Changes in yearling rhesus monkeysТ relationships with their mothers after sibling birth.код для вставкиСкачать
American Journal of Primatology 54:193–210 (2001) Changes in Yearling Rhesus Monkeys’ Relationships With Their Mothers After Sibling Birth B.J. DEVINNEY1*, C.M. BERMAN1,3, AND K.L.R. RASMUSSEN2,3 1 Department of Anthropology, State University of New York at Buffalo, Buffalo, New York 2 Laboratory of Comparative Ethology, National Institute of Child Health and Human Development, Poolesville, Maryland 3 Caribbean Primate Research Center, Punta Santiago, Puerto Rico The birth of a new sibling is believed to signify an abrupt and important transition in a young primate’s relationship with its mother—one that is of potential importance from at least three theoretical perspectives: attachment theory, parent–offspring conflict theory, and dynamic assessment models. This study examines changes in relationships between free-ranging yearling rhesus monkeys (Macaca mulatta) and their mothers concomitant with the birth of the mother’s next infant, and tests predictions derived from each theoretical perspective. We observed 31 yearling rhesus on Cayo Santiago, Puerto Rico, 3 months before and 3 months after their siblings’ births, using focal animal sampling methods. Changes in measures related to mother–yearling interaction and yearling distress were examined using repeated-measures analysis of variance. After sibling birth, mothers and yearlings abruptly reduced amounts of time in contact and increased amounts of time at a distance and out of sight of one another. Mothers and yearlings played approximately equal roles in bringing about decreases in proximity, and yearlings took the primary roles in bringing about decreases in contact. Rates of maternal aggression toward yearlings increased immediately and markedly after birth, possibly providing yearlings with early cues regarding subsequent decreased levels of maternal care. There were no marked increases in overt signs of yearling distress (e.g., vocalizations or tantrums) following the births. We conclude that yearlings generally acquiesced to reduced levels of care, responding behaviorally with increased independence and maturity. In this sense, our study provides preliminary support for dynamic assessment models over attachment theory and parent–offspring conflict theory models. Am. J. Primatol. 54:193–210, 2001. © 2001 Wiley-Liss, Inc. Key words: sibling birth; Macaca mulatta; juveniles; mother–offspring relationships Contract grant sponsor: Wenner-Gren Foundation for Anthropological Research. *Correspondence to: B.J. DeVinney, Laboratory of Comparative Ethology, NICHD, P.O. Box 529, Poolesville, MD 20837. E-mail: email@example.com Received 12 October 2000; revision accepted 30 April 2001 © 2001 Wiley-Liss, Inc. 194 / DeVinney et al. INTRODUCTION The birth of a new sibling is believed to signify an important transition in a young primate’s relationship with its mother. For many species, it is marked by an abrupt decrease in direct forms of maternal care and it often coincides with the complete termination of suckling by the older sibling. As such, it is a transition that is of potential importance from a number of theoretical perspectives, although none address it specifically or exclusively. Attachment theory [Bowlby, 1969; Ainsworth, 1979] would liken the event to the psychological separation of a mother from her infant [Berman et al., 1994], or “partial removal” of the mother [Holman & Goy, 1988]. Hence it might predict similar responses to those seen in studies of physical separation from mothers [reviewed by Mineka & Suomi, 1978], such as increases in behavioral distress followed in some cases by signs of depression, particularly among anxiously attached offspring. Parent–offspring conflict theory [Trivers, 1974] potentially views this transition as a shift in the mother’s distribution of investment from the older immature to the younger. It would predict the increased occurrence of behavioral conflict between mother and yearling as the yearling attempts to use deceptive signaling to elicit more investment from the mother than she is “willing” to give. Alternative “dynamic assessment” approaches to parent–offspring relations propose that rather than conflict, cooperation and compromise between parents and offspring over levels of care may be prevalent [Altmann, 1980; Bateson, 1994]. Bateson’s  approach suggests that during reductions in care, parents provide cues regarding their ability to provide care, offspring signal honestly regarding their condition or need for resources [see Grafen, 1990; Godfray, 1991], and both monitor each other, as well as environmental conditions, and respond dynamically. For example, cat and rat mothers experiencing a restricted diet precociously wean their offspring, and those offspring exhibit accelerations in behavioral development during the weaning process [Bateson et al., 1990; Smith, 1991]. In addition, baboon infants’ weaning tantrums seem to function as honest signals of need to which mothers respond with increased nursing. Tantruming occurs only when weaning foods are unavailable due to seasonal changes in the environment [Barrett & Henzi, 2000]. The relevance of each of these perspectives for the sibling birth transition can not be ascertained, partly because little systematic research has been done documenting the nature of changes in mother–offspring relationships upon the birth of new infants. In this work, we begin to fill this gap by describing sibling birth transitions among free-ranging rhesus monkeys on Cayo Santiago, Puerto Rico, and by testing a few predictions for each perspective. Our goal is not to provide definitive support for one theory over the others, but rather to guide further research into the sibling birth transition. No detailed study of nonhuman primates’ reactions to the birth of a sibling has been conducted outside of captivity other than Lee’s [1983a] study of wild vervet yearlings, and that study included only three yearlings. The sibling birth transition has been studied more thoroughly among humans than among nonhuman primates. The findings to date suggest that a number of parallels exist between the behavioral patterns of human and nonhuman primates in the initial weeks and months after sibling birth. Among humans, older siblings typically experience reduced levels of maternal care or interaction and/or increases in confrontation with their mothers [Dunn & Kendrick, 1980; Dunn et al., 1981]. They also may exhibit signs of increased distress, including tearfulness, clinginess, and increased demandingness and negativity, and signs of disturbance such as withdrawal, passivity, or aimless wandering. Paradoxi- Yearling Rhesus at Sibling Birth / 195 cally, a given child can show both signs of regression (i.e., decreased self-care) and evidence of increased independence in different behaviors [Taylor & Kogan, 1973; Trause et al., 1981; Dunn & Kendrick, 1982]. A few studies of this