Behavior of infant chimpanzees during the night in the first 4 months of life: smiling and suckling in relation to behavioral state.
Yuu Mizuno, Hideko Takeshita, Tetsuro Matsuzawa
This article reports the behavior of 3 newborn chimpanzees in the first 4 months of life, reared by their mothers and living in a community of 14 chimpanzees in a semi-natural enriched environment. We focused on spontaneous activity during the night partly because sleeping behavior constitutes an essential part of the infants' activity. Observation during the night also had the advantage of keeping the influence of the mothers' activity as well as the environmental stimulation constant throughout the observation period. We report several interesting findings. Behavioral states defined through overt features such as open or closed eyes were variable during the night, with the rapid eye movement (REM) and non-REM sleep patterns alternating much as they do in human infants. Although crying is one of the distinctive behavioral states in the case of human infants, the chimpanzee infants did not cry like humans. Suckling behavior was often accompanied by open eyes until the end of the first 2 months. Thereafter, suckling with the eyes closed became more prominent. Although there were no explicit stimuli, the newborns showed neonatal smiling with the eyes closed during REM sleep periods. However, neonatal smiling disappeared within the first 2 months and was replaced by social smiling with open eyes. Taken together, the results suggest a strong similarity between human infants and chimpanzee infants in terms of developmental changes in spontaneous activities at around 2 months of age.
Human infants sleep approximately 16 hr or more every day, from just after birth until the age of 8 months. Sleep periods alternate with waking periods and are distributed throughout the day and night (Parmelee, Wenner, & Schultz, 1964). Later during development, sleep periods become more concentrated during the night while waking periods increase during the day. A classic study by Wolff (1959) involving naturalistic observation of the behavior (including sleep behavior) of human infants found stable and cyclic activity patterns. To describe the activity patterns, he proposed terms such as behavioral states. One can categorize the behavior of infants into several levels or states by focusing on behavioral features such as whether the eyes are open or closed, the presence or absence of eye or limb movements, and so forth. The behavioral states of infants follow stable and cyclic patterns within a day (Prechtl & Beinteima, 1964; Wolff, 1966). Behavioral states have since been used in various studies and have been confirmed as a valuable index in describing the behavior of human infants (Brazelton, 1987; Wolff, 1987). Behavioral states may also be applied to the study of nonhuman animals to provide comparative data based on a unitary scale.
Sleep can be seen as a change in behavioral states indicated by changes in overt behaviors. Spontaneous activity during sleep periods encompasses various kinds of behaviors in addition to the narrowly defined sleeping. The spontaneous activity of human neonates during the sleep period has been described in several important studies (Emde & Koening, 1969; Emde, McCartney, & Harmon, 1971; Korner, 1969; Takahashi, 1974; Wolff, 1959, 1963). For example, human neonates show spontaneous smiling or neonatal smiling, defined as a slow, gentle, sideward and upward pull of the mouth during the rapid eye movement (REM) sleeping phase (Wolff, 1959). This neonatal smiling occurs without known external causes or systematically demonstrable internal ones (Wolff, 1963). Although several researchers have observed neonatal smiling in humans (Freedman, 1965, 1974; Gewirtz, 1965; Messinger et al., 2002; Takahashi, 1973, 1974), there are no comparable data in nonhuman animals. As yet there is no evidence illuminating the evolutionary origins of spontaneous smiling in human neonates.
The study of chimpanzees may constitute the most fundamental approach to elucidating the evolutionary origins of human cognition and behavior. Differences in DNA sequences between humans and chimpanzees are about 1.2% to 1.4% (Fujiyama et al., 2002; International Chimpanzee Chromosome 22 Consortium, 2004). The two species are estimated to have last shared a common ancestor approximately five million years ago. However, there have been very few studies of spontaneous activity during sleep periods in infant chimpanzees. Plooij (1984) described the behavioral development of infant chimpanzees in the first 2.5 years of life based on behavioral observation during the daytime of wild chimpanzees at Gombe, a long-running research site in Tanzania (Goodall, 1986). According to the report, infant chimpanzees showed stable and cyclic activity patterns. The infants had their eyes open during 30% of the daytime observation period in the first week, which rose to 56% in Weeks 2 to 9. Changes in activity patterns did not depend on the mothers’ activity. Plooij (1984) also recorded suckling behavior and confirmed that the infants repeated the suckling-and-pause pattern. Usually, the duration of one bout on the nipple was short, only a few minutes (Mdn = 3 min). However, in some cases the infant would stay on the nipple for up to 10 min. None of the intervals lasted longer than 100 min.
