The signal function of early infant crying

Abstract: In this article I evaluate recent attempts to illuminate the human infant cry from an evolutionary perspective. Infants are born into an uncertain parenting environment, which can range from indulgent care of offspring to infanticide. Infant cries are in large part adaptations that maintain proximity to and elicit care from caregivers. There is not strong evidence for acoustically distinct cry types, however, but infant cries may function as a graded signal. During pain-induced autonomic nervous system arousal, for example, neural input to the vocal cords increases cry pitch. Caregivers may use this acoustic information, together with other cues, to guide care-giving behavior. Serious pathology, on the other hand, results in chronically and severely abnormal cry acoustics. Such abnormal crying may be a proximate cause of adaptive infant maltreatment, in circumstances in which parents cut their losses and reduce or withdraw investment from infants with low survival chances.

An increase in the amount of crying during the first few months of life is a human universal, and excessive crying, or colic, represents the upper end of this normal increase. Potential signal functions of excessive crying include manipulation of parents to acquire additional resources, honest signaling of need, and honest signaling of vigor. Current evidence does not strongly support any of these hypotheses, but the evidence is most consistent with the hypothesis that excessive early infant crying is a signal of vigor that evolved to reduce the risk of a reduction or withdrawal of parental care.

Key Words: colic, crying, early infant crying, honest signaling, infanticide, parental care, parent-offspring conflict, separation call, vocalization.

The baby gazed up at me as ever with wide-open eyes, but whether he was hungry or thirsty or felt some other discomfort I couldn’t tell. He lay with eyes open and expressionless, like a marine plant in the water of the dusk, simply and placidly existing. He demanded nothing, expressed absolutely no emotion. He didn’t even cry.

- Father considering his handicapped son in The Silent Cry, by Kenzaburo Oe.

I. Introduction

Interpretations of early infant crying are paradoxical. For nearly every claim made about the human infant’s cry, the opposite has also been claimed. The infant cry has been characterized as a constellation of acoustically distinct cry types, indicating specific needs such as hunger or sleep, and it has been described as undifferentiated noise. Infant cries can elicit appropriate contact and care-giving behaviors, and they are sometimes the proximate cause of abuse. Infants who cry excessively (i.e. those with colic) have been described as sicker, healthier, and as in no other way different than their counterparts without colic.

In this paper, I critically examine the various proposed signal functions of early infant crying and develop models of infant crying that attempt to resolve these paradoxes. In mapping out the potential signal functions of the early infant cry, I take advantage of a number of theoretical perspectives, all derived from evolutionary reasoning, including attachment theory, parent-offspring conflict theory, and honest signaling. In characterizing the infant cry and examining its potential signal functions, I bring together diverse literatures, including clinical studies of the infant cry, physiological models of cry production, cry bio-acoustics and perception, the child abuse and infanticide literatures, animal studies, and cross-cultural ethnography.

I employ an evolutionary approach to bring together the disparate literatures into a coherent picture of the signal function of the early infant cry. The application of evolutionary theory to human infant crying has increased in recent years (Furlow 1997; Lummaa et al. 1998; Barr 1998a; Barr 1999; Zeifman 2001). Here I consider the breadth of the proposed signal functions, and critically evaluate their ability to explain the amount, the temporal patterning, and the acoustic properties of early infant crying. The scope of this paper is limited in the following respect, however. I examine the signal functions of the infant cry only in the first three months of life (i.e. ‘early infant crying’). At about 3 months of age, infants undergo a developmental shift, in which crying becomes more differentiated, interactive, and intentional (Ostwald & Murry 1985; Barr 1990; Barr and Gunnar 2000).

The theoretical perspective of the paper is introduced in the following section, in which I argue that human parents pursue a complex reproductive strategy, which can range from selfless care of offspring to infanticide. I consider the implications that such a child-rearing environment has for infant signaling, in particular the evolution of honest signaling. The examination of the early infant cry proceeds in four additional sections. The first briefly addresses the infant cry as a separation call fostering protection and care by the mother. Second, I show how particular acoustic characteristics of cries indicate level of current distress and overall health, and develop a model describing how these acoustic properties may influence patterns of infant care and abuse. In the next two sections, I address excessive infant crying (i.e. colic). Following Ronald Barr, I argue that colic is a normal aspect of infant development, albeit in exaggerated form, and in the final section evaluate the signal functions that have been proposed to explain such excessive crying. In the conclusions, I point the way for further research.

Before introducing the relevant evolutionary theory, two important clarifications are in order. First, the evolutionary approach employed here assumes that the behavior expressed by infants and parents has been shaped, in part, by natural selection, and that individual behavioral responses to their environment may be viewed as potential adaptations that tend to increase survival or reproduction (Williams 1966; Barkow et al. 1992). In taking this view, I do not imply that humans or other animals are making conscious fitness calculations or that they are aware of the fitness effects of their behavior. In order to avoid lengthy and awkward sentences, however, I will employ expressions such as “When the chances for infant survival are low, parents may increase overall reproductive output by terminating parental investment in offspring (i.e. engage in ‘adaptive infanticide’)”, which should be taken as shorthand for “Natural selection has favored in parents a psychological mechanism that results in the withdrawal of care from their own offspring, this mechanism is activated under conditions in which the infant is unlikely to survive, and selection has favored this flexibility in parental behavior because in past environments the selective withdrawal of care under inauspicious circumstances resulted, on average, in the production of more offspring over a lifetime.” Second, the moral and legal views on abortion and infanticide vary across time, cultures and individuals. I will advance adaptive explanations for behaviors such as abortion and infanticide in humans, but in doing so I imply neither moral justification nor moral condemnation.

II. Parent-offspring conflict and honest signaling

II.a. The conflict

The most commonly proposed signal function of the infant cry is to alert caregivers of need. This straightforward view of early infant crying presumes that the interests of the offspring and the parent are in agreement. When the infant is hungry, not only does she desire to be fed, but the mother also desires to feed her. But the interests of parents and offspring are not always identical. Parent-offspring conflict theory (Trivers 1974) recognizes that the interests of parents and child are indeed largely overlapping, but it emphasizes where the interests of parents and offspring diverge, and this has important implications regarding the signal function of early infant crying.

The source of parent-offspring conflict is the fact that in sexually reproducing species parents and offspring are not genetically identical, and as a result their genetic interests will be different in important ways (Trivers 1974; Daly 1990). The classic example of this conflict is a mother who maximizes her fitness (i.e. number of surviving offspring) by investing her limited resources equally among her offspring, mirroring her equal genetic relatedness to them. From the genetic point of view of each offspring, however, the self is more important than siblings, and each particular offspring seeks to acquire a disproportionate share of parental resources. Genes expressed in the parent, according to the theory, result in behaviors that limit investment in individual offspring, whereas genes expressed in the offspring result in behaviors that seek to increase such investment.

The above hypothetical example does not imply that natural selection will invariably favor parents to partition resources equally in all offspring. Parents may favor young over old, weaker over stronger, or visa versa, depending on circumstances (Daly 1990; Davis & Todd 1999; Hrdy 1999). Additionally, there is a limit to an offspring’s evolved selfishness because it also has a genetic interest in its siblings (Daly, 1990). Nonetheless, the genetic interests of parents and offspring do not overlap entirely.

Parent and offspring interests diverge most dramatically when parents of a species are prone to withdraw parental investment entirely, or to commit infanticide outright (Daly & Wilson 1988; Daly 1990; Hrdy 1999). When chances for survival of an infant are low, either due to the poor condition of the child or to detrimental circumstances, parents might maximize overall reproductive output by quickly terminating investment in the current offspring. By doing so, they can divert limited resources to other offspring whose expected future fitness is higher than that of the unpromising neonate. For example, parents could divert resources to an existing, older offspring who has already successfully passed through the period of early infant mortality (Daly 1990). Alternatively, parental investment could be diverted to future offspring, when circumstances are more auspicious. By discriminating among offspring in this way, limited parental resources can be targeted towards those offspring most likely to survive and reproduce themselves, resulting in higher lifetime net reproductive output.

Parental neglect, abandonment, killing and sometimes consuming of newborn offspring is known to occur in several animal species when the condition of the offspring is poor, or the conditions for rearing offspring are unfavorable. Shield bug mothers, for example, place their larger, more viable eggs centrally where they can protect them from predators, but place the smaller, less viable eggs in the periphery where they suffer greater predation (Mappes et al. 1997). Such parental discrimination is also found in several monogamous bird species. For example, white stork parents with large clutches kill their smallest, slowest growing chicks (Tortosa & Redondo 1992), and Hermaan’s gull parents reduce large clutches by selectively killing later hatched chicks (Urrutia & Drummond 1990). Among mammals, mice kill and consume offspring when litter sizes are too large or when parental condition is poor (Elwood 1992). Prairie dog mothers in poor physical condition sometimes abandon offspring, make no effort to prevent conspecifics from killing and consuming them, and sometimes participate in the cannibalism themselves (Hoogland 1994; Hrdy 1999).

II.b. Parent-offspring conflict in humans: infanticide, abuse, neglect, and sub-optimal parenting.

The quality of parental care in humans is multiply determined by a variety of influences, including characteristics of parents, characteristics of offspring, and the economic, social and cultural environment (Belsky 1984, 1993; Burgess & Drais 1999). As predicted by parent-offspring conflict theory, human parents often provide sub-optimal care, or reduce or withdraw parental investment from offspring, when the prospects for infant survival are low. Such parental investment manipulations may take many forms, such as transferring the costs of child rearing to others by enlisting the support of kin or wet nurses, or fostering out or abandoning offspring (Hrdy 1994). Alternatively, parents may provide sub-optimal care or terminate investment in offspring without transfer of costs. Such maltreatment may be viewed along a continuum, ranging from sub-optimal parenting, to neglect, abuse, and infanticide. This is not to say that sub-optimal parenting, neglect, abuse and infanticide have identical etiologies (e.g. Chaffin et al. 1996), but the following will show that there is substantial overlap in the risk factors for these phenomena.

Cross-cultural evidence demonstrates that parent-inflicted infanticide in humans occurs overwhelmingly when the child is unlikely to survive (Bugos 1984; Scrimshaw 1984; Daly & Wilson 1984; Daly & Wilson 1988; Daly 1990; Lummaa et al. 1998; Adinkraw 2000; Table 1)¹. The infant’s lack of survival prospects can be due to the poor health of the child itself or, more commonly, to unfavorable circumstances external to the child. The most common rationales for infanticide include the birth of a deformed or ill infant, a lack of paternal support, and inauspicious circumstances such as too many children. Other rationales for infanticide include unwanted sex and ritualistic killings. Results from studies of industrialized societies are similar to the cross-cultural ethnographic sample (Daly & Wilson 1988; Pitt & Bale 1995; Overpeck et al. 1998; Voland & Stephan 2000; Stroud & Pritchard 2001). For example, in a Canadian sample, younger, unmarried women were most likely to kill their own offspring, and infants less than 1 yr of age were at highest risk (Daly & Wilson 1988). Similar risk factors have been found in the United States (Overpeck et al. 1998). The poor condition of an infant is a less likely documented motivation for infanticide in industrialized societies, however, perhaps due to modern medical intervention. But this is not so for abuse and neglect (see below). Additionally, in modern societies, retaliation against a spouse and mental illness of the parent are commonly cited motivations for infanticide. When retaliation or mental illness are contributing factors to infanticide, however, the killing of offspring is more evenly distributed across ages, not selectively targeted toward the young and vulnerable, as in the most likely cases of adaptive infanticide (Dobson and Sales, 2000; Stroud & Pritchard 2001). When mental illness contributes to infant maltreatment, this does not necessarily mean that the behavior is not adaptive, however. For example, the risk factors for postpartum depression are much the same as other risk factors for neglect, abuse and infanticide (e.g. in circumstances inauspicious for successful child-rearing), and may be a psychological mechanism activated during such circumstances that leads to the adaptive withdrawal of parental investment (Hagen 1999; 2002).

Cultural contexts constrain the incidence of infanticide. Rates of infanticide by parents are lower in modern societies than in traditional societies, and rates vary substantially among the latter. For example, of nearly 35 million births in the United States from 1983-1991, the infant death rate from infanticide was only 8 in 100,000 (Overpeck et al. 1998). It should be borne in mind, however, that cases of infant death due to abuse and neglect are undercounted (McClain et al. 1993; Overpeck et al. 1998). In traditional societies, infanticide rates range from zero (or near zero) in many traditional African cultures to a high among the Eipo horticulturists of Papua New Guinea, in which 41% (20/49) of live-born infants were killed by parents from 1974 to 1978, in part driven by parental preferences for males (Hrdy 1994). After missionary contact with the Eipo, the infanticide rate was reduced to 10%. These striking differences in infanticide rates across cultural contexts can be explained, in part, by the reversal of the relative costs of abortion and infanticide (Hrdy 1999). In traditional societies, abortion is costly due to the risk to the mother’s health, while infanticide is often socially sanctioned. In modern societies, abortions are more often sanctioned and less dangerous, while infanticide is routinely outlawed. Abortion has replaced infanticide as a means of selectively eliminating offspring. Indeed, those circumstances that often trigger infanticide also influence the decision to abort, such as poverty, lack of paternal support, and fetal abnormalities (e.g. Lycett and Dunbar 1996; Larsson et al. 2002; Zlotogora 2002).

Child abuse and neglect occur under circumstances similar to those associated with abortion and infanticide (Belsky, 1993). Although child abuse and neglect occur at all child ages, there is a peak in the rate and degree of violence for victims under 1 year of age (Burgess & Drais 1999; Sibert et al. 2002). As in the case of infanticide, common risk factors for abuse and neglect include the poor condition of the infant or child, a lack of spousal or other social support, poverty and unemployment, and large family size (e.g. Frodi 1981; Daly & Wilson 1981, 1988; Belsky 1993; Chan 1994; Vinson et al. 1996; Albarracin et al. 1997; Becker et al. 1998; Kotch et al. 1999; Gonzalvo 2002; Sidebotham et al. 2002). In particular, it has been shown repeatedly that mentally, physically, or behaviorally atypical children are more likely to be neglected and abused. Retarded children, for example, are at exceptionally high risk for abuse, although it could be argued that the retardation is a consequence, rather than a cause, of the abusive behavior (Frodi 1981; Daly & Wilson, 1988). On the other hand, abuse cannot be the cause of premature birth or congenital abnormalities such as spina bifida, cystic fibrosis and Down’s syndrome. Children with these congenital defects are abused anywhere from 2 to 10 times more often than the population at large, and premature infants are abused twice as often (Daly & Wilson 1981; Frodi 1981). As in the case of infanticide, factors unrelated to child survival are also implicated in abuse and neglect, such as a history of abuse, mental illness, substance abuse and the cultural values of perpetrators (e.g. Belsky 1993; Chaffin et al. 1996; Kotch et al. 1999; Klevins et al. 2000; Ferrari 2002). Sub-optimal parenting (e.g. lower levels of affectionate or nurturant contact), while not considered neglectful or abusive, is also related to similar risk factors such as a sick infant, poverty, a lack of social support, substance abuse and mental illness (e.g. Crnic et al. 1984; Mann 1992; Hashima & Amato 1994; Black et al. 1994; Black & Krishnakumar 1999; Newcomb & Loeb 1999; Britton et al. 2001; LeCuyer-Maus, 2003).

