Analysis of rhesus monkey vocalizations Maturation-related sex differences in clear-call frequency.код для вставкиСкачать
Analysis of Rhesus Monkey Vocalizations : Maturation-related Sex Differences in Clear-call Frequency ’ J. ERWIN2 AND G . MITCHELL Department o f Psychology, Department o f Behavioral Biology, and California Primate Research Center, Davis, California 9561 6 ABSTRACT In the first of two studies six like-sexed pairs of rhesus macaques early in their third year of life were observed for two days, then separated for two days, and finally reunited for two days. In the second, about nine months later, the same animals were paired across sex and were separated in a similar fashion to that in the previous study. The number of distress-related clear calls uttered by females remained stable through both studies, but the males showed a decrease with age in this behavior. The results of these two studies are compared with other research in which maturation-related sex differences have become apparent. These comparisons suggested that decreases in the production of clear calls accompany the onset of puberty. Since the females which were involved in our research reached puberty shortly before the time of our first study and the males did so shortly before the second, the decreases in male clear-call rate which we observed across these two studies appears to have been linked with peri-pubertal influences. The dynamics of mother-infant attachment formation and disruption has been a popular research topic for some time (cf. Bowlby, ’58; Elliot and Scott, ’61), and some of the most significant research in this area has involved the use of nonhuman primates as subjects (DeVore, ’63; Hinde, Spencer-Booth, and Bruce, ’66; Jay, ’63; Seay, Hansen and Harlow, ’62; Simons, Bobbitt and Jensen, ’67). Macaques, especially rhesus macaques, have received the most attention. A common research strategy which has been used in the testing of the strength of attachment of mother-infant dyads has been observation of the responses to separation of sociallybonded animals. The most reliable correlate of mother-infant separation has been a distinctive vocal sound emitted by both (Scollay, ’ 7 0 ) , which in its simplest form is tonal and usually of rather long duration relative to most other macaque vocalizations. These sounds vary in form but generally fit into a class of calls designated by Rowel1 and Hinde (’62) as “clear calls” because of their tonal structure. They have also been referred to in the literature as AM. J. PHYS.ANTHROP., 38: 4 6 3 4 6 8 . “coos” (Seay and Harlow, ’65) and “whoo calling” ( Spencer-Booth and Hinde, ’ 6 6 ) , among other names. While clear calls occur in many contexts other than separation, few situations seem to be as closely correlated with these sounds as are those involving separation. Consequently, clear calls have come to be used by many laboratory investigators as indicators of separation distress and therefore of the strength of social-emotional bonds. This measure has been included as part of our continuing examination of social interactions in rhesus macaque dyads of various ages and both sexes in a series of studies in which the incidence of various behavioral elements has been monitored. Some interesting age and sex differences have emerged from this research. 1 The research reported here was supported in part by the following U. S. Public Health Service/NIH grants: MH19760 and MH17425 to G . Mitchell; HD04335 to the Department of Behavioral Biology, School of Medicine; RR00169 to California Primate Research Center (previously The National Center for Primate Biology); and HD06367 to William Mason. ZRequests for reprints of this paper should be addressed to J. Erwin, Department of Psychology, University of California, Davis, California 95616. 463 464 J. ERWIN AND G. MITCHELL In this paper, only the results concerning clear calling are reported from two recent separation studies and are discussed in the light of other investigations conducted in our laboratory and elsewhere. The complete results of these two studies have been reported elsewhere (Erwin et al., ’71; Erwin et al., ’72). Twelve rhesus macaques, six males and six females, were the subjects for our studies; all were laboratory-born. Each subject had experienced two brief experimental separations from its mother in infancy (Scollay, ’70) and was weaned between seven and ten months of age. In addition, each animal had been paired briefly, shortly after weaning, with strangers of the same age and of both like and opposite sex (Stevens and Mitchell, ’72), none of which were also subjects of our study. STUDY ONE - METHODS Subjects When our 12 subjects were 9 to 12 months old, they were matched within the group for age and sex and paired; age differences within pairs ranged from one to ten