transition in cercopithecines have been reported. Among cercopithecines, one sees decreased maternal time in proximity (vervets [Lee, 1983a] and rhesus macaques [Holman & Goy, 1988]), decreased maternal grooming [Lee, 1983a], decreased maternal approaches (male rhesus [Holman & Goy, 1988]), and increased maternal aggression (pigtail macaques and Papio spp. [Bolwig, 1980]) to yearling offspring following the birth of a younger sibling. In addition, at sibling birth, mothers who have not already weaned their older offspring terminate suckling (Japanese macaques [Tanaka, 1992]). A few individual youngsters among vervets [Lee, 1983a] and captive rhesus [Holman & Goy, 1988] may briefly become “depressed” (i.e., they may sit in a typically hunched posture indicative of depression in macaques [Kaufman & Rosenblum, 1967]). Rhesus older siblings (mean age at sibling birth = 2 yr) show increases in interaction with their mothers after sibling birth, apparently due to the older siblings’ interest in the new infant siblings [DiGregorio et al., 1987]. The subjects of DiGregorio et al.’s study were housed with their families in a nuclear family apparatus in which parents and peers were in separate parts of the housing, thereby forcing subjects to make a choice between interacting with family vs. peers. After sibling birth, juvenile cotton-top tamarins, members of a cooperatively breeding species, increase time near parents compared to the period before sibling birth [Achenbach & Snowdon, 1998]. This increase is due to the juveniles’ greater time spent near their parents when the parents are carrying new infants, illustrating the juveniles’ great interest in their infant siblings. Among humans, sex differences in response to sibling birth occur such that withdrawal is more common among boys, while dependence is typical for girls [Dunn & Kendrick, 1980; Nadelman & Begun, 1982]. Captive yearling rhesus males show initial increases in agitation followed by decreases in time spent near the mother, whereas females stay near their mothers and achieve proximity to the new infant as much as they did previously by approaching and grooming their mothers more [Holman & Goy, 1988]. Defining when the age period of infancy ends and when juvenescence begins among nonhuman primates has been a source of controversy and has been based on various criteria, such as the period during which the individual is physically dependent on the mother/caretakers, the age at which the individual can survive the death of its mother, the age at weaning, or the interbirth interval [Walters, 1987]. Each of these criteria potentially points to a different age at which infancy ends, with the interbirth interval being the upper limit [Pereira & Altmann, 1985]. As rhesus monkeys are seasonal breeders, the interbirth interval is minimally about 1 year. Thus, at sibling birth, yearling rhesus are making the transition from infancy to juvenescence. In this study, we ask the following specific questions: 1) to what extent do mothers and yearlings experience abrupt decreases in proximity and contact patterns upon the birth of a sibling; 2) to what extent are these changes due to changes in maternal vs. yearling behavior; 3) to what extent do yearlings show increases in overt distress, contact-seeking behavior, or signs of depression; and 4) to what extent do patterns of change differ for males and females? All three theoretical perspectives predict that mothers and yearlings will show abrupt decreases in contact and proximity at the time of the sibling’s birth. Attachment theory assumes in addition that the agent of separation will not be the yearling itself; hence, it predicts that 1) mothers will be primarily respon- 196 / DeVinney et al. sible for decreases in proximity and contact; and 2) yearlings will respond with increases in behavior related to distress, disturbance, and the reestablishment of contact. Parent–offspring conflict theory similarly assumes that the mother will be the agent of separation as she redirects investment into the new sibling. Hence, it also predicts that 1) mothers will be primarily responsible for decreases in contact and proximity; 2) yearlings will respond with increases in behavior related to distress and the reestablishment of contact; and 3) yearlings will display “regressive behavior,” i.e., reversions to more immature behavior patterns. It sees both reversion and increases in distress and contact-seeking behavior as dishonest attempts to elicit additional investment from mothers who may resist, ignore, or acquiesce to their demands, depending on conditions [Trivers, 1974]. Bateson’s  dynamic assessment approach predicts that 1) mothers will provide a prominent cue to yearlings at the time of the infant’s birth that predicts or accompanies decreases in levels of care. Yearlings in low-risk (e.g., food-rich and predator-free) environments, such as Cayo Santiago, are predicted to respond cooperatively to these cues. For this reason, the dynamic assessment approach predicts that 2) mothers in low-risk environments may not be primarily or directly responsible for actual decreases in proximity or contact with yearlings. Rather, mothers and yearlings may equally share this responsibility or yearlings may even assume the primary role. Similarly, yearlings in low-risk environments are predicted 3) to show few signs of distress, disturbance, or contact-seeking behavior. On the contrary, they are expected to show signs of acquiescence and increased behavioral maturity. METHODS Study Site and Sample This study was conducted between April 1994 and July 1995 on Cayo Santiago, a 15.2-ha island located 1 km from Puerto Rico’s southeast coast. C.R. Carpenter founded the free-ranging rhesus monkey (Macaca mulatta) colony residing there by introducing 409 animals from India in 1938. Since that time, individuals have been added to the population only by birth, although some individuals have been removed in planned culls. The island population during the entire data collection period averaged 897 individuals (range = 812–960). The monkeys were predatorfree and were provisioned with high-protein chow supplied from hog feeders inside quarter-acre corrals. Water was available ad libitum. The monkeys were habituated to human presence, and identification of individual monkeys was facilitated for observers by the monkeys’ tattoos and ear-notches. The population has been censused daily since 1956. Maternal kinship relationships were known for all monkeys. (For more information on Cayo Santiago’s history and terrain, see Rawlins and Kessler  and Sade et al. .) The monkeys organized themselves into species-typical multi-male, multifemale groups [Lee, 1983b]. We studied two social groups: group R, which averaged 237 animals (range = 212–262) and contained three matrilines, and group S, which averaged 125 animals (range = 113–134) and contained two matrilines. Groups R and S formed when group I fissioned during 1985 and 1986. Thus, the oldest females in each