When considering the causes of sub-optimal parenting, infant neglect, abuse and infanticide, it must be emphasized that maltreatment is not likely to be caused by the effects of single variables, but through the interaction of multiple risk factors at various biological and social levels (Wilson-Oyelaran 1989; Belsky 1993). As such, I am not arguing that infant poor health invariably leads to some form of infant maltreatment in humans, or that this is the most important factor contributing to maltreatment. Nor am I arguing that all cases of infant maltreatment represent the adaptive withdrawal of parental investment as a part of an evolved, fitness-enhancing strategy (i.e. that all behavior is adaptive). The evidence clearly demonstrates, however, that the selective reduction or withdrawal of parental care under unfavorable child-rearing circumstances has been a major aspect of parental behavior throughout human history, and that the poor condition of the infant can be one of many factors contributing to the reduction or withdrawal of parental care. The discriminative allocation of parental resources to offspring over evolutionary time represents a major selective pressure on the physiology and behavior of infants, and may explain some characteristics of the infant’s cry.

II.c. Signaling resolutions to parent-offspring conflict

When the goals of parents and infants are divergent in ways described above, how are these conflicts resolved? The simplest case occurs when parents have complete control over the distribution of resources and they can accurately observe the condition of their offspring. In such cases, parents can partition investment selectively among offspring as they choose. When parents cannot directly observe the condition of infants, or some component of their condition, however, then signals from infants to parent may evolve.

The basic problem of communication when interests between parties conflict is honesty. Manipulation was emphasized in early theoretical models of parent-offspring conflict (reviewed by Godfray 1995a; Godfray & Johnstone 2000). In these models, parents monitor the needs of their young but are subject to manipulation because the young exaggerate their needs. These early models assumed that parents and offspring have fixed responses. Parents provide more food to offspring when offspring signal and offspring signal less when fed. Different models incorporated different types of manipulation, such as exaggerated begging regardless of need, or ‘blackmailing’ parents by reducing one’s own fitness until provided with extra food. The typical resolution of these models is intermediate between parent and offspring optima. That is, parents evolve to provide more food to offspring than would yield their highest possible fitness, but each offspring still receives less provisioning than would yield its highest possible fitness. It is unclear if such manipulative signaling would be sustainable, however, if the models did not assume that parents always yield to offspring begging, but allowed the parental response itself to evolve (Godfray & Johnstone 2000).

Later models have examined the conditions under which ‘honest signaling’ can evolve (i.e. behavioral or morphological signals that accurately reflect otherwise unobservable characteristics of the signaler). The first type of honest signaling concerns advertising need for resources (Maynard Smith 1991a; Godfray 1991; Godfray 1995a,b). Consider a situation in which offspring differ in need, but that the parent cannot determine that need. Is it possible for a system of honest signaling to evolve, such that only those in need will advertise to parents, but that those without need will not? The critical element required for such an honest-signaling system is that there be a fitness cost associated with producing the signal. To see why, imagine two types of offspring, those who signal only when very hungry and those who begin signaling at extremely low levels of hunger. Further imagine that conspicuously advertising need to parents has the side effect of increasing the probability of drawing the attention of predators. If producing the signal imposes such costs, then the offspring who signal when relatively sated may not gain an overall net benefit by signaling. That is, the marginal cost of attracting predators may be greater than the marginal benefit of receiving food. For offspring who only signal when truly hungry, on the other hand, the marginal cost of attracting a predator may be lower than the marginal benefit of receiving food. When infants are not truly needy, they stand to gain little from an additional food item, compared to when they are hungry, and natural selection may not favor costly signaling to obtain it. In short, honest signaling of need can evolve when parents cannot reliably detect need directly, producing the signal is costly, and the parental resource is most valuable to the needy.

Animal studies, in particular avian studies, have yielded evidence for communication systems in which offspring honestly signal need to parents. Offspring begging vocalizations increase with need, parents preferentially feed those who beg, and such begging has been shown to be costly (reviewed in Kilner & Johnstone 1997). Recent scholarship has called these data into question, however, because sibling competition, in which offspring rather than parents control the allocation of parental resources, can result in similar findings (reviewed in Royle et al. 2002). There is evidence for honest signaling of need in species that produce one offspring at a time, however, such that the confounding effects of sibling competition are removed. Wilson’s storm-petrels produce single-chick broods, chicks vocalize more when they are in need (low body mass), but only in the presence of their parents, and parents respond by provisioning them with more food (Quillfeldt 2002).

The second type of honest signaling concerns advertising vigorous condition to parents. Consider the situation in which parents are prone to cut their losses and abandon infants when their condition is so poor that they are unlikely to survive, but that parents cannot observe all aspects of infant condition directly. Under these circumstances, it would behoove high quality offspring (with high chances of survival) to advertise their superior status to parents and avoid infanticide. But what can inhibit low quality infants (with low chances of survival) from producing the same signal? Again, honest signaling can evolve if there are costs to signaling (Zahavi 1975, 1977; Grafen 1990a,b; Maynard Smith 1991b; Iwasa et al. 1991). Specifically, the cost of producing the signal must be higher for low quality infants than it is for high quality infants. If conspicuous signaling is costly, in terms of energetic expense, for example, then high quality infants would be better able to afford the energetic expense required to signal, leaving their low quality rivals signaling less, or less vigorously. In this case, parents could use vigor of the signal as a reliable cue to non-observable quality of the infant².

Most of the empirical evidence for honest signaling of quality involves female mate choice (see Andersson 1994). In a variety of species, it has been shown that 1) males possess conspicuous morphological or behavioral signals (e.g. long colorful tails), 2) the signals are costly (e.g. increased predation), 3) the most conspicuous signals are produced by the highest quality males (e.g. those with a genetically based increased resistance to parasites), 4) females preferentially mate with males possessing the most conspicuous signals, and 5) the resulting offspring inherit high quality from the sires. More important for the argument here, there are several animal examples of offspring advertising high quality to parents that results in increased parental investment. In barn swallows, for example, nestling gape color ranges from yellow to red, based on the ingestion of carotenoids. Carotenoids also play a central role in immunostimulation, so that immune-challenged nestlings cannot allocate as much carotenoid to pigmentation. Thus, healthier nestlings can produce brighter, redder gapes, and parents preferentially allocate food to those nestlings with the most brightly colored gapes (Saino et al. 2000). Additionally, American coot chicks possess exaggerated plumage at hatching, parents preferentially feed chicks with the most exaggerated plumage, and neglected chicks with less exaggerated plumage experience lower growth and higher mortality (Lyon et al. 1994).

III. The separation call and mother-infant attachment

The most straightforward signal function of early infant crying is to maintain contact with the caregiver. John Bowlby (1969) proposed a suite of infant attachment behaviors, including infant crying, that serve to establish or maintain proximity to potential caregivers, primarily the mother. While recognizing other important signal functions of crying, such as communicating hunger or discomfort, Bowlby emphasized close physical attachment to mother itself as an instinctive goal of infants independent of other needs. Attachment theory is relevant to early infant crying (0-3 months) only insofar as it functions to maintain proximity to the mother. It is only later in development that attachment theory predicts patterns of maternal response will influence the child’s developing personality and the formation of a mental model of relationships (Bowlby 1969; Ainsworth 1969; Bell & Ainsworth 1972)³. Nevertheless, early infant crying is likely an important means by which infants can maintain contact with the mother because active proximity maintenance behaviors, such as following, are not possible.

Infant cries possess features and produce effects consistent this view. Several studies show that human mothers can recognize the cries of their own infants (Formby 1967; Wiesenfeld et al. 1981; Green and Gustafson 1983), and magnetic resonance imaging of mothers showed that the infant cry stimulus results in brain activity in areas hypothesized to be involved in mammalian parenting behavior (Lorberbaum et al. 2002). Moreover, Bell and Ainsworth (1972) showed that infants (0-4 months) cried more when out of proximity from their mothers, that crying promoted the re-establishment of contact by the mother, and that such contact was effective at terminating the crying. Similarly, Christensson et al. (1995) showed that newborns reliably cried when separated from mothers and stopped crying at reunion. Additionally, cross-cultural evidence shows that indulgent maternal styles are associated with less infant crying compared to maternal styles that do not allow as much physical proximity between infants and caregivers (e.g. Barr et al. 1991; Lee 2000). Infants do not always cease crying when caregiver contact is established, however, as in cases of inconsolable crying (i.e. colic) described later in the paper.

Bowlby speculated that such a signaling system was adaptive during human evolutionary history because maintaining proximity to mothers protected infants from predators. Indeed, the infant cry as a means to maintain proximity with the mother may have a deep evolutionary history. As in human infants, physical separation from the mother also evokes separation calls in a variety of mammalian infants, and the acoustic structure of the human infant cry is similar to that of the separation calls of non-human primate infants (Newman & Symmes, 1982; Newman 1985; Hofer 1996; Panksepp 1998). Moreover, human mothers and infants exhibit a suite of adaptations that are typical of mammalian species that carry their infants, compared to species that tend to cache infants for long periods of time (Blurton-Jones 1972; Zeifman 2001). In caching species, infants are placed in nests or burrows while the mother forages, and feedings are infrequent. In these species, mother’s milk is high in fat and protein, infants have independent thermo-regulatory mechanisms, and infants do not vocalize when separated. In carrying species, including humans, on the other hand, mothers and infants are in more continuous contact and feedings are more frequent. In these species, milk is lower in fat and protein content, independent thermo-regulation is poorly developed at birth, and infants do vocalize when separated from mothers.

IV. The acoustic characteristics of early infant crying: current distress and overall health

Here I review the causes and consequences of the acoustic properties of early infant cries. In doing so, I reconcile several opposing themes in the infant cry literature. With regard to infant cry causation, the cry has been viewed variously as a constellation of cry types, as a graded signal that reveals current level of transient distress, and as an invariant signal associated with chronic illness. With regard to consequent parental reactions, the infant cry has been viewed alternatively as a positive stimulus that triggers altruistic action and as an aversive stimulus that triggers abuse. First, I review models of infant cry production, the acoustic properties of the cries of healthy, stressed and unhealthy infants, and the effects of cries on adult listeners. In the final sections, I develop a model of infant cry acoustics and briefly consider the evolutionary history of mother-infant vocal communication systems.

IV.a. Infant cry production and acoustics

Three component systems, the respiratory system (lungs and trachea), the vocal cords (larynx), and the vocal tract (pharynx, oral and nasal cavities) produce the infant cry (Denes & Pinson 1963; Golub & Corwin 1985; Green et al. 2000). The source of the cry sound is the vibrating vocal cords of the larynx. Variable air pressure from the lungs and alterations in the tension and length of the vocal cords combine to produce vocal cord frequency. The cry is then filtered as it proceeds through the vocal tract and the lips, resulting in the audible cry.

The nervous system innervates the muscles that control the respiratory system, the vocal cords, and the vocal tract. Most models of infant cry production emphasize the role of the 10^th cranial nerve of the parasympathetic nervous system, or the vagus, as the most proximate neural input affecting vocalization acoustics, although other neural inputs are also important (Green et al. 2000). Sympathetic and parasympathetic (vagal) inputs from the autonomic nervous system innervate the laryngeal muscles (Porter et al. 1988). The vagus has an inhibitory effect on the contraction of the laryngeal muscles. A decrease in vagal outflow causes an increase in the fundamental frequency of the infant cry, while an increase in vagal outflow decreases the fundamental frequency.

The nucleus ambiguus of the medulla provides the primary vagal input to various organs including the bronchi, esophagus, pharynx, larynx and heart, regulating and coordinating sucking, breathing, swallowing, vocalizing and heart rate (Porges 1995; 1997). Acute stresses precipitate a parasympathetic withdrawal of vagal output from the nucleus ambiguus, resulting in a cascade of physiological events including an increase in heart rate and an increase in the pitch of vocalizations (Porges 1995). For example, pain pathways distribute information to the autonomic nervous system via the hypothalamus and medulla (Gauriau & Bernard 2002), resulting in decreased vagal tone and increased cry pitch in infants (Porter et al. 1988). Also, structures of the limbic system, in particular the cingulate gyrus and amygdala, communicate with the nucleus ambiguus via the periaqueductal grey area (Larson et al. 1988; Vogt & Barbas 1988; Porges 1995). Since the amygdala is central to the brain circuitry implicated in negative emotions such as fear and anxiety (LeDoux 1996; Panksepp 1998; Mendoza & Ruy 2001), the branch of the vagus originating in the nucleus ambiguus can mediate the rapid expression of emotional state in the infant cry. In addition to mediating acute stress, vagal tone also responds to chronic stress, which results in persistent vagal tone depression, even while sleeping, and has chronic effect on cry acoustics (Porter et al. 1988; Porges 1995).

The current study of infant cry bio-acoustics is problematic. Acoustic analyses of the infant cry emphasize measures of the fundamental frequency, in part because cry production models emphasize the role of the vagus and its relationship to fundamental frequency. Other cranial nerves that innervate the larynx, pharynx, chest and neck also may influence cry acoustics, however, and acoustic properties in addition to those most well-studied, as well as non-cry sounds such as fussing, may also be important (Lester 1987; Gustafson & Green 1989; Green et al. 1995; Green et al. 2000). In addition, acoustic measures are not standardized (Green et al. 2000). Early studies analyzed sound spectrograms, yielding measures such as duration of a single cry, minimum and maximum fundamental frequency, and qualitative changes in frequency contour across the cry (e.g. rising or falling frequency). Computer-based methods, on the other hand, quantify acoustic information of digitized cries, yielding somewhat different measures such as cry duration, mean fundamental frequency across a single cry, harmonic frequencies, and the percentage of time cries are phonated, dysphonated or hyperphonated. Furthermore, computer-based methods differ from one another (Green et al. 1995). Despite the focus on a subset of possible acoustic properties, and the lack of standardization, some consistent findings do emerge with regard to the acoustic properties of cries of healthy, temporarily stressed, and unhealthy infants.