days. Each pair was housed individually in a l m X l m X l m cage and had no physical contact with other animals during a period of about 17 months before the initiation of the separation study described. The six like-sex ( 3 male, 3 female) pairs of subjects ranged in age from 26 to 29 months when they were separated. Apparatus All observations were made in a l m X Im x 2m test cage with three walls of wire mesh and the front one of clear Plexiglas. A sliding translucent Plexiglas partition made it possible to divide the cage into two equal segments the size of the home cage. Observations were recorded by using two clock-counter recording devices described in detail elsewhere (Odom, Mitchell and Lindburg, ’70). Procedure Before data recording began, each pair of animals was placed in the test cage for a one-day adaptation period. A six-day study of each pair following the adaptation period consisted of two days of presepara- tion, two days of separation and two days of reunion. Two observers using clockcounter devices recorded the frequencies and durations of 20 behaviors; each observer scored the behavior of one member of each pair. The behaviors recorded were grouped into classes of five behaviors each; and two five-minute samples of each behavioral class were recorded on each of the six data collection days, which were supplemented by the taking of extensive films and notes. A base rate for comparative purposes was established during the twoday preseparation phase of the study. On the third day the partition was slid into place, dividing the cage and separating the animals, and was left in place until the fifth day of the study when it was removed to allow reunion for two days. After the two-day reunion period, the subjects were removed from the test cage and placed in separate home cages. STUDY ONE - RESULTS The frequency of clear calls increased dramatically at separation, as expected, and decreased at reunion to a level below that of the preseparation phase. During preseparation, separation and reunion phases, respectively, the mean number of “coo”-vocalizations per animal was 11.75, 59.00, and 5.25 ( p < 0.05, ANOVAR). Most of the animals called much more frequently on the first day of separation than on the second day. The intensity of the “coo”-vocalization seemed to be much greater immediately following separation than at any other time. It was common during the period shortly after separation for calls to begin as smooth tonal sounds and to increase in pitch and intensity until they became loud, harsh screeches. There were no apparent sex differences in rates of clear calling in this study. STUDY TWO - METHODS Subjects The same 12 rhesus macaques were used at ages 34 to 37 months as subjects for this later study which was conducted about nine months after the first one. During part of the intervening period, eight of the animals had been involved in another study in which they were paired for two months with infants (Brandt and Mit- 465 RHESUS CALLS: AGE AND SEX DIFFERENCES chell, ’73) ; however, all subjects had been again housed alone for a t least three months before the beginning of this study. Apparatus The same test cages and data acquisition devices were used in study two as in study one. Procedure In this study males and females were paired in a pre-test phase for one, two, or three weeks before separation. Observers recorded data in a manner similar to that described for the first study, except that 50 behaviors grouped in five classes were recorded in one five-minute sampling period per class per day. Again a one-day test-cage adaptation period was followed in order by two-day phases of preseparation, separation, and reunion. Data were also collected in the home cages of the subject pairs for two days during each week of the pre-test-cage phase of the study and for two days per week for two weeks following the test-cage phase. Films and notes supplemented clock-counter data. STUDY TWO - RESULTS Even with the short periods of time allowed in this study for attachment formation, the rate of “coo”-vocalizations increased significantly during the separation phase for all except three animals. The mean number of clear calls per animal during preseparation, separation, and reunion phases was 4.50, 15.58, and 11.67, respectively ( p < 0.05 ANOVAR). It should be noted that two ( a male and a female) of the three animals showing no increase in vocalization at separation were both members of dyads paired for only one week; the female, however, vocalized frequently during the reunion phase. The high reunion mean is therefore a reflection of the behavior of one animal that accounted for more than 80% of the calls occurring during this phase. During this study males vocalized much less often than did females. The mean number of clear calls during the test-cage