Maternal Anxiety in Rhesus Macaques (Macaca mulatta)

Ethology 95, 3 2 - 4 2 (1993) 0 1993 Paul Parey Scientific Publishers, Berlin and Hamburg ISSN 0179-1613

Sub-department of Animal Behaviour, Cambridge

Maternal Anxiety in Rhesus Macaques (Macaca mulatta) 11. Emotional Bases of Individual Differences in Mothering Style

DARIO MAESTRIPIERI MAESTRIPIERI, D. 1993: Maternal anxiety in rhesus macaques (Mucucu rnulatta). 11. Emotional bases of individual differences in mothering style. Ethology 95, 3 2 4 2 .

Abstract This study assessed the extent to which emotional reactivity to the social environment influences maternal behaviour in macaques. Visual monitoring and scratching were used as behavioral indicators of maternal and social anxiety in small captive groups of rhesus macaques. Maternal visual monitoring of the infant and of other individuals proved better predictors of individual differences in maternal protectiveness than did characteristics of the mother-infant dyad such as maternal age, experience and dominance rank, number of immature offspring present in the group, o r sex of the infant. Unlike visual monitoring, maternal scratching was not linearly related to protectiveness. Mothers displaying low o r high rates of scratching ranked low on protectiveness, and the most protective mothers were those who displayed intermediate levels of scratching. Although individual differences in maternal and social anxiety seem to be mediated by maternal age and experience, they might also be influenced by genetics and early experience.

D. MAESTRIPIERI, Yerkes Regional Primate Research Center, Emory University, 2409 Taylor Lane, Lawrenceville, G A 30243, USA.

Introduction Although the pattern of early mother-infant interaction and infant development is generally uniform in Old World monkeys, there are striking quantitative differences among and within species. In a number of species, the primary responsibility for differences in interaction can be traced to differences between mothers rather than infants (BERMAN 1990a). In fact, both in captivity and in natural habitats, primate mothers may differ considerably in the extent to which they nurse and groom their infants, monitor and limit their movements and interactions with other individuals, or actively promote their independence (e.g., HINDE & SPENCER-BOOTH 1971; ALTMANN 1980). Individual differences in maternal behavior may show a significant degree of consistency over successive infant U S Cop!right Clearance Center Code Statement:

0179-1613/93/9501-0032$02.50/0

33

Anxiety and Mothering Style in Macaques

age periods (e.g., in rhesus macaques, they are stable over the first 2.5 years; HINDE& SPENCER-BOOTH1967a) and from one infant to another. The term mothering style has often been used to describe these consistencies (ALTMANN 1980; BERMAN1990a, b). A number of attributes of the mother-infant pair (e.g., maternal age and dominance rank, sex of the infant) as well as of the surrounding environment (e.g., presence of other relatives and risk of predation) have been shown to affect differences in mothering styles (see NASH& WHEELER 1982, and NICOLSON 1987, for reviews). The picture of the influences on mothering style, however, is far from clear and for some variables contradictory effects have been reported even in the same species (NASH& WHEELER 1982; NICOLSON 1987). Moreover, the effects of such variables as the mother’s age and dominance rank, parity, experience, and the size of the family, all tend to be interrelated and it may be difficult to explain or predict differences in maternal behavior by examining any of these variables separately (e.g., WHITE& HINDE 1975). If a generalization can be made, differences in mothering styles seem to be more apparent in individuals and situations where protection for the infants is especially needed. However, since much is likely to depend on the danger as perceived by the mother, individual differences in expectations of and emotional reactions to the external environment may lead to the adoption of different mothering styles even in the same environmental conditions (DUNBAR 1988). Maternal emotionality, in fact, may represent a common pathway through which some of the above mentioned variables exert their effects on maternal behavior. For example, if maternal protective behavior is mainly motivated by a perceived danger for the infant, then young, primiparous, low-ranking mothers can be thought of as being particularly apprehensive and possessive mothers. Maternal emotionality or anxiety, in fact, has often been invoked to explain the effects of maternal age/experience on maternal behavior (e.g. SEAY1966; MITCHELL & STEVENS 1969; ALTMAKN 1980; HOOLEY & SIMPSON1981) or cross-generational continuity in mothering style (FAIRBANKS 1989). Even though maternal characteristics such as dominance rank or age/experience can influence maternal anxiety and, in turn, mothering style, part of the interindividual variability in temperament of the mothers reflected in mothering style may have a genetic basis or be determined in early infancy and only in part be related to the attributes of the mother-infant pair or to social and environmental conditions. For this reason, one may expect that a measure of anxiety will provide a better predictor of mothering style than the variables that have until now been examined. The only two studies in which behavioral indicators of anxiety were related to mothering style, suggested, in fact, that this may be the case. In baboon mothers, visual monitoring (repeated glancing at other individuals) was a better predictor of mothering style than was dominance rank (ALTMANN 1980) and, in Japanese macaques, the rate of maternal scratching (a common primate displacement activity whose frequency increases in anxiety-eliciting situations, see MAESTRIPIERI et al. 1992) was correlated with a possessive mothering style (TROISIet al. 1991). These studies, however, were preliminary as they used small numbers of subjects (in both cases, seven mother-infant dyads) or reported only qualitative observations (ALTMANN 1980). Ethology. Vd. 95 ( I )

