Book - Brain and behavioural development 5

From Embryology
Embryology - 20 Jun 2019    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Dickerson JWT. and McGurk H. Brain And Behavioural Development. (1982) Blackie & Son Ltd., Glasgow.

Brain and Behavioural Development - 1982: 1 Neural Development | 2 Comparative Neural | 3 Malnutrition | 4 Hormones and Growth Factors | 5 Cortical Activity | 6 Functional Asymmetry | 7 Plasticity | 8 Sex Differences

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic Textbook" and "Historic Embryology" appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms and interpretations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

Chapter Five - Cortical Activity in Behavioural Development

Cherry Thompson


The human electroencephalogram (EEG) can be recorded from scalp electrodes and provides a very easy and safe technique for monitoring brain function. Because of the complexity of the neural events within the brain, and because of the great distance away of the recording electrodes, the precise relation between the EEG and brain activity is not known. It is generally believed that the EEG reflects the summation of the excitatory and inhibitory postsynaptic potentials within the dendritic networks of the superficial layers of the cortex, which in turn reflect the probability of neural activation (Creutzfeldt et al , 1966; Creutzfeldt and Kuhnt, 1967; John and Morgades, 1969). The patterns of EEG activity reflect the ongoing bioelectrical events within the cortex and change consistently with different levels of arousal or gross tonic activation, from coma through varying depths of sleep, drowsiness, alertness and to an excited and agitated state. Predictable changes occur in the EEG when psychoactive drugs, toxic chemical and other agencies affect brain function and behaviour. Although some studies have found specific and localized changes in the background EEG that relate to more selective and focused aspects of behaviour, such as attention and perceptual and cognitive activities (Sutton, 1969; Buchsbaum and Fedio, 1970; Sandler and Schwartz, 1971) in general the EEG is reputed to more consistently reflect general, unspecific behaviour states with similar changes in the patterns of electrical activity occurring over large areas of the head.

Single spike

divided into different frequency ranges.

The most noticeable feature of the human EEG is that it contains well organized rhythmic waves which vary in frequency from less than 1 c/sec up to approximately 30 c/sec. The frequencies present in the EEG and the temporal and spatial patterning depend on the age of the subject and his behaviour, and for descriptive and analytical purposes the range of electrical brain activity recordable from the scalp is divided into various frequency bands which over many years of research have been found to have functional and clinical significance (Fig. 5.1). The slowest frequencies seen in the EEG are termed delta rhythms and vary from less than 1 to 3 c/sec. Slightly faster waves between 4 and 7 c/sec lie within the theta frequency band. The alpha rhythm (8 to 13 c/sec) is the most prominent activity that can be recorded from an alert adult, and the highest amplitudes are seen posteriorly over the occipital lobes when the eyes are closed and the individual relaxed. This rhythm was the first to be recorded through the intact skull of man and recognized as electrical brain activity by Hans Berger in 1929. Also present in an alert individual and particularly prominent over the frontal lobes are the much lower voltage fastest frequencies of the EEG, the beta rhythms, which vary between 14 and about 22 c/sec. Throughout life the healthy brain produces a continuous pattern of rhythmic waves which vary in amplitude between 15 and 150 juV. There is a constant daily cycle of changing EEG frequencies which slow during drowsiness and sleep, when theta and delta activity are predominant, while during wakefulness the faster alpha and beta rhythms are recorded. High voltage (greater than 50/iV) beta waves which are faster than 22 c/sec may be recorded, and these are usually associated with either drug effects or one of several clinical conditions. Paroxysmal, high voltage, slow activity and fast sharp transients or spikes are not recorded from a normal brain (Hill and Parr, 1963; Kooi, 1971).

EEG recordings contain a vast amount of information and there is no doubt that the standard procedures of analysis, dividing the EEG into frequency domains, provide a very crude method for quantification of the data. Subtle moment-to-moment changes in the EEG activity, and patterns of changing relationships between the left and right hemispheres and between the various regions of one hemisphere which may relate to more complex cognitive behaviours are frequently ignored, particularly outside the carefully controlled laboratory environment.

There is one other analytical technique that is now universally used which enables the recording of very small evoked potentials generated in specific cortical areas by sensory and voluntary motor events. These potentials are usually buried in the higher voltage activity of the background EEG and can only rarely be recognized by eye. By repetitively evoking, for example, sensory potentials, and sampling and storing the EEG activity in a computer during the sensory stimulation, the small cortical evoked potentials (which are time locked to the stimuli) add together, while the random appearance of positive and negative phases of the background EEG tend to sum to zero. Thus the signal-to-noise ratio of the evoked potentials relative to the EEG is enhanced by a factor which varies in practice from approximately 2 :1 to 4 :1 (Perry and Childers, 1969).

There is one great drawback in this technique of recording evoked potentials— it is necessary to present stimuli or elicit movements a large number of times at random or regular intervals. Evolution has designed a brain which reduces the effectiveness of such repetitive redundant information, and Brazier (1964, 1969) points out that hidden within the recording of an evoked potential is a trend of change, with both central and peripheral habituation. Non-directional variability which may be significant is also lost during a recording. This is of great concern to researchers attempting to build an electrophysiological model which reflects complex perceptual and cognitive behaviours, and careful experimental design can only partly surmount the problem.

Given these broad limitations to the electrophysiological recording of brain activity, nevertheless EEGs and evoked potentials do vary consistently with many changes in brain function that are known to occur, for example, during maturation, ageing and pathological processes. Thus they can provide a useful and reliable sign of certain brain functions. Secondly, information can be obtained about some of the functions of the brain with little or no cooperation on the part of the subject. This means that electrophysiology can provide a unique monitor in young babies and infants where the behavioural repertoire is stereotyped and limited and a lot of the time is spent asleep. The EEG can record maturing patterns of brain activity, emerging cycles of sleep and activity and responsivity to stimulation without interfering with the ongoing processes and thus provide one important view of the changing functions of the cortex in the developing infant.

The EEG in the young premature infant

It is not precisely known when the first signs of electrical activity can be recorded from scalp electrodes, as EEG monitoring of the human foetus is not possible in utero , and records have to be made on very early premature babies before and after they are viable. Reports of the earliest EEGs suggest that very short, disorganized bursts of activity followed by long periods of complete electrical silence can be recorded between 20 and 22 weeks gestational age (Engel, 1964; Robinson and Tizard, 1966; Ellingson et al , 1974). A typical recording from a 27-week gestational age baby is seen in Fig. 5.2. It is not certain whether such activity originates within the cortex (although most researchers imply that it does) since slow ill-defined bursts of electrical activity can also be recorded from hydrancephalic children who have no cortical tissue, and the electrical changes in these cases are presumably emanating from deeper

Figure 5.2 The EEG of a 27-week gestational age baby showing a pattern of irregular bursts of brain wave activity interspersed with periods of electrical silence. Redrawn from Lairy (1975).

brain structures. However, the patterns of total electrical suppression followed by brief bursts of activity are characteristic of the immature cortex as well as other brain structures and have been recorded intracortically in animals, where single cells also follow a pattern of very short bursts of firing followed by long periods of silence (Parmelee et al ., 1969). The electrical cellular events including spike generation are very slow and rapidly fatigued. Thus the primitive process of electrogenesis, which appears very different from a fully differentiated neurone, cannot sustain a repetitive response.

These silent periods in the EEG are never seen in the full term infant nor at any other time of life except when the brain is very close to death, and in the case of the early premature infant reflect the enormous immaturity of the neurones and neural connections within the telencephalon. The brain stem structures are already mature and producing continuous electrical activity (Bergstrom, 1969) while in the cortex the neuroblasts are still dividing and do not reach their full number until the conceptual age of 1\ months. Neural connections have hardly begun and any possible neural activity must be extremely limited (Scheibel and Scheibel, 1971).

One of the most significant features in the early stages of the maturing EEG is that the bursts of irregular electrical activity gradually increase in duration while the periods of silence shorten, paralleling very closely the electrogenesis and early differentiation of the cortical neurones (Schulte et al ., 1972). Between 24 and 27 weeks no EEG activity may be recorded for periods varying from 5 to 10 seconds up to as long as 2 to 3 minutes, and silences of longer than 20 seconds are the usual pattern (Kooi, 1971; Parmelee and Stern, 1972; Ellingson et al , 1974). At this time the activity of the cortex shows none of the features that are characteristic of a fully functioning brain. The electrical activity is of very high voltage (300 fiV) possibly due to the low electrical impedance of the skull and tissues surrounding the brain as well as the large immature potentials generated within single neurones. There are diffuse spike transients which later disappear and have no pathological significance as they do in a more mature brain. The brain wave patterns are disorganized in time and multifocal in location on the scalp with no synchrony between the hemispheres. The waves being generated are within the delta frequencies and are extremely slow, varying between 0.3 and 1 c/sec. There is no sustained rhythmic activity, neither are there any consistent changes with time to herald the beginnings of physiological periodicity, nor is there any change in the EEG in relation to arousal, movement or behaviour. The electrical activity is random, irregular, unresponsive and dissociated (Nolte and Haas, 1978). At this time movement is almost continuous and localized to the extremities of the arms and legs, and eye movements are sparse and ungrouped. Heart rate and respiration are also highly variable and random showing no consistent patterns from time to time (Parmelee and Stern, 1972).

So at this early stage in gestation the immaturity of electrophysiological parameters is apparent. Electrical activity cannot be sustained in the cortex for more than a few seconds, and at a time before intracortical connections have begun to form, the random activity from the various scalp regions is unrelated. Associations between subcortical activating systems and the cortex are yet to develop and there is no arousal response in the EEG or in any of the other electrophysiological parameters. This is associated with a failure of the young premature to produce a behavioural response to stimulation.

Surprisingly, at this stage in development sensory evoked potentials have been recorded from the cortex. Between 24 and 26 weeks, flashed lights and electrical stimulation of the median nerve of the wrist will evoke very localized responses which are confined to a small region over the primary visual and somatosensory projection areas. Later in development the potentials become far more widespread (Robinson and Tizard, 1966; Hrbek et al ., 1973). Auditory evoked potentials have been recorded a little later (between 25 and 27 weeks after conception) and they show a different pattern of distribution, being widespread throughout the scalp, but again with higher amplitudes over the primary projection areas of the temporal lobes (Weitzman and Graziani, 1968; Lairy, 1975). This distribution pattern gradually changes as the cortical mass increases in size and begins to fold, deepening the fissures and overlaying some cortical tissue with more superficial layers. In the temporal lobe the primary auditory areas disappear from view during the first year of life, to become buried deep within the lateral fissure, and the electrical activity from this area becomes inaccessible to scalp electrodes. An auditory evoked potential cannot be recorded from temporal regions once the enfolding process is well advanced and instead a large unspecific vertex response is obtained from the top of the scalp. The vertex potential varies consistently with perceptual changes but the neural connection of this location with the primary auditory pathways is not understood (Gibson, 1976).

A feature seen only in the young premature baby is that no matter what the modality of stimulation the form of the evoked potential is the same. A very large, simple, slow negative potential is evoked some 270 to 300 msec after the receptors have been stimulated. This long delay must in part reflect the immaturity of the peripheral receptors and slow nerve conduction velocities, which at 25 weeks are only 12 metres per second (Thomas and Lambert, 1960), as well as the slow rise times and decay rates of the excitatory and inhibitory postsynaptic potentials of the primitive cortical neurones (Purpura, 1971). The fact that cortical evoked potentials are recordable at all in such an immature brain is remarkable although there is always the problem that the young brain is easily fatigued and cannot respond repetitively. Evoked potentials are frequently not recorded, and the receptors have to be stimulated at a very slow rate compared with children and adults. Stimulation rates have to be longer than once every five seconds (Ellingson, 1960) and Umesaki and Morrell (1970) reported that no response could be obtained unless the interstimulus interval was longer than 10 seconds.