period of the study was 5.42 for males and 17.50 for females ( p < 0.05 Mann-Whitney U ) . Few clear calls occurred during the other parts of the study. A comparison of the overall rates of clear calling in study one, when the animals were a little more than two years old, with the rates i n study two, when the subjects were about three years old, presents a n interesting contrast. Since a n accurate comparison of the results of our two studies required a n adjustment for the different time-periods monitored, the clear-call rates are expressed in figures 1 and 2 as the mean number of vocalizations per animal per minute. The adjusted mean values for twoand three-year-old males were 1.20 and 0.36, respectively, and the respective means for two- and three-year-old females were 1.33 and 1.17 (see fig. 1 ) . STUDIES ONE AND TWO - DISCUSSION The relative rarity of clear calling by adult male rhesus macaques has been rather well established (Rowel1 and Hinde, ’62) although clear calling is common among juvenile males. The data support our guess that there is a decrease in clear call rate among males around the time of the onset of puberty. It is interesting that the two males that vocalized least at three years were also the only ones that succeeded in impregnating their female partners. Although adult females emit clear calls more,often than do adult males, the fullyadult female clear-call rate is much less than that of juveniles. This statement is W t- a W a b 1.0 LT W m 5z 0.5 g =L 0.0 27 36 MEAN AGE I N MONTHS Fig. 1 Clear-call rates of male a n d female rhesus macaques during the third year of life. 466 J. ERWIN AND G. MITCHELL borne out by a recent study of attachment 1 in rhesus consort pairs (Maple, Risse and Mitchell, unpublished) and by a later experiment using ten of our 12 subjects at ages between 42 and 48 months (Erwin, unpublished data). A decrease in male rate of clear-calling conforming to their average age at onset of puberty ( 3 years) was demonstrated in the previously.mentioned experiment which intervened between our two studies (Brandt and Mitchell, in press) involving the pairing with infants of eight of our subjects at ages of approximately 34 to 37 months. The decreases observed in female clear-calling rates were less Fig. 2 Clear-call rates of male and female closely related to pubertal changes, which rhesus macaques during the first three years of occur in females a year earlier than in life. males. The fact that the females were just entering puberty at the time of our first the change of rate of clear-calling with age, study, however, may explain the absence as shown in figures 1 and 2, were approxiin our data of significant sex differences, mately parallel except during most of the since previous studies that included the third year of life, the level of vocalizations same subjects showed definite sex differ- in the study two two-year-old post-pubertal ences in clear-calling rates even at the very females must have already been reduced earliest ages tested ( 2 , 3. 5, or 5 months; to approximately the same level as that of Scollay, ’ 7 0 ) , as well as a n increase with the still pre-pubertal males.4 When correage for both sexes (Stevens and Mitchell, lated with maturation stage, therefore, the ’72). The Stevens and Mitchell study (’72) changes of rate would be consistently pardescribed previously, where the subjects allel for the two sexes. The females, howwere still less than a year old, showed a n ever, seemed to undergo further decrease even stronger sex difference than for Scol- during the fourth year of life, as indicated lay’s young infants, as well as a rate of by their responses during a second separaclear-calling for both sexes higher than at tion from like-aged males. Other evidence supports the point of view that females any other age studied. This result is not surprising since juve- may continue gradually to decrease their nile macaques, both in the field (cf. Lind- emission of calls during early adulthood. burg, ’71) and in the laboratory (Grimm, For example, it has been repeatedly dern’67), have also been shown to clear-call onstrated that primiparous mothers “coo”more frequently than either young infants vocalize more than do multiparous mothers or adults.? The early-age overall increase (Seay, ’66; Mitchell and Brandt, ’70; Scolwith age in infants’ “coo”-calling rates lay, ’70). The age of the mothers was not found by Scollay (’70) corroborated the re- controlled, but primiparous mothers are sults of another study (Mitchell, ’68) in usually younger than their multiparous which the behavior of 32 mother-infant counterparts. This difference may be due dyads during the infants’ first six months primarily to emotional factors directly rewas monitored. No significant sex differ- lated to the amount of maternal experience ences in