American Journal of Primatology 73:516–529 (2011)

REVIEW ARTICLE Emotions, Stress, and Maternal Motivation in

DARIO MAESTRIPIERIÃ Department of Comparative Development, The University of Chicago, Chicago, Illinois Recent research conducted with nonhuman primates confirms that adaptive emotional processes, such as maternal attraction arousability and maternal anxiety arousability, enhance and sustain female motivation to interact with , invest in them, and protect them during the postpartum period. Changes in these emotional processes, and concomitant changes in maternal motivation, facilitate the reduction and eventual termination of maternal investment associated with weaning. Although laboratory studies of and socially deprived rhesus monkeys have suggested that nulliparous females are neophobic and find infant stimuli aversive, recent research indicates that neophobia or aversion to infant stimuli do not occur in females with normal developmental experience. Furthermore, although some and human studies have shown that lactation is accompanied by physiological hyporesponsiveness to stress, other studies of rodents, nonhuman primates, and indicate that are highly vulnerable to stress and that stress-induced dysregulation of emotions can interfere with maternal motivation and behavior. It is possible that some aspects of the emotional and experiential regulation of maternal motivation and parental behavior are different in different mammalian . However, variation in the environments in which subjects are tested and in their developmental experience may also be responsible for the some discrepancies between the results of different studies. Am. J. Primatol. 73:516–529, 2011. r 2010 Wiley-Liss, Inc.

Key words: emotions; arousal; maternal motivation; rhesus monkeys;

INTRODUCTION in raising their litters without any prior maternal experience. In fact, in some of these species, Female is the rule among the over nulliparous females avoid any contact with other 300 species of nonhuman primates, the exceptions females’ pups or kill them and cannibalize them, at being a few socially monogamous species in which least in the laboratory [e.g. Fleming & Orpen, 1986]. carry their infants. Social and In rodents, maternal responsiveness and behavior paternal care are mainly concentrated among New are probably under relatively strong genetic control World monkeys and associated with particular repro- and are regulated by a combination of hormonal ductive or ecological characteristics (e.g. the produc- changes induced by pregnancy and lactation along tion of twins in marmosets and tamarins) [Snowdon with sensory stimulation from the [e.g. & Suomi, 1982]. Among the Old World monkeys and Fleming & Orpen, 1986]. Unlike rodents, Old World the , females typically give to one infant at a monkeys, apes, and humans have relatively long time, feed their offspring with milk for 1 year or more, spans, and in these species, many aspects of maternal and in some cases continue investing in them for the behavior depend, in part, on experience acquired rest of their with little or no contribution from during development. For example, nonhuman pri- males. During the period of infant dependence, mate mothers need experience to learn how to carry maternal care manifests itself mainly as breastfeed- their infants properly, to allow infants to maintain ing, carrying, and protecting the infant. Among great nipple contact long enough to obtain their necessary apes, mothers also play with their infants and share solid food with them. Primate maternal behavior is generally comparable, in terms of its basic character- istics, to that of many other mammalian species, ÃCorrespondence to: Dario Maestripieri, Department of Compara- although primate infants are generally carried more tive Human Development, The University of Chicago, 5730 S. Woodlawn Avenue, Chicago, IL 60637. E-mail: [email protected] by their mothers than other mammalian young. Received 2 July 2010; revised 8 August 2010; revision accepted In mammalian species with a short life span 12 August 2010 (e.g. many rodents), first-time mothers exhibit a full DOI 10.1002/ajp.20882 repertoire of maternal behavior from the moment Published online 24 September 2010 in Wiley Online Library their offspring are born and are generally successful (wileyonlinelibrary.com). r 2010 Wiley-Liss, Inc. Emotions, Stress, and Maternal Motivation / 517

nourishment, to protect infants from unwanted maternal motivation and facilitate the expression of attention from other conspecifics, and to encourage adaptive behavior. Stress-induced dysregulation of infant independence and weaning [e.g. Pryce, 1996]. emotions and their physiological substrates, how- In rhesus monkeys, newborn infants spend a ever, can alter maternal motivation and cause great deal of time nursing or sleeping on their maladaptive parenting behavior. ’s chest. During this time, they are cradled Pryce [1992] proposed a systems model of the and groomed by mothers. Mothers carry their infants regulation of maternal motivation in mammals, in while they travel for several months. During time which emotional processes play a central role (Fig. 1). devoted to resting or social activities, however, The model was based mainly on research conducted infants spend increasing amounts of time out of with laboratory rats, sheep, and socially deprived contact with their mothers, in order to explore the rhesus monkeys. In this model, there are some inputs, environment or to play with other infants. Mothers some outputs, and some intermediary processes. The may restrict and control their infants’ activities by main inputs to the system are infant stimuli. The physically restraining them and by making contact outputs are maternal motivation and maternal beha- with them frequently. During the first few weeks of vior. Crucial intermediary processes are previous infant life, mothers are almost entirely responsible maternal experience, female hormonal state, and an for maintaining contact and proximity with their emotional system called maternal arousability. The infants. For example, infants tend to break contact maternal arousability system comprises four different and walk away from their mothers, while mothers subsystems: maternal attraction arousability, mater- follow them and reestablish contact. During the nal anxiety arousability, maternal aversion arousabil- second or third month of infant life, however, ity, and novelty-fear arousability. Attraction and responsibility for maintaining contact and proximity anxiety arousability have positive influences on ma- shifts to infants. Therefore, mothers frequently ternal motivation, whereas aversion arousability and break contact and walk away from their infants, novelty-fear arousability inhibit maternal motivation. while infants follow their mothers and make contact According to Pryce’s model, immature mamma- with them. Mothers encourage their infants’ inde- lian females are exposed to infant stimuli before they pendence by breaking contact with them frequently are reproductively mature; they store these stimuli in and physically rejecting their infants’ attempts to memory and gain practical experience with them. make contact and gain access to the nipples. The Memory and experience of infant stimuli gained early frequency of maternal rejection increases steadily in a female’s life will facilitate the recognition of these with infant age, peaking when infants are approxi- stimuli later in life and enhance motivation to interact mately 6 months old, when the mating season begins with them via the influence of positive arousability and mothers return into estrus. intermediary processes. Infant stimuli can be passive Although these basic features of maternal (e.g. visual or olfactory characteristics) or active (e.g. behavior are observed in all rhesus monkey females, facial expressions, vocalizations, or motor behavior). individual mothers differ greatly from one another in When females reach puberty and conceive, hormonal the frequency with which they cradle or groom their changes during pregnancy and around the time of infants, make or break contact with them, or restrain parturition enhance maternal motivation and facil- or reject them. Individual differences in maternal itate the expression of maternal behavior. Maternal behavior are consistent over time and across differ- arousability is the most direct intermediary variable ent infants. These differences are accounted for by regulating maternal motivation and is itself regulated sociodemographic characteristics of the mothers (e.g. by three input factors: infant stimuli, previous their age, previous maternal experience, or dom- experience, and neuroendocrine state. inance rank), those of their infants (e.g. age or sex), The maternal attraction arousability subsystem and those of the surrounding environment (e.g. processes passive, visual and olfactory stimuli and is availability of food, risk of predation, risk of aggres- enhanced by sensory recognition of the infant and by sion or infant kidnapping by other group members). neuroendocrine changes occurring during pregnancy Among the mothers’ characteristics that affect their and parturition. These stimuli include the shape of motivation to take care of their infants and the infant faces, the configurations of their bodies, or quality of their maternal care, their personality or their odors. These stimuli are attractive to females temperament are very important. Emotional reactiv- and serve as incentives for maternal motivation and ity, in general, plays a fundamental role in the behavior. The maternal anxiety arousability subsys- regulation of maternal behavior in mammals, and tem processes active, behavioral stimuli (e.g. some this is especially true for primates including humans infant facial expressions and distress vocalizations, [Dix, 1991; Maestripieri, 1999; Pryce, 1992]. Emo- or infant-initiated separation from the mother) and tions can be powerful elicitors of maternal behavior is enhanced by sensory recognition of the offspring, and play a crucial role in mediating the impact of the classical conditioning between auditory and visual surrounding environment on the mother–infant infant stimuli, and by neuroendocrine changes. These dyad. Adaptive emotional processes can enhance stimuli induce anxiety in the mother. Maternal