The formation of cortical evoked potentials appears to indicate that there must be some rudimentary connections between specific projection systems and the primary sensory cortical areas by the 24th to 27th week after conception. The shape of the evoked potential and the intracellular results from animals suggest that the connections are probably axodendritic within the limited network of layer 1 of the cortex (Adinolfi, 1971). The simple slow negative potential is the characteristic early response in all developing mammalian brains and closely reflects the cellular behaviour within the primary areas where sensory stimulation produces delayed, large amplitude excitatory postsynaptic potentials with slow rise times and prolonged fall times, and which result in only one axon spike or at the most two spikes (Purpura, 1971).

The middle months of the premature infant

From 32 weeks conceptual age until full term the greatest changes in the EEG occur, and the most noticeable feature is the gradual disappearance of the silent

Figure 5.3 The EEG of a 30-week gestational age baby. (A) EEG at one day of age. (B) EEG at 8 days of age. Redrawn from Himwich and Himwich (1964).

periods when no electrical activity is recorded. This feature is clearly seen (Fig. 5.3) in the EEGs of a ‘normal’ premature baby at 30 and 31 weeks gestational age. Lairy (1975) suggested that the EEG becomes more or less continuous at 28 weeks conceptual age, and significantly this is associated with the time when the infant first becomes viable. Others place this significant EEG milestone a little later, between 30 and 32 weeks (Robinson and Tizard, 1966; Parmelee et al , 1968; Reisen, 1971; Havlicek et al , 1975). Although the activity of the cortex is continuous after this time, it still retains the primitive pattern of suppression- bursts which remain until approximately two months post-term. During these suppression-burst periods the EEG contains a short run of high voltage slow waves followed by a run of much lower voltage activity (Ellingson et al , 1974). At 33 weeks conceptual age the periods of flattening or suppression (mean duration 11.5 sec) are longer than the high amplitude bursts which have a mean duration of 3.3 secs (Parmelee et al, 1969). At this stage 60 % of the EEG record contains suppression-burst activity and this activity gradually becomes less dominant with increasing age.

From 28 to 30 weeks the EEG also becomes more simple and less random in wave form. It is still primarily delta activity with traces of faster theta waves (4-6 c/sec) which occur in short runs of one to two seconds’ duration. At the same time as the EEG becomes continuous the amplitude of the EEG activity drops significantly to within the range seen in children and adults. Maximum amplitudes occur in the occipital regions. There is occasionally the beginning of some synchrony between the two hemispheres but this waxes and wanes and generally there is still very little relationship between cortical areas or hemispheres, and the EEG has the appearance of a few, independent electrical generators containing the same frequencies. This lack of synchrony between areas is an expected finding since the formation of intracortical connections is largely a postnatal process, and topographical differentiation does not develop until this later maturing process.

Superimposed upon the background EEG, and occurring uniquely in the premature brain from about 30 weeks onwards, is a low voltage fast spindling activity which has been reported to vary between 10 and 14 c/sec (Robinson and Tizard, 1966) and 16 and 20 c/sec (Joseph et al , 1976). This activity is unexpectedly fast and rhythmic in such a young brain and is not related to the much later appearance of sleep spindles or to the sensori-motor rhythms which are associated with motor inhibition. The significance and anatomical basis of the spindles is unknown. Non-specific thalamo-cortical projection systems are not functional at this age so the activity may reflect the influence of other brain stem structures on the cortex (Dreyfus-Brisac, 1964). Alternatively the scalp electrodes may be picking up far-field electrical potentials arising directly from the brain stem which may be particularly prominent during this period when the EEG is generating such low voltage waves. Although such an electrical source is a long way from a scalp recording electrode, in recent years it has been proved that far-field sensory evoked potentials can be recorded from the caudal brain stem in adults and children (Gibson, 1976). Whatever the origin, the significance of the unique spindling activity of the premature brain is not understood.

As the background EEG becomes more continuous and stable, begins to drop in voltage, and develops faster frequencies, another significant feature emerges in the process of maturation. It becomes at times possible to recognize different states or behaviours which some authors report as the beginning of the differentiation between sleeping and waking behaviour, although at such an early stage of development when the parameters which define such states are still very unstable and loosely associated or absent, the use of the terms ‘sleeping’ and ‘waking’ is debatable. Concomitant with the emergence of different physiological states is the development of cycles or rhythms of physiological functions. This feature is first seen at 28-30 weeks gestational age. At first the only reliable criterion is body movement, with the development of periods of inactivity accompanied by a significant reduction in muscle tone. No clear differentiation can be detected at this stage in the EEG, nor in other electrophysiological parameters such as eye movements which still remain sparse and fairly continuous (Parmelee and Stern, 1972; Stern et al , 1973; Werner et al, 1977). Heart rate and particularly respiration remain very irregular and between 24 and 32 weeks apnoeas are a very common feature with sustained periods of no respiration (Lairy, 1975). This point in maturation marks the beginning of an increasing association between various electrophysiological parameters and the initial steps in the temporal organization of CNS systems.

Figure 5.4 Different physiological states in a 35 week gestational age baby. (A) Awake. (B) Active sleep. (C) Quiet sleep, trace alternant.

By 30 to 32 weeks different stages of sleep emerge as seen in Fig. 5.4. One stage is called active sleep , a term applied by Parmelee et ah (1968). It has also been described as irregular sleep (Wolff, 1959), light sleep (Dreyfus-Brisac, 1964) and State 2 (Prechtl, 1968). It appears to correspond to paradoxical or rapid eye movement sleep in children and adults and can be defined as a condition where the eyes are closed and no behavioural responses can be easily elicited to environmental stimuli. There is, however, a considerable amount of phasic motor activity, seen as jerks in the full-term infant and older individuals, and as slow writhing movements in the premature. Muscle tone is maintained in premature and young babies but during the rest of life this state is characterized by total spinal inhibition and loss of muscle tone. In prematures active sleep is frequently accompanied by startle response, grimacing, sucking, frowning, smiling and vocalization. In more mature nervous systems heart rate and respiration rate are fast and irregular, with the frequent occurrence of bursts of rapid eye movements. The EEG usually consists of an activated pattern of fast frequencies similar to the waking EEG. In the premature, physiological functions are poorly correlated and perhaps only one or a few of the signs of active sleep are detectable.

The other recognizable sleep state in young babies is quiet sleep as described by Parmelee et al (1968). This state is also called regular sleep (Wolff, 1959), deep sleep (Dreyfus-Brisac, 1964) and State 1 (Prechtl, 1968) and corresponds with slow wave sleep or sleep stages 1 to 4 in children and adults (Rechtschaften and Kales, 1968). During quiet sleep the individual is relaxed with sustained periods when there is little body movement, although muscle tone is maintained even in adults. Similarly other physiological parameters are quiescent with no eye movements and slow regular respiration and heart rates. During this period in children and adults the EEG increases in amplitude and slows to its lowest frequencies within the delta range, while in the late premature and neonate a unique EEG form emerges called trace alternant which maintains the primitive pattern of suppression-bursts with runs of high voltage delta and theta waves alternating with low amplitude slow waves (Prechtl et al , 1969).

Because the immaturity of the neural mechanisms produces unstable periodicities and a loose association between physiological parameters in the premature, it is inevitable that intermediate states are recorded for a significant period of the time which cannot be described as either one of the two sleep states nor as being typical of waking activity, wakefulness being defined as periods when the eyes are open, some movement is present and the EEG shows a low voltage continuous pattern.

The periods of physiological activation which are manifest as rapid eye movement sleep in later life recur at regular intervals both during the sleeping period when they are easily monitored and during waking behaviour as well (Kales, 1969). The period of the cycle (termed the basic rest activity cycle) is one of the most stable events in mature physiological systems and recurs once every 90 minutes in adults (Kales et al ., 1974). When the periodicity is first apparent at 30 to 32 weeks conceptual age the cycle length is very short and extremely unstable, varying between 7 and 17 minutes. From this time there is a significant correlation between the increasing length of the active sleep or basic rest activity cycle and the conceptual age of the infant, which does not appear to be affected by environmental experiences (Clemente et al ., 1972).

Body movements are the first behaviour to display periods of inactivity and activity (Werner et al , 1977). A little later eye movements become very much more frequent, reaching a peak in activity around 32 weeks, and they also begin to become temporally organized into quiescent and active periods which makes the identification of active sleep easier and more reliable. Petre-Quadens (1969) reports a maximum in eye movement activity occurring a little later in development around 37 weeks after conception, with this event occurring earlier in females than in males. At this stage the premature will spend a significant amount of time without moving (83 %) which reflects the increasing inhibitory control of higher brain centres on the caudal brain stem and spinal cord reflex activities. This early in development the EEG alone cannot identify any of the different states and the most reliable parameters are body and eye movements.

So by 32 to 34 weeks active sleep is the first behaviour to be reliably recognized and it occupies a significant amount of time varying between 60 % and 80% of the day (Killam and Killam, 1976; Werner et a/., 1977). The dominance of active sleep at this stage of development reflects the neurological basis of this state. Active sleep depends on and is controlled by hind brain structures in the pons, particularly the mass of noradrenalin-containing neurones called the locus coeruleus (Jouvet, 1961; Morgane and Stern, 1974). Although midbrain and forebrain areas participate in active sleep phenomena, they are not important in the maintenance of active sleep. Since the hindbrain is mature long before 34 weeks gestation it is capable of sustaining active sleep significantly earlier than other sleep states and sustained wakefulness which require the control of midbrain and forebrain structures. Because the amount of active sleep is highest in the premature and remains high in the full term infant and neonate, declining steadily until adult values are reached in late childhood, active sleep is frequently regarded as a primitive state equivalent to other caudal reflex functions such as sucking and respiration (McGinty, 1971; Himwich, 1974; McGinty et al., 1974).

As the baby matures and active sleep becomes more organized, this state is increasingly associated with significant physiological and CNS activation. At the onset of active sleep there is a sudden rise in blood pressure and a huge increase in the utilization of oxygen within the brain which is associated with a dramatic increase in neural activity, particularly within the sensory systems where neural firing exceeds that recorded during alert wakefulness (Benoit, 1967; McGinty et al ., 1974; Noda and Adey, 1970; Killam and Killam, 1976). It has been suggested that since active sleep is associated with phasic bursts of intense neural activity, it is a mechanism which is important in the processes of neural maturation, particularly synaptogenesis which occurs primarily in the late premature and neonatal stages but is believed to continue throughout at least the first three decades and probably during the whole of a lifetime. Thus active sleep with its neural storms in some way helps determine the functional connections between neurones which may well encode both the innate and experiential components of behaviour. However both active sleep and the functioning of the CNS are quantitatively and qualitatively different in the premature and neonate, and it is by no means certain that the function and effect of active sleep remain constant as other brain areas mature, modify and participate in this early differentiated state.