group were, at one time, members of the same social group. During this study, group S was in the process of fissioning, and group S females again began associating with group R females. Thus, the subjects of this study associated both with animals considered to be in their social group and with another fission-product social group. We observed 31 subjects whose mothers gave birth to viable siblings when the yearlings were between the ages of 40 to 67 wk (average inter-birth interval Yearling Rhesus at Sibling Birth / 197 = 54.36 wk or 380.5 days). Focal subjects were born between 24 November 1993 and 19 April 1994; their younger siblings (10 males, 21 females) were born between 17 December 1994 and 16 April 1995. We observed 19 males (12 in group R, 7 in S) and 12 females (all in group R). All mothers were multiparous. Subjects were balanced as much as possible for mother’s age and dominance rank. The study began with 36 subjects and we had initially planned to compare changes in mother–yearling relationships in families with and without new siblings; however, due to the high fecundity of the mothers (and the deaths of one mother and one infant sibling), only three families without new infant siblings were available for comparison. Data Collection We used focal animal sampling methods [Altmann, 1974] to record the frequencies, durations, and sequences of yearlings’ interactions with all other group members, observing each subject for three 15-min sessions per wk. During the focal-animal sessions, we also collected point (instantaneous) time samples at 3min intervals, recording 1) the identities of animals in contact, within 60 cm, and within 5 m of the subject; 2) the yearling’s and mother’s activities; and 3) the yearling’s activity level and apparent affective state. The behaviors analyzed are listed along with their definitions in Table I. Interobserver reliability was high among observers (Kappa coefficient ≥ .90 [Cohen, 1960]). The day was divided into four time periods: 0700–0930, 0930–1200, 1300–1530, and 1530–1800 AST. The collection of samples was distributed evenly across these time periods within each month of data collection. The sequence of subjects observed within each time period was determined by a random order. After the birth of a younger sibling, data collection was increased to two 15-min sessions per day for 2 wk. We collected the data on hand-held computers (Micropalm 4000/5000), reviewed them for typing errors soon after data collection, and later tabulated them within month-long blocks of time. The month-long blocks were based on the date of the younger sibling’s birth, using that date as the zero point in a time line and counting out by months from that point. The annual trapping period, during which all yearlings on Cayo Santiago were tattooed and ear-notched, and during which our subjects were studied for physiological parameters, occurred between 17 January and 17 February 1995. Capture of animals occurred Monday through Thursday weekly. To minimize the effects of trapping on our data, we collected as much data as possible between Friday and Sunday of each week. In addition, we waited 2 days before resuming observations after each subject was released. Data Analysis The data analysis focused on 6 mo during the yearling life-stage, specifically the 3 mo before and the 3 mo after the younger sibling’s birth. The average total amount of data collected per subject for this study was 24.58 hr (±SD = 31.76 min; range = 23.7–26.3 hr). Repeated-measures ANOVA techniques [SPSS version 10.0, 1999] were used to identify significant changes over time in measures related to the yearlings’ affective state and nursing, and in proximity, contact, and agonism between mothers and yearlings. In the repeated-measures design, we entered yearling sex, mother’s rank, and social group as between-subjects factors, yearling age (in days) as a covariate, and the level of each behavioral measure across each of the 6- 198 / DeVinney et al. TABLE I. Behaviors Analyzed and Their Definitions Behavioral measures related to nipple contact Time on nipple: percentage of point time samples in which the yearling is on mother’s nipple Nipple attempts: frequency/hour of successful and rejected attempts by the yearling to gain access to the mother’s nipple Nipple contacts: frequency/hour of successful attempts to gain access to the nipple initiated by the yearling or mother Rejections: rate per hour of attempts to get on the nipple that are prevented by the mother Behavioral measures related to proximity and contact Time in contact: percentage of point time samples in which the yearling is in physical contact with mother Time < 60cm: percentage of point time samples that the yearling is within 60cm of the mother Time > 5m: percentage of point time samples that the yearling is over 5m away from the mother Time out of sight: percentage of point time samples in which the mother and yearling are out of sight of one another Approaches to mother: frequency per hour of yearling crossing from more than 60cm away from mother to less than 60cm from mother Approaches by mother: frequency per hour of mother crossing from more than 60cm away from yearling to less than 60cm from mother Yearling contacts mother: frequency per hour with which yearling makes physical contact with mother Mother contacts yearling: frequency per hour with which mother makes physical contact with yearling Proportion of mother-initiated contacts: ratio of physical contacts initiated by the mother divided by total physical contacts initiated by both mother and yearling Behavioral measures related to maternal aggression and infant distress Mother threatens: rate per hour of any characteristic threat gesture, e.g., open-mouth threat, head bob, eye-raising, lunging while staying in place Mother lunges: rate per hour of mother moving at least three feet toward yearling, or chasing yearling without making contact with it Gecker: rate per hour of bursts of staccato vocalizations Scream: rate per hour of screaming or shrieking Fear grins/cower: rate per hour of bared teeth grimace and/or lateral flexion of the spine away from mother Time passive: proportion of point time samples in which the yearling was stationary; could be just sitting or self-grooming Time tense/distressed: proportion of point time samples in which the yearling appeared physically tense (i.e., muscles tensed) or exhibited agitated behaviors, such as distress vocalizations Time depressed: proportion of point time samples in which the yearling was sitting in a hunched posture, with head hanging down, yet animal is awake Behavioral measures related to grooming Mother grooms: duration of stereotyped picking through yearling’s fur Yearling grooms: duration of stereotyped picking through mother’s fur, Table II month-long blocks of time (three pre-sibling-birth and three post-sibling-birth) as the within-subject dependent variables. Probability values for F tests were adjusted using the Greenhouse-Geisser