Cortisol and Behavioral Responses to Separation in Mother and Infant Guinea Pigs. ROBYN L. RITCHEY AND MICHAEL B. HENNESSY 1. Department of ...
BEHAVIORAL AND NEURAL BIOLOGY
48, 1-12 (1987)
Cortisol and Behavioral Responses to Separation in Mother and Infant Guinea Pigs ROBYN L.
RITCHEY AND MICHAEL B . HENNESSY 1
Department of Psychology, Wright State University, Dayton, Ohio 45435 In Experiment 1, lactating guinea pigs 3 days postpartum exhibited an elevation of plasma levels of cortisol when their litters were taken from the home cage and placed out of auditory range for 30 rain. If pups were left within auditory range during the period of separation, or if the litter was disturbed but not removed, no increase in plasma cortisol levels was observed. In Experiment 2, 11/12- and 18/19-day-old guinea pig pups placed alone in a novel environment for 30 min displayed levels of plasma cortisol and vocalizations that were greater than those of pups tested in the same environment but with their mother. Pups tested with littermates but not with their mother exhibited plasma cortisol levels that were as great as those of pups tested alone. Pups tested with littermates vocalized much less than pups tested alone but more than pups tested with their mother. These results indicate that brief mother-infant separation can activate the pituitary-adrenal system in mother as well as infant guinea pigs and they provide further evidence for the existence of a reciprocal mother-infant attachment in this species. © 1987AcademicPress. Inc.
In most mammals, the affectional bond between mother and infant is considered to be a special social relationship. In various species (e.g., dogs, rhesus monkeys, and humans), brief involuntary separation of mother and infant elicits vocalizations and other signs of infant, and often maternal, distress (Ainsworth, 1972; Mineka & Suomi, 1978; Scott, 1971). These separation responses are probably the most commonly used measures of attachment between infant and mother (Cairns, 1966; Scott, 1971). Other criteria for attachment include the individual recognition of, and preference for, the other member of the dyad, the tendency of infants to follow their mother, to remain close to her in a threatening 1 These experiments were submitted by R.L.R. to the Department of Psychology, Wright State University, as a senior honor's thesis. The authors thank Dr. George Crampton for assistance throughout the project and Dr. Larry Kurdek for critically reviewing a draft of the manuscript. This research was supported by a grant from the National Institute of Mental Health (MH40905) to M . B . H . R . L . R . is now at the Department of Psychology, University of Illinois, Champaign. Address all correspondence and reprint requests to Michael B. Hennessy, Ph.D. 1
0163-1047/87 $3.00 Copyright © 1987 by Academic Press, Inc All rights of reproduction m any form reserved.
RITCHEY AND HENNESSY
situation, and to use the mother as a "secure base" from which to explore a novel environment (Ainsworth, 1972; Jensen & Tolman, 1962; Rosenblum, 1968; Smith, Van-Toiler, & Boyes, 1966). In addition to behavioral measures, physiological changes which occur during separation have been used to assess separation distress and hence attachment between mothers and infants (Mendoza, Smotherman, Miner, Kaplan, & Levine, 1978; Reite, Short, & Seiler, 1978; Smotherman, Hunt, McGinnis, & Levine, 1979). In squirrel monkeys, brief motherinfant separation produces increased activity of the pituitary-adrenal system, a neuroendocrine sign of heightened arousal (J. Hennessy & Levine, 1979). Thirty minutes following involuntary separation, both infant and mother show markedly elevated plasma levels of cortisol (Coe, Mendoza, Smotherman, & Levine, 1978; Mendoza et al., 1978). In contrast, brief separation of other social partners such as juvenile or adult peers, does not evoke reliable plasma cortisol elevations in this species, presumably because the social relationship, or attachment, between these partners is not as strong as that between mother and infant (Hennessy, Mendoza, & Kaplan, 1982; Hennessy, 1986; Mendoza & Mason, 1986). Among laboratory rodents, the species that exhibits the best evidence of both a specific and reciprocal mother-infant attachment is the guinea pig. Infant guinea pigs vocalize during maternal separation (Pettijohn, 1979) and follow and tend to remain in close proximity to the mother or artificial rearing figure in novel surroundings (Gaston, Stout, & Tom, 1969; Herman & Panksepp, 1978; Seward, 1940) and appear to use the mother/littermates as a "secure base" from which to explore the environment (Porter, Berryman, & Fullerton, 1973). There have been no clear demonstrations that guinea pig infants recognize and prefer their own individual mother, though available evidence suggests that this may be the case (Hennessy & Ritchey, in press). On the other hand, it is clear that guinea pig mothers do recognize and display a preference for their own individual pups (Porter, Fullerton, & Berryman, 1973). Recently, we found that guinea pig infants responded to brief maternal separation with an elevation of plasma cortisol levels (Hennessy & Ritchey, in press). To our knowledge, these results are the first evidence that a brief period of maternal separation can in itself elicit increased pituitaryadrenal activity in a rodent, and they provide additional support for the existence of filial attachment in this species. The purpose of the present study was to extend our observations concerning pituitary-adrenal responsiveness of guinea pigs during separation in two regards. The first experiment investigated the issue of attachment in guinea pig mothers by examining whether they would exhibit a cortisol response to brief separation from their pups. In the second experiment, the behavior and cortisol responses of guinea pig pups during maternal and sibling separation
CORTISOL AND BEHAVIORAL RESPONSES IN GUINEA PIGS
were compared to assess the relative strength of affiliation to these two classes of potential attachment objects. EXPERIMENT 1
In this experiment we measured changes in the plasma cortisol levels of lactating female guinea pigs during brief separation from their litters. The pups either were all removed from the testing room or three were left in auditory range in order to examine the effect of exposure to pup cues during separation. Because guinea pig pups exhibit cortisol responses to a 30-min separation (Hennessy & Ritchey, in press), we tested mothers using this same duration of separation. In this experiment, sample sizes were limited by the number of breeding females in our colony. Therefore, to provide a powerful test with the available number of subjects, a withinsubjects design was used in which each mother served as her own control for resting cortisol levels. Experimental conditions were imposed at 3 days postpartum because lactating guinea pigs have been found to be most responsive to infant distress vocalizations during the first week following birth (Pettijohn, 1977).