Many studies have examined cognitive and behavioral development in infant chimpanzees after birth in captivity (Bard, 1994a, 1994b, 1998, 2000; Bard & Russell, 1999; Fouts, 1997; Hayashi & Matsuzawa, 2003; Hayes, 1951; Ladygina-Kohts, 2002; Myowa-Yamakoshi, Tomonaga, Tanaka, & Matsuzawa, 2003, 2004; Myowa-Yamakoshi, Yamaguchi, Tomonaga, Tanaka, & Matsuzawa, 2005; Redshow, 1989; Takeshita, 2001; Tomonaga et al., 2004). However, few studies have focused on sleeping behavior in infant chimpanzees in the laboratory. Balzamo, Bradley, Bradley, Pegram, and Rhodes (1972) conducted EEG recording during the sleep phase longitudinally. The study revealed that the chimpanzees showed stable and cyclic patterns of sleep and waking phases just like human infants. However, compared with data obtained in the wild, the isolated chimpanzee infants in the laboratory showed longer sleeping phase durations. In general, chimpanzees hand-raised by humans in the laboratory are often left to themselves and receive reduced stimulation. In contrast, mother-raised chimpanzees in the wild cling to their mothers and are embraced by them 24 hr a day. These infants inevitably move around and are exposed to much stimulation during the daytime. Taken together, a consideration of rearing conditions might be essential for understanding spontaneous activity in infant chimpanzees. Comparisons made between human infants raised by their mothers and chimpanzee infants not raised by their biological mothers will provide biased results and are simply unfair. Studies of mother-reared chimpanzees are thus necessary and important when assessing the similarities and differences between the two species (Matsuzawa, 2003).
This study aimed to illuminate behavioral development in infant chimpanzees by examining developmental changes in spontaneous activity during the night. The targets were three mother-infant pairs living in a captive community of chimpanzees. We conducted behavioral observation and video-recording during the night for the first 4 months of the infants’ lives. We focused on sleeping behavior in a broader sense because it comprised an essential part of the infants’ activity. Moreover, observation during the night had the advantage of keeping the influence of the mothers’ activity as well as environmental stimulation constant. It is extremely difficult or almost impossible to carry out nighttime field observation of chimpanzee infants in their natural habitat. It is not easy in the laboratory either. As far as we know, there have been no systematic observations of the spontaneous activity of mother-infant pairs during the night in a laboratory setting. In that sense, this study provides a unique way of obtaining data on infant chimpanzees that is comparable to that for human infants.
The subjects were three mother-infant pairs of chimpanzees (Pun troglodytes verus). One of the infants, Ayumu, was male, and the other two, Cleo and Pal, were female. All three infants were born between April 2000 and August 2000 at the Primate Research Institute of Kyoto University (KUPRI). They were members of a community of 14 chimpanzees of three generations who lived in an enriched environment simulating their natural habitat (Matsuzawa, 2003). Both human infants and chimpanzee infants need intensive maternal care. The major difference is that chimpanzee infants continue to cling to their mothers 24 hr a day, grasping on to body hair by their hands and feet. The grasping power of the feet is relatively weak so that chimpanzee mothers almost always support the infant by embracing it with one hand. In the first 3 months, the mother and her infant are never separated but remain physically attached. Then, the infant starts to stand on its limbs and gradually becomes independent from the mother. In general, chimpanzees are weaned at the age of 4, and the interbirth interval is about 5 years (Boesch & Boesch-Acherman, 2000; Goodall, 1986; Sugiyama, 2004). Maternal care until weaning thus continues much longer than it does in modem human societies.
All three of our infant subjects were raised by their biological mothers. This is not commonly the case among captive chimpanzees. According to records from zoos and other facilities, mothers reject or fail to raise their infants in about 50% of cases (Brent et al., 1996; Udono et al., 1999). In our laboratory, one mother, Ai, succeeded in raising her infant from birth, and the other two mothers, Chloe and Pan, needed human assistance and encouragement to hold their newborns correctly immediately after delivery. Chloe repeatedly refused her infant, Cleo, by pushing away her head when she tried to suckle. However, all three mothers eventually succeeded in rearing their infants. During the day, all three mother-infant pairs of chimpanzees participated in various kinds of cognitive tests, and spent the rest of their time in an enriched outdoor compound with other members of the community (Matsuzawa, 2003). The subjects’ care followed the KUPRI guidelines, and the study adhered to the Guide for the Use and Care for Nonhuman Primates by KUPRI.
Each mother-infant pair was guided to a separate indoor residential room (4 m x 8 m x 3 m) daily at 4:30 p.m., to spend the night alone with her infant, without interference from other members of the community. The room was equipped with a floor-heated bench and two raised wooden platforms. Temperatures inside the room were controlled by a central heating system to remain at about 25°C. Chimpanzees both in the wild and in the laboratory sleep for about 12 hr a day, from sunset until sunrise. The behavior of each mother-infant pair was intensively monitored and video-recorded 24 hr a day during the first 7 days.
This study utilizes data from observation through video monitors and video recordings for 6 hr from 5:OO p.m. to 11:OO p.m., every day during the first 7 days (from Day &Day 6) after birth. Thereafter, recording was conducted four times a week until the end of the second month (from Day 7-Day 60). Recording was reduced to twice a week in the third and fourth months. Thus, we compiled a record of postnatal change in behavior over the first 4 months (120 days) in the lives of each of the three infants
Two video cameras were used for observation, one of which was remotecontrolled and the other of which was fixed. Both were infrared-sensitive video cameras and the room was illuminated by infrared lights. Based on the video data, the first author judged the behavioral states of the infants every 5 min following an instantaneous sampling method. In parallel to the video recording, we also carried out all occurrence sampling based on direct observation: An account was kept describing all the events that occurred during the 6-hr observation period.
