Further Prototypes of Ego Formation—A Working Paper from a Research Project on Early Development

Further Prototypes of Ego Formation—A Working Paper from a Research Project on Early Development

For many years, separately and together, the authors have been conducting research with infants. These investigations have involved detailed observations of infant behavior, electroencephalographic studies, naturalistic studies of mother-infant interaction, and a number of theoretical explorations derived from these and other investigations.

Our recent research efforts have been inspired by Freud's trailbreaking work in Three Essays on the Theory of Sexuality(1905). The principal theoretical statements which have guided us were formulated systematically in three publications of Spitz (1958), (1959), (1961).

The following propositions from an overall framework for our research.

1. There is no aspect, activity, function, or structure of the psyche that is not subject to development.

2. Development is the resultant of the interplay of innate and experiential factors which themselves are often inextricably interwoven.

3. Innate factors include hereditary aspects and aspects pertaining to intrauterine and intrapartum events.

4. The role of the experiential factors in the inception, unfolding, and structuring of the psychic apparatus has two sources: exchanges with the surround and exchanges within the organism. The latter exchanges consist of interactions between incipient psychic operations and the innate physiological prototypes for some subsequently emerging psychic structures. The maturation of these prototypes takes place as an epigenetic unfolding; as a forward moving, irreversible progressive growth and differentiation, programmed as it were by a maturational clock.

We have set ourselves the following tasks:

1. To collect experimental data which will test the correctness of the theoretical statements in question.

2. To construct appropriate models of the development and structure of the emerging psychic system on the basis of these statements. The value of these models should be tested further by systematic observation and experiment.

3. To demonstrate the explanatory power of these models within the framework of the system of psychoanalytic theory.
We would like to illustrate this approach by some initial findings from this rather ambitious program. These findings are psychobiological in nature and involve behavioral units which are common to every individual of the species. Examples from the current phase of our research include the following areas of development: (1) rapid eye movement states (REM states and REM sleep); (2) quiet, "deep" sleep; (3) effects of stress on neonatal sleep; (4) smiling; (5) normal fussiness.

Method and Subjects

The central strategy of our research project includes the weekly or biweekly study of individual infants from before birth through the end of the first year of life. In addition, cross-sectional studies, involving a larger number of infants, are used to explore hypotheses generated by our longitudinal studies. The design of our project includes alternating home visits and visits to our infant neurophysiological laboratory. During home visits, we make naturalistic observations of the infant, obtain a narrative account from the mother of changes since our last visit, and test the infant's responsivity to a variety of standardized stimuli.

The laboratory visits for electroencephalograms and observations are quite different. After the application of EEG and polygraphic leads, a laboratory recording session begins with a feeding of the infant and continues with a transition into sleep and through about 90 minutes of sleep recording. The recording includes direct observation of behavior, as well as continuous electroencephalogram, electro-oculogram, respiration, electromyogram, and evoked response recording.

To date, twenty-two infants have been studied longitudinally. Of these, thirteen infants have been studied intensively during the first three postnatal months, and nine infants have been studied through the first year. Over five hundred have been studied cross-sectionally.

Sleep Investigations

In common parlance, sleep and sleeping are considered equivalent: one is either asleep or awake. At best, the vernacular distinguishes light sleep from deep sleep with dreams ascribed to one or the other. Psychoanalysis has had, from the beginning, a different approach to the problems of sleep. At the outset, Freud (1900), (1901) was primarily interested in the exploration of dreams, which he recognized as the manifestations of the mental life during sleep. He demonstrated that determinism applies to dreams and all psychic activity in the same way as it applies to the physical world. But his differentiation went further than this. Being primarily interested in the problems of dreams and their elucidation, he noted from the very beginning the possible occurrence of dream states outside of the so-called sleep state, in the form of daydreams. He postulated that (1) sleep would be found to be different from the waking state in its physiological functioning and (2) that specific parts of the brain would be found to be involved in the dream process (1900). Furthermore, he dealt in the same period with a tangential but related phenomenon, that of memory. Freud anticipated modern brain physiology with his proposition of a double registration in the memory systems (1900), (1925). It is interesting to note that, following Freud's investigations, Silberer (1912) discussed the importance of recognizing two qualitatively different kinds of sleep.

