. 2. Mean tail-flick latencies for subjects given escape training (EP), yokedinescapable shock (YP), or restraint (RP) 4 hr prior to a session of inescapable tail shock. The other groups received either escape training, then no tail shock 4 hr later (ER), or were not shocked at all (RR).

and YP had reliably longer tail-flick latencies than Group RR, ER, or EP. Furthermore, Groups RP and YP did not differ from one another, and Groups RR, ER, and EP did not differ from one another. The data further demonstrate the’effects of exposure to escapable shock on the long-term analgesic response that occurs following exposure to inescapable shock. Here, exposure to escapable shock prior to a session of inescapable shock completely blocked the analgesic response that would normally have been observed. Further, it can be concluded from Experiments 1 and 2 that the escapability of the shock experienced either prior to or subsequent to inescapable shock determines its immunizing and therapeutic effects. Exposure to inescapable shock in the shuttleboxes either prior to or subsequent to a session of inescapable tail shock in restraining tubes had no effect on the long-term analgesic response, ruling out the possibility that some sort of habituation of the nociceptive mechanism had occurred because of repeated shock exposure. EXPERIMENT 3

The results of Experiments 1 and 2 closely parallel the behavioral results obtained in the immunization and therapy experiments reported by Williams and Maier (1977). The present experiments show that the analgesic response that occurs following inescapable shock can be eliminated by prior or subsequent experience with escapable shock. Similarly, Williams and Maier demonstrated that the deficits in shuttlebox escape/ avoidance performance that occur following inescapable shock could also be mitigated by immunization or therapy treatments. However, the procedures used here differed somewhat from those used by Williams and Maier (1977). Experiment 3 is an attempt to replicate the immunization

142

MOYEETAL.

effect on the long-term analgesic response with a procedure that more closely follows the procedure employed by Williams and Maier (1977). The use of procedures similar to those of Williams and Maier would strengthen statements about possible parallels between the present and earlier findings. In addition, it would indicate that immunization of the analgesic response is not specific to the procedures employed in Experiment 2, thus increasing the generality of the reported findings. The escape training in Experiment 3 employed a wheel-turn task in which subjects were required to turn a small wheel with their paws in order to terminate shock. This escape training took place 24 hr prior to a session of inescapable shock. Thus both the escape task and the temporal parameters were those used by Williams and Maier (1977) and differed from those in Experiments 1 and 2. Also, in order to follow as closely as possible the procedures employed by Williams and Maier, the brief reinstating shocks administered prior to analgesia testing consisted of 5 FR-1 escape trials in a shuttlebox, rather than the brief inescapable shocks employed in Experiment 2. Method Subjects. The subjects were 40 rats of the same sex, strain, and age as in Experiments 1 and 2. Apparatus. The apparatus used was the same as in Experiment 1 with the exception that the initial treatments of either escape training, yoked inescapable shock, or restraint were carried out in three wheel-turn boxes. The dimensions of the chambers were 16.0 x 12.6 x 17.5 cm (L x W x H). The walls, ceiling, and floors were all constructed of clear Plexiglas. A grooved Plexiglas wheel, 6.4 cm in diameter, was attached to the front of each chamber, and extended 1.5 cm into the chamber. The force required to turn the wheel was approximately 50 g, which was sufficient to prevent excess spinning of the wheel. The rat’s tail extended through an opening in the rear of the chamber, and was taped to a 1.Ccm-wide Plexiglas rod. Unscrambled shocks were delivered by three separate shock sources through electrodes coated with electrode paste and taped to the rats’ tails. Each chamber was housed in a separate, sound-attenuating box illuminated by a 7.5-W bulb. Procedure. The subjects were randomly assigned to one of four groups (n = lo/group). On Day 1, Groups EP and YP received 80 shock trials in a wheel-turn box, with shock onset occurring on a variable time 60-set schedule (range 30-200 set). Subjects in Group EP were able to control shock termination by turning the wheel at the front of the box. The first quarter-turn of the wheel occurring more than 0.8 set after shock onset resulted in shock termination. Subjects in Group YP were yoked to subjects in Group EP so that they received shocks of the same duration as EP subjects, but could not control shock termination. If an EP subject did