method when necessary [Greenhouse & Geisser, 1959]. As a post hoc test to examine when changes occurred within the 6-month period for each behavioral measure, we computed paired t-tests, pairing sequential months. For measures related to suckling, paired comparisons were Yearling Rhesus at Sibling Birth / 199 not made for the second and third post-sibling-birth months. Critical values for the paired t-tests were Bonferroni-corrected such that for nursing-related measures the results were considered statistically significant at P < .016 and for all other measures at P < .01. All statistical tests were two-tailed. Because the annual trapping season on Cayo Santiago overlapped with the birth season, it was necessary to account for possible effects of trapping on levels of mother–yearling interaction. We wanted to compare yearlings who had experienced sibling birth and trapping simultaneously with those who had experienced the two events separately. We used the following approach to categorize a yearling’s experience of trapping: if there were 3 or more weeks overlap between trapping season and the sibling’s first month of life, then these subjects were considered to have experienced sibling birth “during” trapping (n = 9). Alternatively, if 3 or more weeks of the sibling’s first month of life were either before or after the trapping season, then these subjects were categorized as having sibling birth “before/after” trapping (n = 12). RESULTS Proximity and Contact Fifteen of 31 (48%) of the yearlings were observed on their mothers’ nipples during the month immediately preceding their siblings’ births, up to within 1 day of sibling birth (seen in two subjects). None was seen on the nipple after the sibling’s birth. Thus, for some subjects, completion of weaning coincided with the sibling’s birth, and among these yearlings the mean percentage of time on the nipple decreased by nearly 50%, from 17% of observation time to 9% of observation across the 3 months preceding sibling birth. For all subjects, time on the nipple (F3,72 = 18.210, P = 0.000; Fig. 1a), nipple attempts (F3,72 = 21.563, P = 0.000), nipple contacts (F3,72 = 21.333, P = 0.000), and maternal rejections per hour (F3,72 = 10.045, P = 0.000) all declined significantly across the pre-birth period (Table II). Yearlings experienced significant reductions in the proportion of time spent in physical contact with mothers (F5,120 = 23.707, P = 0.000; Fig. 1a) and within 60 cm of mothers (F5,120 = 2.723, P = 0.041) over the 6-month period. Figure 1a suggests that mean values of time in contact decreased most rapidly between the first pre-birth month and the first post-birth month, after which they changed little. Similar patterns of change were seen for time at a distance of more than 5 m (F5,120 = 26.726, P = 0.000) and for time out of sight (F5,120 = 24.742, P = 0.000; Fig. 1b) of the mother. Mothers significantly reduced their grooming of yearlings over the 6 months (F5,120 = 8.647, P = 0.000), and yearlings significantly reduced grooming of mothers as well (F5,120 = 2.569, P = 0.030). Changes in Maternal Vs. Yearling Behavior Rates of approaches by yearlings to mothers (F5,120 = 21.079, P = 0.000) and rates of approaches by mothers to yearlings (F5,120 = 11.924, P = 0.000) also declined significantly over the 6 months and appeared to decline most rapidly between the first pre-birth and first post-birth months (Fig. 1c). The proportionate changes in rates of approaches from the first pre-birth to the first post-birth month were similar for mothers and yearlings, suggesting that each contributed approximately equally to changes in their proximity relationships. Rates of initiations of contact by both mothers (F5,120 = 7.019, P = 0.000) and yearlings (F5,120 = 11.447, P = 0.000) also declined significantly over the 6-month period (Fig. 1d). 200 / DeVinney et al. Fig. 1. Means and standard errors for: (a) the proportion of time in contact with the mother and the proportion of time on the nipple, (b) the proportion of time over 5 m from the mother and the proportion of time out of sight of the mother, (c) the rate per hour of approaches by the yearling to the mother and the rate per hour of approaches by the mother to the yearling, and (d) the rate per hour of the yearling making physical contact with the mother and the rate per hour of the mother making physical contact with the yearling. The dashed line in the center of each graph indicates the time of the younger siblings’ births. The time blocks represented along the x-axis are in a format such that “3mPRE” stands for the third month presibling-birth, and so forth. The decrease in mean rates of contact initiation by yearlings was most dramatic (43%) between the first pre-birth month and first post-birth month. On a descriptive level, rates of contact initiation by mothers increased for daughters in the first post-birth month and decreased slightly for sons. The proportion of physical contacts initiated by the mother (calculated as the ratio of mother contacts yearling over mother contacts yearling plus yearling contacts mother) actually increased after sibling birth (F5,90 = 2.464, P = 0.039; Fig. 2a), suggesting that yearlings, not mothers, were primarily responsible for decreases in contact time. Rates of maternal threats (F5,120 = 3.415, P = 0.006) and lunges (F5,120 = 2.734, P = 0.022) directed toward yearlings increased significantly and dramatically over the 6-month period. Like other measures, the largest changes generally coincided with the birth of the sibling (Fig. 2b). Half of the subjects (52%) were threatened by their mothers during the observations in the first week post-sibling-birth, and 71% were threatened during the first post-birth month. The yearlings who were not threatened in the first month were already spending less time near their mothers (within 60 cm) than those who were threatened during that month (Wilcoxon-Mann-Whitney test, W = 94.5, P = 0.029). Yearlings apparently responded to maternal aggression with significantly higher rates of feargrinning at their mothers (F5,120 = 2.783, P = 0.020; Fig. 2c). TABLE II. Repeated Measures ANOVA F-values and Paired t-tests for Month-to-Month Comparisons Behavioral measures related to Repeated measures ANOVA F-values 2mpPRE vs. 1mPRE 1mPRE vs. 1mPOST 1mPOST vs. 2mPOST 2mPOST vs. 3mPOST –2.05 –4.79*** –5.22*** –2.46 –2.25 –0.91 –1.51 –0.10 –3.46** –4.45*** –3.44*** –3.95*** –0.08 1.15 –0.22 –0.61 0.09 1.94 0.29 0.02 0.64 –2.20 –0.82 2.53 3.15** –2.07 –1.63 –0.40 –2.12 –1.36 –4.32*** –1.76 4.49*** 5.02*** –5.43*** –4.32*** –4.92*** 0.38 3.73*** –0.07 –0.21 1.06 0.37 1.71 0.82 0.30 –1.60 –1.73 –0.21 0.03 –0.93 0.63 –0.50 –0.41 –0.16 –2.48 –0.29 –1.21 1.00 1.15 0.92 –3.81*** –0.75 0.39 –1.00 –0.66 0.76 1.31 –0.34 2.24 3.03** 4.21*** 0.64 –1.75 –1.26 0.34 –0.68 –1.24 0.75 –0.66 –0.59 0.60 –0.42 0.13 –0.79 –0.60 –1.13 –2.41 0.12 0.76 –2.85* –1.23 1.03 –1.05 –0.08 1.24 2.17 Note that repeated measures ANOVAs for behavioral measures related to