All video recordings from the infants’ first 120 days of life were subjected to the analysis. However, even with two cameras it was not always possible to record all of the subjects’ behavior in detail. In some cases, infants were fully occluded by the mother’s body. Therefore, for further analysis we used only those scenes in which the face of the infants was clearly visible. Using this criterion, the total recording time for analysis was 306 hr for Ayumu, 291 hr for Cleo, and 324 hr for Pal. The judging of behavioral states and analysis of spontaneous behaviors were conducted as follows.
Behavioral states were judged based on 5-min instantaneous sampling. The definition of each behavioral state followed the classification proposed by Prechtl(l974) and Prechtl and O’Brien (1982). There were five categories in total (Table 1): (a) regular sleep or non-REM sleep, (b) irregular sleep or REM sleep, (c) awake inactivity, (d) awake activity, and (e) crying. Behavioral states were thus judged by focusing on two major behavioral features: first, whether the eyes were open or closed, and second, whether gross movements were present or absent. Two additional features were used for classification: respiratory patterns (regular vs. irregular) and eye movement while the eyes were closed (present or absent). The fifth category, crying, is characterized by crying behavior, although this was not actually observed in chimpanzee infants.
Note. Behavioral states are based on the classification of behavioral features originally proposed by Prechtl(l974) and Prechtl & O’Brien (1982). 1 = regular sleep or non-REM sleep; 2 = irregular sleep or REM sleep; 3 = awake-inactivity; 4 = awake-activity; 5 = crying; na = not available because the presence or absence of the eye movement was checked only when the eyes were closed.
Recording of suckling behavior.
Suckling behavior was recorded by the all occurrence sampling method. The initiation of suckling behavior was defined as the moment when the infant’s mouth first made contact with the mother’s nipple. The termination of the behavior was recorded when the infant ceased nipple contact. The suckling period was thus defined as the period between initiation and termination. We measured the duration of suckling periods, as well as the intervals between them.
Recording of smiling
This article focuses on a particular facial expression (smiling), which was recorded by all occurrence sampling. In short, neonatal smiling (Wolff, 1959) can be defined as a slow, gentle, sideward and upward pull of the mouth without external causes in REM sleep. This facial expression is morphologically close to smiling. However, it must be noted that neonatal smiling is simply a label for a facial expression in the neonatal behavioral repertoire, and it does not imply any functional meaning. Neonatal smiling appears in the sleep phase when the eyes are closed. In contrast, human infants start to show social smiling at around 3 months of age. Social smiling occurs with the eyes open and is directed toward the person in front of the infant. We recorded the occurrence of both neonatal smiling and social smiling in the three chimpanzee infants.
Gross movement is one of the primary state variables. This is defined as the gross motor movement of at least one of the four limbs and does not include the movement of fingers and hands. The infants, keeping eyes open or closed, move the limbs in various ways. Regardless of the direction and volume, the limb movement was recorded as gross movement. This excludes the movement of the respiratory movement of body trunk. The head movement for rooting and suckling is also excluded because of the definition, but is usually accompanied by the limbs’ movement. Once the limbs moved, it was recorded as gross movement.
Reliability of Observation
We tested the reliability of our main observer’s judgment of behavioral state and the other measures through the following procedure. We randomly compiled video scenes from the first 4 weeks to create, for each subject, a sample video 60 min long. Two other observers were asked to view the sample video and make judgments on three criteria: behavioral states (Level 1-5), the presence or absence of eye movement, and the presence or absence of a smile as the subject’s facial expression. The observers recorded their evaluation every 30 sec, thus producing 120 judgments for each infant. The degree of agreement among the three experimenters (the first author and the other two observers) was 95.0% for behavioral states, 79.0% for eye movement, and 95.8% for facial expression.
We observed the behavior of chimpanzee infants during the night in the first 4 months after birth. During the study period, we carried out an average of 72 days of observation on each subject: 70 days on Ayumu, 64 days on Cleo, and 82 days on Pal. During the 6-hr observation period each night (5:OO p.m.-11:00 p.m.), an average of 4.4 hr of video recording in which we were able to obtain a clear image of the infants were collected. Based on the available data, this article addresses the following three major points: behavioral states, suckling behavior, and smiling.
There are five categories of behavioral states. However, we did not observe a single case of the final category, crying, in our subjects. Infant chimpanzees do not cry like human infants. Screaming, a distress call, of course forms part of chimpanzee infants’ vocal repertoire, much like other primate species including humans. One often hears such screaming in isolated chimpanzee infants reared by humans in a laboratory setting. However, infants in our study did not once emit screams; they continued to be embraced by their mothers 24 hr a day. They had no need to cry or scream like human infants, who in fact need to capture their mother’s attention to be embraced or to suckle. In contrast, infant chimpanzees cling to their mother by grasping her hair, root to find the nipples, and suckle by themselves. In some cases, the chimpanzee infants emitted a soft vocalization, a grunt, during the night. This invariably occurred when infants lost physical contact with the mother during sleep. In response to these vocalizations, the mothers tended to retrieve their infants.
|Observation||Behavioral States (%)|
Table 2 shows the ratio of each behavioral state in the first 4 months of life in each subject. We found slight differences among the infants: Cleo showed longer non-REM sleep and shorter REM sleep in comparison to the other two chimpanzees. On average, the infants showed non-REM sleep during 40% of the sampling time, REM sleep during 33%, awake inactivity during 9%, and awake activity during 18%. This means that the infants slept through approximately three quarters (73%) of the night (from 5 p.m.-11:00 p.m.) during the first 4 months.