3

34

DARIOMAESTRIPIERI

The aim of the present study was to evaluate the effectiveness of a behavioral measure of anxiety in predicting mothering style as opposed to some of the most commonly examined characteristics of the mother-infant dyad. Visual monitoring and scratching were used as behavioral indicators of anxiety but analyzed separately because they may represent different components of anxiety and may be related to mothering style in different ways (MAESTRIPIERI 1993). If visual monitoring primarily reflects the anticipation of danger, this behavior should be linearly related to an anxiety-related mothering style. O n the other hand, if scratching primarily reflects a component of maternal anxiety which is associated with a motivational conflict (e.g., over whether or not to retrieve an infant from a situation of potential danger), the relation between scratching and behavior reflective of maternal anxiety might be non-linear. TROISIet al. (1991) in fact suggested that because calm and extremely anxious mothers are unlikely to experience conflict, though for different reasons (the former having a strong tendency not to intervene, the latter toward intervention), the highest levels of scratching should be associated with moderately protective mothers (TROISIet al. 1991). In previous studies on primate maternal behavior (e.g., MITCHELL & STEVENS 1969; ALTMANN 1980), it has been argued that anxiety exhibits itself as much in the way the mother responds to her social environment as in the manner in which she treats her infant. Although the social environment may pose a risk to both members of the mother-infant dyad (MAESTRIPIERI 1993), a mother's emotional concern over a potential threat to her own well-being (i.e., social anxiety) must be distinguished from that relative to her infant (i.e., maternal anxiety; see TROISI et al. 1991). Therefore, a further aim of the present study was to assess the relative importance of social and maternal anxiety in affecting macaque maternal behavior. Methods Details of the study animals, housing conditions and general procedures are given in MAES(1993).

TRIPIERI

Characteristics of the Mother-Infant Dyads

(x

Maternal age. Maternal age at parturition ranged from 4 to 16 yr f SD = 7.41 f 3.53 yr). Maternal dominance rank. Dominance relationships among adult individuals were defined on the basis of the directions of bared-teeth displays (DE WAAL& LUITRELL1985). Rhesus females can

generally be ranked in a linear dominance hierarchy with daughters ranking below their mothers. To make comparisons across the five groups, the mothers' rank was divided into high, middle, or low according to whether they fell into the upper, medium or lower third of the social hierarchy in their own groups. Four mothers were classified high ranking, 7 were middle ranking, 6 were low ranking. Dominance rank remained stable through the period of observation. Maternal experience. Maternal experience in raising infants was estimated from the "Umber of previous offspring successfully raised to the age of 1 yr. This number ranged from 0 to 7 (X f SD = 1.94 f 2.25). Five females were primiparous mothers, 12 were multiparous. Number of immature offspring present in the group. The_number of offspring between 1 and 3 yr of age that mothers had in their group ranged from 0 to 3 (X f SD = 0.76 f 0.90). Three-yearold daughters were included among immature offspring only if they did not give birth in that year. 7 mothers had no other immature offspring, 7 had 1 offspring, 2 had 2 offspring, 1 had 3 offspring.

Anxiety and Mothering Style in Macaques

35

Since some mothers had adult offspring in their own group and all male offspring were removed when they reached 3 4 yr of age, the number of immature offspring present in the group did not necessarily overlap with maternal experience. Sex of the infant. 12 of the infants were male, 5 were female.