Am. J. Primatol. 518 / Maestripieri

Fig. 1. Pryce’s [1992] model of the regulation of maternal motivation in mammals. Figure reprinted with permission. motivation is enhanced in an attempt to reduce Aversion and novelty-fear arousability reduce anxiety. Thus, the maternal anxiety arousability the overall level of maternal arousability and have an subsystem is an example of homeostatic feedback. inhibitory effect on maternal motivation. Maternal

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 519

aversion arousability responds particularly to infant attracted to infants and aroused by them? What olfactory stimuli; habituation with these stimuli is determines infant attractiveness and what modu- necessary to reduce aversion arousability through lates female arousability? Do female arousal and increased sensory recognition of the offspring. attraction to infants enhance maternal motivation Tactile stimulation from infants could also increase and facilitate the expression of adaptive maternal aversion arousability and reduce motivation to behavior? With regard to the maternal anxiety interact with them. Periparturitional neuroendocrine arousability subsystem, what determines maternal changes may contribute to the reduction of aversion anxiety and what are its modulating factors? Does arousability. Novelty-fear arousability processes all maternal anxiety enhance maternal motivation and unfamiliar infant stimuli, passive and active. The maternal behavior, as predicted by the model? outcome of this information processing is infant- Finally, with regard to the maternal aversion and directed fear and inhibition of maternal behavior. the novelty/fear arousability subsystems, do infants According to Pryce’s model, novelty-fear arousability represent novel and aversive stimuli for nulliparous is high in adult females but low in juveniles, thus females and does fear of novelty inhibit maternal allowing them to obtain the necessary maternal motivation, as in rodents? Is lactation associated experience with infant stimuli. In adult females who with lower reactivity to novelty and stress in give birth, both hormonal changes and sensory primates, as in rodents? And finally: can stress recognition of offspring reduce novelty-fear arousa- inhibit maternal motivation and impair the expres- bility. Therefore, the experiential and neuroendo- sion of adaptive maternal behavior? crine state of the parturient female is such that the positive arousability factors override the negative MATERNAL ATTRACTION AROUSABILITY ones, so that the mother is motivated to take care of the infant and maternal behavior is expressed. Female Attraction to Infants and Arousal At the time Pryce’s [1992] model was published, In rhesus , females of all ages are the regulation of maternal motivation in laboratory highly attracted to infants and show visible signs of rats and sheep was already well understood, but the arousal in their presence. They approach other regulation of maternal motivation in nonhuman females’ infants, use particular facial expressions primates was just beginning to be investigated. After and vocalizations to get their attention, and engage almost two decades of additional research on primate in various types of social interactions with them, maternal motivation and behavior, it is important to which are collectively referred to as infant handling assess whether the findings of this research are [Maestripieri, 1994a]. In most cases, infant handling consistent with the assumptions of Pryce’s model or consists of brief touching, holding, or grooming the whether the model needs revision. Although some infant. Rough infant handling (e.g. pulling, hitting, central assumptions of Pryce’s model are based on or dragging the infant) is sometimes referred to as studies of rhesus monkeys conducted in the 1950s infant harassment. Interactions in which infants are and 1960s, these studies were conducted in highly taken away from their mothers and prevented from artificial laboratory conditions, e.g. with socially being reunited with them are referred to as infant deprived and individually housed individuals. Recent kidnapping. research on the neuroendocrine regulation of primate Infant stimuli that seem to drive female attrac- maternal behavior, however, has been conducted with tion include: the shape of the face, body configura- individuals living in social groups in naturalistic or tion and pelage coloration, coo vocalizations, and semi-naturalistic conditions [Saltzman & Maestripieri, possibly olfactory, and motor characteristics [Gerald 2011]. Therefore, it is necessary to assess whether or et al., 2006]. An experimental study involving not the findings of early laboratory studies of rhesus digitized images of faces of newborn and 6-month- monkeys, on which Pryce’s model is based, have been old infants showed that adult rhesus females find confirmed and validated by more recent studies. these faces equally attractive and that the pink The goal of this article is to explore the role of coloration of the newborns’ face is not an additional emotions in the regulation of primate maternal attractive stimulus [Gerald et al., 2006; but see motivation, and in particular the maternal arousa- Higley et al., 1987]. bility system and its subsystems proposed by Pryce In rhesus macaques, a strong sex difference in [1992] in light of new research conducted in the last interest in infants, expressed in rates of infant two decades [Saltzman & Maestripieri, 2011]. The handling, emerges by the end of the first year of life primary focus of the article will be on rhesus and persists through the prepubertal years and macaques, although evidence from other primate beyond [Herman et al., 2003; Lovejoy & Wallen, species will be discussed as well. A number of 1988; Maestripieri, 2005a]. In the first year of life, questions will be raised and addressed about each males and females are not treated any differently by of the four maternal arousability subsystems. their mothers or other group members [Maestripieri, With regard to the maternal attraction arousability 2004]. Furthermore, male and female yearlings subsystem, is it true that primate females are spend similar amounts of time in proximity to their

Am. J. Primatol. 520 / Maestripieri

mothers, and mothers do not encourage or discou- Further correlational evidence of a relation between rage any social activities differentially in their sons infant handling and periparturitional changes in and daughters. Therefore, greater female interest in estrogen, progesterone, and also cortisol has been infants early in life is unlikely to be the result of obtained in Japanese macaques and in socialization and could reflect genetic predispositions [Bardi et al., 2001, 2003a,b,c, 2004; Ramirez et al., or prenatal hormonal exposure [see Berenbaum & 2004]. These findings obtained with Old World Hines, 1992; Gibber & Goy, 1985; Herman et al., monkeys confirmed previous results obtained with 2003; Leveroni & Berenbaum, 1998, for humans]. common marmosets, in which nonpregnant females After approaching other females’ infants and treated with estrogen and progesterone in concen- engaging in visual or tactile interactions with them, trations similar to those of late pregnancy showed a rhesus females of all ages show behavioral signs of significantly higher motivation to interact with arousal. The main one is a dramatic increase in the infant stimuli (they pressed a bar in order to rate of self-scratching. Scratching is the most simultaneously gain visual access to a replica of an common displacement activity in primates and may infant and turn off an audio recording of infant reflect arousal, excitement, or anxiety depending on distress vocalizations) than nontreated females the circumstances [Maestripieri et al., 1992]. Females [Pryce et al., 1993; see also Pryce et al., 1988]. also lip-smack and present their hindquarters to Pregnancy and lactation , such as estrogen, infants. Finally, they emit two specific vocalizations progesterone and , probably do not affect that are likely to reflect increased arousal: grunts maternal motivation directly but do so through their and girneys. Female grunts given in the presence of effects on neurochemical systems [Saltzman & infants have been described as brief bark-like atonal Maestripieri, 2011]. sounds, with an abrupt start and average length of The brain opioid system may provide one of the about 0.2 sec, which occur in a series and with short mechanisms through which females find infant intervals between units, and are produced with lip stimuli attractive and rewarding. The role of the protrusion but negligible lip separation [e.g. Rowell opioid system in regulating maternal motivation has & Hinde, 1962]. Girneys have been described as been investigated in three studies of rhesus maca- soft, low frequency, nasal or ‘‘singing’’ noises that ques producing conflicting results. In one study, the are produced through lip protrusion, minimum lip opioid system was pharmacologically manipulated separation, and rapid lip movement, as if the animal after mother–infant separation and reunion [Kalin is chewing [Green, 1975; Hauser, 1993]. Both grunts et al., 1995]. Morphine decreased clinging with the and girneys are likely to reflect female arousal infant during the first 30 min of reunion, whereas elicited by the sight of an infant or its behavior the opioid receptor blocker naltrexone increased (e.g. walking independently), and are accompanied clinging. In a study of socially living rhesus mothers by other signs of arousal such as tail-wagging and infants, however, naloxone (another opioid [Whitham et al., 2007]. In conclusion, recent studies receptor blocker) reduced both maternal grooming confirm that in rhesus macaques, females of all ages and maternal restraining of the infant, suggesting find other females’ infants attractive, get aroused by decreased rather than increased attachment to the them, and actively attempt to engage in interactions infant [Martel et al., 1993]. In this study, however, with them. the effects of naloxone on affiliative interactions between mothers and other adults were also contrary to the expectations. In fact, the mothers treated with Neuroendocrine Factors Modulating Female naloxone showed reduced number of grooming Attraction Arousability solicitations and reduced amount of grooming A longitudinal study of group-living captive received from other individuals. Finally, in another pigtail macaques reported that the frequency of study of rhesus macaques, naltrexone had no female infant handling increased during late preg- significant effects on any aspects of maternal nancy and peaked the week before parturition behavior, including abusive parenting [Graves [Maestripieri & Zehr, 1998]. The increase in infant et al., 2002]. Although some of these inconsistencies handling during late pregnancy was correlated with may be due to methodological differences between an increase in the concentrations of plasma estradiol studies, further research is needed before any firm and in the estradiol to progesterone ratio. This conclusions can be drawn regarding the relationship correlational evidence that hormones can affect between opioids and maternal motivation. caregiving motivation was corroborated by experi- The neuropeptide is another primary mental manipulations of hormonal variables. Rhesus candidate as a mechanism through which primate females whose ovaries had been removed females find infant stimuli rewarding. The role of increased significantly their frequency of interac- oxytocin in the regulation of maternal motivation in tions with other females’ infants after the adminis- primates, however, is only beginning to be investi- tration of estradiol in doses similar to those of gated. In a pilot experiment conducted with two middle-late pregnancy [Maestripieri & Zehr, 1998]. nulliparous females, Holman and