THERAPY

AND IMMUNIZATION

OF ANALGESIA

143

not respond within 30 set of shock onset, shock was automatically terminated for both the EP and YP subject. Shock intensity was initially set at 0.8 mA, and was increased to 1.0, 1.3, and 1.6 mA, respectively, after 20, 40, and 60 trials had occurred. This was done because pilot work had previously revealed that subjects will typically cease to respond if shock levels are held constant throughout wheel-turn training. Subjects in Groups RP and RR were simply restrained in the wheel-turn boxes for an equivalent amount of time. Twenty-four hours after the end of the session in the wheel-turn box, all subjects were placed in restraining tubes. Groups EP, YP, and RP received 80 inescapable shocks as in Experiments 1 and 2. Group RR was simply restrained in the tubes without shock for the same amount of time as the other groups. Twenty-four hours after the Day 2 treatment, all subjects were exposed to 5 FR-1 escape/avoidance trials in the shuttleboxes, and then immediately tested for analgesia as in Experiment 1. Results and Discussion Nine of the ten subjects in Group EP performed the wheel-turn escape response throughout the session. Average latency to escape shock during the first 10 trials was 11.91 set, whereas average latency to escape during the last 10 trials was 2.5 sec. The subject which failed to perform the escape task was removed from the experiment along with its yoked and nonshocked partners. There was no difference between groups in latency to respond during the 5 FR-1 escape trials immediately prior to analgesia testing [F(3, 32) = 2.251. Mean tail-flick latencies for all groups are presented in Fig. 3. As is apparent, prior experience with escapable shock again blocked the analgesic reaction produced by inescapable shock. Analysis of variance

GROUP

3. Mean tail-flick latencies for subjects given escape training (EP), yokedinescapable shock (YP), or restraint (RP) 24 hr prior to a session of inescapable tail shock. Subjects in the other group (RR) received no shock. FIG.

144

MOYE

ET AL.

indicated a significant effect of Groups [F(3, 32) = 9.601. Subsequent Newman-Keuls comparisons (cu = .05) revealed that Groups RP and YP had significantly longer latencies to tail-flick than either Groups RR or EP. In addition, Groups RP and YP did not differ from one another, nor were Groups RR and EP different from one another. These results once again demonstrate that rats which have first been exposed to escapable shock will not exhibit the analgesia-inducing effects of inescapable shock. Thus, the immunization effect upon analgesia is not specific to subjects that have received escape training on a shuttlebox escape/avoidance task, but is also observed in subjects that received escape training on a wheel-turn task. Furthermore, immunization occurs even if escape training takes place 24 hr prior to the inescapable shock session. Thus the effect is not specific to the temporal parameter employed in experiment 2. Finally, the results indicate that procedures similar to those which mitigate the shuttlebox escape/avoidance deficits that occur following exposure to inescapable shock are also effective in preventing the long-term analgesic response that occurs following exposure to inescapable shock. GENERAL DISCUSSION The purpose of the present experiments was to further explore the parallel between the effect of inescapable shock on escape performance and its effect on pain reactivity. In particular, we wished to determine whether the decreased pain reactivity produced by exposure to inescapable shock could be mitigated by experience with escapable shock prior to or subsequent to the inescapable shock treatment. The results were quite clear. Just as with the escape deficit, the antinociception that results from inescapable shock exposure was completely eliminated by experience with escapable shock either before (Experiments 2 and 3) or after (Experiment 1) inescapable shock treatment. Also, as with escape performance, the “immunization” and “therapy” effects were dependent on the escapability of the treatment and did not occur if the shock was inescapable. Immunization and therapy were not produced by simply experiencing shock on a second occasion. These findings provide a strong parallel between the nociceptive and escape performance changes which follow inescapable shock, and thereby suggest that they are not independent consequences of this treatment. A considerable body of knowledge concerning the anatomical, physiological, and pharmacological nature of pain control systems has been accumulated during the last few years (see Fields & Basbaum, 1978; Liebeskind & Paul, 1977; Sherman & Liebeskind, 1980, for reviews). In contrast, almost nothing is known concerning the psychological and environmental factors which engage and modulate the activity of these systems. This issue would seem to be critical to any understanding of the