Behavioral Observations Measures of mothering style. Monkey mothering style tends to fall along two dimensions that 1967b; can be referred to as protectiveness o r warmth and rejection (HINDE& SPENCER-BOOTH HINDE& SIMPSON1975; ALTMANN1980; SIMPSON& HOWE1980; FAIRBANKS & MCGUIRE1987; TANAKA 1989). O n the basis of the information available in the literature (see references above and & HERRMANN 1977), the following behavioral measures of maternal protectiveness were also HINDE selected: 1) Mother makes contact: percentage of contacts made by mother (contacts made by mother/ contacts made by mother + contacts made by infant x 100). 2) Mother approaches: the same as Mother makes contact but approaches were used instead of contacts. Approaching was defined with reference to a distance of 60 cm between mother and infant. 3) Mother grooms: maternal grooming time for every 1,000 s the pair spent in contact. Grooming was defined as picking through and/or slowly brushing aside the fur of the infant with one o r both hands. A pause of 10 s in grooming was used as the criterion for the end of a bout. Since grooming only occurs when the pair is in contact, the ratio measure insures that this measure is independent of the time mother and infant spend in contact. 4) Mother restrains: the percentage of all attempts by the infant to break contact that were prevented by the mother. Mothers restrained their infants by holding them by the tail o r leg while infants were actively trying to get away. Maternal rejection was assessed through the measure of mother rejects: the percentage of all contacts attempted by the infant that were rejected by the mother. Rejections involved the mother passively preventing the infant’s access to the nipple o r ventrum (e.g. by blocking the chest with an arm or twisting the torso away), holding the infant at a distance with an arm, o r moving away from the infant’s attempts to gain access to her nipple o r ventrum. Measures of anxiety. Maternal visual monitoring and scratching were used as measures of anxiety (MAESTRIPIERI 1993) and recorded with the all-occurrences sampling technique (MARTIN& BATESON1986). Visual monitoring (defined as a quick movement of the mother’s eyes, often but not necessarily accompanied by a movement of the head, directed at another individual) was recorded in terms of rate of glancing at the infant when the infant was away (number of glances per 1,000 s) and rate of glancing at other individuals when the infant was in contact and when it was away (number of glances per 1,000 s). Maternal scratching (defined as a usually repeated movement of the hand o r foot during which the fingertips are drawn across the fur) was recorded in terms of rate of scratching when the infant was in contact and when it was away (number of episodes of scratching per 1,000 5). The rate of maternal visual monitoring (of the infant and of other individuals) and scratching when the infant was away was used as an indicator of maternal anxiety. The rate of scratching and visual monitoring of other individuals when the infant was in contact was used as an indicator of social anxiety (MAESTRIPIERI 1993).

Statistical Analyses Since individual differences in mothering style tend to be consistent over successive infant age periods (see Introduction), all behavioral data were pooled over the 12-wk period of observation. The Spearman’s rank order coefficient of correlation was used to assess whether there was concordance among the different measures of mothering style and anxiety. A stepwise multiple regression procedure was used to determine the relative contribution of maternal and social anxiety (as measured by visual monitoring) and various characteristics of the mother-infant dyad (maternal age, dominance rank and experience, number of immature offspring present in the group, sex of the infant) in explaining the variability in maternal behavior. This technique allows one to examine the effects of any number of quantitative andlor categorical independent variables upon a single dependent variable (PEDHAZUR1973). The relation between scratching and mothering style was analyzed by means of simple and polynomial regression. Stepwise multiple regression was also used to investigate the role of some maternal characteristics (age, experience, and dominance rank) as potential predictors of maternal and social anxiety.

36

DARIOMAESTRIPIERI Table 1: Spearman correlations between measures of mothering style Measures 1 2 3 4 5

r,

5

Mother Mother Mother Mother Mother

makes contact approaches grooms restrains rejects

2

3

4

5

0.82:':'

0.36 0.32

0 ,66::.;:-

-0.33 -0.26 0.02 -0.42

0.49:'

0.49':-

0.05., u:) p 5 0.01.