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 521

Goy [1995] examined whether an intracerebroven- important for the expression of competent maternal tricular injection of oxytocin affected responsiveness behavior in primates. For example, among macaques to infants. The two females were exposed to an living in naturalistic social groups, first-time mothers unfamiliar infant in a cage 10 min after the injection are generally more likely than multiparous mothers of oxytocin or saline. The females sat near the infant, to neglect and abandon their infants or to show and watched, touched, and lipsmacked to the infant clumsy maternal behavior such as carrying infants more frequently following oxytocin compared with upside down [Maestripieri & Carroll, 1998a,b; Schino saline administration. In no case, however, was more & Troisi, 2005]. The probability of infant abandon- intense caregiving behavior observed, perhaps because ment is greatly reduced as females give birth to of the environment in which the animals were tested. successive infants, and the quality of maternal care In another study, CSF levels of oxytocin measured in typically increases as well [Maestripieri & Carroll, ten multiparous rhesus females before, immediately 1998b; Schino & Troisi, 2005]. after parturition, and 7 days postpartum were not Important experience for maternal correlated with mother–infant behaviors such as behavior can be acquired by primate females during contact or grooming [Cooke et al., 1997]. Finally, interactions with their own mothers in infancy, Boccia et al. [2007] reported that the administration through observation of interactions between their of a human uterine oxytocin receptor blocker mothers and younger siblings or between other reduced the frequency of lip-smacking, approaching, females and their infants during the juvenile period, and touching a stimulus infant in one 4-year-old through direct interaction with younger siblings or nulliparous rhesus macaque female. Although the other females’ infants during the juvenile period, effects of oxytocin receptor blockade on infant- and through direct interactions with their own directed behavior in one subject were suggestive of offspring. Young females who are aroused by infants a relation between oxytocin and maternal motivaiton, are highly motivated to interact with them, and the authors of this study acknowledged that other those who do so become better mothers in the future. explanations for their results were also possible. Research with vervet monkeys has shown that interactions with younger siblings or other females’ infants during the juvenile period affect the quality Do Female Attraction to Infants and Arousal of maternal care and the probability of infant Enhance Maternal Motivation and Facilitate survival of first-time mothers. Specifically, first-time the Expression of Adaptive Maternal mothers who had greater infant handling experience Behavior? with infants as juveniles were more likely to be Most, if not, all primate research on female competent mothers and their infants had a higher attraction to infants and arousability has been probability of survival than females with less juvenile conducted using other females’ infants as stimuli. experience with infants [Fairbanks, 1990]. Interest There is virtually no research on maternal arousa- in infants in the juvenile years and motivation to bility in response to a female’s own offspring. Work engage in extensive social interactions with them on postpartum mother–infant bonding in rhesus probably represent a developmental adaptation that macaques, however, is consistent with Pryce’s has been best documented in nonhuman primates [1992] suggestion that arousal and attraction to and humans [Maestripieri & Roney, 2006], but which one’s own infant provide the motivational basis for may be present in other mammalian species as well, postpartum bonding and maternal behavior in the especially those with relatively long lifespans and first few weeks of infant life [Maestripieri, 2001]. complex patterns of maternal care. After the first few postpartum weeks, maternal motivation decreases steadily, in part due to a MATERNAL ANXIETY AROUSABILITY reduction in maternal attraction arousability. This is because there are changes in the value of infant Expression of Maternal Anxiety, Its stimuli (infants probably become less attractive to Determinants, and Modulating Factors their mothers as they grow older) and changes in In primates, maternal anxiety is expressed with maternal neuroendocrine variables that reduce physiological changes (autonomic nervous system responsiveness to these stimuli. Little or no research activation, and release of catecholamines and cortisol) on this topic, however, has been conducted. and with behavioral changes such as increased Evidence that female attraction to infants, in vigilance and increased rates of self-directed activities general, enhances maternal motivation and facilitates such as scratching. Maternal anxiety is triggered by the development and expression of maternal beha- the perception of danger for the offspring. In the vior is provided by the observed relationship between wild, it is likely that the sighting of predators or the infant handling early in life and quality of maternal hearing of alarm calls would trigger maternal behavior in adulthood. Several lines of evidence anxiety. In addition, different aspects of separation suggest that developmental experience with other from the offspring can trigger maternal anxiety such females’ infants and with one’s own offspring is as: (1) forced separation and inability to regain