THERAPY

AND

IMMUNIZATION

OF ANALGESIA

145

functional significance of endogenous pain control systems. Virtually the only relevant experiments are those which have been labeled as stressinduced analgesia and report that exposure to a stressor induces an analgesic reaction. However, even here research has focused on the physiological mediation of the effect rather than on critical psychological variables. The present experiments, along with those reported by Jackson et al. (1979), suggest that the analgesic reaction which follows exposure to a stressor may not be determined by simple exposure to a stressor, but that the controllability of the stressor may be critical. Jackson et al. (1979) found an analgesic reaction upon reexposure to shock 24 hr later only if the original shocks were inescapable. This outcome pointed to the importance of the psychological dimension of control in mediating the analgesic effects of exposure to aversive stimuli. The present results reinforce the critical importance of controllability by demonstrating that either prior or subsequent experience of control over shock will counteract or block the analgesia-inducing effects of exposure to uncontrollable shock. These findings are particularly intriguing because they indicate considerable lability in the operation of pain control systems. Clearly, prior experience with controllable shock must either prevent the uncontrollable shock from having its usual activating effect on analgesia-producing mechanisms or, alternatively, must establish an opposing process. Even more interesting, subsequent experience with controllable shock must be able to restore the system to its baseline condition or establish an opposing process. Thus, the system responsible appears to be highly responsive to the psychological factor of control and is not driven by mere exposure to aversive stimuli. It can be argued that this apparent importance of control makes adaptive sense. If a noxious stimulus is escapable or avoidable it would be most adaptive to escape and avoid it. However, if an aversive stimulus is inescapable and unavoidable it might be most adaptive to withdraw from the situation as much as possible and conserve bodily resources until the time when active coping is possible. Pain inhibition might be especially useful here because it would make it easier to conserve energy. It would be interesting if the analgesia studied here is mediated endorphinergically, because a function of this system might then be implicated. In fact, work by Lewis, Cannon, and Liebeskind (1980) and Maier, Davies, Grau, Jackson, Morrison, Moye, Madden, and Barchas (1980) suggests the involvement of the endorphins in the production of the analgesic reaction studied here, and thus the activation of endogenous opiates may be responsive to control. The implications of these results for learned helplessness are less clear. The parallel between the nociceptive and escape performance deficits produced by inescapable shock suggests a relationship, but does not

146

MOYEETAL.

indicate what the relationship might be. Maier and Jackson (1979), Maier et al., (1979), and Jackson, Alexander, and Maier (1980) have argued that inescapable shock produces both activity and associative deficits, which might be independent. They further argued that the shuttlebox escape performance decrement might primarily reflect activity factors, whereas choice measures such as Y-maze escape performance might be primarily determined by associative factors. The possibility that the activity deficit is partly produced by the analgesic reaction was also noted. Since immunization and therapy effects have previously only been studied in the shuttlebox, it might seem that the demonstration of immunization and therapy with regard to analgesia suggests that immunization and therapy with regard to escape performance operate only through alterations in analgesia-produced inactivity. Thus the data might appear to contradict the sort of mechanism envisaged by the learned helplessness hypothesis. Although this is certainly possible, the data do not compel this conclusion. First, immunization and therapy experiments using a Y-maze or some other measure of choice as the dependent measure have not yet been reported, and an effect on activity does not preclude an effect on associative interference. Both are easily possible. Second, there are not yet data which require the assumption that the activity reduction is produced by the analgesia. They could be separate consequences of an underlying alteration in a neurochemical system. A mode1 that is consistent with the data is that the experience of uncontrollability leads to the learning or the representation of response/outcome independence. This “cognitive state” might produce a neurochemical change and later associative interference. The neurochemical alteration could produce consequences such as analgesia and inactivity. That is, the associative effects of inescapable shock exposure might be produced by the cognitive state and the analgesic and activity effects by the neurochemical changes initiated by the cognitive state. The activity deficit may be produced by the analgesic reaction or be another consequence of the factor that produces the analgesia. Either possibility would produce a correlation between the two effects. Of course, there could also be separate but correlated neurochemical changes which produce the effects. Third, even if the “final cause’ ’ of immunization and therapy effects turned out to be changes in either the analgesic reaction or the substrate which produces the analgesic reaction, it is still necessary to ask why experience with controllable shock counteracts the effects of inescapable shock on the analgesic reaction or neurochemical change. It is difficult to avoid the sort of mechanism indicated by the learned helplessness hypothesis. It is possible that the analgesic reaction or neurochemical change is driven by the kind of cognitive factors which the learned helplessness hypothesis has argued are of importance in immunization and therapy. The present results do not separate these possibilities.

