Edited by: Mehdi Khamassi, UMR7222 Institut des Systèmes Intelligents et Robotiques (ISIR), France
Reviewed by: Giovanni Laviola, Istituto Superiore di Sanità, Italy; Santiago J. Ballaz, Yachay Tech University, Ecuador
†Present address: Eszter Somogyi, Department of Psychology, University of Portsmouth, Portsmouth, United Kingdom
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The aim of this article is to track the fetal origin of infants’ sensorimotor behavior. We consider development as the self-organizing emergence of complex forms from spontaneously generated activity, governed by the innate capacity to detect and memorize the consequences of spontaneous activity (contingencies), and constrained by the sensory and motor maturation of the body. In support of this view, we show how observations on fetuses and also several fetal experiments suggest that the fetus’s first motor activity allows it to feel the space around it and to feel its body and the consequences of its movements on its body. This primitive motor babbling gives way progressively to sensorimotor behavior which already possesses most of the characteristics of infants’ later behavior: repetition of actions leading to sensations, intentionality, some motor control and oriented reactions to sensory stimulation. In this way the fetus can start developing a body map and acquiring knowledge of its limited physical and social environment.
Spontaneous motoneuron activity begins at the same time as motoneuron differentiation. Indeed, motor activity starts as rhythmic bursts of spontaneously generated action potentials correlated across thousands of cells, at a stage when motor neurons are pathfinding and innervating the skeletal muscles. Among other cellular processes, these periodic bursts of action potentials increase the concentration of calcium in the neurons, influencing gene expression and the establishment of cell phenotype (Feller,
Motor patterns most characteristic of the first weeks of gestation are spontaneous startles, general movements (GMs), isolated movements and twitches
As opposed to GMs, isolated movements, which emerge soon after GMs and outnumber them by the 14th week, involve distinctive sequencing of particular body parts (Prechtl,
Twitches are a particular kind of spontaneous motor activity produced during active sleep. Brief contractions of muscles trigger quick extensions or flexions of a limb or the neck. Fetuses start producing twitches at the age of 10–12 weeks, and from 15–16 weeks the frequency of twitches increases substantially. As we will see later, even though they appear during sleep (and according to some authors possibly because they appear during sleep) twitches may have an important role for establishing the brain’s body map (Blumberg et al.,
In general, the number of fetal movements per hour increases until a plateau is reached and decreases from 16 weeks onward (Natale et al.,
Even though early motility appears to be mostly unrelated to sensations, it is difficult to determine precisely when a fetal movement is spontaneously initiated or when it is triggered by sensations, due to movements of the mother or to internal sensations. Reflexive reactions to touch occur almost as early as spontaneous motor behavior. They are first observed in the region around the mouth: for instance, after stroking the perioral region, contraction of the neck muscles on the side opposite the stimulation, making the surface touched move away from the stimulator, has been observed at 7–8 weeks (Hooker,
As the sensory systems develop, non-reflexive responses to stimulations can be observed. The fetus’ environment is often disturbed by sounds, light, and touch and the fetus soon responds to these disturbances by moving (Valman and Pearson,
The first movements of the fetus, general or isolated, give the impression, not only of being spontaneous and not in reaction to sensation, but also of not being aimed at a precise goal, but rather to be randomly distributed across the space around it. We refer to motor babbling when movements seem random (Caligiore et al.,
Movements are not explored equally by the fetus. Early fetal movements are canalized by some constraints, arising from the system itself (characteristics of the articulations, state of development of the nervous system) as well as by the characteristics of the environment. These characteristics change through pregnancy, and due to these changes the contingent effect of the same movements may change. Due to the aquatic environment, arm and leg movements are likely to turn the body around as long as there is enough space and enough amniotic fluid around the fetus. By the end of pregnancy, when space is shrinking as the fetus grows, most arm movements end up not far from the face. The nervous system also is changing. From the beginning, there are two cortico-spinal tracks, one descending directly toward the spinal cord and the periphery (ipsilateral), the other one crossing the corpus callosum and descending on the opposite side of the spinal cord and periphery (contralateral). At first control is ipsilateral but becomes increasingly contralateral as the corpus callosum develops (Malinger and Zakut,
Within these constraints, babbling is extremely variable within and across fetuses. It may result in accidental contacts with the body or with the uterine environment. Such accidental contacts appear to be held in a memory of consequences, in such a way that the fetus soon starts to show a repertoire of “preferred” movements, as we will see in the next paragraph.