Am. J. Primatol. 522 / Maestripieri

contact with the infant, such as the experimental life is associated with a dramatic increase in the separation of mother and infant in a laboratory frequency of maternal scratching and in the rate of setting, or separation due to infant kidnapping by glancing at the infant and at other individuals another group member; (2) infant-initiated sponta- [Maestripieri, 1993a,b, 1995c]. In one study, the rate neous separation, such as for exploration of the of maternal scratching increased when infants were environment. Both visual and auditory stimulation in proximity to individuals who frequently harassed (i.e. observation of the separated infant and hearing and kidnapped them [Maestripieri, 1993a]. Elevated of infant distress calls) can trigger anxiety. When scratching rates were also observed during episodes young infants are out of contact with their mothers, of infant kidnapping, when infants were physically their proximity to and interaction with dangerous prevented from returning to their mothers by other conspecifics can also trigger maternal anxiety adult females [Maestripieri, 1993c]. [Maestripieri, 1993a, 1995b]. Maternal separation anxiety is likely to reflect both fear of infant loss and uncertainty over the Maternal Responses to Forced Separation in a action to be taken. Several findings suggest that maternal separation anxiety has a cognitive com- Laboratory Setting ponent and is not a reflexive, stereotyped response. Experimental separation studies have used Specifically, maternal anxiety appears to reflect both mothers and infants individually housed in a cage the presence of an objective risk for the infant in the or living in a social group. When the subjects are social environment and the mother’s subjective living in a group, separation involves removal of the assessment of such a risk. Both scratching and visual mother or the infant from the group, or removal of monitoring of the infant and of other individuals both mother and infant from the group and while the infant is out of contact decrease steadily subsequent separation. When monkey mothers are with infant age over the first 12 postpartum weeks, forcibly separated from their infants, they typically as infants become less vulnerable to harassment display vocalizations and signs of behavioral agita- and kidnapping [Maestripieri, 1993a]. Furthermore, tion [Levine & Wiener, 1988]. Maternal vocalizations maternal separation anxiety is higher in species of and agitation reflect efforts to locate and regain macaques in which the risk to the infant is higher contact with the infant and they usually cease on [Maestripieri, 1994b,c]. Therefore, the occurrence reunion [Maestripieri & Call, 1996]. Maternal beha- and developmental changes of maternal separation vioral responses to separation are accompanied by anxiety reflect an objective risk of infant injury or physiological responses, such as increased plasma loss and the infant’s vulnerability to such a risk. cortisol levels [squirrel monkeys: see Coe et al., 1985; Although macaque mothers also display signs of Levine & Wiener, 1988, for reviews; rhesus maca- anxiety when they are themselves at risk of aggres- ques: Champoux & Suomi, 1994; Levine et al., 1985; sion, mothers do not simply generalize the concerns but see Smotherman et al., 1979, for negative they have for themselves to their infants. Rather, results]. Altogether, these studies indicate that mothers distinguish between risk to themselves and separation from the infant is an anxiogenic event risk to their infants, and maternal anxiety can be for a monkey mother. Behavioral agitation is distinguished from social anxiety, although the two accompanied by the activation of the hypothalamic– are correlated with each other [Maestripieri, 1993a, pituitary–adrenal axis, and both the behavioral and 1995b]. For example, rhesus mothers did not avoid the physiological responses are generally enhanced adult females lower ranking than themselves but if separation from the infant is associated with displayed anxiety when these individuals approached exposure to novel procedures or environments. and interacted with their infants [Maestripieri, 1995b]. One explanation for this finding could be Maternal Responses to Interruption of Infant that rhesus mothers are aware that their infants’ Contact in Group-Living Monkeys vulnerability to risks can be different from their own Primate mothers and infants continually experi- because of the infants’ smaller size and lack of ence brief and repeated separations as a result of the experience. Alternatively, mothers may have learned process of infant independence. In macaques, in the on the basis of their direct experience that certain first month of infant life, mother–infant contact is social interactions have different consequences for mainly interrupted by maternal attempts to en- themselves and their infants, and elicitation of courage infant locomotion and by infants’ early anxiety would simply reflect these learning experi- attempts to explore their surrounding environment ences. Consistent with this explanation, there is [Maestripieri, 1995a, 1996]. Early separations are evidence that the mother’s age and experience can brief and usually terminated by the mother. Then, influence her assessment of risk for the infant in the the frequency with which infants break contact with environment and her emotional response to separa- their mothers and walk away from them increases tion [Maestripieri, 1993b]. Part of the variability in steadily with age [Hinde & Atkinson, 1970]. Interrup- maternal separation anxiety, however, is not accounted tion of contact with the infant in the first weeks of for by demographic characteristics of mothers and

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 523

infants, suggesting that perhaps the mother’s dependent on mother–offspring contact and proxi- temperament or personality may be important. mity. Therefore, the notion that maternal anxiety Differences in temperament may also be implicated arousability plays a central role in the regulation of in species differences in maternal separation anxiety maternal motivation in mammals [Pryce, 1992] is [see Clarke & Boinski, 1995]. fully legitimate. In primates, as in humans, there is evidence that too little or too much concern over separation may be Does Maternal Anxiety Enhance Maternal associated with dysfunctional parenting. Troisi and Motivation and Maternal Behavior? D’Amato [1984, 1991] presented evidence from a case Consistent with Pryce’s [1992] model, several of infant abuse in Japanese macaques suggesting lines of evidence suggest that maternal anxiety that extreme maternal separation anxiety played an enhances maternal motivation and the expression important role in the occurrence of abuse. Although of protective behaviors. First, behavioral expressions these authors did not measure maternal separation of maternal anxiety are immediately followed by anxiety with scratching or visual monitoring, the maternal protective behaviors. For example, mothers abusive mother alternated abuse with extremely who scratch themselves at high rates when their possessive maternal responses, and abuse occurred infant is in proximity to a dangerous group member when infants attempted to break contact with the [e.g. a female with a history of infant kidnapping) mother or did not respond to her retrieval signals. will immediately retrieve the infant [Maestripieri, Moreover, the administration of an anxiolytic drug to 1995b]. Second, mothers who exhibit high levels of the abusive mother resulted in a reduction of anxiety also have highly protective maternal styles. maternal protectiveness and termination of abuse For example, in rhesus macaques, two indicators of [Troisi & D’Amato, 1991]. A recent study of ten maternal separation anxiety were better predictors group-living rhesus mothers who physically abused of variability in maternal protectiveness than any their infants reported that abusive mothers did sociodemographic characteristics of the mothers. not show the typical separation-related increase The rate of visual monitoring of the infant was in scratching during the first 2–3 weeks of life linearly related to protectiveness, whereas the rate of [Maestripieri, 1998]. The first weeks of infant life scratching showed an inverse U-shaped relation with were also the period with the highest frequency of protectiveness, i.e. the mothers with the highest abuse. In the third month of infant life, however, scratching rates during separation had intermediate when abuse was considerably reduced, maternal levels of protectiveness [Maestripieri, 1993b]. This separation anxiety was higher in the abusive finding was interpreted by suggesting that visual mothers than in the control mothers. Thus, early monitoring primarily reflects the fear component of abuse was associated with low maternal separation maternal anxiety, whereas scratching primarily anxiety, whereas a reduction in abuse and increase reflects the uncertainty component. Therefore, the in protectiveness in the third month were associated most and the least protective mothers have only with increased maternal separation anxiety. The moderate scratching rates because they are strongly developmental changes in abuse frequency, maternal biased toward intervention and non-intervention, separation anxiety, and parenting style suggest that respectively [Maestripieri, 1993b]. A study of Japanese the development of attachment between abusive macaques, which used overall scratching frequency mothers and their infants may be slower or delayed instead of scratching rates during separation, relative to controls. reported a linear relationship between scratching Data on infant abuse in pigtail macaques lend and maternal protectiveness [Troisi et al., 1991]. further support to the notion that lack of concern The relationship between maternal anxiety and over infant separation may result in maladaptive maternal protectiveness is consistent with the notion parenting. In one well-documented case of abuse, that maternal anxiety arousability has an adaptive maternal abusive behavior appeared in strict tem- value in that it promotes proximity to the infant and poral association with the repeated kidnapping and protective behaviors such as retrieval and restraint. severe harassment of the infant by another adult Insofar as maternal separation anxiety activates female in the group, abuse being absent before the behaviors that promote proximity to and protection appearance of the kidnapping episodes and drasti- of the offspring, maternal separation anxiety can cally reduced after their termination [Maestripieri, significantly increase the chances of offspring survi- 1994d]. The abusive mother did not exhibit obvious val and the of the parent. signs of maternal separation anxiety during the Therefore, maternal separation anxiety may be a forced separation from her infant nor did she specific adaptation of parenting, just as infant attempt to prevent further kidnappings by increas- separation anxiety is an adaptation of the developing ing her protectiveness. Abuse occurred at reunion infant. Accordingly, maternal separation anxiety with the infant, often after several hours of forced should be a phenomenon common to all animal separation. Maestripieri [1994d] argued that the species in which offspring survival is crucially mother’s failure to cope with the stressful situation,