Results Concordance among the Different Indicators of Mothering Style

Mother makes contact, mother approaches, and mother restrains were all positively correlated with each other (Table 1). Since the positive correlation between mother makes contact and mother approaches may be due to some overlap between these two measures (e.g., because some approaches to the infant also involved a contact), only mother approaches was used in subsequent analyses. Mother grooms was positively correlated with mother restrains. This correlation, however, was due to a few mothers who often displayed grooming in conjunction with restraining. Therefore mother grooms was also discarded from subsequent analyses. Since the measures of mother approaches and mother restrains provide independent information on the mother's tendency to control her infant's movements, these measures were grouped together and their scores added up to obtain a composite measure (index of protectiveness). Maternal rejection was too infrequent to be analyzed (only three or four mothers rejected their infants). Concordance among Different Indicators of Anxiety

The rate of maternal scratching when the infant was in contact was positively correlated with the rate of scratching when the infant was away (Table 2). The same was true for the rate of glancing at other individuals. Visual monitoring of Table 2:

Spearman correlations between measures of anxiety

Measures 1 Rate of scratching with infant in contact

2 Rate of scratching with infant away 3 Rate of glancing at the infant 4 Rate of glancing at other individuals with infant in contact 5 Rate of glancing at other individuals with infant away

2 0.69't:'

3

4

5

0.41

0.31

0.64":)

0.38

0.25 0.10

0.55'' 0.50" 0.51')

37

Anxiety and Mothering Style in Macaques

the infant was positively correlated with visual monitoring of other individuals when the infant was away but not when the infant was in contact. Taken together, these results suggest that although mothers were consistent in their reactivity to risky situations, maternal and social anxiety were not entirely overlapping. Since the relation between scratching and an anxiety-related mothering style might be nonlinear (see Introduction), scratching and visual monitoring were considered separately in subsequent analyses. a 150 I)

al

-(a)

Q)

m al E

0

-

C

c

C

0

al

=> loo-

.-> 0

al

eP

r

9)

eQ

50-

r

0

0 X

X

al

a

al

0

5

0

I

1

1

I

i

20

40

60

00

100

VIsusl monltorlng of tho Intent (glsncss por 1 . W s)

.-> .I

I=-' :

-

U C

U

I

0

5

10

15

20

VIsusi monltorlng of othsr Indlvlduels when the Intent Is In contect (glenem per 1,000 0 )

Fig. 1: a. Linear regression of the index of

c

protectiveness on the rate of maternal visual monitoring of the infant. b. Linear regression of the index of protectiveness on the rate of 50 r maternal visual monitoring of other individuals 0 when the infant was in contact. c. Binomial X regression of the index of protectiveness on the al U rate of maternal scratching when the infant was away. In all cases, the index of protectiveness Scratching whon the Infant Is sway was obtained by combining together the meas(oplsodr por 1,000 s) ures Mother approaches and Mother restrains. The rates of glancing at the infant and at other individuals were calculated as the number of glances per 1,000 s the infants spent away from their mothers o r in contact. The rate of scratching was calculated as the number of episodes of scratching per 1,000 s the infants spent away from their mothers. Data are pooled over the 12-wk period. See Results for statistics 0 b Q

Predictors of Mothering Style

The rate of glancing at the infant and the rate of glancing at other individuals when the infant was in contact were the best predictors of variability in the index of protectiveness (stepwise regression analysis: glancing at the infant, F (1,15) = 18.31, p < 0.001; Fig. l a ; glancing at other individuals, F (2,14) = 19.46, p < 0.001; Fig. 1 b; see Table 3 for the other variables). Thus, mothers who

38

DARIO MAESTRIPIERI Tuble 3: Variables affecting maternal protectiveness Variables

F (2~4)

R

Maternal age Maternal experience Maternal social rank Sex of the infant Immature offspring Visual monitoring of other individuals with infant away

P

0.15 -0.03 0.16 -0.06 -0.44

0.29 0.01 0.34 0.05 3.17

ns ns ns ns ns

0.15

0.33

ns

I

ranked high on maternal and/or social anxiety were also those who adopted a highly protective mothering style. When the relation between scratching with infant away and protectiveness was analyzed, a binomial regression gave a better fit to the data (F (2,14) = 5.25, p = 0.01; Fig. 1c) than a simple linear regression (F (1,15) = 0.66, ns). Mothers who displayed intermediate levels of scratching were the most protective mothers. Mothers who displayed low or high levels of scratching ranked low on the index of protectiveness. Scratching with infant in contact was not significantly related to protectiveness (linear aggression, F (1,15) = 1.71, ns; binomial regression, F (2,14) = 1.87, ns). Predictors of Anxiety