Method Experimental animals. Experimental animals were 23 lactating female guinea pigs of the Hartley strain. All animals were given water and food ad lib and were maintained on a 13/11 h light/dark cycle (lights on 0600). For breeding, one adult male was housed with four adult females. When noticeably pregnant, females were rehoused individually in clear plastic cages (48.3 x 38.1 x 20.3 cm) with hardwood chips covering the floor. Conditions and test procedures. Three days following birth, the maternal cage was transported to an adjacent test room where the mother was examined under one of three experimental conditions. Under the Disturbance condition (N = 8) all pups were removed from the maternal cage and then immediately returned to it for a 30-min period. This condition provided a control for the disturbance involved under the other two experimental conditions. The Separation condition (N = 8) was identical to the Disturbance condition with the exception that all pups were removed and then placed into a chamber in an adjacent room out of auditory range. Under the Pup Cues condition (N = 7), three pups were placed individually into opaque plastic isolation cages (27.9 x 17.8 x 12.7 cm) located just outside one end wall of the maternal cage for 30 min. Any additional pups in the litter were taken out of auditory range. Thus, under this condition mothers were alone in the cage but were exposed to clear auditory and presumably olfactory cues of three pups. We chose to use three pups in order to maximize pup vocalizations and because guinea pig litters typically contain at least three pups. The number of pup vocalizations per test session under this condition ranged from 1286
RITCHEY AND HENNESSY
to 4842 (~ = 2288). The pup isolation cages remained in place during testing under the Disturbance and Separation conditions. Eight of the 23 subjects were tested with their first litter. Three of these were assigned to the Disturbance condition, two to the Separation condition, and three to the Pup Cues condition. Under all three conditons, each subject had a blood sample collected immediately following the 30-min test session for determination of plasma cortisol levels. Approximately 24 h later, each lactating female was removed from the home cage and a second blood sample was obtained to estimate resting cortisol levels. Blood sampling procedure and cortisol determination. All blood samples were collected under ether anesthesia via cardiac puncture using a heparinized tuberculin syringe. Samples were collected within about 2.5 to 3.5 min, measured from the initiation of disturbance. Although the glucocorticoid response latency of the guinea pig has not, to our knowledge, been documented, available data from rats and mice suggest that the blood was collected rapidly enough to prevent the sampling procedure itself from having any appreciable effect on obtained cortisol values (Coover, Heybach, Lenz, & Miller, 1979; Riley, Fitzmaurice, & Spackman, 1981). All blood samples were collected between 1430 and 1730 h. Samples were centrifuged to separate plasma, which was then extracted and frozen until assayed for cortisol (the primary glucocorticoid secreted by the guinea pig adrenal) using standard radioimmunoassay procedures. Tritiated cortisol, antisera, standards, and general protocol were provided by Radioassay Systems Laboratory. Aliquots were assayed in duplicate and these values were then averaged to provide the final cortisol score for each sample.
Results Blood samples for two animals were lost, reducing the sample size to seven under the Disturbance condition and to six under the Pup Cues condition. Mean (+ SE) cortisol levels (~g/100 ml plasma) were as follows: Disturbance condition, test = 190.1 (+27.8), resting = 175.4 (_+43.2); Separation condition, test = 274.5 (_+67.3), resting = 159.1 (-+44.2); Pup Cues, test = 189.2 (-+61.9), resting = 142.5 (-+35.4). The mean cortisol increase (test value-resting value) for animals under each of the three experimental conditions is depicted in Fig. 1. For the Separation condition, the increase over the resting level was significant [t(7) = 2.64, p < .05]. There was no significant change from resting values under the Disturbance or Pup Cues conditions (ts < 1). In order to determine whether the cortisol response under the Separation condition could be attributed to separation from the litter as opposed to the disturbance involved in the procedure, a direct comparison was made between the cortisol increase under the Separation condition and that under the Dis-
CORTISOL AND BEHAVIORAL RESPONSES IN GUINEA PIGS
E 1 O0 ¢w E oo
///// / / / / / / / / / / / / / / /
/////, ///// ///// /////, /////~ /////
U/Ill /////~ /////.
FIG. 1. Mean increase in plasma cortisol levels from base values for mothers in the Disturbance, Separation, and Pup Cues conditions of Experiment 1. turbance condition. This comparison was significant [t(13) = 1.93, p