Four behavioural states are recognised in the human fetus and are comparable to those of the neonate: 1F (quiet sleep), 2F (active state), 3F (quiet awake), and 4F (active awake). State 5, or crying, is not considered to have a fetal correlate. In a study assessing the effects of exposure to tobacco and cocaine during pregnancy on fetal response and habituation to vibroacoustic stimulation, what appears to be the fetal homologue of crying was observed. These behaviours were seen on ultrasound, and have been captured on video recordings and include: an initial exhalation movement associated with mouth opening and tongue depression, followed by a series of three augmented breaths, the last breath ending in an inspiratory pause followed by an expiration and settling. This is the first report/video documenting these behaviours and suggests the possibility of a state 5F.
- behavioural state
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Five distinct behavioural states are recognised in the neonate: 1, quiet sleep; 2, active sleep; 3, quiet awake; 4, active awake; 5, crying. In the fetus, four distinct behavioural states 1F, 2F, 3F, and 4F have been identified and correspond to the neonatal behavioural states 1–4 (table 1).1,2 State 5, or the crying state, has no reported fetal correlate. We describe what we believe is the first documentation of crying behaviours in the fetus. These behaviours were elicited during vibroacoustic stimulation (VAS), which was performed as a research study assessing the effect of exposure to tobacco and cocaine during pregnancy on fetal response and habituation to VAS.3
Fetal ultrasound assessments were performed on up to three occasions (29–31, 32–35, and 36+ weeks gestation). These assessments were made in a semidarkened room using B mode ultrasonography (Corometrics Aloka 650 ultrasonograph). The ultrasound transducer was positioned for a lateral-oblique view of the fetus. Fetal eye movements and gross body movements were observed in real time by an observer and were recorded on video. Fetal heart rate was measured by continuous wave Doppler (Corometrics Fetal Monitor 145) and recorded continuously on the built in, dual channel, strip chart recorder. Behavioural state was assessed from these sources.
The fetus was observed unperturbed for 20 minutes. After this initial observation period, and when it appeared to be in a stable state (quiet or active sleep), the fetus was challenged with VAS. The VAS was provided by an artificial larynx (model 5c; Western Electric) that emits fundamental tones of about 100 Hz and 95 dB and was mechanically altered to provide exactly 0.5 second of stimulation. This duration was selected because longer stimulus durations have been associated with excessive fetal movements and prolonged tachycardia.4 The physiological intrauterine noise intensity has been reported to be about 85 dB.5 No adverse effects have been reported at this intensity, duration, and frequency.4 After stimulation, the fetus was observed for a blink-startle response for a duration of 10 seconds.6
The study had ethical approval from the Charlotte Medical Center institutional review board, and the parents gave written informed consent for the study.
The mother was an African-American primigravida, who smoked <10 cigarettes a day. She did not abuse cocaine. The female fetus was assessed at 33 weeks gestation when the “cry” behaviours were seen. There were no complications during the pregnancy. At 40 weeks gestation there was spontaneous onset of labour. There was a normal vaginal delivery, and the infant’s birth weight was 3555 g.
The video clip (which can be found at http://adc.bmjjournals.com/supplemental/) and figs 1–3 show what we believe is the fetal homologue of crying. The video begins with the quiet fetus before presentation of the VAS. The fetus is initially seen in an oblique view. The stimulus is presented. The fetus startles and turns her head to a lateral view, and is seen in profile (fig 1). The fetus then initiates behaviours consistent with crying. There is a brief expiration that is followed by a deep inspiratory phase with a subsequent pronounced expiratory phase. This expiration is associated with jaw opening, taut tongue, and chest depression (fig 2). It is immediately followed by three augmented breaths with progressive increase in chest rise and head tilt (fig 3). Each end inspiration is marked by chin quiver. The last augmented breath ends in an inspiratory pause, followed by an expiration and settling. Settling is associated with a turn of the fetal head to the oblique position, mouthing, and swallowing. This pattern of behaviour was seen again when a repeat VAS was presented to the fetus at about one minute after the initial cry. These behaviours are shown in the video and presented in a conceptualised form in fig 4.
This behavioural response was not an isolated case. Once recognised, more episodes of these behaviours were seen during the VAS testing of other fetuses. Ten subjects showed fetal crying behaviour, three of whom showed all of the characteristics on two separate occasions. Fetal crying behaviour was seen three times at the first assessment (gestational age 28–31.9 weeks), four times at the second assessment (32–35.9 weeks), and on six occasions at the third assessment (>36 weeks). Four mothers smoked cigarettes during pregnancy, three smoked and used cocaine, and three neither smoked nor used cocaine. The study population consisted of three groups of mother-fetal dyads: (1) cigarette smokers (n = 54), (2) cocaine users (n = 30), and (3) controls (n = 60). Only one infant had a low Apgar score at one minute and none at five minutes. Six of the infants were female, and four were male. Birth weight and gestation did not differ from the sample as a whole (range 2637–4138 g and 36.7–41.7 weeks).