Previous analyses have shown that the variation in three measures of maternal and social anxiety, namely, visual monitoring of the infant, visual monitoring of other individuals with infant in contact, and scratching with infant away, accounted for a significant fraction of the variation in a composite measure of maternal protectiveness. To assess the effects of maternal characteristics on social and maternal anxiety, stepwise multiple regression analyses were conducted using the above-mentioned measures as dependent variables and maternal age, experience and dominance rank as potential predictors. Maternal experience was the best predictor of variability in the rate of visual monitoring of the infant, while age and rank failed to explain a significant fraction of the remaining variation (Table 4, Fig. 2 a). For both rate of scratching with infant away and rate of glancing at other individuals with infant in contact, maternal age emerged as Table 4: Maternal characteristics affecting measures of anxiety Scratching with infant away Variable

R

Age 0.50 Experience 0.07 Rank -0.46

F

P

5.06 0.06 3.77

< 0.05 ns ns

Visual monitoring of the infant R F P 0.10 -0.47 0.09

0.14 4.22 0.12

ns

< 0.05 ns

Visual monitoring of other individuals with infant in contac R F P -0.63 0.00 0.37

10.02 0.00 2.24

< 0.01 ns ns

I

39

Anxiety and Mothering Style in Macaques c

E=

*I * 30-

lool(a)

Eg .- 9 c r CL c o

20

-

10

-

(b)

- 0

0 .

1

a

5

0

2

4

6

1

8

Mrterrul oxprknco (number of successfully rrlsed offrprlng)

Fig. 2: a. Linear regression of the rate of maternal visual monitoring of the infant on maternal experience. b. Linear regression of the rate of maternal scratching when the infant was away on maternal age. c. Linear regression of the rate of maternal visual monitoring of other individuals when the infant was in contact on maternal age. Rates were calculated as the number of episodes per 1,000 s. Behavioral data are pooled over the 12-wk period. See Results for statistics

-c fi 5ii 58 G E .-

a 10 -

?I fI ,Q

s53

3

(D

5

01 0

I

5

10

15

20

M r t o r r l age

(Y-W

the primary correlate of variation (Table 4; Fig. 2 b, c). Therefore, older and/or more experienced mothers scored lower than younger and/or inexperienced mothers in the measures of maternal and social anxiety.

Discussion The results of the present research provide an empirical confirmation to the hypothesis that individual differences in emotionality play a major role in determining variability in mothering style in nonhuman primates. Rhesus macaque maternal anxiety, as measured by visual monitoring, accounted for variability in a composite measure of maternal protectiveness to a greater extent than did characteristics of the mother-infant dyad such as maternal age, experience and dominance rank, number of immature offspring present in the group or sex of the infant. Although visual monitoring of the infant emerged as the primary correlate of variation in protectiveness, maternal visual monitoring of other individuals when the infant was in contact also explained a significant part of the variation in this dimension of mothering style. This finding lends support to the hypothesis that the manner with which a mother treats her infant is related to the way she responds to her surrounding environment (MITCHELL & STEVENS 1969; ALTMANN 1980) and indicates that both maternal and social anxiety play a role in shaping maternal protectiveness.