Physiological studies of sleep tended to follow two major investigative directions. One direction was that exemplified by the work of von Economo (summarized in 1929) on sleep centers and the neurophysiological work of Hess (1924, 1925), among others. The other direction was that explored for many years by Kleitman (1929). This culminated in the 1953 and 1955 reports, by Aserinsky and Kleitman, which documented in infants the cyclic occurrence of episodes of body motility associated with rapid eye movements. This was soon followed by the reports of Dement (1955) and Dement and Kleitman (1957) indicating that a similar rapid eye movement-bodily activity cycle was seen in adults, and that rapid eye movement (REM) periods normally occurred about every 90 minutes throughout a night's sleep; furthermore, these REM sleep periods were found to be associated with a specific EEG and with dreaming. Much of the emergent physiological and psychophysiological research on sleep and dreaming, and its implications for psychoanalytic theory, has been reviewed and discussed by Fisher (1965).

As a result of these pioneering investigations, we now think of sleep as being composed of a number of different stages (I, II, III, IV, I-REM). Each of these stages is described by multiple criteria, which consist of the presence or absence of eye movements, EEG configuration, muscle tone, and sometimes respiratory and heart rate patterns.1

From the psychoanalyst's point of view it is of primary interest that two of these criteria appear to correspond closely to propositions advanced fifty years earlier by Freud. The first of these is the finding of greatly reduced muscular activity (via active inhibition) during sleep as measured by the chin electromyogram. We remind our readers that Freud (1900) postulated that the inhibition of motility in sleep not only facilitates dreaming, but is also necessary for dreaming to take place. The second of these criteria is the regular occurrence of penile erections during REM sleep—a phenomenon which has been studied by Fisher et al. (1965).

We have been intrigued to find that infant sleep and adult sleep show both significant similarities and important differences. This holds true for behavior as well as physiology. Our investigations of infant sleep were stimulated by these facts in conjunction with the central importance of sleep and dreams to psychoanalysis and the importance of Freud's genetic view.

REM "Sleep" in the Newborn
An initial cross-sectional study resulted in a new conception of neonatal states (Emde and Metcalf, 1970). The combination of detailed behavioral observations and polygraphic recording provided us with information previously not available. In the adult we are accustomed to thinking of rapid eye movements as occurring in sleep. But the matter is not so simple in infancy. In newborns we observe rapid eye movements occurring in a number of different circumstances. Not only do they occur during the sleep cycle when the eyes are closed (sleep REM), but they also occur at times when the eyes are open and glassy (drowsy REM), during times when the infant is engaged in nutritional sucking (sucking REM), and during some times when the infant is fussing or crying (fussy REM and crying REM). We established that two trained behavioral observers reached better than 95 percent agreement in judgments of each of these behavioral states.

The states of drowsy REM, sucking REM, fussing REM, and crying REM have consistent electrophysiological correlates—correlates which are also present in the REM sleep of adults and older children. Even though the infant had his eyes open or was engaging in nutritional sucking, or was fussing or crying, he frequently had the same electrophysiology as the type of "sleep" during which the eyes were closed and rapid eye movements were conspicuous. We carried out an independent minute-by-minute study which compared a behavioral observer's interpretation of state with an EEG specialist's simultaneous EEG and polygraphic findings during the same period of observation. In the analysis of nearly twenty hours of data collection from ten normal newborns, the behavioral observer judged a total of 214 minutes to be either drowsy REM, sucking REM, fussing REM, or crying REM states. In the independent judgments of the EEG and polygraph, the electroencephalographer judged over 98 percent of these (or 210 minutes) to be REM states indistinguishable from REM sleep.

We have used the term "undifferentiated" because drowsy REM, sucking REM, fussy REM, and crying REM disappear over the first three months of postuterine life (Metcalf and Emde, 1969). In addition, neonatal REM sleep shows a relatively high variability in physiological patterning which tends toward stability over the first three months. These changes are concomitant with another major change in the sleep cycle itself; by three months, sleep characteristically begins with a non-REM (NREM) instead of a REM period. This pattern of sleep onset will from here on remain the same. It is the adult pattern.