Crying is a complex, rhythmical series of sounds that requires precise coordination between various motor systems including the musculature of the face, airway, and respiration.7 Fundamental to postnatal cry is vocalisation. Thus newborn and infant crying consists of both a vocal and non-vocal component. Hopkins8 suggests that the non-vocal accompaniments of crying are developed before birth, the vocal component being established with transition to the extrauterine life. The observations recorded from fetal ultrasounds support the concept that the fetus is capable of the complex motor behaviours that accompany the crying state.
These observations may have further developmental implications, as the expression of crying implies more than execution of a motor pattern. Crying requires reception of a stimulus, association of that stimulus with a negative connotation, and incorporation of an appropriate motor response that itself recruits a complex pattern of more primitive motor sequences. The sensory aspect of the cry response—that is, reception of the sound—implies intact afferent processes. Recognition of the stimulus as negative indicates development of more rostral brain sites that mediate affect and further integrate that affect with a motor response. The expression of crying thus indicates a several stage maturation of sensory reception, processing of signals as potentially deleterious, a dimension of affect, and recruitment of an appropriate response. Of this sequence, demonstration of affective integration, which incorporates limbic structures, implies more rostral neural maturation. The ability of the fetus to show affective integration of motor responses is supported by the observation that the fetal crying behaviours were seen only after VAS.
Crying consists of a sequence of vocal and non-vocal behaviours. Although crying is phenomenologically self evident in both the term and preterm infant, there is no unified or accepted definition that incorporates all aspects of the cry behaviour. Well described are the facial accompaniments of cry: grimace or frown, trembling of the chin, swallowing, and eye closure.8 Others describe movement of the extremities.7,9 These behaviours are tightly linked to the respiratory cycle.10 In the newborn, each cry cycle displays a consistent displacement of respiratory volume, suggesting a precise coordination between various motor systems and that of the respiratory system.11 Cries are produced primarily in the expiratory phase; however, the newborn is capable of inspiratory “voicing”.12 Thus Wolff’s13 classic description of infant cry as a rhythmic repetition consisting of an expiratory sound lasting 0.6–1.3 seconds, a brief pause of 0.2 second, and an inspiratory sound or whistle of about 0.1–0.2 second followed by another pause of 0.2 second before the next sound made on expiration, continues to provide the most comprehensive description of the infant cry. The behaviours shown in the video correspond to the neonatal cry with the vocalisation component occurring during the expiratory phases.
By 20 weeks gestation, the fetus possesses the complete motor repertoire necessary for cry behaviours: coordinated breathing efforts, jaw opening, mouthing, chin quiver, tongue extension, and swallowing.14–16 Furthermore, it is well known that the preterm infant of about 24 weeks gestational age is capable of producing crying sounds17,18 and can respond to environmental noise.19 Thus the fetus is capable of responding to sound and other perturbations with highly coordinated movements that mimic the temporal and behavioural components of the extrauterine cry. This phenomenon suggests a prenatal origin of crying, and supports the contention that the fetus does have a comparable state 5—that is, state 5F. This concept provides a developmental continuity between the prenatal and postnatal life.
The method of Nijhuis et al2 for determining fetal state required that the three state variables (body movements, eye movements, and heart rate pattern) are maintained for three minutes. This process follows the strategy used for the assessment of state in the newborn.20 Periods of less than three minutes are referred to as periods of coincidence, and may represent chance occurrence. It may be inappropriate to apply the three minute criterion to 5F, as crying in the neonate is often of shorter duration. In our case, the observed cry behaviours lasted 15–20 seconds.
The fetal “cry” behaviour, or state 5F, was seen in our study only after VAS. This observation may suggest that the “fetal cry” is elicited only when the fetus is disturbed. The behaviours were seen in all gestational ages studied, indicating that the behaviour occurs as early as 28 weeks gestation, and possibly earlier. The behaviour was observed equally in controls (non-smokers and no cocaine use), suggesting that these behaviours are not specific to tobacco or cocaine exposure.
What is already known on this topic
Four behavioural states are recognised in the human fetus and are comparable to those of the neonate: 1F (quiet sleep), 2F (active state), 3F (quiet awake), and 4F (active awake). State 5, or crying, is not considered to have a fetal correlate.
What this study adds
In a study assessing the effects of exposure to tobacco and cocaine during pregnancy on fetal response to vibroacoustic stimulation, we observed what appears to be the fetal homologue of crying. This is the first report/video documenting these behaviours and suggests the possibility of a fetal state 5F.
Finally, what we have observed is similar to the rare, but well authenticated, phenomenon of vagitus uterinus, a term used to describe the audible cry of the fetus in utero.21 Cases of vagitus uterinus are generally found at term, are associated with ruptured membranes that have allowed air to enter the uterus, and some operative intervention usually has occurred which has stimulated the fetus. Our case occurred seven weeks before delivery, and, although the body and facial movements were consistent with crying, an audible cry was not heard.
We thank the Education Media Centre, University of Auckland for the production of the video clip and figs 1–3, and Professor Ron Harper for helpful comments on the manuscript. EAM is supported by the Child Health Research Foundation, Auckland, New Zealand.
Published Online First 27 April 2005
The project was funded by NICHD (National Institutes of Child Health and Development) and NIDA (National Institute on Drug Addiction) (RO1: DA05949).
Competing interests: none declared
To access the video clip mentioned in this paper please ensure that you have QuickTime installed on your computer.