FIGURE 1 Developmental change of the percentage of each behavioral state (or arousal levels) in each subject. Statistical analysis is as follows. Linear regression was calculated. Pearson's correlation coefficients were given in each state of each subject respectively: In Ayumu, +.92 for Level 1, -.53 for Level 2. -35 for Level 3, and -.70 for Level 4. In Cleo, +.I4 for Level 1, +.74 for Level 2, -.72 for Level 3, and -.35 for Level 4. In Pal, +.82 for Level 1, -.47 for Level 2, -.69 for Level 3, and -.39 for Level 4.
Figure 1 shows developmental changes in behavioral states in each infant chimpanzee in the first 4 months. The developmental changes are shown as the increases or decreases in the percentage of each behavioral state in each subject as a function of age. Based on the linear regression, Pearson’s correlation coefficients were calculated (see the details of statistical analysis in Figure 1). Examining differences in behavioral states during sleep shows that Ayumu and Pal’s patterns look similar and show a gradual increase in non-REM sleep along with age (determinant coefficients, R2, are .81 for Ayumu, and .67 for Pal, respectively). However, there was a clear individual difference in Cleo, who showed a lower ratio for the REM sleep phase and a higher ratio for non-REM sleep from the beginning. The linear regressions show that all three infants have the same increase or decrease tendencies of behavioral states except in the case of REM sleep: REM sleep decreased in Ayumu and Pal but increased in Cleo during the first 4 months. This might have been due to frequent rejection of nipple contact and of suckling by the mother especially at the early phase. After struggling to suckle for a while, Cleo often slept deeply (see the following section for a more detailed description).
Suckling is the most important behavior for the neonates' survival. The first observation of suckling behavior was recorded on Day 1 (the day following birth) in the case of Ayumu. The mother, Ai, showed typical and normal behavior during delivery (Matsuzawa & Nakamura, 2004). The neonate clung to the mother and the mother embraced him. The infant was able to root for the nipple and suckle by himself 18 hr after birth. In the case of Pal, the first suckling behavior appeared 29 hr after birth, and for Cleo 51 hr after birth. In the case of Cleo, the mother, Chloe, did not spontaneously embrace the infant and failed to do so even after extensive encouragement from humans: She just watched the infant carefully. Therefore, for the first night, the infant had to be looked after by human caretakers. The next day, Chloe was again encouraged to embrace her infant. The infant screamed repeatedly and the mother approached her closely. Finally, Chloe embraced Cleo just after the infant finally succeeded in grasping the mother’s hair. Chloe continued to embrace the infant thereafter but developed another problem: a tendency to reject the infant’s suckling attempts by pushing her head downward when she tried to make contact with the nipples. Human caretakers helped by providing the infant with additional milk from a bottle (Chloe was given her own bottle of milk at such times).
FIGURE 2 Frequency distribution of suckling duration. The durations of suckling periods were mostly 1 to 5 min, with maxima of less than 10 min in all of the subjects.
We evaluated suckling behavior in terms of duration and interval. Figure 2 shows the frequency distribution of suckling duration. Most suckling periods lasted between 1 and 5 min, with maxima below 10 min in all subjects. We noted slight individual differences: In Ayumu and Pal’s cases mode duration was in the range of 2 to 5 min, whereas Cleo showed relatively shorter suckling duration (mode was 1 min) due to the mother’s frequent rejections. The median duration was about 4 min, 28 sec in Ayumu (n = 247), 5 min, 0 sec in Pal (n = 164), but 1 min, 52 sec in Cleo (n = 191), respectively.
FIGURE 3 Frequency distribution of suckling intervals. An interval in this case was defined as the time from the start of one suckling period to the start of the next suckling period. The intervals between suckling periods were less than 2 hr at maximum, but showed no clear peaks except in the case of Cleo, with a predominance of shorter intervals.
Figure 3 illustrates the frequency distribution of suckling intervals. We defined a suckling interval as the time from the start of one suckling period until the start of the next suckling period. For all three infants, these intervals were less than 2 hr at maximum, but showed no clear peaks except in the case of Cleo. For Cleo, the majority of suckling intervals were short, again reflecting the mother’s rejecting behavior. In sum, the median suckling interval was about 45 min, 9 sec for Ayumu (n = 247), 60 min, 25 sec for Pal (n = 164), and 32 min, 11 sec for Cleo (n = 191), respectively.
FIGURE 4 Developmental changes of eyes open or eyes closed suckling in each infant as a function of age.