The cries of healthy, non-stressed infants have the following acoustic characteristics (see Appendix A for a glossary of acoustic terms). The average duration of a single cry is about 0.5 – 1.5 seconds, the fundamental frequency of a cry falls within the 200–600 Hz range, and the melody is either falling or rising/falling, as opposed to rising, falling/rising, or a flat melody (Table 2). Fundamental frequency is generally, but not necessarily, perceived as pitch. Perceived pitch will not be related to fundamental frequency, for example, when the fundamental is attenuated while higher harmonics are accentuated during filtering in the supra-laryngeal vocal tract, but for simplicity the two terms will be used interchangeably here. In addition, biphonation, glides, furcations, and noise concentrations are rare or absent in normal cries, while glottal rolls are common (Wasz-Hockert et al. 1985; Newman 1985; Furlow 1997).

IV.b. The infant cry as a graded signal of distress

Infant crying is clearly a means by which infants can communicate needs (e.g. hunger, pain, or discomfort) to caregivers, who may be alerted to appropriately satisfy those needs (e.g. by feeding, protecting or soothing). It is a matter of some controversy, however, as to whether there are acoustically distinct cry types (e.g. ‘hunger’ and ‘pain’ cries) to which caregivers can respond specifically without additional contextual cues (reviewed by Gustafson et al. 2000).

Work in the 1960’s by the so-called ‘Scandinavian cry group’ is often cited in support of the "cry types" hypothesis (see Wasz-Hockert et al. 1985; Gustafson et al. 2000). The researchers recorded birth cries, pain cries (during vaccination), hunger cries (4 hrs after feeding), and pleasure cries (after feeding). Listener subjects identified the four cry types better than chance (55% correct versus 25% expected). Gustafson et al. (2000) criticize this work, however, because the best exemplars of each cry type were pre-selected by researchers, and listeners were given the four a priori categories in advance, both of which increase the likelihood of accurate assignment. Additionally, the results were collapsed across all 4 cry types, so that the positive effect could have been due to only the most easily distinguishable cries, such as the contented coos and babbles that constitute the "pleasure cry". In a replication of earlier work, however, exemplars for each cry type were chosen at random, and results for the four cry types were presented separately (Wasz-Hockert et al. 1968). Additionally, they showed that the four cry types differed statistically along several acoustic dimensions, such as the fundamental frequency and melody, although the differences were quantitative rather than qualitative. The accuracy for identifying birth, pain, hunger and pleasure cries was 48%, 63%, 68% and 85%, respectively (grand mean=66%).

Other studies also show that subjects can distinguish between cry types. Wiesenfeld et al. (1981) showed that mothers could identify pain cries (rubber band snap) versus anger cries (taking away pacifier or physical restraint) of their own infants better than chance when given three categories from which to choose (pain, anger or "other"; 66% correct versus 33% expected). Gustafson and Harris (1990) also showed that mothers could identify pain versus hunger cries better than chance when given six cry categories from which to choose (44% correct versus 17% expected).

When subjects are given open-choice tests, however, the evidence for cry types is not as strong. In an early study by Sherman (1927), cries were elicited from babies by late feeding (hunger cry) and by dropping, restraining or pricking with a pin (pain cries). Non-mother subjects behind a screen were asked to judge the “emotional characteristics” of the cries, but there was no agreement among listeners, leading to the conclusion that infant cries were merely “undifferentiated noise” (Gustafson et al., 2000). More recently, Muller et al. (1974) played cries elicited by hunger (pulling the nipple away), pain (rubber band snap), and startling (clap of wooden blocks close to the ears). Again, without a priori categories, subjects could not differentiate the cry types, even of their own children. Participants tended to attribute all cries to hunger. An alternative to the "cry types" model is the view of the infant cry as a graded signal (Murray 1979; Gustafson et al. 2000). According to this view, vocalizations vary quantitatively on some acoustic dimension, such as duration or frequency, and that graded change along the dimension reflects motivational or emotional state.

There is evidence that the acoustic characteristics of the infant cry function as a graded signal of physical pain (Craig et al. 2000). For example, Porter et al. (1986) showed that during the steps of circumcision, which become increasingly invasive and presumably more painful, individual infant cries became shorter and more frequent, the maximum fundamental frequency increased and became more variable, and harmonics became less distinct. The pre-operative maximum fundamental frequency of infant cries was well within the normal range (441 Hz), rose moderately during restraint and preparation (517-640 Hz), and reached the highest levels during the most invasive procedures (707-730 Hz). Post-operatively, the fundamental frequency of infant cries returned to baseline (469 Hz). Similarly, Johnston and Strada (1986) examined infant cries for 1 minute after a painful immunization. The first 1 or 2 cries were relatively long and high pitched (mean 627 Hz), followed by a period of apnea. For the remainder of the first 30 seconds, cries were lower pitched (540 Hz) with some dysphonation. During the last 30 seconds, cries pitch was further reduced (478 Hz), and was rhythmic with a rising / falling tone (i.e. normal crying was re-established). Other studies also show that cries elicited by pain have higher fundamental frequencies, as well as other defining acoustic characteristics, such as increased amplitudes and tenseness (Fuller & Horii 1988; Fuller 1991; Craig et al. 2000).

Current evidence supports the notion that the infant cry acts as a graded signal of current distress, at least with regard to pain, although the existence of cry types should not be ruled out. There may be a few basic types of cry (e.g. cries associated with sudden intense distress such as pain, and those associated with low intensity but growing discomfort such as hunger), and there may be graded variation within each type (Gustafson et al. 2000).

IV.c. Chronically abnormal cry acoustics are linked to severe illness

Above I showed that transient changes in infant cry acoustics reflect acute changes in current levels of distress. Infant cries whose acoustic properties persistently and severely deviate from the norm, however, are associated with chronic poor health (Wasz-Hockert et al. 1985; Newman 1994; Corwin et al. 1996; Furlow 1997; Green et al. 2000). Table 3 summarizes these data. The abnormal infant cry is associated with chromosomal, endocrine, metabolic and neurological disturbances, as well as malnourishment, toxicity, and low birth weight. Infants with acoustically abnormal cries are also at long-term risk. Infants whose cries had the highest fundamental frequencies, for example, performed poorly on cognitive tests at 18 months and 5 years of age, and asphyxiated infants whose cries were most abnormal in terms of high fundamental frequency, variable fundamental frequency, short duration, and the presence of rising or falling/rising melodies, were the most likely to develop neurological sequelae, including epileptic convulsions, mental retardation, and death (Wasz-Hockert et al., 1985; Green et al. 2000).

The most common cry characteristic indicating serious pathology in these studies is an unusually high fundamental frequency (i.e. very high-pitched crying), often up to three times the normal frequency (compare Tables 2 & 3). Biphonation (the simultaneous production of two fundamental frequencies) is also commonly associated with pathology. One mechanism producing biphonated cries is the independent vibration of the left and right vocal folds at unrelated frequencies (Fitch et al. 2002). Biphonation may be an example of 'fluctuating asymmetry' in the voice (Mende et al. 1990; Fitch et al. 2002). Fluctuating asymmetry refers to small departures from bilateral morphologic symmetry, and greater asymmetry is considered an indicator of developmental instability or poor phenotypic quality (e.g. Livshits & Kobyliansky 1991; Moller & Swaddle 1997; Thornhill & Gangestad 1999). For example, human pre-term infants and infants with Down’s syndrome exhibit a variety of bilateral morphologic asymmetries, including unequal left versus right hand breadth (Livshits & Kobyliansky 1991). Other acoustic qualities of the cry associated with poor health are rapid alterations in fundamental frequency (e.g. vibrato and glides), and rising, falling/rising or flat cry melodies (Table 3). Low fundamental frequency is also associated with a few serious disorders (e.g. Down’s syndrome), suggesting that excessively high or low pitch is indicative of poor health.

IV.d. Infant cry acoustics affect adult perceptions and physiology

Infant cry acoustics convey acute and chronic distress in infants, but how do these signals relate to parental responses? The most common study of reactions to infant cries employs questionnaires to obtain adult reactions to tape recordings of infant cries that vary in acoustic structure. Commonly, infant cries of high and variable pitch (usually, but not always, from unhealthy infants, or elicited by a painful stimulus) are variably characterized as ‘aversive’, ‘urgent’, ‘arousing’, ‘grating’, ‘piercing’, ‘distressing’, ‘saddening’ and ‘sick’, compared to cries that are lower and less variable in pitch (Wiesenfeld et al. 1981; Brennan & Kirkland 1982; Boukydis 1985; Frodi 1985; Porter et al. 1986; Zeskind & Marshall 1988; Zeskind & Shingler 1991; Crowe & Zeskind 1992; Lester et al. 1992; Craig et al. 2000; Wood & Gustafson 2001). Studies that manipulated pitch (or other properties) of cries, independently of other acoustic characteristics, confirm these results (Zeskind et al. 1992; Protopapas & Eimas 1997; Dessureau et al. 1998).

A few studies also asked subjects to anticipate their behavioral responses to infant cries. Some studies indicate that high and variable pitched crying would lead to ignoring the infant, abusing the infant, or other non-optimal responses (Frodi 1985; Frodi & Senchak 1990; Zeskind & Shingler 1991). In other studies, however, the cries of high-risk infants and acoustically abnormal cries elicited more positive responses, such as shorter anticipated latency to respond and more "tender and caring" responses (Zeskind 1980; Gustafson et al. 2000; Wood & Gustafson, 2001). Also, Lester et al. (1995) showed that infants whose mothers rated their high-pitched cries as negative scored higher on language and cognitive tests at 19 months, compared to infants of mothers who did not rate high-pitched cries as negative. This result suggests that accurate perception of high-pitched crying may lead to positive responses in mothers.

High or variable pitched cries produce autonomic arousal, as measured by, for example, heart rate or skin conductance (Frodi & Lamb 1980; Wiesenfeld et al. 1981; Frodi 1985; Boukydis 1985; Zeskind 1987; Crowe & Zeskind 1992). Additionally, some studies show that abusers, or those determined likely to abuse, exhibit higher autonomic responses to high pitched cries (Frodi & Lamb 1980; Frodi 1985; Zeskind 1987; Crowe & Zeskind 1992). Donovan and Leavitt (1985) showed that viewers exhibited higher autonomic responses to crying infants compared to smiling infants. It is not clear what specific behaviors such autonomic arousal is likely to motivate. Frodi (1985) argues that increased autonomic response motivates abuse, while Donovan and Leavitt (1985) argue that it motivates help. Zeskind (1987) argues that heart acceleration and deceleration reflect aversiveness and special attention, respectively.

These studies of adult perceptual and physiological responses to infant cries do tell us that the infant cry is emotionally salient and arousing to adult listeners, especially the high-pitched cry. They do not tell us, however, that such crying would elicit any uniform positive or negative response from actual parents. The child abuse and infanticide literature, on the other hand, can yield information on the risk factors associated with abuse and killing of offspring.

IV.e. Cry acoustics and infant maltreatment

The cross-cultural ethnographic evidence presented earlier demonstrates that the poor quality of an offspring is a common risk factor for neglect, abuse and infanticide in humans (section II.b; Table 1). While the relationship between abnormal characteristics of the offspring and maltreatment is clear, it is not clear directly from these literatures that high-pitched or otherwise abnormal crying is a precipitating factor. Several considerations suggest that it may be a factor, however. First, infants with defects that are associated with abuse are the same infants who present abnormal cry acoustics such as high-pitched crying. Second, parents have described crying as the most proximate cause triggering abuse and infanticide (e.g. Frodi 1981). In one study, crying was the proximate cause of fatal abuse in 58% (14/24) of cases (Brewster et al. 1998). In another, 70% (16/23) of mothers whose infants’ cries were excessive, high-pitched and “shrieking” had fantasies of abusing, abandoning or killing their crying infants (Levitzky & Cooper 2000). Third, experimental evidence demonstrates that high pitched crying sometimes, but not always, elicits strong negative emotions in adult listeners (sources above), and these negative emotions are stronger in abusers than non-abusers (Frodi 1985; Crowe & Zeskind, 1992). With regard to the last point, it is possible that the aversion threshold toward acoustically abnormal cries differs across adults and that it originates in adult dispositions prior to parenthood. For example, pregnant women who later rated their infants as “difficult” showed higher autonomic arousal to recordings of a healthy infant cry, and expressed stricter child-rearing attitudes, compared to pregnant women who later rated their infants as “easy” (Frodi et al., 1989).

IV.f. A model of early infant cry acoustics and parental response

Here I propose a model of the signal effects of infant cry acoustics that will require further evidence to confirm (discussed in Conclusions and Prospects). For illustrative purposes, I focus on the fundamental frequency because it has been particularly well-studied, but the same logic can be applied to other acoustic properties. In healthy infants, the acoustic properties of the early infant cry function as a graded signal to adult listeners. The fundamental frequency of the cry conveys salient information regarding degree of distress. The physiological connection between distress and high-pitched crying is mediated by the vagus, in particular the branch of the vagus linking the nucleus ambiguus to the larynx. During acute stress (e.g. pain or fear), the sympathetic nervous system is activated while the parasympathetic nervous system is attenuated. As part of the parasympathetic withdrawal, vagal tone is lowered, fostering a multitude of physiological reactions including an increase in heart rate and a rise in the pitch of vocalizations (Porter et al. 1988; Porges 1995, 1997). According to this view, the cry acts generally to alert caregivers of nonspecific need (e.g. Ostwald & Murry 1985), but the acoustic properties (including, but not limited to, fundamental frequency) vary as a direct reflection of autonomic nervous system activity and emotional state, providing caregivers with information concerning the intensity of infant need or distress. Such acoustic information may influence the latency and urgency of response in caretakers. The actualization of specific responses is likely influenced by a combination of information in addition to the cry sound, such as infant facial expressions and gestures (Johnston & Strada 1986; Gustafson and Green, 1991; Green et al., 1995; Craig et al. 2000), and from the context (e.g. time since feeding; Bernal, 1972).