Am. J. Primatol. 524 / Maestripieri

both emotionally and behaviorally, may have been a abusive mothers contribute to the manifestation of factor precipitating abuse. maternal rejection and abusive parenting in adult- Although failure to show a maternal separation hood. It is possible that experience-induced reduction anxiety response when infants are young and in serotonergic function results in elevated anxiety vulnerable may be associated with neglectful or and impaired impulse control, and that high anxiety abusive parenting, a decrease in maternal separation and impulsivity increase the probability of occur- anxiety as a function of increasing infant age and rence of maternal rejection and abusive parenting reduced infant vulnerability is adaptive and may with one’s own offspring later in life, perhaps in facilitate infant weaning. Thus, the steady reduction conjunction with social learning resulting from in maternal motivation that leads to infant weaning direct experience with one’s own mother or from over the first 6–12 months of infant life is caused not observation of maternal interactions with siblings only by a reduction in maternal attraction arousa- [Maestripieri, 2008]. Although it is likely that bility but also in maternal anxiety arousability. serotonin affects maternal motivation through its At the proximate level, individual differences effects on maternal anxiety arousability, it is also and developmental changes in maternal anxiety are possible that serotonin affects maternal attraction probably the result of the activity of brain neuro- arousability through its effects on oxytocin or transmitter systems that are involved in emotion prolactin release [Insel & Winslow, 1998; Numan & regulation, and especially anxiety [Lonstein, 2007]. Insel, 2003]. A primary candidate is the monoamine neurotrans- In conclusion, research on maternal anxiety mitter serotonin. Early studies of serotonin and arousability and its neurochemical substrates con- maternal behavior in primates reported that monkey ducted in the last two decades confirms Pryce’s mothers with low concentrations of the serotonin [1992] assumption that this system plays an im- metabolite 5-HIAA in their cerebrospinal fluid (CSF) portant adaptive role in the regulation of maternal were more protective and restrictive, and that their motivation in primates. infants spent more time in contact with them, than mothers with high CSF 5-HIAA [Fairbanks et al., MATERNAL AVERSION AROUSABILITY 1998; Lindell et al., 1997]. Cleveland et al. [2004] found no relationship between CSF 5-HIAA and Do Infants Represent Aversive Stimuli for maternal behavior in the first few postpartum days, Nulliparous Primate Females and Does but on postpartum days 15 and 20, females with low Aversion Impair Maternal Motivation? CSF 5-HIAA broke contact and left their infants less Pryce’s [1992] assumption that mammalian frequently than females with high CSF 5-HIAA. females without hormonal priming and/or develop- A preliminary study in our laboratory reported a mental experience with infants find neonates aver- positive correlation between CSF 5-HIAA concentra- sive is mainly derived from laboratory studies of tions measured during pregnancy and maternal rodents and of socially deprived rhesus monkeys. In rejection behaviors in the first postpartum month laboratory rats and other rodents, virgin females find in multiparous females [Maestripieri et al., 2005]. pup odors aversive and will actively avoid or kill Our more recent work involving multiple measure- unfamiliar pups. The idea here is that pup odors ments of CSF 5-HIAA during development, however, elicit negative emotions (perhaps disgust?) and that reported a negative correlation between CSF 5-HIAA such emotions reduce maternal motivation and and maternal rejection among first-time mothers increase avoidance or infanticidal behavior. Thus, [Maestripieri et al., 2007]. research with laboratory rodents has shown that an The serotonin system may also be involved in important component of the elicitation of maternal the intergenerational transmission of infant abuse. behavior through pregnancy, hormonal treatment, We reported that about half of the females who were or sensitization involves the inhibition and suppres- abused by their mothers early in life, whether sion of aversive, avoidant, and infanticidal responses crossfostered or non-crossfostered, exhibited abusive elicited by pups. parenting toward their first-born offspring, whereas Whether nulliparous females’ aversion to infant none of the females reared by nonabusive mothers stimuli is a general phenomenon shared by all did (including those born to abusive mothers) mammals is uncertain. One possibility is that it is a [Maestripieri, 2005b]. Moreover, the abused females, by-product of the reproductive and social systems of both crossfostered and non-crossfostered, who became rodents. Most rodents live in territorial social abusive mothers had lower CSF 5-HIAA concentra- systems in which female–female aggression and tions than the abused females who did not become play an important role [Maestripieri, abusive mothers [Maestripieri et al., 2006]. Since 1992]. Mothers themselves kill and cannibalize their abuse tends to co-occur with high rates of maternal own offspring when their territories have been rejection, our findings suggest that experience- invaded by intruders and they have been unable to induced long-term alterations in serotonergic func- effectively defend their litter [Maestripieri, 1992]. tion in females reared by highly rejecting and Although female territoriality and competitive

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 525

infanticide (i.e. killing other females’ offspring) are interpreted as one of the many aberrations induced not uncommon in other mammals, they appear to by traumatic developmental experience rather than play a relatively small role in the social and the result of a specific deficit in learning mothering reproductive strategies of extant primates, with the skills. Fourth, when primate females are housed possible exception of common marmosets and other alone with an infant in a small cage, they are more New World monkeys [Digby & Saltzman, 2009; likely to avoid or abuse the infant regardless of their Saltzman et al., 2008]. Thus, it is possible that the previous experience, although young females may mechanisms underlying the induction of maternal be more sensitive to this stressful situation than behavior through the suppression of avoidant and older females [Maestripieri & Carroll, 1998a]. Thus, infanticidal tendencies in rodents do not apply to studies of socially deprived monkeys do not offer other mammals such as primates. Another possibility unequivocal evidence that nulliparous females find is that nulliparous females’ aversion to pups is infants aversive or that experience with infants or common in laboratory rodents, which often do not hormonal priming is necessary to develop adequate have normal developmental experiences and live in caregiving skills. highly artificial environments, but less common Nulliparous females with normal developmental in wild rodents and other mammals living in experiences (i.e. who were not socially deprived) do naturalistic environments. When female rodents in not find infant stimuli aversive and do not show the laboratory are exposed to pups and allowed to avoidant or abusive responses to them. When gain experience with them during early develop- macaque females live in their natal social group ment, aversion to infants and the occurrence of surrounded by their family and other conspecifics of infanticide later in life are greatly reduced. Finally, different sex and ages, they usually find newborns in some strains of mice, nulliparous females sponta- very attractive and not aversive. More generally, it is neously exhibit maternal behavior without any hard to imagine the evolutionary reasons why previous juvenile parenting experience. primate females should perceive infant stimuli as Laboratory studies of socially deprived rhesus aversive and not be biologically predisposed to monkeys showed that when females were separated display some basic caregiving skills. Thus, the at birth from their mothers, raised in conditions of existence of a maternal aversion arousability system social deprivation, and individually housed with their in primates is uncertain. Whether this system plays a infant in a small cage, there was a high incidence of central role in the regulation of maternal motivation inadequate maternal behavior including infant abuse in rodents and other mammals is uncertain as well, and neglect [Ruppenthal et al., 1976; Suomi, 1978]. and the possibility should be taken into considera- Furthermore, the proportion of mothers displaying tion that female aversion to infants is a prominent adequate maternal care increased with successive phenomenon in laboratory rodents mainly due to infants, suggesting that maternal behavior improved their abnormal developmental histories and artificial with experience. One study of rhesus monkeys in environments. which individually housed nulliparous females were tested with strange infants showed that females MATERNAL NOVELTY-FEAR avoided any interaction with the infants [Gibber, AROUSABILITY 1986]. Taken together, these findings have led to the notion that, similar to rodents, nulliparous monkey Is Lactation Associated With Lower Reactivity females find newborns aversive and need experience to Novelty and Stress in Primates? and/or hormonal priming to overcome their aversion. In laboratory rodents, nonpregnant and non- These extrapolations, however, need to be made with lactating females are highly neophobic. It has been caution for several reasons. First, one-third of the argued that fear of novelty reduces motivation rhesus mothers studied by Ruppenthal et al. [1976] for maternal behavior and increases avoidance or displayed adequate maternal care with their first- infanticidal behavior [Fleming & Luebke, 1981]. born infants, suggesting that their caregiving Pregnancy and lactation are characterized by skills had not been significantly impaired by their reduced endocrine and behavioral reactivity to developmental history. Second, while the occurrence novelty and stress. Lactation, in particular, makes of infant neglect decreased with successive infants, mothers less anxious and less fearful; this enhances the occurrence of abuse did not. This finding, maternal motivation and facilitates the expression of along with those obtained in group-living macaques maternal care [Lonstein, 2007; Maestripieri & [Maestripieri & Carroll, 1998b], suggests that abuse D’Amato, 1991; Maestripieri et al., 1991]. Whereas does not result from lack of experience with infants the postpartum period has been associated with and is not overcome through such experience. Third, lower physiological reactivity to stress in some most socially deprived mothers displayed maladap- studies of rodents [e.g. Tu et al., 2006; but not in tive behavior in all behavioral interactions with others such as Deschamps et al., 2003, in which rat conspecifics [Suomi, 1978], suggesting that their mothers showed high physiological reactivity to maladaptive infant-directed behavior could be best stress when tested in presence of their pups and