THERAPY

AND IMMUNIZATION

OF ANALGESIA

147

It is appropriate to close with a caution. We have assumed throughout that changes in speed of responding on a tail-flick test represent changes in pain perception or reactivity. We make this assumption because the tail-flick is one of the most widely accepted and popular assessment techniques used by researchers studying pain, The basis for this acceptance is that tail-flick latencies change appropriately when the organism is subjected to manipulations thought to alter pain reactivity or perception. Nevertheless, tail-flick responding might be influenced by other processes, and the changes that we observed might thus be a secondary consequence of alterations in a process not directly related to pain. Such a possibility can only be evaluated by further work utilizing procedures known to alter pain reactivity. REFERENCES Akil, H., Madden, J., Patrick, R. L., & Barchas, J. D. Stress-induced increase in endogenous opiate peptides: Concurrent analgesia and its partial reversal by naloxone. In H. Kosterlitz (Ed.), Opiufes and endogenous opiate peptides. Amsterdam: Elsevieri North-Holland, 1976. Amir, S., & Amit, Z. Endogenous opioid ligands may mediate stress-induced changes in the affective properties of pain related behavior in rats. Life Sciences, 1978,23, 1143-I 152. Amir, S., & Amit, Z. The pituitary gland mediates acute and chronic pain responsiveness in stressed and nonstressed rats. Life Sciences, 1979, 24, 439-448. Bodnar, R. J., Kelley, D. D., & Glusman, M. Stress-induced analgesia: Time course of pain reflex alterations following cold water swims. Bulletin of the Psychonomic Society, 1978, 11, 333-336. Bodnar, R. J., Kelly, D. D., Spiaggia, A., Ehrenberg, C., & Glusman, M. Dose-dependent reductions by naloxone of analgesia induced by cold water stress. Pharmacology, Biochemistry and Behavior, 1978, 8, 667-672. Bodnar, R. J., Kelly, D. D., Spiaggia, A., & Glusman, M. Biphasic alterations of nociceptive thresholds induced by food deprivation. Physiological Psychology, 1978, 6, 391395. Fields, H. L., & Basbaum, A. I. Brainstem control of spinal pain-transmission neurons. Annual Reviews of Physiology. 1978, 40, 217-248. Hayes, R. L., Bennett, G. J., Newlon, P. G., & Mayer, D. J. Behavioral and physiological studies of non-narcotic analgesia in the rat elicited by certain environmental stimuli. Brain Research, 1978, 155, 69-90. Jackson, R. L., Alexander, J. H., & Maier, S. F. Learned helplessness, inactivity, and associative deficits: Effects of inescapable shock on response choice escape learning. Jdurnal of Experimental Psychology: Animal Behavior Processes, 1980, 6, I-20. Jackson, R. L., Maier, S. F., & Coon, D. J. Long-term analgesic effects of inescapable shock and learned helplessness. Science, 1979, 206, 91-93. Lewis, J. W., Cannon, J. T., & Liebeskind, J. C. Opioid and nonopioid mechanisms of stress analgesia. Science, 1980, 208, 623-625. Liebeskind, J. C., & Paul, L. A. Psychological and physiological mechanisms of pain. Annual Reviews of Psychology, 1977, 28, 41-60. Madden, J., IV, Akil, H., Patrick, R. L., & Barchas, J. D. Stress-induced parallel changes in central opioid levels and pain responsiveness in the rat. Nature (London), 1977, 265, 358-360. Maier, S. F., Anderson, C., & Lieberman, D. A. Influence of control of shock on subsequent shock-elicited aggression. Journal of Comparative and Physiological Psychology, 1972, 81, 94-100.

148

MOYE ET AL.

Maier, S. F., Coon, D. J., McDaniel, M. A., Jackson, R. L., & Grau, J. The time course of learned helplessness, inactivity, and nociceptive deficits in rats. Learning and Motivation, 1979, 10, 467-487. Maier, S. F., Davies, S., Grau, J. W., Jackson, R. L., Morrison, D. H., Moye, T., Madden, J., IV, & Barchas, J. D. Opiate antagonists and the long-term analgesic reaction induced by inescapable shock. Journal of Comparative and Physiological Psychology, 1980, 94, 1172-l 183. Maier, S. F., & Jackson, R. L. Learned helplessness: All of us were right (and wrong): Inescapable shock has multiple effects. The Psychology of Learning and Motivation, 1979, 13, 155-218. Seligman, M. E. P., & Maier. S. F. Failure to escape traumatic shock. Journal of Experimental Psychology, 1967, 14, l-9. Seligman, M. E. P., Maier, S. F., & Geer, J. The alleviation of learned helplessness in the dog. Journul of Abnormal and Social Psychology, 1%8, 73, 256-262. Sherman, J. E., & Liebeskind, J. C. An endorphinergic. centrifugal substrate of pain modulation: Recent findings, current concepts, and complexities. In J. J. Bonica (Ed.), Pain. New York: Raven Press, 1980. Williams, J. L., & Maier, S. F. Transsituational immunization and therapy of learned helplessness in the rat. Journal of Experimental Psychology: Animal Behavior Processes, 1977, 3, 240-253. Received August 20, 1980 Revised December 5, 1980