When infants are chronically ill or stressed, on the other hand, then crying is chronically abnormal, characterized by extremely high pitch, variable pitch and the presence of biphonation (Table 3). Although the same mechanisms that mediate transient changes in cry pitch during acute stress may also operate when infants are chronically ill, the acoustic abnormalities associated with severe pathology are different than transient acoustic changes associated with acute stress in two important ways. First, the acoustic abnormalities are chronic. Low birth weight children, for example, persistently exhibit mean vagal tones 2.2 times lower than the norm, even while sleeping (Porges 1995), and diminished vagal input to the larynx produces higher fundamental frequencies (Porter et al. 1988; Porges 1995; Green et al. 2000). Second, acoustic deviations associated with severe pathology are greater in degree than acoustic deviations associated with transient distress. Specifically, the fundamental frequency of normal cries is 200 – 600 Hz, while the fundamental frequencies observed in transient pain cries, for example, are only slightly outside this range (up to 730 Hz). The fundamental frequencies of cries associated with severe pathology, on the other hand, are usually over 1000 Hz (Table 3). Finally, biphonated cries, a possible instance of fluctuating asymmetry resulting from developmental instability, are rare or absent in normal infants but are common in infants with pathology.

In contrast to infant cries within the normal range described above, including those associated with transient distress, the more chronically and severely abnormal infant cries are, the more likely they are to be a proximate cause of adaptive neglect, abuse or infanticide. In pre-modern environments, severely ill infants would have had very low chances of survival, and theory predicts that selective withdrawal of investment can be adaptive in such circumstances. Moreover, evidence demonstrates that severe illness in infants can precipitate maltreatment, severe illness is precisely what chronically abnormal crying reveals, and infant crying can be the proximate cause of abuse. The persistently abnormal cry (along with other infant characteristics such as listlessness or small size) may activate psychological mechanisms in caregivers that include converting the infant cry from a motivator of sympathetic altruism (e.g. Hoffman 1975; Murray 1985) to a motivator of abuse or infanticide (e.g. Frodi & Lamb 1980; Frodi 1985; Mann 1992). This view of dynamic psychological dispositions in parents is supported by the fact that high pitched crying is often viewed as aversive, and, more importantly, that abusive parents react more strongly and negatively to high pitched cries than do non-abusive parents (e.g. Frodi 1985; Crowe & Zeskind, 1992). Also, there is one study showing that high-pitched crying was associated with fantasies of infant maltreatment (Levitzky & Cooper 2000).

The proposed effects of infant cry acoustics can be integrated into the multi-level model of parenting determinants described earlier. According to the multi-level model, factors such as infant and parental characteristics, household dynamics and economy, and the wider community and culture jointly influence patterns of parental care and abuse. The most important point is that maltreatment is multiply determined by many factors, and that single variables are unlikely to be necessary or sufficient determinants of maltreatment (Belsky 1993). The persistent and severely abnormal cry may be one of many factors influencing parenting choices. For example, a lack of spousal and other social support could create conditions where a new parent may withdraw care from a burdensome infant, in particular if cultural and parental values do not rule out infanticide or abandonment. Under these circumstances, the poor condition of the child, gleaned in part by the persistently abnormal acoustic quality of crying, may tip the scales from acceptance to rejection in some parents. This is not to say that persistent and severely abnormal crying will invariably result in infant abandonment or maltreatment. Under the same circumstances, but with strong spousal and social support, for example, even very sick infants may not be maltreated. Additional scenarios could be produced in which healthy infants may or may not be maltreated, depending on circumstances. The model does predict the following, however. All things being equal, normal acoustic variation in the infant cry, including that associated with autonomic arousal due to transient distress, is that most likely to elicit positive care-giving. On the other hand, the more acoustically abnormal cries are, both in terms of their persistence and distance from the normal range, the more likely that crying will result in the reduction or withdrawal of parental care, including sub-optimal care, neglect, abuse, abandonment and infanticide.

IV.g. The evolution of the early infant cry and mother-infant communication

The comparative anatomy of living animals suggests that the larynx evolved vocalization capabilities very late relative to the integrated and evolutionarily conserved autonomic nervous system that innervates it. The autonomic nervous system is developed in all vertebrates and is evolutionarily conserved, particularly in tetrapods (Butler & Hodos 1996). In primitive vertebrates (air-breathing fish), however, the larynx functions as a protective valve involved in feeding and swallowing (Butler & Hodos 1996; Hofer 2002). As air-breathing increased in importance, the vertebrate larynx strengthened to accommodate additional functions such as optimizing air-flow and protecting the lungs from foreign matter. Only later in vertebrate evolution was the larynx further modified for additional functions including vocalizing (Kirchner 1993; Armstrong and Netterville 1995; Laitman & Reidenberg 1997; Hofer 2002). This evolutionary history demonstrates that the vagal regulation of conserved laryngeal functions such as swallowing and breathing was already in place when the derived laryngeal function of vocalizing evolved. As part of the overall autonomic stress response, therefore, vagal input to the larynx regulating conserved functions would have concurrently affected vocalization acoustics as well. The vagus also regulates the stress response in other organs such as the bronchi, esophagus, pharynx and heart. Acoustically distinct early infant crying during acute and chronic stress is likely a necessary by-product of integrated autonomic nervous system mobilization, and parents may have evolved to react to this acoustic information.

The following simple evolutionary sequence gives an idea of how natural selection may have elaborated laryngeal functions and the behavior of mothers to form communication systems. Hofer (2002) separated rat pups from their mothers and showed that consequent ultrasound separation vocalizations involved "laryngeal braking" (increased air-pressure when the larynx closes on escaping air), and that this phenomenon regulated recovery from hypothermia. Crying in human infants also retains a thermoregulatory function by generating heat (Lester, 1985). Hofer argued that laryngeal breaking first evolved in mammalian infants because it facilitated recovery from hypothermia when maternally separated, and that the behavioral complex involving separation calls and retrieval came into being when mothers later evolved to respond to those ultrasound emissions by retrieving their separated infants. To this it could be added that early mammalian infant vocalizing also varied acoustically as a function of autonomic nervous system arousal, and that mothers later evolved to respond differentially to infant cries according to these varying acoustic characteristics.

As Furlow (1997) has argued, if parents evolved the ability to determine the level of infant distress based on the acoustic properties of cries, then these properties can be considered honest signal of need. Crying is metabolically costly (Rao et al. 1997), the autonomic arousal producing acoustic changes (e.g. in fundamental frequency) imposes additional costs (Nelson 2000), and there is evidence that the fundamental frequency of cries accurately reflects the degree of physical pain. Because such cry acoustics are likely unalterable manifestations of autonomic nervous system activity, however, they may not necessarily result in increased care from parents⁴. As I argued earlier, cries resulting from transient distress are more likely to increase care, but under some circumstances chronically and severely abnormal cries resulting from serious pathology may inhibit care.

V. Excessive crying and colic

V.a. The colic syndrome

Excessive infant crying is the most common clinical complaint of mothers with infants under three months of age (Forsyth et al. 1985). The core symptom of colic is excessive, inconsolable crying that begins and ends without warning (Gormally & Barr 1997; Lehtonen et al. 2000). Colic is most commonly defined according to Wessel’s operational "rule of 3’s", in which an infant cries for more than three hours a day, for more than three days a week, for more than three weeks (Wessel et al. 1954). Definitions of colic based the amount of crying, namely, Wessel’s rule of 3’s and its variations (Sijmen et al. 2001), are collectively known as ‘Wessel’s colic’. There are also definitions of colic that are qualitatively different, however, which include symptoms in addition to the amount of crying (e.g. Lester 1997). Such additional symptoms include 1) high-pitched crying, 2) hypertonia (e.g. clenched fists, flexed legs, grimacing, distended abdomen), and 3) flatulence, regurgitation, vomiting or diarrhea. I will refer to excessive crying meeting Wessel’s criteria but with any of these additional symptoms as ‘Wessel’s plus’ colic (following Gormally & Barr 1997). Colic is also characterized by two temporal crying patterns (Barr 1990; Gormally & Barr 1997; Lehtonen et al. 2000). Developmentally, crying increases from 2 weeks after birth, peaks in the middle of the second month, and resolves by the fourth month. Diurnally, crying bouts cluster in the late afternoon and early evening hours, and this clustering pattern is most marked during the sixth week peak in overall crying (Barr 1990). Prevalence rates vary widely (ranging from 2 – 40%), depending on the definition used and whether or not the infants are from a referred clinical population or a community sample (Barr 1999; Sondergaard et al. 2000; Sijmen et al. 2001; Canivet et al. 2002).

In what follows I argue that what is labeled as colic represents the extreme cases of an otherwise normal and universal increase in crying during the first few months of life. I go on to discuss factors that may contribute to variation in the amount of infant crying, such as maternal style and infant temperament. I evaluate the potential signal functions of excessive crying in the final section of the paper.

V.b. Colic is part of the normal infant crying curve

An n-shaped curve of increased crying during the first three months of life is typical feature of human development, and those with colic comprise the extreme end of this distribution. Brazelton (1962) first detected the infant crying curve from data gleaned from the diaries of 80 mothers. Since then, various methodologies have shown that the crying pattern of infants is characterized by an overall increase in crying until about 6 weeks of age, followed by a gradual decrease in crying until 3 or 4 months, after which it remains relatively stable (Barr 1990; Lehtonen et al. 2000). In addition, crying undergoes a diurnal rhythm, with clustering in the late afternoon and early evening hours, and this diurnal clustering is most marked during the 6^th week peak in overall crying (Barr 1990). These same developmental and diurnal patterns are found in both "normal" infants and infants with colic (St. James-Roberts & Halil 1991; St James-Roberts et al. 1994; Lehtonen et al. 2000).

The crying curve manifests itself across different cultural contexts. The !Kung (South African hunter-gatherer society) exhibited a crying curve similar to a Dutch sample (Barr et al. 1991). Both Manili and London infants experienced a 6^th week crying peak, and the diurnal cluster of early evening crying (St. James-Roberts et al. 1994). Finally, similar amounts of crying, the same evening cry clustering, and similar incidences of persistent crying characterized Denmark, United Kingdom and North American infants (Alvarez & St James-Roberts 1996). Taken together, the data indicate that the crying curve in infants is a common property of neonates in a wide variety of cultural contexts, although it should be remembered that within populations there is always extensive individual variation. Excessive criers labeled with ‘colic’ are simply those individuals who show the most extreme form of an otherwise normal and universal crying curve.

V.c. An organic etiology of colic is rare

In a minority of cases colic has known organic etiology, but in these cases infants exhibit symptoms in addition to excessive crying. Gormally and Barr (1997) and Lehtonen et al. (2000) estimate that about 5%-10% of colic cases are due to organic disorder. The most common of such etiologies are cow’s milk protein intolerance, and (rarely) other ailments such as fructose intolerance and infantile migraine. There is weak evidence for other organic causes such as reflux esophagitus and lactose intolerance. Most importantly, colicky infants with organic etiology are those who have symptoms in addition to excessive crying (i.e. Wessel’s plus colic), such as high pitched crying, arched back, regurgitation, vomiting, and diarrhea (Gormally & Barr 1997; Barr 1999; Lehtonen et al. 2000). Part of the reason behind the perception that colic is associated with disease is that those studies showing such an association are based on highly selective samples of excessive criers with additional symptoms of organic disease (Gormally & Barr 1997).

Some recent studies have continued to show that colic is associated with poor health, such as low fetal birth weight and maternal smoking (Sondergaard et al. 2000, 2001). In these cases, too, however, "colic" infants possessed symptoms in addition to excessive crying, such as flexed legs, distended abdomen, and excessive flatus. Kanabar et al. (2001) showed that lactase treated feed reduced crying by more than 45% in colicky infants, indicating the specific organic etiology of lactose intolerance, but the definition of colic included spasm, lower limb flexure, and diarrhea. In short, when colic simply refers to excessive crying (i.e. Wessel’s colic), then it is not attributable to or associated with organic pathology, but when it refers to that subset of cases in which there are additional symptoms (i.e. Wessel’s plus colic), then it is more likely attributable to organic causes.

V.d. Colic cries are not acoustically abnormal

If most cases of colic are not due to illness, then cries of infants with colic should not be high-pitched cries, which are associated with illness (Table 3). However, Lester et al. (1992) and Lester (1997) showed that infants with colic had higher and more variable pitched cries than controls. High-pitched crying itself was part of the definition of colic in both studies, however, so it is not surprising that those subjects did present acoustic abnormalities including, of course, higher pitched cries. Moreover, infants in both samples had symptoms in addition to excessive and high-pitched crying, such as premature birth or hypertonia. These infant subjects were drawn from that small subset of colicky infants with physical ailments (Wessel’s plus colic). Even so, the mean fundamental frequency of their cries was well within the normal range (492 Hz for "colic" infants and 414 for controls; see Table 2).

Zeskind and Barr (1997) showed that cries of colic infants (without additional symptoms) did have higher mean fundamental frequencies than controls, but these cries were within the normal range (mean 591 Hz for colic, and 498 for controls). St. James-Roberts (1999) found no significant difference between the mean fundamental frequency of Wessel’s colic cries (491 Hz) and controls (458 Hz), however. Finally, infants with Wessel’s colic and controls did not show differences in autonomic nervous system balance (which affects cry pitch) during the time of colic or after its resolution (Kirjavainen et al. 2001). For the most part, infants with colic produce higher pitched cries than their non-colicky counterparts only when they have Wessels’ plus colic, and there is no evidence that infants with colic (by any definition) have fundamental frequencies outside the normal range.

V.e. Maternal style affects early infant crying

Despite the cross-cultural validity of the early infant crying curve, cultural and individual variation in maternal style can attenuate the crying curve. In particular, increased responsiveness to fussing or crying infants lessens the overall amount of infant crying. For example, the number of crying bouts was the same for both !Kung and western infants, but the total duration of !Kung infant crying was ½ that of the western infants (Barr et al. 1991). Unlike most western mothers, !Kung mothers carry their babies continuously (> 80% in the daytime), engage in continuous feeding (4 times/hr), and are highly responsive to frets (92% response within 15 sec), usually with breast-feeding (Konner & Worthman 1980; Barr et al. 1991). This contrasts with the normative western care-giving style in which infants are not in constant contact with mothers and crying infants are often deliberately ignored up to 40% of the time (Barr 1999). Generally, traditional societies are characterized by the indulgent maternal style (Barr 1999; Ziefman 2001).

Within western societies, variation in maternal style also has been shown to affect the amount of early infant crying. Crying decreased when mothers were asked to increase infant carrying (Hunziker & Barr 1986), eliminating the 6-week peak in crying, but not the diurnal clustering. St. James-Roberts et al. (1995) replicated this methodology, however, and found that increased infant carrying did not decrease crying amount. In another study comparing La Leche League (an avid pro-breastfeeding group) and standard mothers, frequent nursing reduced the number of crying episodes at 2 months of age (Barr & Elias 1988).