Am. J. Primatol. 526 / Maestripieri

pups were in potential danger] and humans [Altemus motherhood and low social status are stressful for et al., 1995, 2001; but see Heinrichs et al., 2001, who females, and that the stress of motherhood is greater showed that this is a transient effect that occurs only for low-ranking than for high-ranking females. Thus, during breastfeeding], pregnancy and the postpar- as with the maternal aversion arousability system, tum period in primates are characterized by high the existence of a novelty-fear arousability system in emotional instability and reactivity. primates is uncertain. Recent data obtained with free-ranging rhesus macaques on the island of Cayo Santiago strongly suggest that lactation and motherhood is character- MOTHERHOOD AND STRESS IN PRIMATES ized by heightened, not diminished, endocrine Although some human studies have shown that reactivity to stress. This 2-year study involved, as moderately high cortisol levels in the early postpar- subjects, 70 adult females between 6 and 26 years of tum period can be associated with greater sensitivity age. We observed their behavior as well as collected to infant cues and greater maternal responsiveness fecal samples for cortisol assays on a weekly basis. [Fleming et al., 1987, 1997], a large body of evidence Females were captured once a year, and blood indicates that extremely high or chronically elevated samples were collected for plasma cortisol assays. cortisol levels due to stress can impair maternal Of the 70 females, 30 were captured in the first year motivation and result in maladaptive parenting of the project, 18 were captured in the second year, behavior. Specifically, it has been shown in humans and 22 were captured in both years. Fecal that stress is a major risk factor for postpartum metabolite concentrations were used as a measure of depression and for neglect and abuse; in baseline cortisol levels while plasma cortisol concen- addition, stressed mothers are more likely to find trations measured after capture and overnight infant stimuli (e.g. cries) aversive [e.g. Adamakos individual housing were used as indicators of endo- et al., 1986; Besser et al., 2002; Wolfe, 1987]. crine responsiveness to stress. Data from the 22 Preliminary evidence that dysregulation of females captured in both years indicated that there emotion due to stress can impair maternal motiva- were significant positive correlations between mea- tion and result in maladaptive parenting also exists sures of plasma cortisol concentrations in the 2 years for nonhuman primates. In a study of captive [Hoffman et al., 2010]. Moreover, individual differ- lowland , stress due to social disruption and ences in plasma cortisol responses to stress were abnormal housing conditions was associated with positively correlated with individual differences in high concentrations of cortisol in fecal samples, and baseline cortisol concentrations measured in fecal high cortisol, in turn, was associated with infant samples [Hoffman et al., in preparation]. Both neglect [Bahr et al., 1998]. In pigtail macaques, between- and within-subject analyses showed that maternal abuse of offspring is often precipitated plasma cortisol concentrations in response to the by socially stressful events [Maestripieri, 1994d; stress of capture were higher when females were Maestripieri & Carroll, 1998c]. In addition, rhesus lactating than when they were cycling [Hoffman macaque abusive mothers have neurochemical et al., 2010]. Moreover, higher cortisol concentra- profiles similar to those of humans with PTSD, tions in lactating females were also found with the suggesting chronic stress [Maestripieri et al., 2005]. fecal hormonal data [Hoffman et al., in preparation]. In captive baboons, Brent et al. [2002] reported that Higher baseline cortisol and plasma cortisol responses the postpartum period was associated with a sig- to stress are unlikely to reflect metabolic processes nificant increase in stress-related behavior, and in a associated with lactation because, in humans, it has related study, the same investigators showed that been demonstrated that cortisol levels return to pre- stressed mothers who showed abnormal behaviors conception levels within 24–48 hr after giving birth also had higher levels of urinary cortisol [Bardi et al., (the same is true for corticosterone levels in rodents). 2004]. Finally, there is a recent experimental Instead, higher cortisol levels in lactating females are evidence that exogenous administration of elevated likely to reflect social stress associated with mother- doses of cortisol in lactating female common marmo- hood. In fact, motherhood and lactation are periods sets impairs maternal motivation and interferes with of high stress and increased stress reactivity in the expression of appropriate maternal behavior primates and humans due to maternal concerns for [Saltzman & Abbott, 2009]. infant safety, disruption of sleep and other main- Although the relationship between primate tenance activities, and stressful changes in social maternal behavior and stress-sensitive physiological relationships with others. In our rhesus monkey systems such as the HPA axis or the brain study, plasma cortisol levels were also higher among serotonergic system needs to be better understood low-ranking than among high-ranking females. [Saltzman & Maestripieri, 2011], it is becoming Moreover, low-ranking females had greater increases increasingly clear that motherhood is associated with in cortisol from the cycling condition to the lactating increased stress, and that stress can impair maternal condition than high-ranking females [Hoffman et al., motivation. Therefore, the effects of stress need to be 2010]. Taken together, these results suggest that taken into consideration in a model of the regulation

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 527

of maternal motivation in mammals, along with the Bardi M, Shimizu K, Barrett GM, Borgognini-Tarli S, possibility that the relationship between lactation and Huffman MA. 2003c. Peripartum sex steroids and maternal stress may be different in different mammalian species style in rhesus and Japanese macaques. General and Comparative Endocrinology 133:323–331. or in animals tested in the laboratory vs the wild. Bardi M, French JA, Ramirez SM, Brent L. 2004. The role In conclusion, evidence that emotions play a of the endocrine system in maternal behavior. central role in the regulation of maternal motivation Biological Psychiatry 55:724–732. and parental behavior of mammals is overwhelming. Berenbaum SA, Hines M. 1992. Early are related to Many assumptions and predictions of Pryce’s [1992] childhood sex-typed toy preferences. Psychological Science 3:203–206. model have been confirmed by recent studies of Besser A, Priel B, Wiznitzer A. 2002. Childbearing depressive nonhuman primates. At the same time, these studies symptomatology in high- risk pregnancies: the roles of have suggested that certain extrapolations from working models and social support. Personal Relationships studies of laboratory rodents and socially deprived 9:395–413. monkeys, particularly with the regard to the existence Boccia ML, Goursaud AS, Bachevalier J, Anderson KD, Pedersen CA. 2007. Peripherally administered non-peptide and importance of a maternal aversion arousability oxytocin antagonist, L368,899, accumulates in limbic brain and a maternal novelty-fear arousability subsystem, areas: a new pharmacological tool for the study of social need to made be caution. Future studies are motivation in non-human primates. Hormones and Beha- necessary to assess whether the lack of strong vior 52:344–351. evidence for these subsystems in other Brent L, Koban T, Ramirez S. 2002. Abnormal, abusive, and stress-related behaviors in baboon mothers. Biological than laboratory rodents is due to differences in the Psychiatry 52:1047–1056. regulation of maternal motivation between rodents Champoux M, Suomi SJ. 1994. Behavioral and adrenocortical and other mammals, or due to the abnormal develop- responses of rhesus macaque mothers to infant separation mental histories of laboratory rodents and the in an unfamiliar environment. Primates 35:191–202. characteristics of their rearing and living environ- Clarke AS, Boinski S. 1995. Temperament in nonhuman primates. American Journal of Primatology 36:1–23. ment. Further research is also needed to better Cleveland A, Westergaard GC, Trenkle MK, Higley JD. 2004. understand the effects of stress on maternal motiva- Physiological predictors of reproductive outcome and tion so that this variable can be effectively incorpo- mother-infant behaviors in captive rhesus macaque females rated in future models of mammalian parenting. (Macaca mulatta). Neuropsychopharmacology 29:901–910. Coe CL, Weiner SG, Rosenberg LT, Levine S. 1985. Endocrine and immune responses to separation and maternal loss in ACKNOWLEDGMENTS nonhuman primates. In: Reite M, Field T, editors. The psychobiology of attachment and separation. New York: I thank Dorothy Fragaszy, Wendy Saltzman, Academic Press. p 163–200. Karen Bales, and two anonymous reviewers for Cooke B, Higley JD, Shannon C, Lindell SG, Higley HM, helpful comments on this manuscript. My research Suomi SJ, Linnoila M. 1997. Rearing history and CSF described in this article adhered to the American oxytocin as predictors of maternal competency in rhesus macaques. American Journal of Primatology 42:102. of Primatologists’ principles for the ethical Deschamps S, Woodside B, Walker C-D. 2003. Pups’ presence treatment of nonhuman primates. eliminates the stress hyporesponsiveness of early lactating females to a psychological stress representing a threat to the REFERENCES pups. Journal of Neuroendocrinology 15:486–497. Digby LJ, Saltzman W. 2009. Balancing cooperation and Adamakos H, Ryan K, Ullman DG, Pascoe J, Diaz R, Chessare J. competition in callitrichine primates: examining the relative 1986. Maternal social support as a predictor of mother-child risk of infanticide across species. In: Davis LC, Ford SM, stress and stimulation. Child Abuse and Neglect 10: Porter LM, editors. The smallest anthropoids: the marmo- 463–470. set/Callimico radiation. Berlin: Springer. p 135–153. Altemus M, Deuster PA, Galliven E, Carter CS, Gold PW. Dix T. 1991. The affective organization of parenting: adaptive 1995. Suppression of hypothalamic-pituitary-adrenal axis and maladaptive processes. Psychological Bulletin 110: responses to stress in lactating women. Journal of Clinical 3–25. Endocrinology and 80:2954–2958. Fairbanks LA. 1990. Reciprocal benefits of allomothering for Altemus M, Redwine LS, Leong YM, Frye CA, Porges SW, female vervet monkeys. Animal Behaviour 40:553–562. Carter CS. 2001. Responses to laboratory psychosocial stress Fairbanks LA, Melega WP, McGuire MT. 1998. CSF 5-HIAA is in postpartum women. Psychosomatic Medicine 63:814–821. associated with individual differences in maternal protec- Bahr NI, Pryce CR, Dobeli M, Martin RD. 1998. Evidence from tiveness in vervet monkeys. American Journal of Prima- urinary cortisol that maternal behavior is related to stress tology 45:179–180. in gorillas. and Behavior 64:429–437. Fleming AS, Luebke C. 1981. Timidity prevents the virgin Bardi M, Shimizu K, Fujita S, Borgognini-Tarli S, Huffman MA. female rat from being a good mother: emotionality diffe- 2001. Hormonal correlates of maternal style in captive rences between nulliparous and parturient females. Physio- macaques (Macaca fuscata and M. mulatta). International logy and Behavior 27:863–868. Journal of Primatology 22:647–662. Fleming AS, Orpen G. 1986. Psychobiology of maternal Bardi M, Shimizu K, Barrett GM, Borgognini-Tarli S, behavior in rats, selected other species, and humans. In: Huffman MA. 2003a. Peripartum cortisol levels and Fogel A, Melson G, editors. Origins of nurturance. Hillsdale, mother-infant interactions in Japanese macaques. American NJ: Erlbaum. p 141–208. Journal of Physical Anthropology 120:298–304. Fleming AS, Steiner M, Anderson V. 1987. Hormonal and Bardi M, Shimizu K, Borgognini-Tarli S. 2003b. Mother-infant attitudinal correlates of maternal behavior during the early relationships and maternal estrogen metabolite changes in postpartum period in first-time mothers. Journal of Repro- macaques (Macaca fuscata, M. mulatta). Primates 44:91–98. duction and Infant Psychology 5:193–205.