nipple contact for the third, second, and first pre-sibbirth months and the first post-sib-birth month, df=3,75; for all other repeated measures ANOVAs df=5,120, except for proportion of mother-initiated contacts df=5,90; for paired t-tests, df=30. Critical values for paired t-tests are Bonferroni corrected. *P < 0.05; **P < 0.01; ***P < 0.001. Yearling Rhesus at Sibling Birth / 201 Nipple contact Time on nipple 18.210*** Nipple attempts 21.563*** Nipple contacts 21.333*** Rejections per hour 10.045*** Proximity and contact Time in contact 23.707*** Time < 60cm 2.723* Time > 5m 26.726*** Time out of sight 24.742*** Approach to mother 21.079*** Approach by mother 11.924*** Yearling contacts mother 11.447*** Mother contacts yearling 7.019*** Proportion of mother2.464* initiated contacts Maternal aggression and infant distress Mother threatens 3.415* Mother lunges 2.734* Time passive 4.765** Fear grin 2.783* Gecker 7.820*** Scream 3.467* Grooming Mother grooms yearling 8.647*** Yearling grooms mother 2.569* t-values 3mPRE vs. 2mPRE 202 / DeVinney et al. Fig. 2. Means and standard errors for: (a) the proportion of physical contacts that were initiated by the mother, (b) rate per hour of maternal threats and maternal lunges, (c) the rate per hour of females’ fear grins and males’ fear grins, and (d) the rate per hour of geckering and screaming. The dashed line in the center of each graph indicates the time of the younger siblings’ births. The time blocks represented along the x-axis are in a format such that “3mPRE” stands for the third month pre-sibling-birth, and so forth. Distress and Disturbance Contrary to expectations, overt signs of distress (e.g., screaming and geckering) decreased significantly over the 6 months (screaming: F5,120 = 3.467, P = 0.006; geckering: F5,120 = 7.820, P = 0.000; Fig. 2d). The mean proportion of time yearlings were observed in apparently tense or depressed states was ≤1% and did not change across the study. Two male yearlings were observed in depressed postures at one time-point sample each. For one male, this occurred 2 days after sibling birth, and represented 1.35% of time during the first week after sibling birth, and, for the other male, the depressed posture was observed 4 days after sibling birth and represented 1.25% of the first week’s observation time. What distress was shown apparently was ineffective in eliciting additional care from mothers. In fact, there was a significant negative correlation between rates of yearlings’ geckering and rates of being approached by mother in the first post-birth month (Table III). Similarly, there was a significant negative correlation between geckering and time within 5 m of the mother and a nonsignificant trend toward a negative relationship between geckering and time in contact with the mother. There was no relationship between geckering and these measures of care during the pre-birth period. Time in a passive state, a possible indicator of mild distress, significantly increased across these 6 months (F5,120 = 4.765, P = 0.001), particularly between the first pre-birth month and the first post-birth month. Yearling Rhesus at Sibling Birth / 203 TABLE III. Partial Correlations for Geckering vs. Measures of Mother-Yearling Interaction, Controlling for Age* Geckering and Mother approaches Time < 5m Time in contact 3mPRE 2mPRE 1mPRE 1m POST 2mPOST 3mPOST r= 0.09 (P=0.62) –0.13 (P=0.50) 0.15 (P=0.42) r= 0.03 (P=0.87) 0.01 (P=0.97) 0.17 (P=0.38) r= –0.29 (P=0.12) 0.02 (P=0.90) 0.20 (P=0.16) r= –0.40 (P=0.03) –0.36 (P=0.05) –0.35 (P=0.06) r= –0.08 (P=0.69) –0.20 (P=0.29) 0.36 (P=0.05) r= –0.28 (P=0.22) –0.45 (P=0.01) 0.33 (P=0.08) *P values are given in parentheses. Abruptness of Change The fact that many measures of behavior seemed to change most rapidly between the first pre-birth month and the first post-birth month suggests abrupt change around the time of the sibling’s birth. This notion is supported by the results of the paired t-tests between sequential months (Table II). Fourteen of 21 measures (66%) showed highly significant changes between the first pre-birth and the first post-birth month. No significant changes occurred after that time. In addition to the changes at sibling birth, some measures showed changes earlier. These included: 1) measures related to the dynamics of nipple contacts (nipple attempts and nipple contacts), which significantly decreased between the third and second pre-birth months; 2) time out of sight of the mother, which significantly increased between the second and first pre-birth months; and 3) geckering, which significantly decreased between the third and second pre-birth months. Sex Differences Across the 6 months of analysis, male yearlings spent significantly more time at a distance of over 5 m from mothers than did female yearlings (F1,24 = 7.191, P = 0.013), while female yearlings spent significantly more time within 60 cm of their mothers than did male yearlings (F1,24 = 6.320, P = 0.019). Female yearlings approached their mothers more than males did (F1,24 = 5.034, P = 0.034) and female yearlings groomed their mothers more than males did (F1,24 = 11.912, P = 0.002). Females, but not males, showed a nonsignificant tendency to fear-grin at their mothers more frequently after the births of their siblings (F1,24 = 0.299, NS). Fear-grinning to the mother may have functioned as an appeasement gesture, facilitating the females’ efforts to stay near the mother. There were no significant two-way interactions between the factor sex and the various behavioral measures, suggesting that patterns of change over time did not vary significantly by sex. Rank and Group Differences Yearlings of low-ranking mothers spent more time in contact with their mothers (F1,24 = 5.010, P = 0.035) and initiated physical contact more frequently with their mothers (F1,24 = 5.582, P = 0.027) than did yearlings of high-ranking mothers. Low-ranking mothers themselves made contact with their yearlings more frequently than did high-ranking mothers (F1,24 = 5.256, P = 0.031). Before sibling birth, highranking mothers groomed their yearlings more than did low-ranking mothers, and after sibling birth, low-ranking mothers groomed their yearlings more than did highranking mothers (F1,24 = 6.063, P = 0.021). Across all 6 months, yearlings of low- 204 / DeVinney et al. ranking mothers groomed their mothers more than did yearlings of high-ranking mothers (F1,24 = 13.584, P = 0.001). Group S yearlings made more nipple attempts (F1,24 = 6.893, P = 0.015), made more nipple contacts (F1,24 = 8.585, P = 0.007), and geckered more (F1,24 = 8.390, P = 0.008) than did Group R yearlings. Trapping Effects Contrary to expectations, we found no evidence of increased distress when trapping coincided with sibling birth (i.e., for the “during” monkeys) compared to when sibling birth occurred outside the trapping period (“before/after”). Yearlings Without Younger Siblings Our sample included only three families in which no new sibling was born. As a