V.f. Infants with colic do not have difficult temperaments

When the excessive crying diagnostic of colic resolves, no long-term differences persist between infants with colic and those without it, although the parents of infants with colic may perceive that they do (Gormally & Barr 1997; Barr & Gunnar 2000; Lehtonen et al. 2000) In some studies claiming that a difficult temperament and colic are associated with one another, measures of temperament were recorded before colic resolved, and the excessive crying and fussing diagnostic of colic also contributed to the difficult temperament designation (e.g. Carey 1972). Using this methodology, infants with colic have difficult temperaments by definition (Barr & Gunnar, 2000).

One way to avoid this circularity is to examine older children with and without histories of colic to see if difficult temperaments persist after the excessive crying has resolved. In general, such studies show that parents may perceive their formerly colicky offspring as "difficult", but actual behavioral or physical differences are not apparent (Lehtonen et al. 2000; Barr & Gunnar 2000). In 4 colic follow-up studies summarized by Barr (1998b), for example, mothers of excessive criers often perceived their infants as difficult, and were themselves distressed, after the colic resolved. In the majority of cases, however, there was a reduction in infant crying, little maternal distress, and normal attachment relationships. Additionally, St. James-Roberts et al. (1998) found infant negative behaviors at 6 weeks did not predict behavior at 15 months of age, and Canivet et al. (2000) showed no long-term differences between ex-colics and controls in behavior, growth, weight, height, or days in the hospital. In both studies, however, parents rated their formerly colicky children as more emotional or difficult.

V.g. Infant responsivity affects early infant crying

Infant characteristics are also likely to contribute to variation in the amount of early infant crying. Above I argued that colic is not a manifestation of infant temperament. Barr and Gunnar (2000) offered a related explanation (also see Blum et al. 2002), the ‘transient responsivity’ hypothesis, according to which infants differ in their responsivity to similar stimuli, and that these differences are manifest as variable amounts of crying. The key difference between the ‘temperament’ hypothesis and the ‘transient responsivity’ hypothesis is that according to the former, individual differences among infants are permanent features of the individual’s constitution, while according to the latter differences among infants are transient.

Supportive evidence for the ‘transient responsivity’ hypothesis is that the pattern of early infant crying itself (i.e. the crying curve) is transient, as described above. Crying undergoes a developmental change at around three months of age, in which it becomes less frequent, but more context-dependent, intentional and communicative (Ostwald & Murry 1985; Barr 1990; Barr & Gunnar 2000). Furthermore, sleep patterns, attention to visual stimuli, and cortisol response to immunizations also undergo a similar developmental curve. Finally, measures of temperament and emotional regulation in early infancy are not predictive of later stages of development, but measures taken after 3 months are (Barr & Gunnar 2000).

The increased responsivity of infants with colic may be due, in part, to an opioid-dependent mechanism regulating crying behavior (Panksepp et al. 1988; Barr & Gunnar, 2000). Sucrose is an opioid-dependent calming agent of both crying infants and maternally separated non-human primates. The initial suckling stimulus and the sucrose itself exert independent calming effects, with the sucrose effect lasting longer and mediated by opioids. The crying of infants with colic is regulated just as much as controls in the first minute after sucrose exposure, but infants with colic return to crying sooner than controls during the late opioid-dependent phase. This suggests that that variation in amounts of infant crying may be due in part to variation in the opioid-mediated regulation of crying, and that those infants characterized as having colic may be those least regulated.

VI. The Signal Functions of Excessive Infant Crying

VI.a. Excessive crying as a manipulative signal

Lummaa et al. (1998) and Barr (1999) argued that early excessive crying in infants may, in addition to serving other functions, be a manipulative signal by which infants attempt to gain a disproportionate share of parental resources. Specifically, excessive crying prolongs the period of investment for an individual offspring by increasing breast-feeding, maintaining a longer state of lactational amenorrhea, and thereby postponing the next pregnancy. Indeed, in traditional societies, inter-birth intervals are often long (up to 4 years), they are in large part due to continuous breast-feeding and resultant lactational amenorrhea, and crying is often a proximate cause of breast-feeding (Konner & Worthman 1980; Blurton-Jones 1986; Barr et al. 1991, 1999; Zeifman 2001).

Further consideration of parent-offspring conflict, however, contradicts this ‘manipulative’ view of the signal function of early infant excessive crying. The conflict between parents and offspring should increase and become most pronounced at the end of the period of parental investment. During the first three months of life, disagreements over the flow of resources should be lowest. As the period of investment proceeds, however, and the production of a new child becomes increasingly advantageous for the mother, then parent-offspring conflict should increase. In fact, in hunter-gatherer societies, it has been argued that inter-birth intervals that are too short are disadvantageous for mothers’ lifetime reproductive output because of the cost of carrying infants and food at the same time (Blurton-Jones 1896; 1987; Winterhalder & Smith 2000). Additionally, close birth spacing is one of the principal rationales for infanticide in the cross cultural survey (Table 1).

If the amount of infant crying reflects the degree of parent-offspring conflict over the length of investment, therefore, then it should start at low levels in early infancy, increase over the duration of investment and peak at weaning, when conflict is at its greatest. The crying curve of early infancy, however, shows the opposite temporal pattern. Crying peaks in the second month, and is greatest in the first three months of life, compared to later infancy and toddlerhood. This is not to say that crying and other behaviors are not used in later development as a means to prolong investment. I am merely arguing that this is an unlikely explanation for increased crying in early infancy.

Manipulative crying need not be narrowly construed as only functioning to prolong investment and postpone the next birth, but could function to increase attention and care throughout early infancy. Individual variation in the amount of crying could be explained by differences in parental styles. Indulgent parental styles result in less manipulative crying while less indulgent styles result in more manipulative crying. The temporal pattern of early excessive crying is difficult to reconcile with this notion, however. Manipulative crying that exaggerates need for attention or breast-feeding, for example, cannot be expected to exhibit the 6^th week peak or the early evening peaks that are observed.

VI.b. Excessive crying as an honest signal of need

Another possible signal function is that excessive early infant crying is an honest signal of need, according to which individual variation in crying amount honestly reflects different resource requirements. One prediction of this hypothesis is that there be costs associated with crying. Rao et al. (1997) documented that infant crying is energetically costly, with a 13.2% increase in metabolic rate compared to resting. Another prediction is that excessive crying should be associated with either transient distress or permanent poor health. There is no clinical evidence that excessive crying is associated with immediate, transitory needs, such as distress or hunger, however. In fact, transitory stresses cannot be expected to follow the distinct temporal patterns of early infant excessive crying, such as the 6^th week peak or the diurnal pattern of early evening clustering. Neither does excessive crying reflect overall poor health. Only in a minority of cases, when additional symptoms are involved (i.e. Wessel’s plus colic), is excessive crying associated with illness. In short, there is no support that excessive crying observed in early infancy is associated with an increased need for parental resources.

VI.c. Excessive crying as an honest signal of vigor

Barr (1998a) and Lummaa et al. (1998) argued that excessive crying or colic may be an honest signal of vigor to avoid infanticide. This argument does not require that infant variation in quality is the only factor influencing cry amount. For example, I argued that maternal style explains some of the variation in the amount of infant crying. An indulgent maternal style may explain the higher average amount of early infant crying in the !Kung compared to western societies. Within each population, however, there is still extreme individual variation in the overall amount of early infant crying, and these within-population differences may be explained by variation in infant quality.

Current evidence is more consistent with this view than the first two hypotheses discussed above. First, such signals are adaptations to those specific environments in which parents withdraw investment or kill infants when the prospects for infant survival are low. Abundant cross-cultural evidence indicates that such selective withdrawal of parental investment is a normative aspect of our species’ behavioral repertoire. Second, in order for such signals to be honest, the cost of crying has to be higher for low quality infants than it is for high quality infants. Two studies demonstrate this key prediction. Rao et al. (1997) showed that infant crying is energetically costly, and, more importantly, demonstrated that metabolic expenditure was higher for low quality (low weight and preterm) infants than it was for high quality (normal weight, full term) infants. Similarly, White et al. (2000) showed that infants with colic cried longer, more intensely, and more inconsolably than infants without colic in a mock physical examination. Despite increased crying, however, infants with colic did not show any differences in heart rate, vagal tone, or cortisol secretion during the examination. Furthermore, daily salivary cortisol levels of excessive criers were also indistinguishable from controls, despite the fact that they cried for more hours per day. These results suggest that colicky infants experience no more physiological stress as a result of excessive crying compared to infants who cry less.

Third, if excessive crying is a signal of high quality then infants with colic should have higher survival or better developmental outcomes compared to other infants. In this instance, the supportive evidence is not strong, however. The fact that crying is less costly for infants with colic compared to controls does suggest that these infants may be of higher current quality than infants who cry less. But there is no clinical evidence showing that infants with colic are in better health or have better outcomes than infants who cry less. Of course, infants who are severely ill may cry very little or not at all, so that the lower bound crying amount (i.e. little or none), compared to all other cry amounts, is related to infant survival in the way predicted by the signaling vigor hypothesis. But it is unclear if the remainder of the variation in cry amount is also related to health or survival. It is possible that modern medical environments mask most health differences among infants who cry in different amounts. The differences in survivability between excessive criers and those who cry less may only become apparent in more variable environments, when conditions are sometimes exceptionally poor. Supportive evidence for this view comes from a study on the Massai, traditional African subsistence herders (deVries 1984). Infants with difficult and fussy temperaments, measured before a drought, were more likely to survive the drought (5/6=83%) compared to those infants with easy temperaments (2/7=29%). It should be noted that the infants were past 4 months of age, however, when the normal infant crying curve has normally resolved.

Finally, the temporal pattern of excessive early infant crying mirrors the expected and observed temporal pattern of the probability of infanticide in humans. Theory predicts that psychological mechanisms influencing the continuation of parental investment in offspring should be activated soon after birth, so that losses can be cut early if the prospects for infant survival are poor. Indeed, the observed incidence of infanticide by parents is highest in the first three months of life, and then decreases over the rest of the first year (Daly & Wilson 1988; Overpeck et al. 1998). Figure 1 shows that the amount of infant crying over the first year mirrors this temporal pattern of infanticide risk (Barr 1990; Alvarez & St. James-Roberts 1996).

This pattern is consistent with the interpretation that excessive crying has evolved to signal vigor to avoid the withdrawal of parental care, but it is also consistent other interpretations. First, the causal relationship could be reversed, such that excessive infant crying is itself a proximate cause of infanticide. Two studies are relevant here. Brewster et al. (1998) examined 32 cases of infanticide among US Air Force personnel, and found that crying was the proximate trigger of the fatal abuse in 58% (14/24) of cases, and that colic was indicated in the medical records of 35% (14/24) of cases. However, no definition of colic was available, there was no control group for comparison, the average age of death was about 5 months (past the usual resolution of colic), and only 10% (2/20) of parents reported colic to their physicians. The authors conclude that colic was overestimated by physicians due to crying at visits that resulted from non-fatal physical injuries prior to the infanticides (12/22=55% of cases). In another study, mothers of infants with ‘colic’ had fantasies of abuse and infanticide (Levitzky & Cooper 2000). The infants had ‘Wessel’s plus’ colic, however, which included symptoms in addition to excessive crying including high-pitched crying, “shrieking” and hypertonia. Earlier, I suggested that the acoustic quality, not the quantity of crying, was the trigger for abuse fantasy (section IV.e) At present there is no strong evidence showing the how excessive (or acoustically abnormal) crying influences the immediate abusive responses of caregivers.

Second, infanticide and excessive crying may not be causally related to one another but each may be independently related to early development. Infanticide may be most likely at younger ages because it benefits parents to cut losses early if the infant has diminished survival chances, as I have argued. At the same time, increases in crying during early infancy may be part of developmental processes unrelated to infanticide probability, such as sleep-wake cycles specific to the first three months of life (Lester, 1985; Zeifman, 2001). Unlike the previous two hypotheses, however, the pattern conforms to expectations from the signaling vigor hypothesis. In conclusion, while the evidence in support of the signaling vigor hypothesis is not strong, the evidence is more consistent with it than the signaling need or manipulative signaling hypotheses.

Figure 1:

Relationship between number of infanticides and amount of infant crying in the first year of life. Both infanticide and infant crying are highest in the first three months of life, and decrease throughout the year. Crying data in lower panel are for 590 infants in Denmark (redrawn from data in Figure 1 from Alvarez & St. James-Roberts 1996). Infanticide data in the upper panel are from 2776 infanticides in the United States (Overpeck et al. 1998; Figure 1). Upper panel copied with permission. Copyright 1998 Massachusetts Medical Society. All rights reserved.

VII. Conclusions and Prospects

The human infant cry literature is fraught with controversy. Here I evaluated evolutionary approaches that may illuminate the signal functions of the early infant cry. Overwhelming cross-cultural evidence suggests that the reduction or withdrawal of optimal care when child-rearing circumstances are unfavorable is a prominent feature of human evolutionary history. Such an environment of uncertain parental care constitutes a strong selective pressure on the human infant, including its cry signal, which can influence the behavior of potential caregivers.

Of course, the most obvious signal function of the early infant cry is to maintain proximity and elicit care from caregivers, and I do not dispute that essential role for infant crying. There is not strong evidence for acoustically distinct cry types that reflect specific needs, however, but there is evidence that the acoustic properties of infant cries function as a graded signal, at least with regard to level of pain. To further investigate the communicative content of infant crying, it will be necessary to identify the set of acoustic parameters that contain the most salient information. The universe of potentially important cry acoustics is large, and the choice of the most important acoustic measures should continue to be developed based on physiological models of cry production. In addition, acoustic measurement needs to be standardized. With such acoustic properties and measurements in hand, statistical techniques (e.g. discriminant analysis) can be used to determine the extent to which cries resulting from different causes can be categorized into acoustically distinct types, or to further show how graded acoustic variation reflects degree of distress. Naturalistic observations and perceptual experiments on parents (preferably using the cries of their own infants) can determine if and how parents attend to these acoustic properties of cries.