EEG Development in Quiet Sleep
It is evident that our EEG studies relate at every step to our behavioral studies. The EEG is used here as one tool in a psychobiological, interdisciplinary research program based on the principles we have already stated. The study of the human electroencephalogram is recent, dating from the work of the psychiatrist Hans Berger in the 1930's. Less well known is the fact that Berger quickly recognized that the human EEG undergoes progressive change throughout life. This is particularly evident from birth to three years. Because of this marked developmental change, the EEG provides excellent opportunity for the study of central nervous system maturation and development. Many workers, notably Ellingson (1967) in recent years, have contributed to knowledge in this domain; it has been our particular area of interest for some time.

In the neonatal period, although awake and sleep EEGs can often be distinguished from each other, this differentiation is sometimes difficult and unreliable, because REM sleep and undifferentiated "nonsleep" REM states share the same electrophysiology. The two main sleep stages, on the other hand, "active" REM sleep and "quiet" NREM sleep, show clearly different EEGs.

The EEG of REM sleep is one of extensive cortical activation and in many ways resembles the EEG associated with the alert, awake state. It shows a low amplitude pattern whose rhythms appear to be fast and irregular. During this state there is activation of many physiological systems as evidenced by extreme variability of pulse, respiration, and blood pressure, by a rise in brain oxygen utilization by REMs, by penile erections, and by increased body motility. Even the lack of muscle tonus during the REM state is not a passive phenomenon, but is a result of CNS activation; muscle tonus is reduced through the mechanism of active inhibition.2 It is noteworthy that REM-state physiology, for all its inherent variability, changes very little throughout life once its characteristic integrated patterning, loosely present at first, becomes established at about three months.

Quiet sleep also shows many important changes during the first three months, and, in contrast to REM sleep, shows continued development throughout infancy and childhood. The quiet NREM sleep EEG shows waveforms that are more regular, slower, and of higher amplitude than those of REM sleep. During quiet sleep there are no REMS, respirations are very regular, thresholds to arousal are high, and infants are generally motionless, except for occasional spontaneous body jerks or startlelike movements. It can, indeed, be surprisingly difficult to awaken an infant from this state. At about age three weeks, the quiet sleep EEG becomes more differentiated as a result of increased regularity of slow rhythms which replace the previously somewhat chaotic picture of this stage. Four to seven days later (at about five weeks), a momentous EEG change occurs, namely the onset of "sleep spindles." These EEG wave sequences are easily recognizable. We have shown that this seemingly maturational step can be accelerated by the impact of experience (Metcalf, 1969).

At about eight to twelve weeks, quiet sleep shows a further differentiation, the emergence of definable stage II sleep. Stage II sleep is marked by the presence of well-formed sleep spindle bursts and a more characteristic, regular EEG appearance. The onset of sleep spindles probably indicates a complex maturational and developmental step. Existing and partially functioning excitatory and inhibitory brain systems become capable of integrated, self-regulating interaction. This step may have primary CNS integrative significance. Part of our research is concerned with searching for behavioral manifestations or correlates of this physiological development.

After the establishment of Stage II sleep at about three months, another kind of quiet sleep begins to emerge. We see here the beginning of adult types of deep, quiet sleep. This form of quiet sleep may be labeled "Stage III/IV." After this integrative patterning is complete, EEG development proceeds in a slower and apparently smooth fashion until about five or six months. When, as part of these developments, sleep Stages, II, III, and IV become clearly distinguishable, three different kinds of deep sleep are formed out of what had been one undifferentiated kind of deep sleep. This is an oft-overlooked and important developmental step, in the course of which a further major EEG pattern emerges, namely spontaneous K-complexes. The K-complex is a specific brief series of waveforms, consisting of conspicuous deviations from the ongoing EEG tracing. K-complexes are similar to sensory evoked responses (Cobb, 1963), and can probably be elicited by sensory stimulation at earlier ages, but they do not occur spontaneously during sleep until after five months (Metcalf et al., 1970). When elicited at older ages, there is evidence that the patterned waveform of the K-complex varies according to the psychological meaning of the sensory stimulation (Oswald, 1962).