No clear developmental changes were apparent within the first 4 months in either the duration or the interval of suckling period. However, we noted an interesting behavioral difference over the course of development. Examining suckling behavior in relation to behavioral states, we found a significant relationship with eye opening (Figure 4). We checked whether the eyes were open or closed when the suckling behavior started. In all three subjects, suckling behavior was almost always (in more than 98% of instances) accompanied by open eyes in the first and the second months, whereas in the third and fourth months suckling behavior occurred with the eyes closed. The statistical analysis by Fisher’s exact probability test showed p < .001 in all three subjects. This shows that in the later stages performing suckling behavior does not require visual search and may therefore be easier for the infant to control.
Neonatal smiling was observed in two of our subjects, Ayumu and Pal (Figure 5). We recorded 23 cases of neonatal smiling in Ayumu and 37 cases in Pal. In all cases, the infants showed a slow, gentle, sideward and upward pull of the mouth without any obvious external cause, while their eyes remained closed.
There was a very clear correlation between the occurrence of neonatal smiling and behavioral state. In all 60 cases where neonatal smiling was observed in infant chimpanzees, it always occurred during REM sleep. It is interesting to note that rapid eye movement is highly correlated to dreaming in humans. Neonatal smiling did not occur during non-REM sleep. Cleo did not show any evidence of neonatal smiling. This may have been linked to the relatively low ratio of REM sleep phase in this subject.
FIGURE 5 Neonatal smiling in an infant chimpanzee, Pal, 16 days after birth.
Developmental changes in neonatal smiling were also very clear (Figure 6). The following is the statistical analysis for the age change in neonatal and social smiling. We split the ages in half (0-7 weeks, 8-15 weeks) and computed the mean smiling (frequency/hour) and did a 2 x 2 analysis of variance (ANOVA) with smiling type (neonatal vs. social) and age (young vs. old) as the two factors. It showed that neonatal smiling was frequent in infants less than 7 weeks old and social smiling was frequent in infants after 8 weeks old (p < .001). Ayumu showed neonatal smiling for the first time on Day 0 and Pal on Day 6. Interestingly, neonatal smiling occurred only in the first month in Ayumu and in the first 2 months in Pal, after which it disappeared completely in the third and fourth months. In contrast to the disappearance of neonatal smiling, social smiling4haracterized by smiling with the eyes open-began to increase dramatically at the age of 2 to 3 months. The transition from neonatal smiling to social smiling is also found in human infants at around 3 months of age.
FIGURE 6 Frequencies of neonatal smiling and social smiling as a function of age in weeks.
In this article we present empirical data on the behavior of infant chimpanzees during the night in their first 4 months of life. This study may be important for the following three reasons. First, the target of our observation was unique: We recorded the behavior of infants reared by their mothers in a seminatural enriched environment. Second, our study of behavior during the night brought forth new findings so far neglected in the literature, such as changes in behavioral states, suckling behavior at night, and neonatal smiling later replaced by social smiling. Third, these data clearly showed drastic developmental changes in behavior within the first 4 months of chimpanzees’ lives. In the following sections, we discuss each of these three points in turn.
The Study of Mother-Reared Chimpanzee Infants
This study focused specifically on infants being reared by their biological mothers from birth. Their behavior was not affected by human artifacts such as isolating them from the mother. For example, human rearing would have deprived them of physical contact and various kinds of stimulation, would have provided a time anchor through bottle-feeding, and so forth.
The mother-reared infants clung to their mother’s body at all times. However, human-reared chimpanzee infants are usually kept in a horizontal position most of the time, prone or supine, just like human infants. Chimpanzee infants are not stable in the supine posture and make spontaneous efforts to turn over by moving cross-lateral limbs alternatively. In turn, they may become exhausted from struggling and sleep longer than mother-reared infants: The proportion of REM sleep may be reduced, as suggested by this study. This may be the reason neonatal smiling in infant chimpanzees has not been detected in any previous laboratory work.
Spontaneous Behavior During the Night
Developmental changes in behavioral states during the night in our chimpanzee subjects were fundamentally comparable to those found in humans. Non-REM sleep periods increased with age while waking periods with eyes open decreased. One major difference between the two species concerned crying behavior, which was completely lacking in chimpanzee infants. The infant chimpanzees neither screamed nor cried like human infants. The chimpanzee mothers continued to embrace their infants at all times, even during the night, such that the infants did not need to struggle to gain access to the nipple. They occasionally emitted soft vocalizations called effort grunts, but they never cried in the way human infants do at night.
This study was also the first to record infant chimpanzees’ suckling behavior during the night. Suckling duration was 2 to 5 min on average (in median). Suckling occurred one to two times an hour on average, at intervals of no longer than 2 hr. The data on suckling durations and intervals were comparable to those obtained in the daytime in the wild. Moreover, we also found an interesting correlation between suckling behavior and behavioral states at night. During the first 2 months infants suckled with their eyes open, but later on they began to suckle with the eyes closed. This may indicate that the infants were becoming increasingly more skillful at rooting for and finding the nipples, without the need for actual visual search. As a result of their early efforts in the first 2 months, infants may become highly skilled at locating the nipples. The innate behavioral chain of clinging, rooting, and suckling by the infant becomes part of a coordinated interaction between mother and infant.