I proposed a model describing how the acoustic characteristics of infant cries may influence patterns of parental care and abuse. Some of the most important evidence regarding this proposal is precisely that which is unavailable, namely, the immediate and cumulative parental responses to infant cries depending on the acoustic qualities of cries. In the case of negative responses such as neglect, abuse and infanticide, this is not surprising, because the most proximate precursors to such maltreatment cannot be directly observed and such parents are not enthusiastic participants in research. Future work on maltreatment, in particular when infant health is compromised, should include information on the acoustic properties of infant cries, questionnaires or interviews of parents, and studies of cry perception, to determine the relationship of cry quality to maltreatment potential, maltreatment fantasy, and actual maltreatment. With regard to variation in parenting that does not involve maltreatment, naturalistic home audio-visual recordings could yield information on acoustic analysis of cries and parental responses. Additionally, experiments on the perception of infant cries that vary in quality should be continued to determine the physiological and anticipated behavioral responses of parents, in particular to cries of their own offspring. Finally, the perception of crying as particularly aversive in abusers could result from the cumulative effects of acoustically abnormal crying, unfavorable child-rearing circumstances, or due to prior dispositions in abusive parents. Longitudinal studies of parents will be necessary to understand how parental characteristics, child-rearing circumstances, and the acoustic qualities of cries influence adverse parental responses to cries.

There was no strong support for any of the proposed signal functions of excessive early infant crying, although the data were most consistent with the signaling vigor hypothesis. To further test these ideas, the immediate and cumulative effect of crying amount on parental behaviors needs to be investigated, in much the same way as I proposed to investigate parental responses to cry acoustics. Additionally, further light may be shed on the relationship between excessive infant crying and infant health by investigating infants in traditional societies where environments may be more variable than in industrialized societies.

Much remains to be learned about how infant crying contributes to the infant-caregiver relationship. I hope this evaluation of the evolutionary approach to understanding the early infant cry will stimulate further scientific inquiry into this fundamental aspect of the human experience.

References

Adinkraw, M. (2000) Maternal infanticides in Fiji. Child Abuse and Neglect 24: 1543-1555.

Ainsworth, M.D.S. (1969) Object relations, dependency, and attachment: a theoretical review of the infant-mother relationship. Child Development 40: 969-1025.

Albarracin, D., Repetto, M.J., and Albarracin, M. (1997) Social support in child abuse and neglect: support functions, sources, and contexts. Child Abuse and Neglect, 21: 607-615.

Alvarez, M. & St. James-Roberts, I. (1996) Infant fussing and crying patterns in the first year in an urban community in Denmark. Acta Paediatrica, 85: 463-466.

Andersson, M. (1994) Sexual Selection. Princeton University Press.

Armstrong, W.B. & Netterville, J.L. (1995) Anatomy of the larynx, trachea, and bronchi. Otolaryngologic Clinics of North America, 28: 685-699.

Barkow J.H., Cosmides, L., & Tooby, J. (1992) The Adapted Mind: Evolutionary Psychology and the Generation of Culture. Oxford University Press.

Barr R.G. (1990) The normal crying curve: what do we really know? Developmental Medicine and Child Neurology, 32: 256-362.

Barr R.G. (1998a) Reflections on measuring pain in infants: dissociation in responsive systems and "honest signalling". Archives of Disease in Childhood, 79: 152-156.

Barr R.G. (1998b) Crying in the first year of life: good news in the midst of distress. Child: Care, Health and Development, 24: 425-439.

Barr R.G. (1999) Infant crying behavior and colic: an interpretation in evolutionary perspective. In: Evolutionary Medicine, ed. Trevathan, W., McKenna, J.J., & Smith, E.O. Oxford University Press.

Barr, R.G. & Elias, M.F. (1988) Nursing interval and maternal responsivity: effect on early infant crying. Pediatrics, 81: 529-536.

Barr, R.G. & Gunnar, M. (2000) Colic: the ‘transient responsivity’ hypothesis. In: Crying as a Sign, a Symptom and a Signal: Clinical, emotional and developmental aspects of infant and toddler crying, ed. Barr, R.G., Hopkins, B., & Green, J.A. Mac Keith Press, London.

Barr, R.G., Konner, M., Bakeman, R., & Adamson, L. (1991) Crying in !Kung San infants: a test of the cultural specificity hypothesis. Developmental Medicine and Child Neurology, 33: 601-610.

Becker, J.C., Liersch, R. Tautz, C., Schlueter, B., and Andler, W. (1998) Shaken baby syndrome: report on four pairs of twins. Child Abuse and Neglect, 22: 931-937.

Bell, S.M. & Ainsworth, M.D.S. (1972) Infant crying and maternal responsiveness. Child Development, 43: 1171-1190.

Belsky, J. (1984) The determinants of parenting: a process model. Child Development, 55: 83-96.

Belsky, J. (1993) Etiology of child maltreatment: a developmental-ecological analysis. Pychological Bulletin, 114: 413-434.

Bernal, J. (1972) Crying during the first ten days of life, and maternal responses. Developmental Medicine and Child Neurology, 14: 362-372.

Black, M.M., Hutcheson, J.J., Dubowitz, H., & Berenson-Howard, J. (1994) Parenting style and developmental status among children with nonorganic failure to thrive. Journal of Pediatric Psychology, 19: 689-707.

Black, M.M. & Krishnakumar, A. (1999) Predicting longitudinal growth curves of height and weight using ecological factors for children with and without early growth deficiency. Journal of Nutrition, 129: 539S-543S.

Blum, N.J., Taubman, B., Tretina, L., & Heyward, R.Y. (2002) Maternal ratings of infant intensity and distractibility: relationship with crying duration in the second month of life. Archives of Pediatric and Adolescent Medicine, 156: 286-290.

Blurton-Jones, N. (1972) Comparative aspects of mother-child contact. In: Ethological Studies of Infant Behavior, ed. Blurton-Jones, N. Cambridge University Press.

Blurton-Jones, N. (1986) Bushman birth spacing: a test for optimal interbirth intervals. Ethology and Sociobiology, 7: 91-105.

Blurton-Jones, N. (1987) Bushman birth spacing: direct tests of some simple predictions. Ethology and Sociobiology, 8: 183-203.

Boukydis, C.F.Z. (1985) Perception of infant crying as an interpersonal event. In: Infant Crying: Theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Bowlby, J. (1969 [1982]) Attachment and Loss. Volume 1. Attachment. Second edition. Basic Books, New York.

Brazelton, T.B. (1962) Crying in infancy. Pediatrics, 29: 579-588.

Brennan, M. & Kirkland, J. (1982) Classification of infant cries using descriptive scales. Infant Behavior and Development, 5: 341-346.

Brewster, A.L., Nelson, J.P., Hymel, K.P., Colby, D.R., Lucas, D.R., McCanne, T.R., & Milner, J.L. (1998) Victim, perpetrator, family, and incident characteristics of 32 infant maltreatment deaths in the United States Air Force. Child Abuse and Neglect, 22: 91-101.

Britton, H.L, Gronwaldt, V., Britton, J.R. (2001) Maternal postpartum behaviors and mother-infant relationship during the first year of life. Journal of Pediatrics, 138: 905-909.

Bugos, P.E. (1984) Ayoreo infanticide: a case study. In: Infanticide, ed. Hausfater, G. & Hrdy, S.B. Aldine, New York.

Burgess, R.L. & Drais, A.A. (1999) Beyond the "Cinderella effect": Life history theory and child maltreatment. Human Nature, 10: 373-398.

Butler, A.B. & Hodos, W. (1996) Comparative Vertebrate Neuroanatomy: Evolution and Adaptation. Wiley-Liss, New York.

Canivet, C., Jakobsson, I., & Hagander, B. (2000) Infantile colic. Follow-up at four years of age: still more "emotional". Acta Paediatrica, 89: 13-17.

Canivet, C., Jakobsson, I., & Hagander, B. (2002) Colicky infants according to maternal reports in telephone interviews and diaries: a large Scandinavian study. Developmental and Behavioral Pediatrics, 23: 1-8.

Carey, W.B. (1972) Clinical applications of infant temperament measures. Behavioral Pediatrics, 81:823-828.

Chaffin, M., Kelleher, K., and Hollenberg, J. (1996) Onset of physical abuse and neglect: psychiatric, substance abuse, and social factors from prospective community data. Child Abuse and Neglect, 20: 191-203.

Chan, Y.C. (1994) Parenting stress and social support of mothers who physically abuse their children in Hong Kong. Child Abuse and Neglect, 18: 261-269.

Chisholm, J.S. (1996) The evolutionary ecology of attachment organization. Human Nature, 7: 1-38.

Christensson, K, Cabrera, T., Christensson, E., Uvnas-Moberg, K., & Winberg J. (1995) Separation distress call in the human neonate in the absence of maternal body contact. Acta Paediatrica, 84: 468-473.

Corwin, M.J., Lester, B.M., & Golub H.L. (1996) The infant cry: what can it tell us? Current Problems in Pediatrics, 26: 325-334.

Craig, K.D., Gilbert-MacLeod, C.A., & Lilley, C.M.. (2000) Crying as an indicator of pain in infants. In: Crying as a Sign, a Symptom and a Signal: Clinical, emotional and developmental aspects of infant and toddler crying, ed. Barr, R.G., Hopkins, B., & Green, J.A. Mac Keith Press, London.

Crnic, K.A., Greenberg, M.T., Robinson, N.M., & Ragozin, A.S. (1984) Maternal stress and social support: effects on the mother-infant relationship from birth to eighteen months. American Journal of Orthopsychiatry, 54: 224-235.

Crowe, H.P. & Zeskind, P.S. (1992) Psychophysiological and perceptual responses to infant cries varying in pitch: comparison of adults with low and high scores on the child abuse potential inventory. Child Abuse and Neglect, 16: 19-29.

Daly, M. (1990) Evolutionary theory and parental motives. In: Mammalian Parenting: biochemical, neurobiological, and behavioral determinants, ed. Krasnegor, A. & Bridges, R.S. Oxford University Press.

Daly, M. & Wilson, M. (1981) Abuse and neglect of children in evolutionary perspective. In: Natural Selection and Social Behavior, ed. Alexander, R.D. & Tinkle, D.W. Chiron, New York.

Daly, M. & Wilson, M. (1984) A sociobiological analysis of human infanticide. In: Infanticide, ed. Hausfater G. & Hrdy S.B. Aldine, New York.

Daly, M. & Wilson, M. (1988) Homicide. Aldine de Gruyter, New York.

Davis, J.N. & Todd, P.M. (1999) Parental investment by simple decision rules. In: Simple Heuristics that Make Us Smart, ed. Gigerenzer, G., Todd, P.M., & the ABC Research Group. Oxford University Press.

Denes, P.B. & Pinson, E.N. (1963) The Speech Chain. Baltimore, Waverly Press.

Dessureau, B.K., Kurowski, C.O., & Thompson, N.S. (1998) A reassessment of the role of pitch and duration in adults’ responses to infant crying. Infant Behavior and Development, 21: 367-371.

deVries M.W. 1984. Temperament and infant mortality among the Masai of East Africa. American Journal of Psychiatry, 141: 1189-1194.

Dobson, V. & Sales, B. (2000) The science of infanticide and mental illness. Psychology, Public Policy, and Law, 6: 1098-1112.

Donovan ,W.L. & Leavitt, L.A. (1985) Physiology and behavior: parents’ response to the infant cry. In: Infant Crying: Theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Elwood, R. (1992) Pup-cannibalism in rodents: causes and consequences. In: Cannibalism: ecology and evolution in diverse taxa, ed. Elgar, M.A & Crespi, B.J. Oxford University Press, Oxford.

Ferrari, A.M. (2002) The impact of culture upon child rearing practices and definitions of maltreatment. Child Abuse and Neglect, 26: 793-813.

Fitch, W.T., Neubauer, J., & Herzel, J. (2002) Calls out of chaos: the adaptive significance of nonlinear phenomena in mammalian vocal production. Animal Behaviour, 63: 407-418.

Formby, D. (1967) Maternal recognition of infant’s cry. Developmental Medicine and Child Neurology, 9:293-298.

Forsyth, B.W., Leventhal, J.M., & McCarthy, P.L. (1985) Mothers’ perceptions of problems of feeding and crying behaviors: a prospective study. American Journal of Diseases of Childhood, 139: 269-272.

Frodi, A.M. (1981) Contribution of infant characteristics to child abuse. American Journal of Mental Deficiency, 85: 341-349.

Frodi, A.M. (1985) When empathy fails: aversive infant crying and child abuse. In: Infant Crying: Theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Frodi, A.M., Bridges, L., & Shonk, S. (1989) Maternal correlates of infant temperament ratings and of mother-infant attachment: a longitudinal study. Infant Mental Health Journal, 10: 273-289.

Frodi, A.M. & Lamb M.E. (1980) Child abusers’ responses to infant smiles and cries. Child Development, 51: 238-241.

Frodi, A. & Senchak, M. (1990) Verbal and behavioral responsiveness to the cries of atypical infants. Child Development, 61: 76-84.

Fuller, B.F. (1991) Acoustic discrimination of three types of infant cries. Nursing Research, 40: 156-160.

Fuller, B.F. & Horii, Y. (1988) Spectral energy distribution in four types of infant cry. Journal of Communication Disorders, 21: 251-261.

Furlow, F.B. 1997. Human neonatal cry quality as an honest signal of fitness. Evolution and Human Behavior, 18:175-193.

Gauriau, C. & Bernard, J.F. (2002) Pain pathways and parabrachial circuits in the rat. Experimental Physiology, 87: 251-258.

Godfray, H.C.J. (1991) Signalling of need by offspring to their parents. Nature, 352: 328-330.

Godfray, H.C.J. (1995a) Evolutionary theory of parent-offspring conflict. Nature, 376:133-138.

Godfray, H.C.J. (1995b) Signaling of need between parents and young: parent-offspring conflict and sibling rivalry. American Naturalist, 146: 1-24.

Godfray, H.C.J. & Johnstone, R.A. (2000) Begging and bleating: the evolution of parent-offspring conflict. Philosophical Transactions of the Royal Society of London, Series B, 355: 1581-1591.

Golub, H.L. & Corwin, M.J. (1985) A physioacoustic model of the infant cry. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.G.Z. Plenum Press, New York.

Gonzalvo, G.O. (2002) Maltreatment of children with disabilities: characteristics and risk factors. Annales Espanoles de Pediatrica, 56: 219-223.

Gormally, S. & Barr, R.G. (1997) Of clinical pies and clinical cues: proposal for a clinical approach to complaints of early crying and colic. Ambulatory Child Health, 3: 137-153.

Grafen, A. (1990a) Sexual selection unhandicapped by the Fisher process. Journal of Theoretical Biology, 144: 473-516.

Grafen A. (1990b) Biological signals as handicaps. Journal of Theoretical Biology, 144: 517-546.

Green, J.A., & Gustafson, G.E. (1983) Individual recognition of human infants on the basis of cries alone. Developmental Psychobiology, 16: 485-493.