Am. J. Primatol. 528 / Maestripieri

Fleming AS, Steiner M, Corter C. 1997. Cortisol, hedonics, and Lovejoy J, Wallen K. 1988. Sexually dimorphic behavior in maternal responsiveness in human mothers. Hormones and group-housed rhesus monkeys (Macaca mulatta) at 1 year of Behavior 32:85–98. age. Psychobiology 16:348–356. Gerald MS, Waitt C, Maestripieri D. 2006. An experimental Maestripieri D. 1992. Functional aspects of maternal aggres- examination of female responses to infant face coloration in sion in mammals. Canadian Journal of Zoology 70: rhesus macaques. Behavioural Processes 73:253–256. 1069–1077. Gibber JR. 1986. Infant-directed behavior of rhesus monkeys Maestripieri D. 1993a. Maternal anxiety in rhesus during their first pregnancy and parturition. Folia Prima- macaques (Macaca mulatta). I. Measurement of anxiety tologica 46:118–124. and identification of anxiety-eliciting situations. Gibber JR, Goy RW. 1985. Infant-directed behavior in young 95:19–31. rhesus monkeys: sex differences and effects of prenatal Maestripieri D. 1993b. Maternal anxiety in rhesus macaques androgens. American Journal of Primatology 8:225–237. (Macaca mulatta). II. Emotional bases of individual diffe- Graves FC, Wallen K, Maestripieri D. 2002. Opioids and rences in mothering style. Ethology 95:32–42. attachment in rhesus macaque abusive mothers. Behavioral Maestripieri D. 1993c. Infant kidnapping in group-living 116:489–493. rhesus macaques: Why don’t mothers rescue their infants? Green S. 1975. Variation of vocal pattern with social situation Primates 34:211–216. in the Japanese monkey (Macaca fuscata): a field study. Maestripieri D. 1994a. Social structure, infant handling, and In: Rosenblum LA, editor. Primate behavior: developments mothering styles in group-living Old World monkeys. in field and laboratory research, vol. 4. New York: Academic International Journal of Primatology 15:531–553. Press. p 1–102. Maestripieri D. 1994b. Mother-infant relationships in Hauser MD. 1993. The role of articulation in the production of three species of macaques (Macaca mulatta, M. nemestrina, rhesus monkeys, Macaca mulatta, vocalizations. Animal M. arctoides). I. Development of the mother-infant Behaviour 45:423–433. relationship in the first three months. Behaviour 131: Heinrichs M, Meinlschmidt G, Neumann I, Wagner S, 75–96. Kirschbaum C, Ehlert U, Hellhammer DH. 2001. Effects Maestripieri D. 1994c. Mother-infant relationships in three of suckling on hypothalamic-pituitary-adrenal axis responses to species of macaques (Macaca mulatta, M. nemestrina, psychosocial stress in postpartum lactating women. Journal M. arctoides). II. The social environment. Behaviour 131: of Clinical Endocrinology and Metabolism 86:4798–4804. 97–113. Herman RA, Measday MA, Wallen K. 2003. Sex differences in Maestripieri D. 1994d. Infant abuse associated with psychoso- interest in infants in juvenile rhesus monkeys: relationship cial stress in a group-living pigtail macaque (Macaca to prenatal . Hormones and Behavior 43:573–583. nemestrina) mother. American Journal of Primatology 32: Higley JD, Hopkins WD, Hirsch RM, Marra LM, Suomi SJ. 41–49. 1987. Preferences of female rhesus monkeys (Macaca Maestripieri D. 1995a. First steps in the macaque world: do mulatta) for infantile coloration. Developmental Psycho- rhesus mothers encourage their infants’ independent 20:7–18. locomotion? Animal Behaviour 49:1541–1549. Hinde RA, Atkinson S. 1970. Assessing the roles of social Maestripieri D. 1995b. Assessment of danger to themselves partners in maintaining mutual proximity, as exemplified and their infants by rhesus macaque (Macaca mulatta) by mother-infant relations in rhesus monkeys. Animal mothers. Journal of 109:416–420. Behaviour 18:169–176. Maestripieri D. 1995c. Maternal responsiveness to infant Hoffman CL, Ayala JE, Mas-Rivera A, Maestripieri D. 2010. distress calls in stumptail macaques. Folia Primatologica Effects of reproductive condition and dominance rank on 64:201–206. cortisol responsiveness to stress in free-ranging female Maestripieri D. 1996. Maternal encouragement of infant rhesus macaques. American Journal of Primatology locomotion in pigtail macaques (Macaca nemestrina). Ani- 72:559–565. mal Behaviour 51:603–610. Holman SD, Goy RW. 1995. Experiential and hormonal Maestripieri D. 1998. Parenting styles of abusive mothers in correlates of care-giving in rhesus macaques. In: Pryce CR, group-living rhesus macaques. Animal Behaviour 55:1–11. Martin RD, Skuse D, editors. Motherhood in human and Maestripieri D. 1999. The biology of human parenting: nonhuman primates: biosocial determinants. Basel: Karger. insights from nonhuman primates. Neuroscience and p 87–93. Biobehavioral Reviews 23:411–422. Insel TR, Winslow JT. 1998. Serotonin and neuropeptides in Maestripieri D. 2001. Is there mother-infant bonding in affiliative behaviors. Biological Psychiatry 44:207–219. primates? Developmental Review 21:93–120. Kalin NH, Shelton SE, Lynn DE. 1995. Opiate systems in Maestripieri D. 2004. Maternal behavior, infant handling, mother and infant primates coordinate intimate contact and socialization. In: Thierry B, Singh M, Kaumanns W, during reunion. Psychoneuroendcrinology 20:735–742. editors. Macaque : a model for the study of social Leveroni CL, Berenbaum SA. 1998. Early androgen effects on organization. Cambridge: Cambridge University Press. interest in infants: evidence from children with congenital p 231–234. adrenal hyperplasia. Developmental Neuropsychology 14: Maestripieri D. 2005a. Effects of early experience on female 321–340. behavioural and reproductive development in rhesus maca- Levine S, Wiener SG. 1988. Psychoendocrine aspects ques. Proceedings of the Royal Society of London B 272: of mother-infant relationships in nonhuman primates. 1243–1248. Psychoneuroendocrinology 13:143–154. Maestripieri D. 2005b. Early experience affects the interge- Levine S, Johnson DF, Gonzalez CA. 1985. Behavioral and nerational transmission of infant abuse in rhesus monkeys. hormonal responses to separation in infant rhesus monkeys Proceedings of the National Academy of Sciences USA 102: and mothers. Behavioral Neuroscience 99:399–410. 9726–9729. Lindell SG, Higley JD, Shannon C, Linnoila M. 1997. Low Maestripieri D. 2008. The role of the brain serotonergic levels of CSF 5-HIAA in female rhesus macaques predict system in the origin and transmission of adaptive and mother-infant interaction patterns and mother’s CSF maladaptive variations in maternal behavior in rhesus 5-HIAA correlates with infant’s CSF 5-HIAA. American macaques. In: Bridges R, editor. Neurobiology of the Journal of Primatology 42:129. parental brain. Amsterdam: Elsevier. p 163–173. Lonstein JS. 2007. Regulation of anxiety during the postpartum Maestripieri D, Call J. 1996. Mother-infant communication in period. Frontiers in Neuroendocrinology 28:115–141. primates. Advances in the Study of Behavior 25:613–642.