Several weeks after active sleep is first identified, short, irregular, unstable periods of quiet sleep can be recognized. The best early identification of this behaviour is periods of regular respiration associated with quiescence (Lairy, 1975). The EEG does not begin to correlate with sleep state until 36 to 38 weeks (Lairy, 1975; Werner et al , 1977). Quiet sleep first emerges between 35 and 36 weeks gestation (Parmelee and Stern, 1972; Lombroso, 1979), and at this time active sleep occupies 60 % of the time and quiet sleep 21 %, while the rest of the activity is ill-defined and termed ‘intermediate-stage’. After the first appearance of quiet sleep there is a progressive increase in the amount of time spent in this stage which is concomitant with a gradual decline in the amount of active sleep. This change in dominance is correlated closely with the conceptual age of the premature baby (Parmelee and Stern, 1972). Interestingly, whereas active sleep is maintained by caudal brain structures, the midbrain, limbic, thalamic and particularly basal forebrain areas are essential for maintaining quiet sleep (Morgane and Stern, 1974). Like complex appetitive behaviours, quiet sleep is integrated at many levels of the neuroaxis and its appearance probably signals the beginning of functional connections with the forebrain areas (McGinty et al , 1974). The increasing dominance of quiet sleep and later of course of prolonged wakefulness reflects the developing inhibitory and excitatory control of the cortex over brain stem mechanisms. If anything goes wrong with cortical development, particularly at these early stages but to some extent throughout life, it is reflected in an abnormally low amount of quiet sleep (Magnes et al ., 1961).

So during the eighth month of pregnancy significant electrophysiological events occur. Transient high voltage spikes and periods of electrical silence disappear from the EEG. There are periods of increasing length when the baby lies still with no movement. Eye movements increase and begin to cluster into early embryonic cycles of activity and inactivity. Respiration becomes more regular and rhythmic and is associated with the increasing stability of heart rate which remains throughout prematurity much faster than the heart rate recorded in full-term infants. Quiet sleep and active sleep emerge from the primitive undifferentiated state. It is during this period that the huge expansion of the telencephalon begins with the appearance of secondary sulcation (Lemire et al ., 1975; Yakovlev, 1976). Viability of the premature baby improves significantly although mortality is still some two to ten times greater than that in the full term infant (Behrman et al ., 1971). There is also a sudden maturation in the response to many neurological tests, and sensory evoked potentials change significantly (Graziani et al ., 1968).

From a time when evoked potentials can first be recorded there is a linear decrease in the latency of the cortical response which is significantly correlated with the conceptual age and does not appear to bear any relationship with birth weight or extra-uterine experience (Graziani et al , 1968; Ganoti et al , 1980). Some sexual differences have been reported, with shorter latencies occurring in females than in males, and shorter latencies have also been recorded in negroes who usually mature earlier in terms of electrophysiological parameters than Caucasian children, although the reason is not known (Engel, 1965). As seen in the visual evoked potentials recorded from five infants in Fig. 5.5, between 35 and 37 weeks gestational age the primitive single negative evoked potential becomes more complex and is preceded by a faster positive component. In animals a similar positive wave is closely confined to the primary sensory areas and is believed to reflect specific basilar axosomatic connections between the sensory pathways and the cortical neurones. The development of the positive component is associated with increasing responsiveness to external stimuli, manifest in visual fixation and pursuit. Visual electrophysiological arousal which coincides with visual classical conditioning and visually guided behaviour occurs much later. Dark-reared cats with abnormal cellular development in the visual areas of the cortex and defective vision have reduced positive components. There is some evidence in man that similar reductions in the positivity of visual evoked potentials are associated with visual defects (Thompson, 1978).

Figure 5.5 A representation of the typical developmental changes in the visual evoked potential from (A) a premature baby to (E) an adult.

At the same time that the sensory evoked potential begins to develop a more complex morphology, there is a sudden acceleration in the curve of reducing latency. Somatosensory evoked potentials mature earlier than auditory and visual potentials, and by 35 weeks consist of three waves, so that the wave form and the latencies are closer to adult values than the other sensory modalities (Hrbek et al , 1973). At 30 weeks gestational age the latency of the visual evoked potential is 210 to 250msec (Engel and Butler, 1969; Ellingson, 1960; Umezaki and Morrell, 1970). Between 35 and 36 weeks the latency decreases to 200 to 210 msec and by 40 weeks, when the myelination of the optic nerve has begun, the latency of the cortical response to flashing lights is between 155 and 190 msec. The cortical distribution of the evoked potentials remains different from that recorded in adults with a more localized distribution confined to the primary sensory cortices (Ellingson, 1960).

During the last month until full term the eleetrophysiological changes begun earlier continue to become more stable and organized. The periods of active sleep and quiet sleep lengthen and are less likely to be disrupted (Parmelee and Stern, 1972). There is an increasing association between eleetrophysiological parameters (Nolte and Haas, 1978) and an increasing amount of time is spent in quiet sleep (Lombroso, 1979).

Before 38 weeks there is no significant difference in the power of the EEG frequencies during different behaviour states, but in the last weeks the EEG begins to develop different patterns of activity. Between 37 and 40 weeks periods of wakefulness become clearly recognizable and there is a noticeable increase in frequency with more theta activity between 4 and 6 c/sec (Lairy, 1975). The pattern of fast frequencies varying between 16 and 28 c/sec which is unique to the premature baby is still present. By 40 weeks gestational age the EEG has clearly differentiated and become more closely associated with different behaviours and patterns of physiological activity, a significant milestone according to Dreyfus-Brisac (1967) associated with a dramatic improvement in the survival chances of the baby and heralding the imminent birth. In the late premature baby, brief periods of wakefulness are vssociated with a low voltage mixture of theta waves, active sleep is accompanied by a low voltage mixture of delta and theta waves between 0.5 and 6 c/sec, and during quiet sleep either high voltage delta or the pattern of trace alternant is recorded with an alternating pattern of high voltage and low voltage delta waves. The length of the suppression periods has been gradually reducing during the last month and these are by now shorter than the runs of high voltage slow waves—the mean duration of the suppressions is 4.4 sec, the mean duration of the bursts is 5.9 sec (Werner et al , 1977). The density of the eye movements begins to fall by the 40th week and the amount of quiet sleep has increased to 38 % of the time, while active sleep has fallen to 52% (Parmelee and Stern, 1972). Because of the increasing improvement in the temporal organization and differentiation of electrophysiological parameters only 3 % of the activity is classified as an intermediary stage.

Several authors have reported that it is possible at this stage to record a diffuse general activation response in the EEG to stimulation but that it is extremely inconsistent (Dreyfus-Brisac, 1964; Kooi, 1971). In animals this electrophysiological event is associated with the development of functional connections between tl ? ascending reticular activating system, the diffuse thalamic projection and t ^rtical neurones, and is associated with a recognizable behavioural response to stimuli. It is primarily a postnatal process in animals (Creutzfeldt and Kuhnt, 1967), and is likely to be similar in man since such an inconsistently evoked response may be the result of the misinterpretation of data. Many authors report that no diffuse activation response can be seen in prematures, and EEG arousal is only characteristic of the post-term infant (Havlicek et al., 1975; Lairy, 1975).

It can be seen that during the second half of gestation electrophysiological parameters follow a rapid and precise developmental pattern which can provide a useful index of conceptual age. The majority of authors report that the EEG is correlated with conceptual age but is not as closely associated with birth weight and extrauterine experience (Dreyfus-Brisac, 1962,1964,1966; Ellingson, 1967; Parmelee and Stern, 1972; Dittrichova, 1969). Lairy (1975) claims that between 24 and 41 weeks the EEG changes so rapidly that the conceptual age can be accurately evaluated within two weeks and the addition of other electrophysiological parameters inevitably improves the estimates. It is certainly not possible to be as accurate at any other time of life. The premature EEG contains lower power spectra of theta and delta waves than that of the full-term infant. There are still traces of fast spindling activity and the periods of suppression which occur in quiet sleep are longer than in the full-term baby. However, there is some evidence that the experience of the premature baby may slightly accelerate the developmental processes although the effect seems to be very small and not often reported. There may be longer suppression periods in full-term infants compared with premature babies of the same gestational age (Parmelee and Stern, 1972), and the earlier occurrence of a postnatal milestone, that of the appearance of sleep spindles, by three to four weeks in premature babies (Metcalf, 1969).

Changes after birth—the first year of life

Whatever the gestational age of the baby it appears that CNS maturation progresses in a fixed sequence during the first 40 to 50 weeks after conception, with the programmed unfolding of anatomical and biochemical events altering in a predictable fashion the electrophysiological activity of the brain. It is generally believed that at first the neonate is functioning primarily at a subcortical level. The evidence for this is various; for example, motor defects are not apparent until about the second month post term. Primitive reflexes such as the Moro and Babinski reflex are present at birth. An ancephalic child is indistinguishable from a normal child during the first month, and volitional activity is not possible in the neonate (Dreyfus-Brisac and Blanc, 1957; Pritchard, 1964). However the importance of myelinization, which is essentially a postnatal process within the cortex, in controlling the functional capacity of the brain is frequently overemphasized, and electrophysiological evidence clearly suggests that the cortex is having an important modifying effect on brain stem activity before and at full term, since for example an ancephalic child or other babies with cortical damage have different sleep patterns, sleep/wake cycles, muscle tension, etc., and hemiparesis can be recognized from an early age (Robinson and Tizard, 1966).

The brain is still developing very rapidly during the first year of life as seen in the large increase in brain weight. During the first month the rapid expansion and elaboration of the association areas of the frontal, temporal and parietal regions begins, increasing the convexities of these brain areas and deepening the primary and secondary fissures. Myelinization of the cortex begins at term, and synaptogenesis which is also primarily a postnatal process within the cortex reflects environmental influence as well as innate programming (Himwich, 1974).

There is no striking pattern or consistency in the EEG of the newborn during the short periods of waking that they are able to maintain. The activity is diffuse with no apparent differentiation either between the different regions of the cortex or between the hemispheres. The waves are of low voltage (<50/rV), random and arrhythmic, and there is a constantly shifting pattern of asymmetries and asynchronies. The variability of the neonatal EEG is very high with low correlations between repeated EEGs (Ellingson et a/., 1974; Kellaway and Peterson, 1964; Lairy, 1975; Werner et a/., 1977). The dominant frequency during the first three months is within the delta range 3 to 4 c/sec, mixed with diffuse low voltage slow theta waves. Interestingly, some rhythmic activity can be recorded over the central regions of the scalp, and the frequencies arising from this region are also faster. From about the third week post term for the whole of the first year the central regions produce this more mature EEG pattern, followed some three to five months later by similar changes in the occipital lobes, and later still by changes in the temporal, parietal and frontal regions (Hill and Parr, 1963; Hague et al ., 1972). These EEG changes follow closely the maturation sequence of the cortex which involves a huge increase in the arborization of the neural plexus, an increase in the dendritic connections and the myelinization of intracortical and the thalamocortical connections. This occurs first within the sensori-motor cortex, later in the visual and auditory areas and last in the association areas (Yakovlev, 1976).

During the first three months there is no individualization in the waking EEG, and no sex differences have been detected (Hague et al, 1972). After this time the EEG begins to mature from a slow, random disorganized pattern to one of faster, regular rhythmic activity, greater differentiation between the various cortical regions, an increasing association with different behaviours and a developing responsiveness to stimulation. During the early months it is generally agreed that there is either an inconsistent response or no response at all to increased arousal and attention to stimuli, and no change in the EEG when the eyes are opened or closed. During the third month there emerges more rhythmic activity with the eyes closed and an activation or desynchronized response of faster EEG rhythms when the eyes are opened or when the baby alerts to stimuli, although this reaction is much less consistently evoked than in adults (Kellaway and Peterson, 1964; Havlicek et a/., 1975; Lairy, 1976). This change in EEG reactivity coincides with a noticeable increase in a child’s ability to remain awake. Before 8 weeks the waking periods are usually brief and specifically because of hunger or physical discomfort. By three months activities such as sucking, fussiness and crying decline and the increasing periods of wakefulness are used for other activities; social interaction increases and attentive behaviour to external stimuli is more often present (Weitzman and Graziani, 1968; Sterman et al, 1977).