Green, J.A., Gustafson, G.E., Irwin, J.R., Kalinowski, L.I., & Wood, R.W. (1995) Infant crying: acoustics, perception and communication. Early Development and Parenting, 4: 161-175.

Green, J.A., Irwin, J.R., & Gustafson, G.E. (2000) Acoustic cry analysis, neonatal status and long-term developmental outcomes. In: Crying as a Sign, a Symptom and a Signal. Clinical, emotional and developmental aspects of infant and toddler crying, ed. Barr, R.G., Hopkins, B., & Green, J.A. Mac Keith Press, London.

Gustafson, G.E., & Green, J.A. (1989) On the importance of fundamental frequency and other acoustic features in cry perception and infant development. Child Development, 60: 772-780.

Gustafson, G.E., & Green, J.A. (1991) Developmental coordination of cry sounds with visual regard and gestures. Infant Behavior and Development, 14: 51-57.

Gustafson, G.E., Wood, R.M., & Green, J.A. (2000) Can we hear the causes of infants’ crying? In: Crying as a Sign, a Symptom and a Signal. Clinical, emotional and developmental aspects of infant and toddler crying, ed. Barr, R.G., Hopkins, B. & Green, JA. Mac Keith Press, London.

Hagen, E.H. (1999) The functions of postpartum depression. Evolution and Human Behavior, 20:325-359.

Hagen, E.H. (2002) Depression as bargaining: the case postpartum. Evolution and Human Behavior, 23: 323-336.

Hashima, P.Y. & Amato, P.R. (1994) Poverty, social support, and parental behavior. Child Development, 65: 394-403.

Hofer, M.A. (1996) Multiple regulators of ultrasonic vocalization in the infant rat. Psychoneuroendocrinology, 21: 217.

Hofer, M.A. (2002) Unexplained infant crying: an evolutionary perspective. Acta Paediatrica, 91: 491-496.

Hoffman, M.L. (1975) Developmental synthesis of affect and cognition and its implications for altruistic motivation. Developmental Psychology, 5:607-622.

Hoogland, J.L. (1994) Nepotism and infanticide among prairie dogs. In: Infanticide and Parental Care, ed. Parmigiani, S. & vom Saal, F.S. Harwood Academic Publishers, Chur, Switzerland.

Hrdy, S.B. (1994) Fitness tradeoffs in the history and evolution of delegated mothering with special reference to wet-nursing, abandonment, and infanticide. In: Infanticide and Parental Care, ed. Parmigiani, S. & vom Saal, F.S. Harwood Academic Publishers, Chur, Switzerland.

Hrdy, S.B. (1999) Mother Nature: maternal instincts and how they shape the human species. Ballantine Books, New York

Hunziker, U.A. & Barr, R.G. (1986) Increased carrying reduces infant crying: a randomized controlled trial. Pediatrics, 77: 641-648.

Irons, W. (1998) Adaptively relevant environments versus the environment of evolutionary adaptedness. Evolutionary Anthropology, 6: 194-204.

Iwasa, Y., Pomiankowski, A., & Nee, S. (1991) The evolution of costly mate preferences II. The "handicap" principle. Evolution, 45: 1431-1442.

Johnston, C.C. & Strada, M.E. (1986) Acute pain response in infants: a multidimensional description. Pain, 24: 373-382.

Johnstone, R.A. (1997) The evolution of animal signals. In: Behavioural Ecology: an Evolutionary Approach, eds. Krebs, J.R. & Davies, N.B. Blackwell Science, London.

Johnstone, R.A. & Grafen, A. (1993) Dishonesty and the handicap principle. Animal Behaviour, 46: 759-764.

Kanabar, D., Randhawa, M., & Clayton, P. (2001) Improvement of symptoms in infant colic following reduction of lactose load with lactase. Journal of Human Nutrition and Dietetics, 14: 359-363.

Kilner, R. & Johnstone, R.A. (1997) Begging the question: are offspring solicitation behaviours solicitations of need? Trends in Ecology and Evolution, 12:11-15.

Kirchner, J.A. (1993) The vertebrate larynx: adaptations and aberrations. Laryngoscope, 103: 1197-1201.

Kirjavainen, J., Jahnukainen, T., Huhtala, V., Lehtonen, L., Kirjavainen, T., Korvenranta, H., Mikola, H., & Kero, P. (2001) The balance of the autonomic nervous system is normal in colicky infants. Acta Paediatrica, 90:250-254.

Klevens, J., Bayon, M.C., Sierra, M. (2000) Risk factors and context of men who physically abuse in Bogota, Columbia. Child Abuse and Neglect, 24: 323-332.

Konner, M.J. & Worthman, C. (1980) Nursing frequency, gonadal function, and birth spacing among !Kung hunter gatherers. Science, 207: 788-791.

Kotch, J.B., Browne, D.C., Dufort, V., Winsor, J., and Catellier, D. (1999) Predicting child maltreatment in the first four years of life from characteristics assessed in the neonatal period. Child Abuse and Neglect, 23: 305-319.

Laitman, J.T. & Reidenberg, J.S. (1997) The human aerodigestive tract and gastroesophageal reflux: an evolutionary perspective. The American Journal of Medicine, 103: 2-8.

Lamb, M., Thompson, R., Gardner, W., & Charnov, E. (1985) Infant-Mother Attachment: The origins and developmental significance of individual differences in strange situation behavior. Lawrence Erlbaum, New Jersey.

Larson, C.R., Ortega, J.D. & DeRosier, E.A. (1988) Studies of the relation of the midbrain periaqueductal grey, the larynx and vocalization in awake monkeys. In: The Physiological Control of Mammalian Vocalization, ed. Newman, J.D. Plenum Press, New York.

Larsson, M., Aneblom, G., Odlind, V., & Tyden, T. (2002) Reasons for pregnancy termination, contraceptive habits and contraceptive failure among Swedish women requesting an early pregnancy termination. Acta Obstetricia et Gynecologica Scandinavica, 81: 64-71.

LeCuyer-Maus, E.A. (2003) Stress and coping in high-risk mothers: difficult life circumstances, psychiatric-mental health symptoms, education, and experiences in their families of origin. Public Health Nursing, 20: 132-145.

LeDoux, J. (1996) The Emotional Brain: The mysterious underpinnings of emotional life. Simon and Schuster, New York.

Lee, K. (2000) Crying patterns of Korean infants in institutions. Child: Care, Health and Development, 26: 217-228.

Lehtonen, L., Gormally, S., & Barr R.G. (2000) ‘Clinical pies’ for etiology and outcome in infants presenting with early increased crying. In: Crying as a Sign, a Symptom and a Signal. Clinical, emotional and developmental aspects of infant and toddler crying, ed. Barr, R.G., Hopkins, B. & Green, J.A. Mac Keith Press, London.

Lester, B.M. (1985) Introduction: there’s more to crying than meets the ear. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.G.Z. Plenum Press, New York.

Lester, B.M. (1987) Developmental outcome prediction from acoustic cry analysis in term and preterm infants. Pediatrics, 80: 529-534.

Lester, B.M.. (1997) Definition and diagnosis of colic. In: Colic and excessive crying: report of the 105^th Ross Conference on Pediatric Research, ed. Sauls, H.S. & Redfern, D.E. Ross Production Division, Abbott Laboratories, Columbus.

Lester, B.M., Boukydis, C.F.Z., Garcia-Coll, C.T., Hole, W., & Peucker, M. (1992) Infantile colic: acoustic cry characteristics, maternal perception of cry, and temperament. Infant Behavior and Development, 15: 15-26.

Lester, B.M., Boukydis, C.F.Z., Garcia-Coll, C.T., Peucker, M., McGrath, M.M., Vohr, B.R., Brem, F., & Oh, W. (1995) Developmental outcome as a function of the goodness of fit between the infant’s cry characteristics and the mother’s perception of her infant’s cry. Pediatrics, 95: 516-521.

Levitzky, S. & Cooper, R. (2000) Infant colic syndrome – maternal fantasies of aggression and infanticide. Clinical Pediatrics, 39: 395-400.

Livshits, G. & Kobyliansky, E. (1991) Fluctuating asymmetry as a possible measure of developmental homeostasis in humans: a review. Human Biology, 63: 441-466.

Lorberbaum, J.P., Newman, J.D., Horwitz, A.R., Dubno, J.R., Lydiard, R.B., Hammer, M.B., Bohning, D.E., and George, M.S. (2002) A potential role for thalamocingulate circuitry in human maternal behavior. Biological Psychiatry, 51: 431-445.

Lummaa, V., Vuorisalo, T., Barr, R.G., & Lehtonen, L. (1998) Why cry? Adaptive significance of intensive crying in human infants. Evolution and Human Behavior, 19:193-202.

Lycett, J.E., & Dunbar, R.I.M. (1996) Abortion rates reflect the optimization of parental investment strategies. Proceedings of the Royal Society of London, Series B, 266:2355-2358.

Lyon, B.E., Eadle, J.M., Hamilton, L.D. (1994) Parental choice selects for ornamental plumage in American coot chicks. Nature, 371: 240-243.

Mann, J. (1992) Nurturance or negligence: maternal psychology and behavioral preference among preterm twins. In: The Adapted Mind: Evolutionary psychology and the generation of culture, ed. Barkow, J.H., Cosmides, L., & Tooby, J. Oxford University Press.

Mappes, J, Mappes, T, Lappalainen, T. (1997) Unequal investment in offspring quality in relation to predation risk. Evolutionary Ecology, 11: 237-243.

Maynard Smith, J. (1991a) Honest signalling: the Philip Sidney game. Animal Behaviour, 42: 1034-1035.

Maynard Smith, J. (1991b) Theories of sexual selection. Trends in Ecology and Evolution, 6: 146-151.

McClain, P.W., Sacks, J.J., Froehlke, R.G., & Ewigman, B.G. (1993) Estimates of fatal child abuse and neglect, United States, 1979 through 1988. Pediatrics, 91: 338-343.

Mende, W., Herzel, H., & Wermke, K. (1990) Bifurcations and chaos in newborn infant cries. Physics Letters A, 145: 418-424.

Mendoza, S.P. & Ruy, J.D. (2001) The beginning of an alternative view of the neurobiology of emotion. Social Science Information, 40: 39-60.

Michelsson, K., Eklund, K., Leppanen, P., & Lyytinen, H. (2002) Cry characteristics of 172 1- to 7-day-old infants. Folia Phoniatrica et Logopaedica, 54: 190-200.

Moller, A.P. & Swaddle J. (1997). Asymmetry, Developmental Stability and Evolution. Oxford University Press.

Muller, E., Hollien, H., Murry, T. (1974) Perceptual responses to infant crying: identification of cry types. Journal of Child Language, 1: 89-95.

Murray, A.D. (1979) Infant crying as an elicitor of parental behavior: an examination of two models. Psychological Bulletin, 86:191-215.

Murray, A.D. (1985) Aversiveness is in the mind of the beholder. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Nelson, R.J. (2000) An Introduction to Behavioral Endocrinology. Sinauer Associates, Sunderland.

Newcomb, M.D. & Loeb, T.B. (1999) Poor parenting as an adult problem behavior: general deviance, deviant attitudes, inadequate family support and bonding, or just bad parents? Journal of Family Psychology, 13: 175-193.

Newman, J.D. (1985) The infant cry of primates: an evolutionary perspective. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis C.F.Z. Plenum Press, New York.

Newman, J.D. (1994) Crying in infants. In: Principles and Practice of Pediatrics. 2^nd Edition, ed. Oski, F.A., DeAngelis, C.D., Feigin, R.D., McMillan, J.A., & Warshaw, J.B. Lippincott Company, Philadelphia.

Newman, J.D. & Symmes, D. (1982) Inheritance and experience in the acquisition of primate acoustic behavior. In: Primate Communication, ed. Snowdon, C.T., Brown, C.H., & Petersen, M.R. Cambridge University Press.

Ostwald, P.F. & Murry, T. (1985) The communicative and diagnostic significance of infant sounds. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Overpeck, M.D., Brenner, R.A., Trumble, A.C., Trifletti, L.B., & Berendez, H.W. (1998) Risk factors for infant homicide in the United States. New England Journal of Medicine, 339:1211-1216.

Panksepp, J. (1998) Affective Neuroscience: The foundations of human and animal emotions. Oxford University Press.

Panksepp, J., Normansell, L., Herman, B., Bishop, P., & Crepeau, L. (1988) Neural and neurochemical control of the separation distress call. In: The Physiological Control of Mammalian Vocalization, ed. Newman, J.D. Plenum Press, New York.

Pitt, S.E. & Bale, E.M. (1995) Neonaticide, infanticide, and filicide: a review of the literature. Bulletin of the American Academy of Psychiatry and Law, 23: 375-386.

Porges, S.W. (1995) Cardiac vagal tone: a physiological index of stress. Neuroscience and Biobehavioral Reviews, 19: 225-233.

Porges, S.W. (1997) Emotion: an evolutionary by-product of the neural regulation of the autonomic nervous system. Annals of the New York Academy of Sciences, 807:62-77.

Porter, F.L., Miller, R.H., & Marshall, R.E. (1986) Neonatal pain cries: effect of circumcision on acoustic features and perceived urgency. Child Development, 57: 790-802.

Porter, F.L., Porges, S.W., & Marshall, R.E. (1988) Newborn pain cries and vagal tone: parallel changes in response to circumcision. Child Development, 59: 495-505.

Protopapas, A. & Eimas, P.D. (1997) Perceptual differences in infant cries revealed by modifications of acoustic features. Journal of the Acoustical Society of America, 102: 3723-3734.

Quillfeldt, P. (2002) Begging in the absence of sibling competition in Wilson’s storm-petrels, Oceanitus oceanicus. Animal Behaviour, 64: 579-587.

Rao, M., Blass, E.M., Brignol, M.M., Marino, L., & Glass, L. (1997) Reduced heat loss following sucrose ingestion in premature and normal human newborns. Early Human Development, 48:109-116.

Royle, N.J., Hartely, I.R., & Parker, G.A. (2002) Begging for control: when are offspring solicitation behaviours honest? Trends in Ecology and Evolution, 17: 434-440.

Saino, N., Ninni, P., Calza, S., Martinelli, R., De Bernardi, F., & Moller, A.P. (2000) Better red than dead: carotenoid-based mouth coloration reveals infection in barn swallow nestlings. Proceedings of the Royal Society of London, Series B, 267: 57-61.

Scrimshaw, C.M. (1984) Infanticide in human populations: societal and individual concerns. In: Infanticide. ed. Hausfater, G. & Hrdy, S.B. Aldine, New York.