To repeat a movement, the fetus must know the connections between motoneurons and muscles, in other words it must have some sort of sensorimotor mapping. Scientists increasingly believe that sensorimotor mapping emerges progressively from spontaneous movements. Indeed, there are no movements without sensory consequences (the reverse being not true since sensory stimulations are not always followed by movements). Even twitches, produced on a background of muscle atonia (during sleep), are believed to play a fundamental role in the self-organization of spinal and supraspinal sensorimotor circuits and body mapping (Blumberg et al.,
Thus, fetal sensory stimulations arise from several sources, from endogenously-triggered spontaneous movements as well as from other sources, to stimulation arising from fetal or outside environment. All are likely to contribute to the development of somatosensory cortex and to the formation of cortical body maps (Milh et al.,
With isolated movements, fetuses soon seem to increasingly prefer those parts that are richly innervated. Starting at 10–12 weeks, face contacts are seen very often, which is interesting knowing that the trigeminal, which innervates the face, is an important source of tactile and proprioceptive sensations (Kurjak et al.,
At first, hands move independently. At 20–22 weeks, fetuses can be seen touching one hand with the other or crossing hands. They may also grasp the umbilical cord when they accidently contact it, thanks to the grasping reflex (Piontelli,
If the fetus increasingly aims its movements toward the more sensitive body parts, this means that it progressively selects these movements that induce interesting sensory feedback. Indeed, recent observations suggest that the fetus is capable of anticipating the consequences of its movements, which may be a first step toward action planning. For instance, two studies showed that fetuses anticipate their movement toward the mouth by opening the mouth before the hand arrives (Myowa-Yamakoshi and Takeshita,
Fetal behavior already presents some social characteristics observed in neonates. We have already mentioned the specific response to the mother’s touching her abdomen (Marx and Nagy,
In conclusion, the fetus’ motility is no longer seen as a purely reflexive behavior, or as simply emerging from motor primitives hardwired in the spinal cord or brainstem. And development itself is no longer considered as the results of increasing cortical control over lower reflexes through an unfolding program. Rather, development is now considered as the self-organizing emergence of complex forms from the spontaneously generated activity inherent in individuals with a nervous system, from the sensory constraints due, for instance, to the non-uniform distribution of tactile sensors, and from the capacity to detect and memorize the consequences of spontaneous activity (see, for instance Yamada et al.,
As isolated movements change along pregnancy, the fetus’ sensorimotor behavior comes to possess some of the characteristics later observed in the child’s behavior: curiosity or intrinsic motivation to explore surrounding space and the body, detection of contingencies, repetition of actions leading to sensations, reaction to sensory inputs, intentionality, goal-directed movements and some motor control. It is noteworthy that fetuses already display habituation, namely the decrease in reaction as a repeated stimulus loses its novelty, with vibroacoustic stimulation, speech sequences, and tones: such habituation has been observed in fetal heart changes and body movements (Lecanuet et al.,
There is some continuity between fetal and neonate motor behavior in the sense that all movements observed in fetal life are present in neonatal life (Kurjak et al.,
During the first 2 months the infant adapts to a new environment, new feeding, rhythm of day and night, it frequently has digestive problems, so that even though it has moments of clear awakening and interaction with people, most of its time is shared between feeding, crying, sleeping. This leaves little time for exploring the world, including itself. The emotional reactions of the baby interacting with its social partners are the most significant behaviors at that age. Then, around 2 or 3 months, infants are seen exploring their own body frequently.
Self-touch re-appears shortly after birth with little variations from right before birth, except for hand to mouth which increases and hand to knee which decreases (Sparling et al.,
Before being able to grasp an object and manipulate it—which infants start to do between 3 months and 5 months—they may use their legs or arms to produce interesting effects on their environment. Conjugate reinforcement studies with the legs (Rovee and Rovee,
After the post-natal period when pre-reaching can be observed only occasionally (Trevarthen,
Once the infant is able to grasp objects in a broad array of situations, it begins manipulating them in various ways. The main action of the first 6 months is mouthing, but this behavior decreases over the next few months (McCall,
Micro-analysis on observations of mother-infant interactions suggest that playful situations between mothers and infants emerge very early in life and play an important role in day-to-day interactions. Around 2- to 3 months infants begin to show contingent facial expressions (smiles) to the mother in situations of dyadic interactions (Trevarthen,
Taken together, these studies suggest that interacting develops from a very early age through repetitive and structured interactions.
Fetal movements have often been interpreted as a way for the fetus to practice and exercise its emerging motor system, participating in its maturation. For instance, when pathology creates a relative absence of GMs, the limbs do not develop normally and show deformities (Piontelli,
JF wrote the first draft of the article. All the other authors contributed equally to the final writing.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
1Since the mid-late 1970s ultrasound recordings allows observing the fetal behavior for several minutes in a row. More recently, 4D ultrasound produces computerized reconstructions of the fetus in motion. Off-line analyses are made frame by frame, and any observable action or reaction, when the fetus is in an alert state, is coded. This leads to a classification of fetal movements largely based on Prechtl’s pioneering work.
2Fetal movements appear, increase in frequency, sometimes reach a plateau and decrease (in general). Not all authors agree on the precise timing of these changes in occurrences. The difference may be due to the quality of the ultrasound recording, the number of fetuses observed and the number of observations for each fetus, the variability of movements between and within fetuses, as well as to the length of time of each observation (15–60 min). However, the order of appearance of movements is quite similar from one author to the next.
3A model of light transmission from the external environment to the uterus seems to indicate that illumination of the cavity, although variable, allows for some visual experience before birth (Del Giudice,