Neonatal smiling has so far been believed to be uniquely human. However, this study has provided evidence that this behavior is also characteristic of chimpanzees. Follow-up studies may reveal that neonatal smiling is rooted in a common primate ancestor beyond the human-chimpanzee lineage. The occurrence of neonatal smiling was also perfectly correlated with behavioral states: Infants showed neonatal smiling only in the REM sleep phase. Neonatal smiling disappeared within 2 months after birth, giving way to social smiling with open eyes. An overall scenario regarding the development of smiling may be proposed as follows. Infants have an innate tendency to alternate sleeping and waking phases uncoupled with the day-night cycle, and initially they show neonatal smiling in REM sleep during both day and night. However, as they continue to be exposed to the natural day-night cycle after birth, they begin to sleep more during the night. Moreover, smiling with the eyes open takes on a functional meaning in the context of mutual gaze, or eye-to-eye contact, with the mother during the day. Social smiling is thus reinforced and increases in frequency, whereas neonatal smiling during the night decreases and disappears. There may exist an underlying brain mechanism in which the reflexlike behavior controlled at the subcortical level becomes suppressed along with the maturation of the cortical level.
Developmental Shift at 2 Months of Age
The three mother-infant pairs in this study also served as subjects for a series of experimental observations on chimpanzee cognitive development. The observations were carried out in the daytime, overlapping with the period of this study (Tomonaga et al., 2004). The tests focused on topics such as neonatal imitation, face recognition, gaze preference, and mutual gaze.
Human neonates 12 to 21 days old can imitate human facial gestures without visual feedback of their own behavior (Meltzoff & Moore, 1977, 1983). Some researchers suggest that neonatal imitation in humans disappears or declines at approximately 2 to 3 months of age (e.g., Fontaine, 1984). Myowa-Yamakoshi et al. (2004) provided evidence that neonatal chimpanzees could discriminate between, and imitate, human facial gestures (tongue protrusion and mouth opening) at less than 7 days old. However, by the time they reached 2 months of age, the chimpanzees no longer imitated the gestures. The one exception was mouth opening: A demonstrator performing mouth opening continued to elicit the same action in the infants. This may have been caused by mouth-opening behavior resembling a form of social smiling, that is, a play face.
Studies on the development of face or nonface discrimination in human infants, using visual preference techniques, demonstrated that newborns preferentially track facelike patterns over nonface patterns (e.g., Goren, Starty, & Wu, 1975; Johnson, Dziurawiec, Ellis, & Morton, 1991). Myowa-Yamakoshi et al. (2005) assessed developmental changes in face recognition by the three infant chimpanzees in the first 4 months of life. The study used preferential looking procedures that measured the infants’ eye and head tracking of moving stimuli. They prepared photographs of the mother of each infant and an “average” chimpanzee face using computer graphics technology. Prior to 1 month of age, the infants showed few tracking responses and no differential responses. Between 1 and 2 months of age, they paid greater attention to their mother’s face. From 2 months onward, they again showed no differences, but exhibited frequent tracking responses. This study suggests that, through frequent face-to-face interaction, the chimpanzees learned to recognize their mother’s face at the age of 1 to 2 months.
Among facial features, the eyes are a significant area of focus for humans and nonhuman animals. By the age of 4 months, human infants are able to discriminate between direct and averted gaze (Fmoni, Johnson, Brockbank, & Simon, 2000; Vecera & Johnson, 1995). Myowa-Yamakoshi et al. (2003) studied gaze perception in the three infants, using a two-choice preferential-looking paradigm. The infants were presented with two photographs of human faces: (a) with eyes open or closed, and (b) with a direct or an averted gaze. They found that the chimpanzees preferred looking at the direct-gaze face over the averted-gaze face at the age of 2 months old and thereafter. Mutual gaze time between mother and infant chimpanzees also increased at around 2 months old (Bard et al., 2005).
This study has also uncovered various changes in the chimpanzees’ behavior at around 2 months of age: the shift from suckling with the eyes open to suckling with the eyes closed, and the shift from neonatal smiling to social smiling. These changes are thus correlated with those in neonatal imitation, face recognition, gaze preference, and mutual gaze as previously described. Johnson and Morton (1991) proposed that there was a developmental shift in visual processing from the subcortical visual pathway (called CONSPEC) to a second mechanism that appears in the plastic cortical visual pathway (called CONLERN), and that this shift occurs at around 2 months of age in human infants. The shift from the subcortical to the cortical level may not be limited to visual processing but could potentially be generalized to cognitive functions in other domains. This study, as well as related studies testing the same chimpanzee infants, may lend support to the notion that a similar shift of the early social-cognitive abilities along with cortical maturation is operating in chimpanzees.