method to assess whether measures of mother-infant interaction for the no-sib- Fig. 3. Means (squares) and 95% confidence intervals of the mean for the with-sibling group and means (filled circles) and raw data points (open circles) for the three no-sibling yearlings for: (a) proportion of time on the nipple, (b) proportion of time in contact with mother, (c) proportion of time out of sight of the mother, and (d) rate of yearlings’ approaches to mother. Asterisks along the top of each panel indicate those months in which the three data points for the no-sibling group fell outside the 95% confidence interval of the mean for the with-sibling group. Yearling Rhesus at Sibling Birth / 205 ling families differed from those for families with new siblings, we plotted the 95% confidence interval of the mean for measures of proximity and physical and nipple contact for the with-sibling group, and then examined whether the data points for the three no-sibling focals fell outside that range. The mean age at sibling birth (54.4 wk) was used to determine where to divide the data into months for the nosibling yearlings. For the no-sibling focals, 95/117 (81.2%) of the comparisons (3 infants × 3 mo × 13 measures) before sibling birth and 104/117 (88.9%) of the comparisons after sibling birth were outside the confidence interval. The yearlings without siblings generally exhibited higher levels of association with their mothers across these 6 months, with the differences between groups being stronger in the post-sibling-birth period, as illustrated for four behaviors in Fig. 3. DISCUSSION Rhesus mother–yearling relationships changed markedly during the 6-month period surrounding the siblings’ birth. Levels of maternal care decreased across this period in many ways, including the termination of nursing and the reduction of maternal grooming, contact, and proximity (the latter being “care” in that proximity to the mother can provide protection from conspecific aggression [Maestripieri, 1993] and opportunities for observational learning [Pereira & Altmann, 1985]). Since these are developmental data, at the outset one would anticipate some behavioral change over time as the subjects became more independent of their mothers across these 6 months. What these data seem to point to, however, is a discontinuity, or abrupt transition, toward accelerated independence from the mother associated with the birth of a younger sibling. Most measures showed their largest and most consistent changes between the first pre-sibling-birth month and the first post-sibling-birth month (Table II). A comparison between yearlings that did and did not experience sibling birth could provide further suggestive evidence that the changes observed in yearlings with new siblings represented abrupt transitions associated specifically with the sibling’s birth rather than simple age-related shifts toward independence. Unfortunately, in this well-provisioned and relatively unmanipulated colony, our sample yielded only three families without new siblings to compare with the yearlings with siblings, and of course we were unable to control for possible differences in quality [Clutton-Brock, 1991] or responsiveness [Berman et al., 1993] between mothers who did and did not give birth again. Nevertheless, in our sample the yearlings without siblings exhibited higher and more constant levels of association with the mother compared to yearlings with siblings across these 6 months. Attachment Theory From the view of attachment theory, the birth of a younger sibling is a normal yet stressful transition in which older siblings experience a reduction in maternal care to which they often respond with behavioral distress [Teti et al., 1996], particularly if they are insecurely attached. Evidence for increases in yearling distress and/or disturbance was lacking or, at best, equivocal. All measures of overt distress decreased rather than increased with the birth of the sibling. Time spent in a passive state increased in the month after sibling birth, but the extent to which passivity represents mild distress vs. a simple reorganization of yearlings’ time budgets is unclear. Finally, the two males who assumed depressed postures [Kaufman & Rosenblum, 1967] did so only fleet- 206 / DeVinney et al. ingly and only during the first week after the sibling’s birth. Hence, while some yearlings may have been disturbed briefly, in general yearlings showed only mild or no disturbance and distress in response to their siblings’ births. Since we did not attempt to assess the security of the mother–yearling attachment, we cannot evaluate whether the low levels of distress observed at sibling birth are related to yearlings being securely attached to their mothers or, alternatively, whether the theory’s prediction of distress at this “partial removal” of the mother should be considered unsupported. Attachment theory also would predict that the mother will be the agent of separation, meaning that she would be responsible for decreases in proximity and contact, and that yearlings should exhibit behavior related to trying to reestablish contact. Although mothers behaved more aggressively and punitively after the sibling’s birth, they did not take primary responsibility for decreases in proximity and contact. Indeed, mothers actually increased the proportion of contacts they initiated toward yearlings, and increased the rate of contact initiations with daughters. Thus, attachment theory predictions regarding the agent of change in proximity and contact were not supported. Parent–Offspring Conflict Parent–offspring conflict theory would view sibling birth as a reallocation of maternal investment by mothers from the yearling to the new sibling. Even if we assume that the changes in care provided by mothers to year-old offspring before and after sibling birth represent changes in maternal investment that could theoretically trigger genetic conflict between mothers and yearlings [Trivers, 1974], we are unable to provide strong support for this perspective. As discussed above, two key predictions for this perspective were not confirmed by the results of this study: 1) mothers did not take primary responsibility for bringing about decreases in proximity and contact, and 2) yearlings showed little or no increases in behavior related to distress or contact-seeking. What distress was shown apparently did not serve to elicit additional care from the mothers. Indeed, there were negative correlations between both the rates of yearlings’ geckering and of being approached by their mothers in the first postbirth month and time within 5 m of the mother, suggesting that geckering may have been a reaction to decreases in maternal care rather than a mechanism for soliciting it, or that mothers actively resisted attempts to elicit “extra” care. Finally, we saw no signs of “regression” to an earlier state of development. Trivers  suggested that youngsters might mimic or imitate earlier stages of development in order to