As well as not being able to jecognize any arousal response in the EEG, it is also impossible in the first few months to identify any change in the EEG as the baby becomes drowsy (Kellaway and Peterson, 1964; Samson-Dolfuss et al, 1964). This suggests that early in development either the EEG is unable to reflect the changing behaviours of the baby, or that at this age more subtle changes in behaviour have not developed and the transition from waking to sleep is extremely rapid (Lairy, 1975). Sleep can occur almost immediately and has more the appearance of reflex response than the characteristics of an appetitive behaviour (McGinty, 1971). Another unique feature is seen as the neonate falls asleep. Instead of slow wave sleep always being at the beginning of a sleep period, which is the case throughout most of life, the mode of onset is variable and the first sleep activity may be that of quiet sleep or more frequently active sleep (Prechtl et al, 1969; Curzi-Dascalova, 1977).

During the first three months of life probably the most dramatic changes occur in sleep behaviour, which over the next few years or so gradually develop into one of the most outstanding biological constants in terms of individual differences in behaviour (Morgane and Stern, 1974). The full term baby sleeps for long periods of time, although individual differences are enormous (mean total sleep time 16.6 hours, range of variation 10.5 to 23 hours). There is a gradual decline in sleep time with maturity so that by three weeks post term the mean sleep time has already declined to 14.5 hours (Sterman and Hoppenbrouwers, 1971; Parmelee and Stern, 1972; Sterman, 1972). In the premature baby and during the first two months of life rhythmic patterns of activity occur up to seventeen to twenty times every 24 hours. The duration of the sleep/wake and active sleep/quiet sleep alternation is short, irregular and very sensitive to such disruption as body movements and crying (Lairy, 1975). The baby will initially wake every couple of hours but there is a gradual increase in the duration of the sleep periods. At first these longer sleeping times can occur at any time of the day, but between three and five weeks of age a circadian rhythm begins to emerge with the longest sleep period and more sleep occurring at night (57% total sleep occurring between 8 p.m. and 8 a.m.). By six weeks of age a baby will usually sleep for five to six hours, and after twelve weeks there is a well established diurnal rhythm with a sustained nightly sleep period lasting between eight to nine hours, which is about 70 % of the total sleep time. Day time sleep also becomes consolidated with increasing periods of wakefulness occurring every three to four hours.

The body’s circadian rhythm is fully established by the sixth month and following the sleep/wakefulness cycle the body’s physiology also develops a diurnal periodicity. This is first recognized for body temperature by two to three weeks after birth. Rhythms of urine excretion, heart rate variation and electrolyte metabolism develop between four and twenty weeks post term, while patterns of hormone secretion do not become associated with circadian and sleep periodicities until the rhythms are well established.

Large individual differences in sleep behaviour are apparent from birth and can be seen in the total sleep time, number of rapid eye movements and facial and body movements in active sleep, the frequency of respiration in quiet sleep, vocalization, crying and sucking. These differences persist with considerable stability into later life (Dittrichova et al ., 1976). Some of the differences are probably genetic, others are related to early experience and chronic subclinical states, for example nutrition and hormonal anomalies. Environmental factors may well be important since it has been shown that animals raised in isolated environments sleep far less (sleep is reduced by 40 %) while novel environments can increase the subsequent sleep time by 25% (McGinty, 1971). Stress and psychological function may also relate closely to evolving sleep patterns (Kales, 1969; Sterman, 1972), and since infants spend so much time asleep it is possible that more sleep research could provide a better measure of both individual differences and the progress of development than many waking measures.

One of the most interesting characteristics of sleep behaviour in both the premature and young infant during the first three months is that active sleep is independent of quiet sleep, whereas in the child and adult it is embedded within the long night’s sleep and only occurs after a prolonged period of slow wave sleep. In the young baby active sleep occurs frequently at sleep onset or during waking, particularly when the child is fussing, crying and sucking. At this time there is still a loose association of physiological parameters and heart and respiration rate do not yet show a clear acceleration at active sleep onset. The most significant feature of all is that there is no loss of muscle tone and spinal inhibition, which is a cardinal feature of the mature nervous system found in all mammals.

The percentage of active sleep is still high in the full term infant (between 40 % to 50 % of sleep, mean duration 25 minutes) but from birth there is a very rapid decline in active sleep dominance until it occupies about 35 % of the sleep time at three months (Parmelee and Stern, 1972). Other authors suggest a slightly slower fall to between 40% and 42% with a mean duration of 14 minutes (Parmelee et al , 1968; Stern et al , 1969; Dittrichova et a/., 1976). For the rest of the first year the amount of active sleep continues to decline more gradually. The amount of active sleep at the different post term stages is reputed to be reduced in mongol babies, in microcephale and in cases of placental insufficiency (Petre- Quadens, 1969). It has also been reported that active sleep is frequently associated with feeding and sucking behaviour in the early months and occurs for longer periods with breast feeding, possibly because this may involve more handling of the baby, longer periods of rocking and more frequent feeding (Metcalf and Jordan, 1972). At present, the consequences of this suggested relationship can only be guessed at.

The EEG during active sleep consists of a mixture of low voltage (15 to 30 fiV) irregular theta and delta waves, interspersed with higher amplitude delta waves. The dominant frequency is around 4 c/sec (Havlicek et al , 1975). During the first half of the year the frequencies increase to vary between 2 and 6 c/sec and the amplitude decreases slightly so that all the brain activity in active sleep is an irregular low voltage mixture. During the second half of the year the pattern becomes increasingly closer to that of the waking pattern of activity (Ellingson et al ., 1974). The appearance of so-called ‘saw-toothed’ waves during the second to third month post term may reflect the manifestation within the cortex of the ponto-geniculate-occipital spikes recorded during the neural storms in animals (Curzi-Dascalova, 1977).

Beyond a slight shift in frequency there is very little further change in the EEG of active sleep during the first year, which provides a significant contrast to the dramatic EEG changes that occur during quiet sleep corresponding with the early postnatal maturation of the cortex.

As the amount of active sleep becomes less with increasing conceptual age, so there is an increasing amount of time spent in quiet sleep, and this increase is dramatic during the first eight months post term (Dittrichova et al , 1976; Werner et al, 1977). Dittrichova (1969) studying ten full-term babies found that the mean duration of a quiet sleep period was 11 minutes at term and 13.8 minutes at two weeks. This increased to a mean duration of 22 minutes by twelve weeks and was accompanied by a parallel increase in the total percentage of quiet sleep experienced which was particularly dramatic during the second to thirty-fourth week. By the eighth month twice as much quiet sleep was recorded as active sleep with 55% of the sleep period being involved in quiet sleep processes. Parmelee et al (1968) reported a close correlation between conceptual age post term and the total amount and duration of quiet sleep periods, and this EEG feature parallels the change in basal forebrain control of limbic and brain stem structures. The increasing amount of quiet sleep is associated with the increasing ability of the cortex to maintain longer and longer periods of wakefulness, accompanied by a reorganization of the temporal patterns of the body’s activity, the processes and significances of which are poorly understood but are certainly related to CNS maturity and not chronological age (Purpura, 1971). Apathetic and unresponsive babies, infants with delayed milestones and hyperactive children have been reported to show a significant slowing in the increase of quiet sleep with age and they also have less active sleep (Kales, 1969; Himwich, 1974; Weitzman, 1974).

The familiar suppression-burst EEG pattern of the premature is manifest as the trace alternant pattern of quiet sleep in the neonate. It is seen as bursts of high voltage delta waves (1 to 3 c/sec) lasting some four to five seconds and interspersed with lower voltage slow waves. The amount of trace alternant declines rapidly during the first weeks of life and it is generally agreed that it disappears some four to five weeks post term (Robinson and Tizard, 1966; Dittrichova, 1969; Metcalf, 1969; Hague, 1972; Ellingson et al , 1974; Werner et al , 1977; Lombroso, 1979). It is replaced by runs of continuous high voltage delta activity (amplitudes greater than 50 fiW) which increases significantly in abundance during the first two months, reflecting the increasing intra-cortical connectivity which begins to synchronize the activity of large populations of neurones (Mizuno et al , 1969). At the beginning of the third month, therefore, slow high voltage delta waves begin to dominate quiet sleep (Gibbs and Knott, 1949; Hague, 1968; Parmelee et al, 1969; Lairy, 1975).

Several weeks after the disappearance of the trace alternant pattern an important sleep activity appears whose significance is not understood. This is the phenomenon of sleep spindles which consist of bursts of medium voltage waves whose frequency varies around 14 to 16 c/sec. The spindles are very distinctive and characteristic of all normal sleep, but they have been reported to be absent in some clinical conditions such as in epileptic and hormone deficient children (Howe et al, 1974; Sterman et al, 1977). Rudimentary spindles may be recognized as early as five to six weeks post term (Robinson and Tizard, 1966; Sterman and Hoppenbrouwers, 1971; Hague, 1972), and are clearly present between the end of the second and third months (Katsurador, 1965; Sterman and Hoppenbrouwers, 1971; Hague, 1972). During the next two months there is a significant increase in the amplitude and duration of the sleep spindles, until they are almost continuous in quiet sleep and far exceed the spindle activity seen in adults. There is then a decline in the spindle duration around the age of six to seven months and by eight months this sleep activity begins to cluster into specific quiet sleep periods (Hague, 1972).

The appearance of spindle activity coincides with the beginning of the myelination of the non-specific thalamic projection system and the formation of contacts between its ascending processes and cortical neurones (Himwich, 1974; Yakovlev, 1976). It has been suggested that specific nuclei within the lateral thalamus which are associated with the reticular formation are responsible for the characteristic hypersynchronous runs of rhythmic waves always seen within the cortex (Andersen and Andersen, 1968). This certainly includes spindle activity which may also depend on the integrity of the corpus callosum (Kooi, 1971; Scheibel and Scheibel, 1971), and the appearance of sleep spindles closely coincides with the increasing amount of quiet sleep, sustained sleep periods and maintained wakefulness. Sleep spindles may also be associated with the sensorimotor rhythm and thus be important in inhibitory cortical function particularly in controlling skilled motor behaviour (McGinty et al, 1974). However, the functional significance of this sleep phenomenon must remain speculative until more research is completed.

The arousal response is often most clearly seen in sleep and stimuli can evoke midline phenomena such as the ‘K’ complex, a high amplitude series of waves with an initial sharp negative followed by a slow positive/negative complex, and often succeeded by runs of spindle activity or a reduction of amplitude and an increase in the frequencies of the EEG which heralds a lightening of sleep or perhaps an awakening. It has been suggested that the shape of the ‘K’ complex depends on the significance of the signal (Oswald, 1962). Another similar wave form is the ‘parietal hump’ or ‘vertex sharp wave’ which is again maximal at the vertex but simpler in form. This arousal response consists of repetitive high amplitude sharp negative waves which are particularly abundant early in sleep. These sleep phenomena are absent in the first three months of life which may be due to detection problems in the slow diffuse sleep EEG of the young, yet vertex sharp waves are easy to recognize within the slow activity of adults and children (Metcalf and Jordan, 1972).

So during the first three months of life the sleep of the young infant is different from the child and adult. Unusual behaviour such as sucking and fussing occur frequently, the physiological parameters are as yet not stable, and there are no clear hormonal changes (Sterman and Clemente, 1974). The physiological rest/ activity rhythm, still much shorter than that seen in adults, becomes increasingly regular with a period of about 50 minutes. The periods of activation, which easily become locked on to external cues, frequently coincide with feeding demands (Globus, 1966). This periodicity changes very little during the first year (Weitzman and Graziani, 1968), then gradually lengthens to a period of seventy minutes by the age of ten years or later (Sterman and Clemente, 1974). The timing of this maturation pattern is however really a matter of speculation since data is sparse after the first months of life and others report more rapid changes in the basic rest/activity cycle within the first year of life (Sterman and Hoppenbrouwers, 1971).