40

DARIOMAESTRIPIERI

Although maternal scratching and visual monitoring showed a similar sensitivity to potentially risky situations (MAESTRIPIERI 1993), they were related to maternal protectiveness in different ways. Whereas visual monitoring was linearly related to maternal protectiveness, scratching and protectiveness were associated according to an inverse U-shaped relation, the most protective mothers being those displaying intermediate levels of scratching. These findings seem to support the hypothesis that visual monitoring and scratching reflect different components of anxiety (see MAESTRIPIERI 1993) and in particular, that the former primarily reflects anticipation of danger whereas the latter is primarily an indicator of uncertainty o r motivational conflict. The present results, however, support only in part the hypothesis advanced by TROISI et al. (1991; see Introduction). O n the one hand, the fact that the most protective mothers displayed only intermediate levels of scratching confirms that highly protective mothers may not necessarily experience high motivational conflict. On the other hand, the fact that mothers who displayed the highest levels of scratching did not rank high on the Index of Protectiveness may also suggest that extreme uncertainty may inhibit some mothers and frequently prevent them from intervening on behalf of their infants. Although maternal and social anxiety are likely to reflect a true risk posed by 1993), individual the social environment to the mother-infant dyad (MAESTRIPIERI differences in anxiety seem to be mediated by maternal characteristics such as age and experience. In fact, primiparous macaque mothers have often been described as more anxious than multiparous ones (SEAY1966; MITCHELL & STEVENS 1969; HOOLEY & SIMPSON1981). The larger amount of aggression received by the former was insufficient to account for this difference because, for example, in one study (SEAY1966), primiparous mothers were anxious despite the fact that they were living alone. Therefore, HOOLEY & SIMPSON (1981) suggested that the lack of experience may be enough to cause anxiety in the mothers initially, and stress from the social environment and threats from others to both themselves and their infants may perpetuate it. Although macaque mothers seem to differ considerably in the extent to which they react to potential dangers present in their social environment, the present data d o not prove that individual differences in maternal emotional responsiveness necessarily reflect differences in baseline emotionality among individuals. In fact, females with similar emotional baselines may react differently to motherhood (TROISIet al. 1991). It is worth considering, however, that individual differences in temperament, especially along the dimensions of confidence and timidity, have shown some degree of consistency both over time and across generations, in nonhuman primates (e.g., STEVENSON-HINDE & SIMPSON 1981) as well as in other animals (e.g., LYONSet al. 1988). In fact, evidence from some nonprimate mammals indicates that individual differences in timidity or anxiety can be influenced by genetics and early experience (e.g., LYONSet al. 1988; MAESTRIPIERI & D’AMATO1991) and the same may hold true for nonhuman primates (e.g., SUOMI1987; FAIRBANKS 1989). Further studies using physiological indicators of anxiety to compare pre- and post-partum emotional responsiveness to environmental challenges may prove indispensable to answer the question of

Anxiety and Mothering Style in Macaques

41

whether differences emerging during motherhood are matched by corresponding differences in other stages of the reproductive cycle.

Acknowledgements I would like to thank Barry KEVERNE for allowing me to use the facilities available at the SubFran MARTEL, department of Animal Behaviour of the University of Cambridge; Barry KEVERNE, and Michael SIMPSONfor support and encouragement throughout the study and Claire NEVISON for his care of the monkeys; Filippo comments on an earlier draft of the manuscript; David RAYMENT for helpful discussion; Kelly CARROLLfor improving the style of the AURELIand Alfonso TROISI manuscript. The monkey colony is supported by a grant from the Medical Research Council. I was supported by a grant from the University of Rome La Sapienza.