elicit the higher levels of care typically given to younger individuals. While there is some support for this interpretation of children’s regressive behaviors [Stewart et al., 1987], we found no evidence for this tactic among year-old monkeys. Dynamic Assessment Bateson’s alternative parent–offspring dynamic assessment perspective would predict that transitions in parent–offspring relationships would be determined by mutual honest signaling regarding the mother’s ability to provide care and the offspring’s need for care. Offspring in low-risk environments would be expected to respond cooperatively to mother’s signals and to accelerate rates of development when mothers are unable to continue high levels of care. Although our data do not provide a definitive test of this perspective, we believe our find- Yearling Rhesus at Sibling Birth / 207 ings point to its potential usefulness for guiding future research in this area. Certainly, two key predictions were supported: 1) yearlings played roles equal to or larger than mothers in bringing about decreases in proximity and contact, and 2) yearlings showed little distress or disturbance at the sibling’s birth. A third key prediction was that mothers would provide yearlings with prominent cues indicating that the period of greatest maternal investment was over [Bateson et al., 1990]. Although further research is needed, we would like to raise the hypothesis that the abrupt increases in maternal aggression after the sibling’s birth may have represented such cues, although other changes, such as the appearance of the new sibling itself, or more subtle cues from the mother may also be worthy of consideration. Maternal aggression was almost entirely in the form of mild threats and lunges rather than contact aggression. Hence, it provided conspicuous stimuli to yearlings with little potential for actual injury. The increase in this kind of maternal aggression at the time of the sibling’s birth was abrupt and dramatic; threats in particular increased by a factor of more than 3 and remained elevated during the first 3 post-birth months. Yearlings apparently responded with increased fear-grinning directed towards mothers and with decreased attempts to remain near and make contact with them, but not with increased distress, disturbance, or contact-seeking. In this sense, mothers may have instigated the transition in their relationships with their yearlings, but rather than protesting or attempting to cling, yearlings cooperated with mothers in bringing about changes in patterns of proximity and contact more typical of juvenile rhesus monkeys. This is in contrast to reactions by Cayo Santiago rhesus infants when their mothers resume mating, an event that takes place approximately 6–9 months earlier than sibling birth [Berman et al., 1993, 1994]. At that time, rates of maternal punishment (defined as hits, bites, slaps, and pushes) increase abruptly, and the infants’ time in proximity and contact with the mother decrease abruptly. However, changes in the relationship are due primarily to decreases in mothers’ propensities to seek contact, and the infants respond with dramatic increases in distress and contact-seeking behavior. Thus, compared with youngsters approximately 6–9 months younger, we suggest that the yearlings’ behavior represented an acquiescence to their mothers’ increased threats and an abrupt increase in behavioral maturity rather than an attempt to restore contact or elicit more care. One potential concern with this interpretation is that Holman and Goy’s  study of captive rhesus reported no increase in maternal aggression with the birth of a sibling. Further research is needed to determine whether this discrepancy in findings indicates that 1) maternal aggression is not the salient cue, 2) a dynamic assessment approach is not useful in interpreting the sibling transition, or 3) different cues or different theoretical perspectives may apply in different environments. Barrett and Henzi’s  analysis of weaning tantrums in infants of nonseasonally-reproducing wild baboons also provides preliminary support for a dynamic assessment model over parent–offspring conflict theory, in that tantrums appeared to be honest signals of need to which mothers responded with increased nursing. However, Barrett and Henzi describe a primarily passive role for mothers in which signaling takes place primarily from infant to mother. Infants are hypothesized to take the initiative to reduce levels of care before mothers need to reallocate care to younger siblings. We suggest that this scenario is not likely to occur in food-rich, seasonally-reproducing species because mothers operate within a more constrained reproductive schedule. Rich food resources open the possibility of yearly reproduction provided mothers are able to conceive during the mat- 208 / DeVinney et al. ing season following their current infant’s birth. However, this often requires initiating reductions in levels of care both when they resume mating [Gomendio, 1989; Berman et al., 1993] and when the new sibling is born. Failure to do so is likely to be costly in that females will have to delay reproduction for an additional year. Hence, one would expect to find signaling by both infants and mothers in these circumstances. In younger infants, active care reduction also appears to be necessary. Sex Differences The sex differences in behavior we observed were consistent across the 6 months of observation. Males spent more time at a distance greater than 5 m from their mothers than did females, and females spent more time within 60 cm of their mothers and approached their mothers more than did males. Holman and Goy  found sex differences in rhesus mother-yearling proximity, but only after sibling birth; thus they suggested that the males’ “disengagement” from their mothers was triggered by sibling birth. Berman et al.  found that when Cayo Santiago rhesus infants are approximately 6 months old and their mothers return to estrus, infant sons suddenly spend more time at a distance (>5 m) from their mothers; however, it is unclear whether this pattern persisted from the time of first estrus until the 3 months preceding sibling birth. The sex differences in behavior observed in these yearlings foreshadow the pattern of male peripheralization typical of rhesus social organization [Colvin, 1983]; however, it is unknown why this pattern would begin earlier in free-ranging groups than in captive groups. In summary, after sibling birth yearling rhesus monkeys experienced marked reductions in maternal care and increases in maternal aggression, yet exhibited little distress. We suggest that maternal aggression served as a cue regarding decreased levels of care, and that yearlings acquiesced to the reduction in care, exhibiting increased independence. These findings are in concert with the predictions of dynamic assessment approaches to the sibling birth transition, more so than with the predictions of attachment theory or parent–offspring