In the third postnatal month some important milestones are established in sleep behaviour which are closely correlated with increased wakefulness, social interaction and complex responses to stimuli. Firstly, sleep becomes consolidated into a sustained sleep period of six to eight hours at night. At the same time there is an increasing differentiation between the brain activity of quiet and active sleep which in the neonate are very similar (Havlicek et al , 1975). By three months, active sleep activity has reduced in amplitude and increased in frequency, while quiet sleep shows increasing dominance of high voltage delta waves and spindle activity. Also at this time active sleep, which up to this point had been independent of sleep behaviour and frequently occurred at sleep onset and during wakefulness, becomes embedded in quiet sleep and during the rest of life follows a fixed, significant period of quiet sleep before being triggered by quiet sleep mechanisms (Jouvet, 1961). Thus by three months there is no active sleep at the beginning of a sleep period (Graziani, 1974; Werner et a/., 1977). Phasic activity reduces significantly and the physiological parameters are now more closely associated, with heart and respiration rate increasing and becoming irregular in active sleep, while they slow and become regular in quiet sleep. Finally, the last significant change in active sleep at this time is the beginning of the normal pattern of profound spinal inhibition and raised arousal thresholds, whereas before in the more immature system muscle tension is facilitated in active sleep and the infant can be easily aroused (Pompeiano, 1969; Kales, 1969).

During these first three to four months of life the infant is uniquely vulnerable to various forms of mild stress including a disruption of the normal routine. Only a small adaptation response to sleep deprivation has been reported by Anders and Roffwarg (1973) in a group of full-term infants some twenty-four to ninety-six hours old. They lost one sleep period of three to four hours and in their subsequent sleep showed some increase in total sleep time and the percentage of active sleep, but there was no reported change in quiet sleep which is the usual pattern in older individuals. Sterman et al (1977) suggested that sleep adaptation to stress is inadequate in the immature, and in young animals at the equivalent CNS maturity as the first three months in man, there is evidence of an increased number of apnoeas with mild sleep deprivation (Baker and McGinty, 1972).

The age of three months, with the consolidation of sleep into prolonged time periods, decreased muscle tone and arousability and poor adaptation to mild stress, is also the time of the greatest number of reported cases of sudden infant death syndrome. Typically between two and four months young infants die silently in sleep and as yet there is no clearly recognized specific cause. Sudden infant death syndromes are often preceded by a disruption in daily routine and interrupted sleep schedules and are typically seen in lower birth weight babies (Weitzman and Graziani, 1974; Kraus et al , 1977). Maintained levels of muscle tension throughout all sleep stages in the very young baby may be protective, to allow easy arousability if an imbalance occurs in the respiratory mechanisms of the brain stem which are still physiologically immature in the first months. Apnoeas are common in the neonate and premature baby and are exacerbated by upper respiratory tract infection, stress and sleep deprivation. With maturity, increasing control and association of physiological parameters should occur at the same time as reduced arousability develops in active sleep. Thus the possibility is raised of a maturational mismatch as sleep is consolidated with a limited response or complete failure to arouse during sleep to correct prolonged apnoeas, which instead results in increasing acidosis and possible respiratory arrest (Sterman and Clemente, 1974).

During the first year of life, and particularly during the first months, the immature cortex cannot sustain high frequency repetitive neuronal discharges and the common pathological EEG pattern of high voltage spikes and slow waves is rare in young babies. Transient spikes in the EEG do not have clinical significance (Ellingson et al , 1974). However, as with premature babies, a good indicator of abnormal development is delayed EEG milestones with an immature EEG within the first year indicating a poor prognosis (Samson- Dollfus et al , 1964). Thus the presence of trace alternant in quiet sleep beyond the first four to five weeks and a delay or failure in appearance of sleep spindles is seen in hypothyroidism, hypoxia and brain damage (Parmelee and Stern, 1972). Malnutrition and hypoglycaemia which are associated with defective myelination and retarded neurocellular growth result in an immature EEG and very disturbed sleep patterns. Quiet sleep is particularly poorly developed. It is reduced in amount, few sleep spindles are recorded and respiration is also less regular. Eye movements may be significantly reduced in active sleep (Dobbing, 1960; Schulte et al, 1972).

Changes into childhood

From three months after birth the EEG begins rapidly to acquire the features which are so characteristic of the adult EEG.

Firstly, the frequencies present in the waking EEG gradually become faster. As can be seen from Table 5.1, delta is the dominant frequency in the early part of the first year. More complex analyses of contemporary data have confirmed the patterns reported by Lindsley (1939) and Smith (1941). By six months theta activity is beginning to predominate, and is clearly established by twelve months. It is mixed with traces of alpha activity, which increases during the second and third year to supersede the slower theta and delta waves during the fourth year.

Table 5.1 Changes in the EEG average frequency recorded from occipital electrodes: postnatal development from three months to fifteen years

Frequency c/sec

Age Lindsley Smith

(months) (1939) (1941)














































Data adapted from:

Lindsley, D. B. (1939) Longitudinal study of the occipital alpha rhythm in normal children: frequency and amplitude

standards. J. Genet Psychol. , 55, 197-213. Smith, J. R. (1941) The frequency growth of the human alpha rhythms during normal infancy and childhood. J. Psychol, 11, 177-198.

The rate of change is very rapid in the first two years of life but slows after this until adult values are reached in the mid teens (Robinson and Tizard, 1966; Kooi, 1971). During the second half of the first year differences between the male and female EEG begin to emerge and these persist until puberty with a ‘more mature’, faster pattern of frequencies being produced earlier by females (Hague, 1968; Hague et al , 1972). The young EEG is always typified by the presence of low voltage, slow waves, mixed with the faster dominant rhythms, particularly in posterior regions of the scalp.

Several other changes in the EEG progress with the frequency changes. The brain activity becomes more rhythmic and less random and diffuse with sustained runs of clearly recognizable single frequency waves. This rhythmicity also becomes confined firstly to the central regions and later, during the second half of the first year of life, to the occipital regions reflecting the maturation of the various cortical regions. Thus there is a trend of increasing topographical differentiation of the EEG which begins in the first year and is clearly established by the fourth year, where electrical activity arising over the frontal, temporal, parietal and occipital areas has its own characteristic pattern of amplitude, frequency and rhythmicity. Also late in the first year a relationship begins to develop between the electrical activity of the two hemispheres. The EEG becomes increasingly synchronous while hemispheric differences in amplitude and wave form become apparent, although they are not stable until the seventh to tenth year (Hague et al , 1972; Werner et al , 1977).

Early in the first year the arousal response of the EEG is sometimes recognizable although inconsistently evoked. When the eyes are open, low voltage fast frequencies are recorded, while when the eyes are closed and the individual is relaxed, high voltage, rhythmic runs of slower alpha activity appear, particularly over the posterior regions of the head. This desynchronized pattern with eyes open is also evoked during arousal and attention to stimuli. The response is poor in the second year but is more consistently evoked in the third and fourth years.

This slow progressive maturation of the waking EEG correlates well with changes within the brain. The cortex is still expanding rapidly until the end of the second year and the individual patterns of the tertiary sulcation become well established at this time, continuing more slowly until the end of the first decade (Yakovlev, 1976). Also myelination, synaptogenesis and the elaboration of the cortical dendritic processes proceed at a rapid pace for the first two years then slow to continue into the second and perhaps even into the third decade of life. Hand in hand with this the most rapid changes in the post-term EEG occur within the first two years. Thereafter the maturation rate of the EEG slows, and as there is a greater response to environmental events and more significant changes relating to the behaviour of the individual, so it becomes increasingly difficult to define a normal EEG and specify the timing of EEG milestones. This is further exacerbated by the greater lability of the young EEG, and much larger changes are seen from moment to moment and day to day than is acceptable in the adult. There is therefore a diminishing relationship between the EEG and age, and after the first few years it generally provides a very poor index of maturity except within broad categories.

One of the last dramatic milestones in EEG development occurs between six and eight months with the appearance of a feature unique in the young infant. Drowsiness becomes associated with a specific pattern of high amplitude, very rhythmic theta waves; such activity is termed hyper synchronous. These waves increase in frequency and amplitude during the rest of the first year and become more or less continuous as the child quietens and falls asleep (see Fig. 5.6). Later in the second and third years this activity becomes less marked and decreases in amplitude; it is rarely seen in the fourth year (Kellaway and Petersen, 1964; Samson-Dollfus et al , 1964; Kooi, 1971).

Figure 5.6 Changes in the EEG during drowsiness in a 20-month-old child.

At the same time as the EEG develops several different patterns of activity during the states of arousal, quiet wakefulness and drowsiness, quiet sleep becomes differentiated into sleep stages which are similar to those of the adult. Arousal phenomena such as the vertex sharp waves and ‘K’ complex (see p. 152) are clearly recognizable after five or six months, and become increasingly prominent during the first two years, although the final wave form is not complete until much later in life and the response matures into a very individual pattern. Why the ‘K’ complex develops so slowly when evoked potentials can be recorded in prematures and neonates is not clear. Metcalf (1969) suggested that since the ‘K’ complex alters with the type of stimulus evoking it, this sleep phenomenon may well reflect information processing and therefore develops slowly with the increasing capabilities of the child. At the end of the first year sleep spindles have decreased in duration and become clustered into the lighter periods of quiet sleep, and sleep takes on its mature appearance with four recognized sleep stages, Stage 1 with low voltage theta waves and vertex sharp waves, Stage 2 with lower frequencies, 6 K’ complexes and sleep spindles, and Stages 3 and 4 with increasing amounts of high voltage delta activity.

By the end of one year sleep is well consolidated, a circadian rhythm is established and 90 % of infants do not wake habitually during the night (Webb, 1969). Total sleep time falls rapidly during the first year and then follows the pattern of other physiological changes, decreasing more slowly in later years. The mean total amount of sleep is 10.2 hours for three to five year olds and this falls to an average of 9.8 hours in the ninth and tenth years. Webb (1969) reported an enormous variation in the amount of sleep needed in young teenagers but could not relate these differences to school achievement, personality variables or other psychological characteristics. He did not look at younger children. Active sleep does not change very dramatically post term except that its portion of sleep gradually declines over the years to finally reach a stable value of about 20 % in the early twenties. Quiet sleep slowly changes as well until it stabilizes in the fourth year. From the age of two to five years the high voltage delta activity of Stage 4 becomes increasingly prominent. It may be partly for this reason that young children are very slow to arouse from sleep, and of course sleep walking, which occurs in Stage 4, is more common in the young (Kales, 1969).