Sherman, M. (1927) The differentiation of emotional responses in infants. II. The ability of observers to judge the emotional characteristics of the crying of infants, and of the voice of an adult. Journal of Comparative Psychology, 7: 335-351.

Sibert, J.R., Payne, E.H., Kemp, A.M., M. Barber, Rolfe, K., Morgan, R.J.H., Lyons, R.A., and Butler, I. (2002) The incidence of severe physical child abuse in Wales. Child Abuse and Neglect, 26: 267-276.

Sidebotham, P., Heron, J., Golding, J., The ALSPAC Study Team. (2002) Child maltreatment in the “Children of Nineties:” deprivation, class, and social networks in a UK sample. Child Abuse and Neglect, 26: 1243-1259.

Sijmen, A.R., Brugman, E., & Hirasing, R.A. (2001) Excessive infant crying: the impact of varying definitions. Pediatrics, 108: 893-897.

Sirvio, P. & Michelsson, K. (1976) Sound-spectrographic cry analysis of normal and abnormal newborn infants. Folia Phoniatrica, 28:161-173.

Sondergaard, C., Skajaa, E., & Henriksen, T.B. (2000) Fetal growth and infantile colic. Archives of Disease in Childhood, 83: 44-47.

Sondergaard, C., Henriksen, T.B., Obel, C., & Wisborg, K. (2001) Smoking during pregnancy and infantile colic. Pediatrics, 108: 342-346.

St. James-Roberts, I. (1999) What is distinct about infants’ ‘colic’ cries? Archives of Disease in Childhood, 80: 56-62.

St. James-Roberts, I., Conroy, S., & Wilsher, C. (1998) Stability and outcome of persistent infant crying. Infant Behavior and Development, 21: 411-435.

St. James-Roberts, I., Bowyer, J., Varghese, S., & Sawdon, J. (1994) Infant crying patterns in Manali and London. Child: Care, Health and Development, 20: 323-337.

St. James-Roberts, I. & Halil, T. (1991) Infant crying patterns in the first year: normal community and clinical findings. Journal of Child Psychology and Psychiatry, 32: 951-968.

St. James-Roberts, I., Hurry, J., Bowyer, J., & Barr, R.G. (1995) Supplementary carrying compared with advice to increase responsive parenting as interventions to prevent persistent infant crying. Pediatrics, 95: 381-388.

Stroud, J. & Pritchard, C. (2001) Child homicide, psychiatric disorder and dangerousness: a review and an empirical approach. British Journal of Social Work, 31: 249-269.

Thornhill, R. & Gangestad, S.W. (1999) Facial attractiveness. Trends in Cognitive Sciences, 3: 452-460.

Tortosa, F.S. & Redondo, T. (1992) Motives for parental infanticide in white storks, Ciconia ciconia. Ornis Scandinavica, 23: 185-189.

Trivers, R.L. (1974) Parent-offspring conflict. American Zoologist, 14: 249-264.

Urrutia, L.P. & Drummond, H. (1990) Brood reduction and parental infanticide in Heerman;s gull. Auk, 107: 772-774.

Vinson, T., Baldry, E. & Hargreaves, J. (1996) Neighbourhoods, networks, and child abuse. British Journal of Social Work, 26: 523-543.

Vogt, B.A. & Barbas, H. (1988) Structure and connections of the cingulate vocalization region in the rhesus monkey. In: The Physiological Control of Mammalian Vocalization, ed. Newman, J.D. Plenum Press, New York.

Voland, E. & Stephan, P. (2000) "The hate that love generated" – sexually selected neglect of one’s own offspring in humans. In: Infanticide by Males and its Implications, ed. van Schaik, C.P. & Janson, C.H. Cambridge University Press.

Wasz-Hockert, O., Lind, J., Vuorenkoski, V., Partanen, T., & Velanne, E. (1968) The Infant Cry: a spectrographic and auditory analysis. Clinics in Developmental Medicine, No. 29. Suffolk, Spastics International Medical Publications.

Wasz-Hockert, O., Michelsson, K., & Lind, J. (1985) Twenty-five years of Scandinavian cry research. In: Infant Crying: theoretical and research perspectives, ed. Lester, B.M. & Boukydis, C.F.Z. Plenum Press, New York.

Wessel, M.A., Cobb, J.C., Jackson, E.B., Harris, G.S., & Detwiler, A.C. (1954) Paroxysmal fussing in infancy, sometimes called "colic". Pediatrics, 14: 421-434.

White, B.P., Gunnar, M.R., Larson, M.C., Donzella, B., Barr, R.G. (2000) Behavioral and physiological responsivity, sleep, and patterns of daily cortisol production in infants with and without colic. Child Development, 71:862-877.

Wiesenfeld, A.R., Malatesta, C.Z., & DeLoach, L.L. (1981) Differential parental response to familiar and unfamiliar infant distress signals. Infant Behavior and Development, 4: 281-295.

Williams, C.G. (1966) Adaptation and Natural Selection. Princeton University Press, Princeton, New Jersey.

Wilson-Oyelaran, E.B. (1989) The ecological model and the study of child abuse in Nigeria. 1989. Child Abuse and Neglect, 13: 379-387.

Winterhalder, B. & Smith, E.A. (2000) Analyzing adaptive strategies: human behavioral ecology at twenty-five. Evolutionary Anthropology, 9:51-72.

Wolff, P.H. (1969) The natural history of crying and other vocalizations in early infancy. In: Determinants of Infant Behavior, Volume 4, ed. Foss, B.M. Methuen & Co., London.

Wood, R.M. & Gustafson, G.E. (2001) Infant crying and adults’ anticipated caregiving responses: acoustic and contextual influences. Child Development, 72: 1287-1300.

Zahavi, A. (1975) Mate selection – a selection for a handicap. Journal of Theoretical Biology, 53: 205-214.

Zahavi, A. (1977) The cost of honesty (further remarks on the handicap principle). Journal of Theoretical Biology, 67: 603-605.

Zeifman, D.M. (2001) An ethological analysis of human infant crying: answering Tinbergen’s four questions. Developmental Psychobiology, 39: 265-285.

Zeskind, P.S. (1980) Adult responses to cries of low and high risk infants. Infant Behavior and Development, 3: 167-177.

Zeskind, P.S. (1987) Adult heart-rate responses to infant cry sounds. British Journal of Developmental Psychology, 5: 73-79.

Zeskind, P.S. & Barr, R.G. (1997) Acoustic characteristics of naturally occuring cries of infants with "colic". Child Development, 68: 394-403.

Zeskind, P.S., Klein, L., & Marshall, T.R. (1992) Adults’ perceptions of experimental modifications of durations of pauses and expiratory sounds in infant crying. Developmental Psychology, 28: 1153-1162.

Zeskind, P.S. & Marshall, T.R. (1988) The relation between variations in pitch and maternal perceptions of infant crying. Child Development, 59: 193-196.

Zeskind, P.S. & Shingler, E.A. (1991) Child abusers’ perceptual responses to newborn infant cries varying in pitch. Infant Behavior and Development, 14: 335-347.

Zlotogora, J. (2002) Parental decisions to abort or continue a pregnancy with an abnormal finding after an invasive prenatal test. Prenatal Diagnosis, 22: 1102-1106.

Table 1: Rationales for infanticide in a cross-cultural sample.

Rationale	Number of mentions^a
Infant quality	21
Deformed or very ill	21

Paternal support in question	41
Adulterous conception, nontribal sire, sired by mother’s first husband	20
Mother unwed	14
No male support	6
Quarrel with husband	1

Inauspicious circumstances	35
Twins	14
Birth too soon or too many children	11
Mother died	6
Economic hardship	3
Wrong season	1

Other (e.g. female infant, rituals, incest)	15

^a Number of times rationales for infanticide were mentioned across a randomly chosen sample of 35 ethnographies in which the circumstances surrounding infanticide were discussed (adapted from Table 3.1 in Daly & Wilson 1988).

Table 2: Normal infant cry characteristics

Acoustic parameter	Values	Source

Duration (sec)	0.4 – 0.9 1.4 ± 0.6	Newman 1985 Michelsson et al. 2002
Fundamental frequency range (Hz)	200 – 600 250 – 450 300 – 600 400 – 600 450 – 600 496 ± 95	Porter et al. 1988 Wolff 1969 Furlow 1997 Sirvio & Michelsson 1976 Crowe & Zeskind 1992 Michelsson et al. 2002
Melody	Falling or rising / falling fundamental frequency	Wolff 1969; Furlow 1997; Michelsson et al., 2002

Table 3: Pathological conditions of infants and associated acoustic characteristics of infant cries.

CONDITION	Fundamental frequency (Hz)	Bi-phonation	Melody	Glide	Other

Asphyxiation – peripheral^a	High pitch mean max: 1000	>20%	Rising, or Falling / Rising >30%	>10%
Asphyxiation – central^a	High pitch Mean max: 1460	>20%	Rising, or Falling / Rising >30%	>10%
Bacterial meningitis ^a,b,c	High pitch Mean max: 750-1000	49%	Rising, or Falling / Rising 24%	11%
Cardio-pulminary disorder^b	High pitch	Present
Chromosome 13 & 18 trisomy^b	Low pitch		Flat
Chromosome 4 & 5 abnormalities^b	High pitch
Congenital hypo-thyroidism^a	Low Pitch Mean min: 270 Mean max: 470
Cri-du-chat syndrome ^a,c	High pitch, Mean: 600-1000		Flat (36%) Rising (23%)
Cocaine exposure in utero^b	High pitch
Colic^b	High pitch
Diabetic mother ^a,b	High pitch Mean max: 1480
Down’s syndrome^a,b	Low pitch Mean min: 270 Mean max: 510		Flat (63%)		Long duration mean: 4.5 sec
Krabbe’s disease (mal-formation)^a	High pitch Mean min: 590 Mean max: 1120		Rising/ Falling (27%)
Herpes simplex viral encephalitis ^a,b	High pitch	Present		Present	noise con-centration
Hydro- cephalus ^a,b,c	High pitch Mean min: 430 Mean max: 750-970	14%	Flat	8%
Hyper-bilirubinemia^a	High pitch Mean min: 960 Mean max: 2120	49%			Furcation
Hypo-glycemia^a,b	High pitch mean max: 1600	67%		18%	Vibrato
Lead Exposure^b	High pitch
Low birth weight ^b	High pitch
Metabolic disorders^b	High pitch
Neurological symptoms^b	High pitch
Obstetric/ delivery complications^b	High pitch				Short duration
Premature birth ^a	High pitch mean min: 570 mean max: 1360
SIDS^d	High pitch

a: Wasz-Hockert et al. 1985; b: Furlow 1997; c: Newman 1994; d: Corwin et al. 1996.

Appendix A: Glossary of acoustic terms^a

Term	Definition
Cry	Vocalization during a single expiration
Duration	Total time of vocalization during single expiration
Fundamental frequency (F₀)	Lowest frequency component of the cry, generally perceived as pitch, measured in hertz
Maximum F₀	Highest hertz value of a cry
Minimum F₀	Lowest hertz value of a cry
Mean F₀	Average hertz value across a cry
Harmonics	Component frequencies of the cry higher than the fundamental frequency
Rising Melody	F₀ rises across cry
Falling Melody	F₀ falls across cry
Flat Melody	F₀ does not change across cry
Biphonation	Simultaneously produced fundamental frequencies (also: bifurcation)
Dysphonation	Aperiodic vibration of vocal cords; turbulence generated at vocal cords
Hyperphonation	High F₀ (1000-2000 Hz)
Furcation	F₀ of strong cry splits into >1 weak cries each with its own pitch
Glide	Rapid rise or fall in F₀ (600 Hz / 0.1 sec)
Vibrato	Rapid falling and rising F₀
Glottal roll	Low pitched sound at end of cry
Noise concentration	High energy peaks in cry (2000-2500 Hz)

a: From Denes & Pinson 1963; Corwin et al. 1996; Furlow 1997; Green et al. 2000.

Footnotes

¹ Throughout the paper I emphasize cross-cultural ethnography. The way of life in traditional societies, in particular hunter-gatherer societies, is the best extant representation of the environment in which most of human evolutionary history unfolded, the so-called ‘environment of evolutionary adaptedness’, or EEA (Bowlby 1969; Barkow et al. 1992). This is not to say that traditional societies of modern times are perfect representations of all earlier human societies. The concept of the EEA has been criticized for painting a too homogeneous picture of the 2 million years that it spans, and for not recognizing the possibility of more recent evolutionary change in humans (Irons 1998). Nevertheless, a cross-cultural approach offers the most inclusive view of the human condition, including that of our evolutionary past, and I follow that approach here.

² In the "condition-dependent handicap" described in the text, low quality infants simply cannot produce the signal, or cannot produce it as conspicuously, as high quality infants. In the related "strategic choice handicap", low and high quality infants can produce any level of the signal, but the marginal cost at the same level of signal is higher for low quality infants than it is for high quality infants. For a low quality infant, then, the marginal cost of producing a certain level of signal may be higher than the marginal benefit of reducing the probability of infanticide, and so no increase in the signal level would be favored. For a higher quality infant, on the other hand, the marginal cost of producing the same level of signal may be less than the marginal benefit of reducing the probability of infanticide, and an increase in signaling would be favored (see Grafen 1990b). In both scenarios, parents can use the signal to evaluate infant quality. Additionally, there can still be some cheating in ‘honest signaling’ systems, as emphasized in the ‘manipulation’ models described in the text, but signaling is reliable if most signals are honest (Johnstone & Grafen 1993; Johnstone 1997).

³ According to this view, children develop secure attachments when mothers are available, responsive and accepting, and develop (possibly different forms of) insecure attachment when mothers are not, although child temperament may also contribute to the attachment process. These different forms of attachment themselves may be adaptive responses to specific rearing environments (Lamb et al. 1985; Chisholm 1996; Hrdy 1999; Zeifman 2001).

⁴ Infant crying that mimicked need in the absence of autonomic arousal could be a manipulative signal to gain extra parental care or attention. The role of acute and chronic stress in determining early infant cry acoustics dominates the empirical literature, however, so it is unclear what amount of variation in infant cry acoustics, if any, remains unexplained by such autonomic arousal.

Ackowledgements

I benefited from commentary on an initial draft by Ronald G. Barr, Barbara DeVinney, John D. Newman, Joseph Manson, and Stephen Suomi, and from enlightening discussions with Ronald G. Barr. I thank eight referees including Johannes Lehtonen, Virpi Lummaa, and John D. Newman, whose comments resulted in substantial improvements to the paper. I received funding from an Intramural Training Award at the National Institutes of Health.