the infants’ coos were observed in of primiparous mothers relative to multithe latter, however, perhaps because the parous mothers. Clearly there are many other factors pairs were not separated as were Scollay’s. In general, these comparisons show other than sex and chronological age/ sharp increases in clear-call frequency dur- maturation stage associated with clear-call ing the first year of life and sharp de3 Nishimura (’72) reported similar age-related difin rates of clear calling among Japanese macreases during the second year for both ferences caques (Macaca fuscata). sexes. with the decrease for males continu4 Nishimura (personal communication) also found rate. of clear calling among female Japanese maing during their third year (fig. 2). Since that caques decreased at an earlier age than among males. RHESUS CALLS: AGE AND SEX DIFFERENCES production in rhesus macaques and related species. However, it seems obvious that some of the most influential factors are those that are operative in many situations. Sex and stage of maturation seem to be two of these factors influencing even such capricious behaviors as rhesus macaque clear calls. Through our studies and comparisons with similar research, we have demonstrated that, at least under conditions of separation, females consistently clear call more frequently than do males, except during the time-period when females have already reached puberty and males have not. LITERATURE CITED Bowlby, J. 1958 The nature of the child's tie to his mother. Int. J. Psycho-anal., 39: 350-373. Brandt, E., and G. Mitchell 1973 Pairing preadolescents with infants (Macaca mulatta). Developmental Psychol., 8: 14-19. DeVore, I. 1963 Mother-infant relations in free-ranging baboons. In: Maternal behavior in mammals. H. L. Rheingold, ed. Wiley, New York, Chap. X, pp. 305-335. Elliot, O., and J. P. Scott 1961 The development of emotional distress reactions to separation, in puppies. J. of Genet. Psychol., 99: 3-22. Erwin, J., E. Brandt and G. Mitchell The development of attachments in three-year-old heterosexually naive rhesus monkeys. Paper presented at the Annual Meeting of the Animal Behavior Society, Reno, Nevada, 1972. Erwin, J., J. Mobaldi and G. Mitchell 1971 Separation of juvenile rhesus monkeys of the same sex. J. Abnorm. Psychol., 78: 134-139. Grimm, R. J. 1967 Catalogue of sounds of the pigtailed macaque (Macacn nemestrina). J. Zool. Lond., 152: 361-373. Hinde, R. A., Y. Spencer-Booth and M. Bruce 1966 Effects of 6-day maternal deprivation on rhesus monkey infants. Nature, 210: 10211023. Jay, P. 1963 Mother-infant relations in langurs. In: Maternal behavior i n mammals. H . L. Rheingold, ed. Wiley, New York, Chap. IX, pp. 287-304. 467 Lindburg, D. G. 1971 The rhesus monkey in North India: A n ecological and behavioral study. In: Primate behavior: Developments in Field and Laboratory research. L. A. Rosenblum, ed. Academic Press, New York, Vol. 11, Chap. I, pp. 1-106. Maple, T., G. Risse and G. Mitchell 1972 Separation of adult male from adult female rhesus monkeys (Macaca mulatta). Unpublished manuscript. Mitchell, G. D. 1968 Attachment differences in male and female infant monkeys. Child Development, 39: 611-620. Mitchell, G., and E. Brandt 1970 Behavioral differences related to experience of mother and sex of infant in the rhesus monkey. Developmental Psychol., 3: 149. Nishimura, A. 1972 Age changes in the behavior of Japanese monkeys with special reference to vocalization. In: Sym. Proc. 4th Int. Cong. Primatol., Portland, Ore., 15-18 Aug. 1972. E. W. Menzel, ed. S. Karger, Basel/New York, Vol. I, in press. Odom, B., G. Mitchell and D. Lindburg 1970 A device to record primate social behavior. Primates, 11: 93-96. Rowell, T. E., and R. A . Hinde 1962 Vocal communication by the rhesus monkey (Macaca mulatta). Proc. Zool. SOC.Lond., 138: 279-294. Scollay, P. 1970 Mother-infant separation in rhesus monkeys (Macaca m u l a t t a ) . Unpublished Ph.D. thesis, Department of Anthropology, University of California, Davis. Seay, B. 1966 Maternal behavior in primiparous and multiparous rhesus monkeys. Folia Primat., 4: 146-168. Seay, B., E. Hansen and H. F. Harlow 1962 Mother-infant separation in monkeys. J. Child Psychol. Psychiat., 3: 123-132. Seay, B., and H. F. Harlow 1965 Maternal separation in the rhesus monkey. J . Nerv. and Ment. Dis., 140: 434-441. Simons, R. C., R. A. Bobbitt and G. D. Jensen 1968 Mother monkeys' (Macaca nemestrina) responses to infant vocalizations. Percept. Mot. Skills, 27: 3-10. Spencer-Booth, Y., and R. A. Hinde 1966 The effects of separating rhesus monkey infants from their mothers for six days. J. Child. Psychol. Psychiat., 7: 179-198. Stevens, C., and G. Mitchell 1972 Birth order effects, sex differences, and sex preferences in the peer-directed behaviors of rhesus infants. Int. J . Psychobiol., 2: 117-128.