Cortical evoked potentials, established so early in the young brain, are quite well developed at birth. The wave form which becomes more complex in late pregnancy doer not change dramatically and this is particularly so for somatosensory and auditory evoked potentials (Barnet et al , 1975; Desmedt, 1977). The most significant change post term is a decrease in the latency of the response as the nerve conduction velocities, which are only 50 % of adult values at birth, reach their final values by the age of four years (Cracco et al , 1979). Again the change in latency is most rapid during the first year; Ellingson (1960) reported a very sudden, short period of acceleration in the changing latency curve for visual potentials at about seven or eight weeks post term, which he associated with the maturing of macula function and the appearance of focused eye movements together with the beginnings of visual attention. Exact latency changes are reported in the literature (Ellingson, 1964, 1966a, 1967; Desmedt, 1977; Barnet et al, 1975) and it is frequently suggested that the latency of evoked potentials provides a useful and reliable index of age and maturity post term (Barnet et al, 1975). It is generally the early components of the evoked potential which are used in latency measurements and later components are often absent in the early weeks. This has been reported for all sensory modalities, and since it is the late components which reflect changes in attentive behaviour and information processing this finding suggests that although the sensory signals are arriving in the cortex, further elaborate signal processing is not carried out until the fourth to sixth week of life (Ellis and Ellingson, 1973; Desmedt, 1977). This finding correlates with the scarcity and diffuseness of any behavioural response before this age. It has been reported that the late components of the auditory response are poorly developed in Down’s syndrome and may be a measure of abnormal mental function and some forms of brain damage (Barnet, 1971). Harter et al (1977) used a black and white checkerboard pattern of various sizes to evoke visual potentials in ten infants between the ages of six and forty-five days. The smallest pattern subtended a visual angle of eleven minutes of arc and would only evoke a response within the macula region, while the largest patterns provided effective stimuli for peripheral vision. The authors found that all the check sizes produced a significant change in the visual evoked potential even in the youngest babies which would indicate that vision is better than 20/220 in the first week. Behavioural discrimination was not seen however until the twenty-seventh to fortieth day post term, and interestingly it was at this stage that the late components of the cortical response became more prominent and their presence and amplitudes were highly correlated with the percentage fixation time. The authors raised the possibility that these late components reflected cortical processing and their appearance marked the transition from passive to active visual discrimination. Krulisova and Figar (1979) could find no change in heart rate when stimuli were presented to young babies until the sixth to eighth week. From this time heart rate was consistently elevated and correlated significantly with attentive behaviour.

The young infant’s evoked potential is of much higher amplitude than the responses recorded in adults. This may be due in part to contamination from the high amplitude waves of the background EEG and because the infant’s thinner skull does not attenuate the activity so much. Less well developed inhibitory processes and more accessible generators, which become remote as the cortex continues to expand and enfold, also play a part (Ellingson, 1964; Thompson, 1978). Infant evoked potentials are also closely confined over the primary sensory projection areas. This distribution is seen during the first three months; then, at about the same time as the late components of the evoked potential develop, the distribution becomes more widespread and can be recorded over association areas of the cortex.

Throughout the first year the sensory systems are easily fatigued and evoked potentials cannot be recorded at high rates of stimulation. This is particularly so in the visual system where cortical potentials are only clearly evoked at flash rates slower than one every second. The rate of response improves to only 4 flashes per second by the end of the first year (Ellingson, 1964).

The wave form of cortical evoked potentials alters predictably with changes in state, for example, changes in attention and distraction. Because of the lability of such behaviours in the infant and young child, evoked potentials are far more variable both from moment to moment and from day to day than in adults. Only very large changes in the shape, latency and amplitude of the evoked potential can be used as indicators of, for example, the age of the child, his behaviour or of the presence of brain pathology.

There is no foolproof method for assessing mental function in children particularly when they are small, and many researchers have turned to the records of brain activity in the hope of finding an objective and reliable index of the ability of a child. Controversy has reigned over the last twenty to twenty- five years as to the usefulness of the EEG and new and better techniques of analysis seem only to have added fuel to the arguments.

Many claim that the patterns of activity in the EEG are related to intellectual development with the dominant EEG frequency and hemispheric associations being the two most significant features. Faster EEG frequencies have been recorded in more intelligent children although very often their controls have been the mentally retarded and brain-damaged individuals. Less regional differentiation and less marked asymmetries in the main frequencies arising from the two hemispheres together with significant slowing of the EEG have been cited as signs of limited cerebral processing. Fast, high voltage beta frequencies, focal slow waves and spikes and paroxysmal activity have frequently been associated with cognitive difficulties (Monnier, 1956; Vogel and Broverman, 1964,1966; Vogel et al , 1968; Nelson, 1969).

Evoked potential data is even more controversial. Ertle and his colleagues have argued that the latency of these responses relates to the efficiency of cortical processes. Thus shorter latencies correlated significantly with high IQ scores, whereas slow potentials were recorded in the dull and mentally retarded (Chalke and Ertle, 1965; Ertle, 1968,1971; Nodar and Graham, 1968; Ertle and Schafer, 1969). Rhodes et al (1969) found different features of the visual evoked potential correlated with intelligence. Bright children who scored between 124 and 140 on the WISC had greater asymmetry in the potentials evoked in the right and left hemispheres compared with children who only obtained a WISC score between 70 and 90, and this was most marked in the late components. Conflicting results of hemispheric differences have been reported by other authors (Richlin et al , 1971), while Martineau et al (1980) could only find amplitude differences between the evoked potentials of normal children compared with autistic and mentally retarded children, with greater amplitudes in the late components being found once again to correlate with higher intelligence.

Many investigators have failed to find any consistent feature of brain activity that can provide a valid measure of mental efficiency (Lindsley, 1940; Subirana et al, 1959; Netchine and Lairy, 1960; Netchine, 1967, 1968, 1969; Ellingson, 1966b; Petersen and Eeg-Olofsson, 1971). Netchine (1967) and Ellingson (1966 b) point out that normal individuals are frequently compared with mentally retarded patients manifesting clear evidence of brain damage. Differences in focal and paroxysmal slow and spike activity are then identified to differentiate the groups. This difference is a qualitative not a quantitative one and cannot be used as an index of differences in mental ability amongst clinically normal individuals. Ellingson further points out that the criteria used to judge both normal behaviour and brain activity have varied between authors, controls have been poor and methods deficient, and data are rarely reported on the reliability of the EEG. At best the correlations are low and although statistically significant are not large enough to be useful in identifying single individuals. It is quite possible that inappropriate information is being used in the EEG and techniques of analysis are often very simple. Moreover, investigations of the relationship between brain activity and behaviour have been almost exclusively cross-sectional in nature. Longitudinal studies would provide more informative data on such critical issues as individual differences in maturation and developmental outcomes.

Investigations of evoked potentials have also failed to produce any reliable index of intelligence and psychological function (Schenkenberg, 1970; Schenkenberg and Dustman, 1971; Thompson, 1978; Lowe et al, 1979). Individual variability, particularly in children, is so large and the reported differences in latency, amplitude and asymmetry of the waves of the evoked potentials so small, that significant differences can easily be lost and individual findings cannot be identified. Symmes and Eisengart (1971) draw attention to the problems of the huge variation in evoked potentials, not only between children but also during the recording of one child, and argue that extraneous variation due to lapses in attention presents more problems when working with children than with adults. Also eye blinks and eye movements occur frequently in the young and can produce a consistent artifact at the same latency as the late components of the evoked response to visually presented stimuli (Shelburne, 1973).


In spite of the fact that we have been able to record electrical activity through the intact skull for many years, very little is as yet understood about the exact relationship between the EEG and neuronal, biological and psychological function. At best we can record changes or signs within the EEG which are known to occur consistently at the same time as some aspect of behaviour or some change in physiological function. The data on these EEG signs are useful in studies of the premature infant and the neonate. Thus early in development, EEG changes appear to relate to universal innate maturational patterns whereas later in development individual differences become manifest and environmental and psychological factors complicate the findings. The research data on later developmental changes are fragmentary and the suggested significance of the EEG changes can only be tentative. Such psychological milestones as those of sensori-motor development and the emergence of speech have been relatively neglected in EEG and related research. Moreover, in the bulk of EEG and related research to date, data on brain activity in infants and young children have tended, implicitly or explicitly, to be evaluated against criteria derived from studies of adult subjects. Often this has resulted in findings being interpreted as manifesting a lack of, or a reduced level of, function with respect to some adult characteristic. This orientation has identified some fascinating and important associations between electrophysiological activity and maturation patterns in the premature infant and the neonate. On the other hand, it is an orientation which tends to overlook the potential uniqueness of many processes to the infant and young child. Recognition of such uniqueness has been the occasion for significant advances in other areas of developmental psychology. It is perhaps in this direction that the most promising future of developmental psychophysiology lies.


Adinolfi, A. M. (1971) ‘Postnatal development of synaptic contacts in cerebral cortex’, in Brain Development and Behaviour (eds. M. B. Sherman, D. J. McGinty and A. M. Adinolfi), Academic Press, 73—89.

Anders, T. F. and Roffwarg, H. P. (1973) The effects of selective interruption of sleep in the human newborn. Devel. Psychobiol, 1, 79-91.

Andersen, P. and Andersen, S. A. (1968) Physiological Basis of the Alpha Rhythm. Appleton Century Crofts, New York.

Baker, T. and McGinty, D. J. (1972) Characteristics of apneic episodes during sleep in kittens. Sleep Res., 3, 70.

Barnet, A. B. (1971) EEG audiometry in children under three years of age. Acta Otolaryngol. , 72, 1-13.

Barnet, A. B., Ohlrich, E. S., Weiss, I. P. and Shanks, B. (1975) Auditory evoked potentials during sleep in normal children from 10 days to 3 years of age. Electroenceph. Clin. Neurophysiol ., 39, 29-41.

Behrman, R. E., Babson, G. S. and Lessel, R. (1971) Fetal and neonatal mortality in white middle class infants. Amer. J. Dis. Child, 121, 486-489.

Benoit, O. (1967) Influences toniques et phasiques par le sommeil sur l’activite de la vie visuelle. J. Physiol. (Paris) 59, 295-317.

Berger, H. (1929) Uber das Elektrenkephalogram des Menschen. Arch. Psychiatr ., 87, 527-570.

Bergstrom, R. M. (1969) ‘Electrical parameters of the brain during ontogeny’, in Brain and Early Behaviour (ed. R. J. Robinson), Academic Press, 223-231.

Brazier, M. A. B. (1964) Evoked responses recorded from the depths of the human brain. Ann. N.Y. Acad. Sci., 112, 33-39.

Brazier, M. A. B. (1969) Electrophysiological studies of the hippocampus in man with average response computations. Acta Physiol, Acad. Sci. Hung., 26,107-116.

Bronson, G. (1969) Vision in infancy structure function relationships’, in Brain and Early Behaviour (ed. R. J. Robinson), Academic Press, 207-210.

Buchsbaum, M. and Fedio, P. (1970) Hemispheric differences in evoked potentials to verbal and non-verbal stimuli in the left and right visual fields. Physiol. Behav., 5, 207-210.

Chalke, F. C. R. and Ertle, J. (1965) Evoked potentials and intelligence. Life Sci. 4, 1319-1322.

Clemente, C. D., Purpura, D. D. and Mayer, F. E. (1972) Sleep and the Maturing Nervous System, Academic Press.

Cracco, J. B., Cracco, R. Q. and Stolove, R. (1979) Spinal evoked potentials in man: a maturational study. Electroenceph. Clin. Neurophysiol., 46, 58-64.

Creutzfeldt, O. D., Watanabe, S. and Lux, H. D. (1966) Relations between EEG phenomena and potentials of single cortical cells. Electroenceph. Clin. Neurophysiol., 20, 1-18.

Creutzfeldt, O. D. and Kuhnt, U. (1967) The visual evoked potential: physiological, developmental and clinical aspects. Electroenceph. Clin. Neurophysiol., 26, 29-41.

Curzi-Dascalova, L. (1977) EEG de veille et de sommeil du nourisson normal avant 6 mois d’age. Rev. EEG Neurophysiol., 7, 316-326.

Desmedt, J. E. (1977) Visual Evoked Potential in Man: New Developments. Oxford University Press.

Dittrichova, J. (1969) ‘Development of sleep in infancy’, in Brain and Early Behaviour (ed. R. J. Robinson), Academic Press, London, 193-201.