conflict theory. Dynamic assessment approaches to parent–offspring relationships have only recently been applied to studies of primates; thus, future work should address whether this approach has value in addressing the sibling birth transition in other environments and species, and address the relevance of maternal aggression as a cue regarding reductions in maternal care among nonhuman primates. ACKNOWLEDGMENTS We thank John Berard and Matt Kessler for permission to work on Cayo Santiago. We thank the Wenner-Gren Foundation for Anthropological Research (grant to B.J.D.) and the National Institute of Child Health and Human Development through the Laboratory of Comparative Ethology, Division of Intramural Research. This investigation was supported in part by an Animal Resources Branch Program Award, RR-03460, from the National Center for Research Resources, National Institutes of Health and the University of Puerto Rico, Medical Sciences Campus. We are grateful to Kristin Abbott, Jennifer Rodger, and Steve Ross for assistance with the behavioral observations. Many thanks to those who wrote the computer programs used in this research: Helen Ball, Michael Hulett, and Ken Whang. We thank Stephen Suomi and three anonymous reviewers for their helpful suggestions regarding this manuscript. Yearling Rhesus at Sibling Birth / 209 REFERENCES Achenbach GG, Snowdon CT. 1998. Response to sibling birth in juvenile cotton-top tamarins (Saguinus oedipus). Behaviour 135: 845–862. Ainsworth MDS. 1979. Attachment as related to mother-infant interaction. Adv Study Behav 9:2–52. Altmann J. 1974. Observational study of behavior: sampling methods. Behavior 49: 227–267. Altmann J. 1980. Baboon mothers and infants. Cambridge, MA: Harvard University Press. 242 p. Barrett L, Henzi SP. 2000. Are baboon infants Sir Phillip Sydney’s offspring? Ethology 106:645–658. Bateson P, Mendl M, Feaver J. 1990. Play in the domestic cat is enhanced by rationing of the mother during lactation. Anim Behav 40:514–525. Bateson P. 1994. The dynamics of parent-offspring relationships in mammals. Trends Ecol Evol 9:399–402. Berman CM, Rasmussen KLR, Suomi SJ. 1993. Reproductive consequences of maternal care patterns during estrus among free-ranging rhesus monkeys. Behav Ecol Sociobiol 32:391–399. Berman CM, Rasmussen KLR, Suomi SJ. 1994. Responses of free-ranging rhesus monkeys to a natural form of social separation. I. Parallels with mother-infant separation in captivity. Child Dev 65:1028– 1041. Bolwig N. 1980. Early social development and emancipation of Macaca nemestrina and species of Papio. Primates 21:357–375. Bowlby J. 1969. Attachment and loss. Vol. I. Attachment. London: Hogarth Press. 428 p. Clutton-Brock TH. 1991. The evolution of parental care. Princeton, NJ: Princeton University Press. 325 p. Cohen J. 1960. A coefficient of agreement for nominal scales. Educ Psychol Measure 20:37–46. Colvin JD. 1983. Influences of the social situation on male emigration. In: Hinde RA, editor. Primate social relationships: an integrated approach. Oxford: Blackwell Scientific Publishing. p 160–171. DiGregorio G, Suomi SJ, Eisele CE, Chapman SA. 1987. Reactions of nuclear-family-reared rhesus macaques to the births of younger siblings. Am J Primatol 13:231–253. Dunn J, Kendrick C. 1980. The arrival of a sibling: changes in patterns of interaction between mother and first-born child. J Child Psychol Psych 21:119–132. Dunn J, Kendrick C, MacNamee R. 1981. The reaction of first-born children to the birth of a sibling: mothers’ reports. J Child Psychol Psych 22:1–18. Dunn J, Kendrick C. 1982. Siblings: love, envy, and understanding. Cambridge, MA: Harvard University Press. 289 p. Godfray HCJ. 1991. Signalling of need by offspring to their parents. Nature 352: 328–330. Gomendio M. 1989. Suckling behavior and fertility in rhesus macaques (Macaca mulatta). J Zool Lond 217:449–467. Grafen A. 1990. Biological signals as handicaps. J Theor Biol 144:517–546. Greenhouse SW, Geisser S. 1959. On methods in the analysis of profile data. Psychometrika 24:95–112. Holman SD, Goy RW. 1988. Sexually dimorphic transitions revealed in the relationships of yearling rhesus monkeys following the birth of siblings. Int J Primatol 9:113–133. Kaufman IC, Rosenblum LA. 1967. The reaction to separation in infant monkeys: anaclitic depression and conservation-withdrawal. Psychosom Med 29:648–675. Lee PC. 1983a. Effects of parturition on the mother’s relationship with older offspring. In: Hinde RA, editor. Primate social relationships: an integrated approach. Oxford: Blackwell Scientific Publishing. p 134–139. Lee PC. 1983b. Species, study sites and methods. In: Hinde RA, editor. Primate social relationships: an integrated approach. Oxford: Blackwell Scientific Publishing. p 8–16. Maestripieri D. 1993. Maternal anxiety in rhesus macaques (Macaca mulatta). I. Measurement of anxiety and identification of anxiety-eliciting situations. Ethology 95:19–31. Mineka S, Suomi SJ. 1978. Social separation in monkeys. Psychol Bull 85:1376–1400. Nadelman L, Begun A. 1982. The effect of the newborn on the older sibling: mothers’ questionnaires. In: Lamb ME, SuttonSmith B, editors. Sibling relationships: their nature and significance across the lifespan. Hillsdale, NJ: Lawrence Erlbaum Assoc. p 13–37. Pereira ME, Altmann J. 1985. Development of social behavior in free-living nonhuman primates. In: Watts ES, editor. Non-human primate models for human growth and development. New York: Alan R. Liss. p 217–309. Rawlins RG, Kessler MJ. 1986. The Cayo Santiago macaques: history, behavior, and biology. Albany, NY: State University of New York Press. 306 p. Sade DS, Chepko-Sade BD, Schneider JM, Roberts SS, Richtsmeier JT. 1985. Basic demographic observations on free-ranging rhesus monkeys. New Haven, CT: Human Relations Area Files. 98 p. 210 / DeVinney et al. Smith EF. 1991. The influence of nutrition and postpartum mating on weaning and subsequent play behavior of hooded rats. Anim Behav 41:513–524. SPSS. 1999. SPSS Advanced Models, version 10.0. Chicago: SPSS Inc. Stewart RB, Mobley LA, Van Tuyl SS, Salvador MA. 1987. The firstborn’s adjustment to the birth of a sibling: a longitudinal assessment. Child Dev 58:341–355. Tanaka I. 1992. Three phases of lactation in free-ranging Japanese macaques. Anim Behav 44:129–139. Taylor MK, Kogan KL. 1973. Effects of birth of sibling on mother-child interactions. Child Psych Hum Dev 4:53–58. Teti DM, Sakin JW, Kucera E, Corns KM. 1996. And baby makes four: predictors of attachment security among preschool-age firstborns during the transition to siblinghood. Child Dev 67:579–596. Trause MA, Voos D, Rudd C, Klaus M, Kennell J, Boslett M. 1981. Separation for childbirth: the effect on the sibling. Child Psych Hum Dev 12:32–39. Trivers RL. 1974. Parent-offspring conflict. Am Zool 14:249–264. Walters J. 1987. Transition to adulthood. In: Smuts BB, Cheney DL, Seyfarth RM, Wrangham RW, Struhsaker TT, editors. Primate societies. Chicago: University of Chicago Press. p 358–369.