This study was financially supported by the following grants from the Ministry of Education, Culture, Sports, Science, and Technology, Japan: #07 102010, 12002009, and 16002001 to Tetsuro Matsuzawa. Further support was provided by grants from the 21st Century Center of Excellence program of biodiversity (A14) and the Japan Society for the Promotion of Science core-to-core program HOPE (Primate Origins of Human Evolution]. We are grateful for the assistance and continuous support given by our colleagues, especially Masaki Tomonaga and Masayuki Tanaka. We also thank the staff and students participating in the project of cognitive development of chimpanzees 2000: Masako Myowa-Yamakoshi, Satoshi Hirata, Tomomi Ochiai, Claudia Sousa, An Ueno, Chisato Douke, Noe Nakashima, Midori Uozumi, Sanae Okamoto, Tomoko Imura, Misato Hayashi, Toyomi Matsuno, Nobuyuki Kawai, Akihiro Izumi, and Sumiharu Nagumo. Without their efforts, we would not have been able to obtain the data on the three infant chimpanzees. Thanks are also due to the following people for providing daily care and veterinary help for the chimpanzees: Kiyoaki Matsubayashi, Shunji Gotoh, Juri Suzuki, Akino Kato, Kiyonori Kumazaki, and Norihiko Maeda. We also thank Dora Biro for English corrections to the article. Finally, we appreciate the collaboration of the three mother-infant pairs of chimpanzees: Ai and Ayumu, Chloe and Cleo, and Pan and Pal.
- Balzamo, E., Bradley, R. J., Bradley, D. M., Pegram, G. V., & Rhodes, J. M. (1972). Sleep ontogeny in the chimpanzee: From birth to two months. Electmencephalography and Clinical Neurophysiology, 33,4146.
- Bard, K. A. (1994a). Evolutionary roots of intuitive parenting: Maternal competence in chimpanzees. Early Development and Parenting, 3, 19-28.
- Bard, K. A. (1994b). Similarities and differences in the neonatal behavior of chimpanzee and human infants. In G. Eider, E. Kaiser & F. A. Kinf (Fils.), The role of the chimpanzee in research (pp. 43-55). Basel, Switzerland: Karger.
- Bard, K. A. (1998). Socio-experimental contributions to imitation and emotion in chimpanzees. In S. Braten (Ed.), Inrersubjective communication and emorion in early ontogeny (pp. 208-227). Cambridge, UK: Cambridge University Press.
- Bard, K. A. (2000). Crying in infant primates: Insights into the development of crying in chimpanzees. In R. G. Barr, B. Hopkins, & J. A. Green (Fds.), Crying as a sign. a symptom, and a signal: Clinical emotional and developmental aspects of infant and toddler crying (pp. 157-175). London: Mac Keith Press.
- Bard, K. A,, Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., Costall, A,, & Matsuzawa, T. (2005). Group differences in the mutual gaze of chimpanzees (Pan tmglodytes). Developmental Psychology, 41,613--624.
- Bard, K. A,, & Russell, C. L. (1999). Evolutionary foundations of imitation: Social, cognitive, and developmental aspects of imitative processes in non-human primates. In J. Nadel & G. Butterworth (Eds.), Imitation in infancy (pp. 89-123). Cambridge, UK: Cambridge University Press.
- Boesch, C., & Boesch-Acherman, H. (2000). The chimpanzees of the Tai forest: Behavioral ecology and evolution. Oxford, UK: Oxford University Press.
- Brazelton, T. B. (1987). Neonatal behavioral assessment scale. In J. D. Osofsky (Ed.), Handbook of infanr development (2nd ed., pp. 780-817). New York: Wiley.
- Brent, L., Williams-Blangero, S., & Stone, A. M. (1996). Evaluation of the chimpanzee breeding program at the Southwest Foundation for Biomedical Research. Laboratory Animal Science, 46,405409.
- Emde, R. N., & Koening, K. L. (1969). Neonatal smiling and rapid eye movement states. Journal of American Academy of Child Psychiatry, 8, 57-67.
- Emde, R. N., McCartney, R. D., & Harmon, R. J. (1971). Neonatal smiling in REM states: IV. Premature study. Child Development, 42, 1657-1661.
- Farroni, T., Johnson, M. H., Brockbank, M., & Simon, F. (2000). Infants' use of gaze direction to cue attention: The importance of perceived motion. Visual Cognition, 7, 705-71 8.
- Fontaine, R. (1984). Imitative skills between birth and six months. Infant Behavior and Development, 7,323-333.
- Fouts, R. (1997). Next of kin. New York: Morrow.
- Freedman, D. (1965). Hereditary control of early social behavior. In B. M. Foss (Ed.), Determinants of infant behavior (pp. 149-159). London: Methuen.
- Freedman, D. (1974). Human infancy: An evolutionary perspective. New York: Wiley.
- Fujiyama, A., Watanabe, H., Toyoda, A,, Taylor, T., Itoh, T., Yaspo, M., et al. (2002). Construction and analysis of a human-chimpanzee comparative clone map. Science, 295, 131-134.
- Gewirtz, J. L. (1965). The course of infant smiling in four child-rearing environments in Israel. In B. M. Foss (Ed.), Determinants of infant behavior (pp. 205-248). London: Methuen.
- Goodall, J. (1986). The chimpanzees of Gombe: Patterns of behavior. Cambridge, MA: Harvard University Press.
- Goren, C., Starty, M., & Wu, P. (1975). Visual following and pattern discrimination of face-like stimuli by newborn infants. Pediatrics, 56, 544-549.
- Hayashi, M., & Matsuzawa, T. (2003). Cognitive development in object manipulation by infant chimpanzees. Animal Cognition, 6,225-233.
- Hayes, C. (1951). The ape in our house. New York Harper & Brothers. International Chimpanzee Chromosome 22 Consortium. (2004). DNA sequence and comparative analysis of chimpanzee chromosome. Nature, 429,382-388.