Literature Cited ALTMANN, J. 1980: Baboon Mothers and Infants. Harvard Univ. Press, Cambridge. BERMAN,C. M. 1990a: Consistency in maternal behavior within families of free-ranging rhesus monkeys: An extension of the concept of maternal style. Am. J. Primatol. 22, 159-169. _ _ 1990 b: Intergenerational transmission of maternal rejection rates among free-ranging rhesus monkeys. Anim. Behav. 39, 329-337. DE WAAL,F. B. M. & LUTTRELL,L. M. 1985: The formal hierarchy of rhesus macaques: An investigation of the bared-teeth display. Am. J. Primatol. 9, 73-85. DUNBAR,R. I. M. 1988: Primate Social Systems. Croom-Helm, London. FAIRBANKS, L. A. 1989: Early experience and cross-generational continuity of mother-infant contact in vervet monkeys. Dev. Psychobiol. 22, 6 6 9 4 8 1 . _ _ & MCGUIRE,M. T. 1987: Mother-infant relationships in vervet monkeys: response to new adult males. Int. J. Primatol. 8, 351-366. HINDE, R. A. & HERRMANN, J. 1977: Frequencies, durations, derived measures and their correlations in studying dyadic and triadic relationships. In: Studies in Mother-Infant Interaction. (SCHAFFER, H . R., ed.) Acad. Press, London, pp. 1 9 - 4 6 . _ _ & SIMPSON,M. J. A. 1975: Qualities of mother-infant relationships in monkeys. In: ParentInfant Interaction, Ciba Found. Symposium Vol. 33, Elsevier, Amsterdam, pp. 3 9 - 6 7 . _ - & SPENCER-BOOTH,Y. 1967a: The behaviour of socially living rhesus monkeys in their first two and a half years. Anim. Behav. 15, 169-196. _ _ & - - 1967b: The effect of social companions on mother-infant relations in rhesus monkeys. In: Primate Ethology. (MORRIS,D., ed.) Weidenfeld & Nicholson, London, pp. 267-286. _ _ & - - 1971 : Towards understanding individual differences in rhesus mother-infant interactions. Anim. Behav. 19, 165-173. HOOLEY, J. M. & SIMPSON,M. J. A. 1981: A comparison of primiparous and multiparous motherinfant dyads in Mucacu muluttu. Primates 22, 379-392. LYONS,D. M., PRICE,E. 0. & MOBERG,G. P. 1988: Individual differences in temperament of domestic dairy goats: Constancy and change. Anim. Behav. 36, 1323-1333. MAESTRIPIERI, D. 1993: Maternal anxiety in rhesus macaques (Mucucu muluttu). I. Measurement of anxiety and identification of anxiety-eliciting situations. Ethology 95, 19-31. _ _ & D’AMATO,F. R. 1991 : Anxiety and maternal aggression in house mice (Mus musculus): A look at interindividual variability. J. Comp. Psychol. 105, 295-301. - -, SCHINO,G., AURELI,F. & TROISI, A. 1992: A modest proposal: Displacement activities as an indicator of emotions in primates. Anim. Behav. 44, 967-979. MARTIN,P. & BATESON,P. 1986: Measuring Behaviour. Cambridge Univ. Press, Cambridge. C. W. 1969: Primiparous and multiparous monkey mothers in a mildly MITCHELL, G. D. & STEVENS, stressful social situation: first three months. Dev. Psychobiol. 1, 28C-286.

42

DARIOMAESTRIPIERI, Anxiety and Mothering Style in Macaques

NASH,L. T. & WHEELER, R. L. 1982: Mother-infant relationships in non-human primates. In: Child H . E., Nurturance. Vol. 3, Studies of Development in Nonhuman Primates. (FITZGERALD, MULLINS, J. A. & GAGE,P., eds.) Plenum Press, New York, pp. 21-61. NICOLSON,N. A. 1987: Infants, mothers, and other females. In: Primate Societies. (SMUTS,B. B., CHENEY,D. L., SEYFARTH, R. M., WRANGHAM, R. W. & STRUHSAKER, T. T., eds.) Univ. of Chicago Press, Chicago, pp. 33G-342. PEDHAZUR, E. J. 1973: Multiple Regression in Behavioural Research. Holt, Rinehart & Winston, New York. SEAY,B. 1966: Maternal behavior in primiparous and multiparous rhesus monkeys. Folia primatol. 4, 146168. SIMPSON, M. J. A. & HOWE,S. 1980: The interpretation of individual differences in rhesus monkey infants. Behaviour 73, 127-155. STEVENSON-HINDE, J. & SIMPSON,M. J. A. 1981 : Mothers’ characteristics, interactions, and infants’ characteristics. Child Dev. 52, 1 2 4 6 1 2 5 4 . SUOMI,S. J. 1987: Genetic and maternal contributions to individual differences in rhesus monkey biobehavioral development. In: Perinatal Development: A Psychobiological Perspective. W., eds.) Acad. Press, New York, (KRASNEGOR, N., BLASS,E., HOFER,M. & SMOTHERMAN, pp. 3 9 7 - 4 1 9 . TANAKA, I. 1989: Variability in the development of mother-infant relationships among free-ranging Japanese macaques. Primates 30, 4 7 7 - 4 9 1 . TROISI,A,, SCHINO, G., D’ANTONI,M., PANDOLFI,N., AURELI,F. & D’AMATO,F. R. 1991: Scratching as a behavioral index of anxiety in macaque mothers. Behav. Neur. Biol. 56, 307-313. WHITE, L. E. & HINDE,R. A. 1975: Some factors affecting mother-infant relations in rhesus monkeys. Anim. Behav. 23, 527-542. Received: August 14, 1992 Accepted: April 7, 1993

u. Brockmann)