Am. J. Primatol. Emotions, Stress, and Maternal Motivation / 529

Maestripieri D, Carroll KA. 1998a. Child abuse and neglect: jacchus): development and application of an operant system usefulness of the animal data. Psychological Bulletin 123: with maternal reinforcement. Journal of Comparative 211–223. Psychology 107:99–115. Maestripieri D, Carroll KA. 1998b. Risk factors for infant Ramirez SM, Bardi M, French JA, Brent L. 2004. Hormonal abuse and neglect in group-living rhesus monkeys. Psycho- correlates of changes in interest in unrelated infants across logical Science 9:143–145. the peripartum period in female baboons (Papio hamadryas Maestripieri D, Carroll KA. 1998c. Behavioral and environ- anubis sp.). Hormones and Behavior 46:520–528. mental correlates of infant abuse in group-living pigtail Rowell TE, Hinde RA. 1962. Vocal communication by the rhesus macaques. Infant Behavior and Development 21:603–612. monkey (Macaca mulatta). Procedings of the Zoological Maestripieri D, D’Amato FR. 1991. Anxiety and maternal Society of London 138:279–294. aggression in house mice (Mus musculus): a look at Ruppenthal GC, Arling GL, Harlow HF, Sackett GP, Suomi SJ. interindividual variation. Journal of Comparative Psycho- 1976. A 10-year perspective of motherless-mother monkey logy 105:295–301. behavior. Journal of Abnormal Psychology 85:341–349. Maestripieri D, Roney JR. 2006. Evolutionary developmental Saltzman W, Abbott DH. 2009. Effects of elevated circulating psychology: contributions from comparative research with cortisol concentrations on maternal behavior in common nonhuman primates. Developmental Review 26:120–137. marmoset monkeys (Callithrix jacchus). Psychoneuroendo- Maestripieri D, Zehr JL. 1998. Maternal responsiveness crinology 34:1222–1234. increases during pregnancy and after estrogen treatment Saltzman W, Maestripieri D. 2011. The neuroendocrinology in macaques. Hormones and Behavior 34:223–230. of primate maternal behavior. Progress in Neuro-Psycho- Maestripieri D, Badiani A, Puglisi-Allegra S. 1991. Prepartal pharmacology and Biological Psychiatry, in press. chronic stress increases anxiety and decreases aggression Saltzman W, Liedl KJ, Salper OJ, Pick RR, Abbott DH. 2008. in lactating female mice. Behavioral Neuroscience 105: Post-conception reproductive competition in cooperatively 663–668. breeding common marmosets (Callithrix jacchus). Hormones Maestripieri D, Lindell SG, Ayala A, Gold PW, Higley JD. and Behavior 53:274–286. 2005. Neurobiological characteristics of rhesus macaque Schino G, Troisi A. 2005. Neonatal abandonment in Japanese abusive mothers and their relation to social and maternal macaques. American Journal of Physical Anthropology behavior. Neuroscience and Biobehavioral Reviews 29: 126:447–452. 51–57. Smotherman WP, Hunt LE, McGinnis LM, Levine S. 1979. Maestripieri D, Higley JD, Lindell SG, Newman TK, Mother-infant separation in group-living rhesus macaques: McCormack K, Sanchez MM. 2006. Early maternal rejection a hormonal analysis. Developmental Psychobiology 12: affects the development of monoaminergic systems and 211–217. adult abusive parenting in rhesus macaques. Behavioral Snowdon CT, Suomi SJ. 1982. Paternal behavior in primates. Neuroscience 120:1017–1024. In: Fitzgerald HE, Mullins JA, Gage P, editors. Child Maestripieri D, Lindell SG, Higley JD. 2007. Intergenerational nurturance: psychobiological perspectives. New York: Plenum transmission of maternal behavior in rhesus monkeys and Press. p 63–108. its underlying mechanisms. Developmental Psychobiology Suomi SJ. 1978. Maternal behavior by socially incompetent 49:165–171. monkeys: neglect and abuse of offspring. Journal of Maestripieri D, Schino G, Aureli F, Troisi A. 1992. A modest Pediatric Psychology 3:28–34. proposal: displacement activities as an indicator of emotions Troisi A, D’Amato FR. 1984. Ambivalence in monkey in primates. Animal Behaviour 44:967–979. mothering: infant abuse combined with maternal posses- Martel FL, Nevison CM, Rayment FD, Simpson MJA, siveness. Journal of Nervous and Mental Disease 172: Keverne EB. 1993. Opioid receptor blockade reduces 105–108. maternal affect and in rhesus monkeys. Troisi A, D’Amato FR. 1991. Anxiety in the pathogenesis of Psychoneuroendocrinology 18:307–321. primate infant abuse: a pharmacological study. Psychophar- Numan M, Insel TR. 2003. The neurobiology of parental macology 103:571–572. behavior. New York: Springer. Troisi A, Schino G, D’Antoni M, Pandolfi N, Aureli F, Pryce CR. 1992. A comparative systems model of the D’Amato FR. 1991. Scratching as a behavioral index of regulation of maternal motivation in mammals. Animal anxiety in macaque mothers. Behavioral and Neural Biology Behaviour 43:417–441. 56:307–313. Pryce CR. 1996. Socialization, hormones, and the regulation of Tu MT, Lupien SJ, Walker C-D. 2006. Measuring stress in maternal behavior in nonhuman primates. Advances in the postpartum mothers: perspectives from studies in human Study of Behavior 25:423–473. and animal populations. Stress 8:19–34. Pryce CR, Abbott DH, Hodges JH, Martin RD. 1988. Maternal Whitham JC, Gerald MS, Maestripieri D. 2007. Intended behavior is related to prepartum urinary estradiol levels in receivers and functional significance of grunt and girney red-bellied tamarin monkeys. Physiology and Behavior vocalizations in free-ranging female rhesus macaques. 44:717–726. Ethology 113:862–874. Pryce CR, Dobeli M, Martin RD. 1993. Effects of sex steroids Wolfe DA. 1987. Child abuse. Implications for child development on maternal motivation in the common marmoset (Callithrix and psychopathology. Newbury Park: Sage Publications.

Am. J. Primatol.