Dittrichova, J., Parul, K. and Vondracek, J. (1976) Individual differences in infants’ sleep. Develop. Med. Child Neurol, 18, 182-188.

Dobbing, J. (1966) The effect of undernutrition on myelination in the central nervous system. Biol Neonatorium, 9,132-147.

Dreyfus-Brisac, C. (1962) The electroencephalogram of the premature infant. Wld. Neurol, 3, 5-15.

Dreyfus-Brisac, C. (1964) ‘The electroencephalogram of the premature infant and the fullterm newborn’, in Neurological and Electroencephalographic Studies in Infancy (eds. P. Kellaway and I. Petersen), Grune and Stratton, New York, 186-207.

Dreyfus-Brisac, C. (1967) ‘Ontogenese du sommeil chez le premature humain: etude polygraphique’, in Regional Development of the Brain in Early Life (ed. A. Minkowski), Proceed. Symposium, Blackwell Sci. Pub., Paris, 437-457.

Dreyfus-Brisac, C. (1968) Sleep ontogenesis in early human prematurity from 24-27 weeks of conceptual age. Develop. PsychobioL, 1,162-169.

Dreyfus-Brisac, G. et Blanc, C. (1957) ‘Aspects EEG de la maturation cerebrale pendant la premiere annee de la vie’, in Conditionnement et Reactivite en Electroencephalographs, Masson, Paris, 432-440.

Eisengart, M. A. and Symmes, D. (1971) Effect of eyeblink on the visual evoked response in children. Electroenceph. Clin. Neurophysiol., 31, 71-75.

Ellingson, R. J. (1960) Cortical electrical responses to visual stimulation in the human infant. Electroenceph. Clin. Neurophysiol., 12, 663—677.

Ellingson, R. J. (1964) Studies of the electrical activity of the developing human brain’, in The Developing Brain (eds. W. Himwich and H. E. Himwich), Proc. Brain Res., 9, 26-53.

Ellingson, R. J. (1966a) ‘Method of recording cortical evoked responses in the human infant’, in Regional Maturation of the Nervous System in Early Life (ed. A. Minkowski), Blackwell, New York, 413-429.

Ellingson, R. J. (19666) Relationship between EEG and intelligence: a commentary. Psychol. Bull., 65,91-98.

Ellingson, R. J. (1967) The study of brain electrical activity in infants. Adv. Child Develop. Biol., 3, 53-97.

Ellingson, R. J., Dutch, S. J. and Mclntire, M. S. (1974) EEG of prematures: 3-8 year follow-up study. Develop. PsychobioL, 7, 529—538.

Ellis, R. R. and Ellingson, R. J. (1973) Responses to electrical stimulation of the median nerve in the human newborn. Develop. Biol, 6, 235-244.

Engel, R. (1964) Abnormal electroencephalogram in the newborn period and their significance. Amer. J. Ment. Defic., 69, 341-346.

Engel, R. (1965) Race and sex differences shown in the electroencephalographic pattern of maturation in the newborn period. Electroenceph. Clin. Neurophysiol., 19, 414-416.

Engel, R. and Butler, B. V. (1969) Individual differences in neonatal photic responses in the light of test performance at 8 months of age. Electroenceph. Clin. Neurophysiol., 26, 237.

Ertle, J. P. (1968) Evoked potentials, neural efficiency and I.Q. Proceed. Internat. Sympos. Biocybernetics.

Ertle, J. P. (1971) Fourier analysis of evoked potentials and human intelligence. Nature, 230, 525-526.

Ertle, J. P. and Schafer, E. W. P. (1969) Brain response correlates of psychomotor intelligence. Nature, 223, 421-422.

Ganoti, D., Rossini, P. M., Albertini, G., Soliazzo, D., Torriolo, M. G. and Poliduvi, G. C. (1980) Follow-up of visual evoked potential in fullterm and pre-term control newborns and in subjects who suffered from perinatal respiratory distress. Electroenceph. Clin. Neurophysiol., 48, 509- 516.

Gibbs, F. A. and Knott, J. R. (1949) Growth of the electrical activity of the cortex. Electroenceph. Clin. Neurophysiol., 1, 223-229.

Gibson, W. P. (1976) Evoked Response Audiometry, Oxford Univ. Press.

Globus, G. G. (1966) Rapid eye movement cycle in real time, implications for a theory of D state. Arch. Gen. Psychiat., 15, 654-659.

Graziani, L. J. (1974) The maturation and interrelations of EEG patterns and auditory evoked responses in premature infants. Electroenceph. Clin. Neurophysiol., 36, 367-375.

Graziani, L. J., Weitzman, E. D. and Velasco, M. S. A. (1968) Neurologic maturation and auditory evoked responses in low birth weight infants. Pediatrics, 41, 483-494.

Hague, I. (1968) ‘Development of waking EEG in normal infants during first year of life’, in Clinical Electroencephalography of Children (eds. P. Kellaway and J. Petersen), Almquist and Wiksell, Stockholm, 97-118.

Hague, I. (1972) Development of the EEG in normal infants during the first year of life. Acta, paediat. Scand., 232, 5-24.

Hague, I., Persson, J, Magnusson, R. and Petersen, I. (1972) ‘Spectral analysis via fast Fourier transform of waking EEG in normal infants’, in Automation of Clinical EEG (eds. P. Kellaway and I. Petersen), Raven Press, New York, 3-48.

Harter, M. R„ Deaton, F. K. and Odon, J. V. (1977) Maturation of evoked potentials and visual preferences in 6-45 day old infants, effects of check size, visual acuity and refractive error. Electroenceph. Clin. Neurophysiol, 42, 595-607.

Havlicek, V., Childiaeva, R. and Chernick, V. (1975) EEG frequency spectrum characteristics of sleep states in full term and pre-term infants. Neuropadiatrie, 6, 24-40.

Hill, J. D. N. and Parr, G. (1963) (eds.) Electroencephalography. McDonald and Co.

Himwich, W. A. (1974) ‘Developmental neurobiology’, in Biological Foundations of Psychiatry, Vol. 2, (eds. R. G. Grenell and S. Gabay), Raven Press, New York, 2-18.

Howe, P. M., Rayer, P. H. W., Williams, J. W. and Rudd, B. T. (1974) Growth hormone secretion during sleep in short children: a continuous sampling study. Arch. Dis. Child, 49, 246-251.

Hrbek, A., Karlberg, P. and Olsson, T. (1973) Development of visual and somatosensory evoked responses in preterm newborn infants. Electroenceph. Clin. Neurophysiol., 34, 225-232.

John, E. R. and Morgades, P. P. (1969) The pattern and anatomical distribution of evoked potentials and multiple unit activity elicited by conditioned stimuli in trained cats. Communications in Behavioural Biol, 3, 181-207.

Joseph, J. P., Lesevre, N., Dreyfus-Brisac, C. (1976) Spatio-temporal organisation of EEG in premature infants and fullterm newborns. Electroenceph. Clin. Neurophysiol, 40, 153-168.

Jouvet, M. (1961) ‘Telencephalic and rhombencephalic sleep in the cat’, in The Nature of Sleep (eds. G. E. W. Westenholme and M. O’Connor), Brown and Co., Boston, 188-208.

Kales, A. (ed.) (1969) Sleep Physiology and Pathology. Lippincott Co., Philadelphia and Toronto.

Kales, A., Bixler, E. O. and Kales, J. D. (1974) ‘Role of sleep research and treatment facility: diagnosis, treatment and education’, in Advances in Sleep Research, Vol. 1, (ed. E. D. Weitzman), Spectrum Pub. Inc., New York, 391-416.

Katsurada, M. L. (1965) Electroencephalogic study of sleep in infants and young children. Ann. Paediat. Jap., 11, 104-105.

Kellaway, P. and Petersen, I. (eds.) (1964) Neurological and Electroencephalographic Correlative Studies in Infancy, Grune and Stratton, New York.

Killam, E. K. and Killam, R. (eds.) (1976) ‘Physiological correlates of EEG’, in Handbook of Electroencephalography and Clinical Neurophysiology, Vol. 7, (ed.-in-chief A. Remond), Elsevier Pub. Co., Amsterdam.

Kooi, K. A. (1971) Fundamentals of Electroencephalography . Harper and Row.

Kraus, A. N., Solomon, G. E. and Auld, P. A. M. (1977) Sleep state, apnea and bradycardia in preterm infants. Develop. Med. Child Neurol, 19, 160-168.

Krulisova, H. and Figar, S. (1979) Development of heart rate response in early infancy. Acta. Ner. Super., 19, 274-275.

Lairy, G. C. (ed.) (1975) ‘The normal EEG throughout life’, Part B., ‘The evolution of the EEG from birth to adulthood’, in Handbook of Electroencephalography and Clinical Neurophysiology (ed.-in- chief A. Remond), Elsevier Pub. Co., Amsterdam.

Lemire, R. J., Loeser, J. D., Leech, R. W. and Alvord, E. C. (eds.) (1975) Normal and Abnormal Development of the Human Nervous System. Harper and Row.

Lindsley, D. B. (1940) Bilateral differences in brain potentials from the two cerebral hemispheres in relation to laterality and stuttering. J. Exp. Psychol, 55, 197-213.

Lombroso, C. T. (1979) Quantified electrographic scales on 10 preterm healthy newborns followed up to 40-43 weeks of conceptual age by serial polygraphic recordings. Electroenceph. Clin. Neurophysiol., 46, 460-474.

Lowe, M. D., Rogers, L. J., Purves, S. J. and Dunn, H. G. (1979) ‘Spontaneous and evoked cerebral electrical activity and localisation of language function in children with minimum cerebral dysfunction’, in Human Evoked Potential Applications and Problems (eds. D. Lehmann and E. Callaway), Plenum Press, New York.

Martineau, J., Laffont, F., Bruneau, N., Roux, S. and LeLord, G. (1980) Event related potentials evoked by sensory stimulation in normal, mentally retarded and autistic children. Electroenceph. Clin. Neurophysiol., 48, 140-153.

McGinty, D. J. (1971) ‘Encephalisation and the control of sleep’, in Brain Development and Behaviour (eds. M. B. Sterman, D. J. McGinty and A. M. Adinolfi), Academic Press, 335-358.

McGinty, D. J., Harper, R. M. and Fairbanks, M. (1974) ‘Neuronal unit activity and the control of sleep states’, in Advances in Sleep Research, Vol. 1, (ed. E. D. Weitzman), Spectrum Pub. Inc., New York, 173-214.

Metcalf, D. R. (1969) The effect of extrauterine experience in the ontogenesis of EEG sleep spindles. Psychosomatic Medicine, 3, 393-399.

Metcalf, D. R. and Jordan, K. (eds.) (1972) EEG Ontogenesis in Normal Children in Drug, Development and Cerebral Function. C. Thomas, Springfield, Illinois.

Mizuno, N., Clemente, C. and Sauerland, E. R. (1969) Projections from the orbital gyrus in the cat. II.

To telencephalic and diencephalic structures. J. Comp. Neurol., 136, 127- 142.

Monnier, M. (1956) Le Probleme de Stade en Psychologie de FEnfant. P.U.F., Paris.

Morgane, P. J. and Stern, W. C. (1974) ‘Chemical anatomy of brain circuits in relation to sleep and wakefulness’, in Advances in Sleep Research, Vol. 2, (ed. E. D. Weitzman), Spectrum Pub. Inc., New York, 1-131.

Nelson, G. K. (1969) A neuro-psychological study of the children of the Knysna Forest workers. Psychol affric., 12, 143-171.