- Johnson, M. H., Dziurawiec, S., Ellis, H., & Morton, J. (1991). Newborns' preferential tracking of face-like stimuli and its subsequent decline. Cognition, 40, 1-19.
- Johnson, M. H., & Morton, J. (1991). Biology and cognitive development: The case of face recogni tion. Oxford, UK: Blackwell.
- Korner, A. F. (1969). Neonatal startles, smiles, erections, and reflex sucks as related to state, sex, and individuality. Child Development, 40, 1039-1053.
- Ladygina-Kohts, N. N. (2002). Infant chimpanzee and human child: A classic 1935 comparative study of ape emotions and intelligence. New York: Oxford University Press.
- Matsuzawa, T. (2003). The Ai project: Historical and ecological contexts. Animal Cognition, 6, 199-21 1.
- Matsuzawa, T., & Nakamura, M. (2004). Caregiving: Mother infant relations in chimpanzees. In
- M. Bekoff (Ed.), Encyclopedia of animal behavior (pp. 196-203).Westport, CT: Greenwood.
- Meltzoff, A. N., & Moore, M. K. (1977). Imitation of facial and manual gestures by human neonates. Science, 198, 75-78.
- Meltzoff, A. N., & Moore, M. K. (1983). Newborn infants imitate adult facial gestures. Child Development, 54,702-709.
- Messinger, B., Dondi, M. G., Nelson-Goens, C., Beghi, A., Fogel, A,, & Simion, F. (2002). How sleeping neonates smile. Developmental Science, 5,48-54.
- Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., & Matsuzawa, T. (2003). Preference for human direct gaze in infant chimpanzees (Pan troglodytes). Cognition, 89, B53-B64.
- Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., & Matsuzawa, T. (2004). Imitation in neonatal chimpanzees (Pan troglodytes). Developmental Science, 7,4371142.
- Myowa-Yamakoshi, M., Yamaguchi, M., Tomonaga, M., Tanaka, M., & Matsuzawa, T. (2005). Development of face recognition in infant chimpanzees (Pan troglodytes). Cognitive Development, 20,4943.
- Parmelee, A. H., Wenner, W. H., & Schultz, H. R. (1964). Infant sleep patterns from birth to 16 weeks of age. Journal of Pediatrics, 65,576-582.
- Plooij, F. (1984). The behavioral development of free-living chimpanzee babies and infants. Nonvood, NJ: Ablex.
- Prechtl, H. F. R. (1974). The behavioral states of the newborn infants: A review. Brain Research, 76, 184-212.
- Prechtl, H. F. R., & Beintema, D. J. (1964). The neurological examination of thefull term newborn infant. London: Spastic Society & Heinemann.
- Prechtl, H. F. R., & O'Brien, M. J. (1982). Behavioral states of the full-term newborn: The emergence of a concept. In P. Stratton (Ed.), Psychobiology of the human newborn (pp. 53-73). New York Wiley.
- Redshow, M. E. (1989). A comparison of neonatal behavior and reflexes in the great apes. Journal of Human Evolution, 18, 191-200.
- Sugiyama, Y. (2004). Age-specific birth rate and lifetime reproductive success of chimpanzees at Bossou, Guinea. American Journal of Primatology, 32, 31 1-318.
- Takahashi, M. (1973). Smiling responses in newborn infants: Relations to behavioral state, spontaneous movements, and the tactile stimulus (In Japanese with English summary). The Japanese Journal of Psychology, 44,465 I .
- Takahashi, M. (1974). The longitudinal study of infants' smiling responses: In relation to neonatal spontaneous smiles (In Japanese with English summary). The Japanese Journal of Psychology, 45, 256267.
- Takeshita, H. (2001 ). Development of combinatory manipulation in chimpanzee infants (Pan troglodytes). Animal Cognition, 4, 335-345.
- Tomonaga, M., Tanaka, M., Matsuzawa, T., Myowa-Yamakoshi, M., Kosugi, Y., Mizuno, Y., et al. (2004). Development of social cognition in infant chimpanzees (Pan troglodytes): Face recognition, smiling, gaze, and the lack of triadic relations. Japanese Psychological Research, 46, 227-235.
- Udono, T., Teramoto, M., & Hayasaka, I. (1999). A review of 100 pregnancies in chimpanzees at the Sanwa Kagaku Kenkyusho (In Japanese with English summary). Primate Research, 15,243-250.
- Vecera, S. P., & Johnson, M. H. (1995). Gaze detection and the cortical processing of faces: Evidence from infants and adults. Visual Cognition, 2, 59-87.
- Wolff, P. H. (1959). Observations on newborn infants. Psychosomatic Medicine, 21, 1 10-1 18.
- Wolff, P. H. (1963). Observations on the early development of smiling. In B. M. Foss (Ed.), Determinants of infant behavior (pp. 113-134). London: Methuen.
- Wolff, P. H. (1966). The causes, controls and organization of behavior in the newborn. Psychological Issues, 5, 1-105.
- Wolff, P. H. (1 987). The development of behavioral states and expressions of emotions in early infancy. Chicago: University of Chicago Press.