Netchine, S. (1967) Apports de la methode longitudinale a l’etude de la maturation de l’electro- encephalogramme et de ses relations avec le developpement psychologique chez l’enfant normal. Enfance, 20, 347-351.

Netchine, S. (1968) Maturation de 1’electroencephalogramme et developpement psychologique. Rev. Neuro-psychiat. infant, 16, 249-256.

Netchine, S. (1969) ‘L’activite electrique cerebrale chez l’enfant normal de 6 a 10 ans’, in Croissance de Tenfant, Genese de Fhomme, Vol. 2, (ed. R. Zarzo), P.U.F., Paris, 246 pp.

Netchine, S. et Lairy, G. C. (1960) Ondes cerebrates et niveau mental. Quelques aspects de 1’evolution genetique du trace EEG suivant le niveau mental. Enfance, 4-5, 427-439.

Noda, H. and Adey, W. R. (1970) Firing variability in cat associated cortex during sleep and wakefulness. Brain Res., 18, 513-526.

Nodar, R. H. and Graham, J. T. (1968) An investigation of auditory evoked responses of mentally retarded adults during sleep. Electroenceph. clin. Neurophysiol., 25, 73-76.

Nolte, R. and Haas, G. (1978) A poly graphic study of bioelectric brain maturation in preterm infants.

Develop. Med. Child Neurol, 2, 167—182.

Oswald, I. (1962) Sleeping and Waking, Elsevier, Amsterdam.

Parmelee, A. H., Franz, M. D., Schulte, J., Akiyama, Y., Wenner, W. H, Schultz, M. A. and Stein, E. (1968) Maturation of EEG activity during sleep in premature infants, Electroenceph. clin. Neurophysiol., 24, 319-329.

Parmelee, A. H., Akiyama, Y., Stern, E. and Harris, M. A. (1969) A periodic cerebral rhythm in newborn infants. Exp. Neurol, 25, 575.

Parmelee, A. H. and Stern, E. (1972) ‘Development of states in infants’, in Sleep and the Maturing Nervous System (eds. C. D. Clemente, A. H. Parmelee and S. Mayer), 199-215.

Perry, N. W. and Childers, D. G. (1969) The Human Visual Evoked Response Method and Theory.

C. C. Thomas, Springfield, Illinois.

Petersen, I. and Eeg-Olofsson, O. (1971) The development of the EEG in normal children from the age of 1 through 15 years: non-paroxysmal activity. Neuropadiatric, 3, 277-304.

Petre-Quadens, O. (1969) Contribution a l’etude de la phase dite paradoxale. Acta med. Belg., Bruxelles, 134-150.

Pompeiano, O. (1969) ‘Sleep mechanisms’, in Basic Mechanisms of the Epilepsies (ed. H. H. Jasper), Little Brown, Boston, 453-473.

Prechtl, H. F. R. (1968) ‘Polygraphic studies in new born infants’, in Ontogenesis of the Brain (eds. L.

Jilek and S. Trojan), University Karbua Praha, 437-438.

Prechtl, H. F. R, Weinmann, H. and Akiyama, Y. (1969) Organisation of physiological parameters in normal and neurologically abnormal infants. Neuropaediatrie, 1, 101-129.

Prichard, J. B. (1964) ‘The character and significance of epileptic seizures in infancy’, in Neurological and Electroencephalographic Correlative Studies in Infancy (eds. P. Kellaway and I. Petersen), Grune and Stratton, 273-285.

Purpura, D. P. (1971) ‘Synaptogenesis in the mammalian cortex’, in Brain Development and Behavior (eds. M. B. Sterman, D. J. McGinty and A. M. Adinolfi), Academic Press, New York, 23-41. Rechtschaffen, A. and Kales, A. (1968) A Manual of Standardised Terminology Techniques and Scoring System for Sleep Stages of Human Subjects, U.S. Government Printing Office, Washington, D. C.

Reisen, A. H. (1971) ‘Problems of correlating behavioural and physiological development’, in Brain Development and Behavior (eds. M. B. Sterman, D. J. McGinty and A. M. Adinolfi), Academic Press, New York, 59-70.

Rhodes, L. E., Dustman, R. E. and Beck, E. C. (1969) The visual evoked response, a comparison of bright and dull children. Electroenceph. Clin. Neurophysiol, 27, 364-372.

Richlin, M., Weinstein, S., Weisinger, M. (1976) Development of neurophysiological indices of retardation: interhemispheric asymmetry of the visual evoked cortical response. Int. J. Nenrosci., 6,257-261.

Richlin, M., Weisinger, M., Weinstein, S. and Giannin, I. M. (1971) Interhemispheric asymmetries of evoked cortical responses in retarded and normal children. Cortex, 7, 98-105.

Robinson, R. J. and Tizard, J. P. M. (1966) The central nervous system in the newborn. Brit. Med. Bull, 22,49-55.

Samson-Dollfus, D., Forthomme, J. and Capron, E. (1964) ‘EEG of the human infant during sleep and wakefulness during the first year of life’, in Neurological and Electroencephalographic Correlative Studies in Infancy (eds. P. Kellaway and I. Petersen), Grune and Stratton, New York, 208-229.

Samson-Dollfus, D., Nogues, B., Verdue-Ponssin, A. et Malleville, F. (1977) Electroencephalo- gramme du sommeil de l’enfant normal entre 5 mois et 3 ans. Rev. EEG Neurophysiol., 7, 335-345.

Sandler, L. S. and Schwartz, M. (1971) Evoked responses and perception—stimulus content versus stimulus structure. Psychophysiol., 8, 727-739.

Scheibel, M. E. and Scheibel, A. B. (1971) ‘Selected structural-functional correlations in postnatal brain’, in Brain Development and Behavior (eds. M. B. Sterman, D. J. McGinty and A. M. Adinolfi), Academic Press, New York, 1-21.

Schenkenberg, T. (1970) Visual, auditory and somatosensory evoked responses in normal subjects from childhood to senescence. Unpub. Ph.D. Dissertation, Univ. of Utah.

Schenkenberg, T. and Dustman, R. E. (1971) Visual, auditory and somatosensory evoked responses related to age, hemisphere and sex. Proc. Amer. Psychol. Assoc., 78, 183-184.

Schulte, F. J., Hirze, G. and Schrempf, G. (1972) ‘Maternal toxemia, foetal malnutrition and biochemical brain activity in the newborn’, in Sleep and the Maturing Nervous System (eds. C. D. Clemente, D. D. Purpura and F. E. Mayer), Academic Press, New York, 419-442.

Shelburne, S. A. (1973) Visual evoked responses to language stimuli in normal children. Electroenceph. Clin. Neurophysiol., 34, 135-143.

Sterman, M. B. (1972) ‘The basic rest-activity cycle and sleep’, in Sleep and the Maturing Nervous System (eds. C. D. Clemente, D. D. Purpura and F. E. Mayer), Academic Press, New York, 175-197.

Sterman, M. B. and Clemente, D. C. (1974) ‘Forebrain mechanisms for the onset of sleep’, in Basic Sleep Mechanisms (eds. O. Petre-Quadens and J. D. Schlag), Academic Press, New York, 54-71.

Sterman, M. B. and Hoppenbrouwers, T. (1971) ‘The development of the sleep-waking and rest- activity patterns from fetus to adult in man’, in Brain Development and Behavior (eds. M. B. Sterman, D. J. McGinty and A. M. Adinolfi), Academic Press, New York, 203-225.

Sterman, M. B., Harper, R. M., Havens, B., Hoppenbrouwers, D. R., McGinty, D. R. and Hodgman, J. E. (1977) Quantitative analysis of infant EEG development during quiet sleep. Electroenceph. Clin. Neurophysiol., 43, 371-385.

Sterman, M. B., McGinty, D. J. and Adinolfi, A. M. (1971) Brain Development and Behavior, Academic Press, New York.

Stern, M., Fram, D. H., Wyatt, R., Grinspoon, L. and Tursky, B. (1969) All night sleep recordings of acute schizophrenics. Arch. Gen. Psychiat ., 20, 470-477.

Stern, M., Parmelee, A. H. and Harris, M. A. (1973) Sleep state periodicity in prematures and young infants. Develop. Psychobiol, 6, 357-365.

Subrana, A., Corominas, J., Oller-Daurell, L., Maso-Subrana, E. et Hermandez, A. (1959) Donnes EEG apportees par 1’influence de la maturation sur la dominance hemispherique’. Rapp, au XVIIeme Congres de f Association des Ped. de Langue Francaise, Montpellier, 149-204.

Sutton, S. (1969) ‘The specification of psychological variables in an average evoked potential experiment’, in Average Evoked Potentials: Methods, Results and Evaluations (eds. E. Donchin and D. B. Lindsley), NASA, Washington, D.C., 237-297.

Symmes, D. and Eisengart, M. A. (1971) Evoked response correlates of meaningful visual stimuli in children. Psychophysiol., 8, 769-779.

Thomas, J. E. and Lambert, E. H. (1960) Ulnar nerve conduction velocity and H. reflex in infants and children. J. App. Physiol, 15,1-9.

Thompson, C. R. S. (1978) The development of flash visual evoked potential techniques for the diagnosis of visual disorders with dense opacities of the optic media. Unpub. Ph.D. Diss., Univ. of Aston, Birmingham, England.

Umezaki, H. and Morrell, F. (1970) Developmental study of photic evoked responses in premature infants. Electroenceph. Clin. Neurophysiol., 28, 55-63.

Varner, J. L., Ellingson, R. J., Danaby, T. and Nelson, B. (1977) Interhemispheric amplitude symmetry in EEGs of full term infants. Electroenceph. Clin. N euro physiol, 43, 846-852.

Vogel, W. and Broverman, D. M. (1964) Relationship between EEG and test intelligence: a critical review. Psychol Bull, 62, 132-144.

Vogel, W. and Broverman, D. M. (1966) A reply to ‘Relationship between EEG and test intelligence’: a commentary. Psychol. Bull, 65, 99-109.

Vogel, W., Broverman, D. M. and Klaiber, E. L. (1968) EEG and mental abilities. Electroenceph. Clin. Neurophysiol., 24, 166-175.

Wagner, A. L. and Buchthal, F. (1972) Motor and sensory conduction in infancy and childhood: a reappraisal, Develop. Med. & Child Neurol, 14, 189-216.

Webb, W. B. (1969) ‘Twenty-four hour sleep cycling’, in Sleep and Sleep Pathology (ed. A. Kales), Lippincott, 53-65.

Weitzman, E. D. (ed.) (1974) Advances in Sleep Research, Vol. 2, Spectrum Pub. Inc.

Weitzman, E. D. and Graziani, L. J. (1968) Maturation and topography of the auditory evoked response of the prematurely born infant. Electroenceph. Clin. Neurophysiol., 23, 82-83.

Werner, S. S., Stockard, J. E. and Bickford, R. G. (1977) Atlas of Neonatal Electroencephalography, Raven Press, New York.

Yakovlev, P. I. (1976) ‘Morphological criteria of growth and maturation of the nervous system in man’, in Biological Foundation of Psychiatry (eds. R. G. Grenell and S. Gabay), Raven Press, New York, 169-178.

Dickerson JWT. and McGurk H. Brain And Behavioural Development. (1982) Blackie & Son Ltd., Glasgow.

Brain and Behavioural Development - 1982: 1 Neural Development | 2 Comparative Neural | 3 Malnutrition | 4 Hormones and Growth Factors | 5 Cortical Activity | 6 Functional Asymmetry | 7 Plasticity | 8 Sex Differences

Cite this page: Hill, M.A. (2019, June 20) Embryology Book - Brain and behavioural development 5. Retrieved from

What Links Here?
© Dr Mark Hill 2019, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G