The K-complex may be considered to be an electrophysiological indicator of CNS information processing. This view is supported by many basic neurophysiological investigations such as those of Dawson (1958) and Barlow (1964). We are intrigued by the fact that spontaneous K-complexes are not seen before five to six months. This may be taken to indicate a new and important step in the capacity of the CNS to respond to itself; i.e., to respond in a patterned and perhaps selective way to processes arising in one part of the CNS and acting on other parts. Therefore, the emergence of spontaneous K-complexes may be connected with the increasing systematic manipulation of memory traces.
Somewhat apart from, but still connected with the preceding, is the question of drowsiness. Up to three months, the observational impression of drowsiness is not paralleled by characteristic changes in the EEG. The latter appear for the first time around three months. It would appear that drowsiness, an important psychological condition, has now acquired a specific EEG pattern. This linkage becomes clearer after six months and continues into the latency period.
Thus it appears that the age of three months is a period of critical developmental importance. EEG and physiological patterns become more clearly organized and systematically integrated with certain behaviors such as drowsiness, sleep behaviors, and sleep cycles. A further, rather abrupt change in the direction of increased differential organization is the development of spontaneous K-complexes during NREM quiet sleep at about five to six months.
Thus our attention is drawn to three nodal points in early EEG development. Our work, in broader perspective, utilizes the totality of EEG development in relation to a variety of behavioral developments, and focuses on these nodal points as "anchor points" of assumed special CNS maturational significance. The three points are aspects of quiet sleep development. They are the development of sleep spindles at four to six weeks, the development of Stage II sleep at about three months, and the development of spontaneous K-complexes at five to six months. A fourth nodal point (resting on the REM-state EEG) is the more complex integration of behavioral, physiological, and EEG manifestations, whereby at about three months undifferentiated REM states disappear and sleep-onset REM is replaced by mature sleep-onset NREM.
Effects of Stress on Neonatal Sleep
Normal hospital routine requires blood drawing from the newborn during the first three days of life. This is performed by pricking the infant's heel with a sharp stylette. We observed that this blood drawing procedure is frequently followed by long, relatively motionless NREM sleep periods. Curiosity about this led us to explore the effects of such a stress-producing (painful) event at this age (Emde et al., 1970). Hospital circumcision, done routinely without anesthesia, was chosen as an independent variable with sleep patterns following circumcision as dependent variables. Circumcision, done by plastibel technique, results in a gradually developing ischemic necrosis of the foreskin which could be expected to produce a continual bombardment of stimulation of pain pathways for many hours.
Results in two observational studies were dramatic. In the twelve hours following circumcision, most infants showed a primary increase in NREM or deep sleep. In a subsequent polygraphic study, normal male infants were observed for a ten-hour period on each of two successive nights. Continuous recordings consisted of electroencephalogram, eye movement recording, electromyogram, respiration recording, and behavioral observations. Twenty infants were studied on two successive nights beginning at twenty-four hours of age. One half of the infants (control group) did not have circumcisions or heel pricks during the period of study; the other half (experimental group) were circumcised during the beginning of the second night's recording and observation. Eight out of ten circumcised infants showed a major increase in NREM or quiet sleep on the night following circumcision. Percentage increases of this deep NREM sleep varied from 41 to 121 percent. In contrast to this, in the undisturbed control group, the total amount of NREM sleep varied little from the first to the second night—the greatest single increase was less than 3 percent. These results are significant at well beyond the .01 level of confidence.
The cause of this phenomenon is not immediately understandable. A "common sense" guess about the effects of a continual disruptive stimulation at this age is that an infant would sleep less and cry more. Our results showed the opposite. Cirmumcised infants slept more and were awake less, and cried for the same amount of time as before circumcision. These results suggest an inborn adaptive mechanism which responds to stress by producing a quiescent state with high sensory thresholds. We are currently in the process of studying individual differences in regard to this phenomenon, as well as the endocrinological basis for it.
As derived from the Genetic Field Theory of Ego Formation(Spitz, 1959) our research project considers that there are certain periods during the first year when physiological and behavioral development progresses at an increased pace and in a definite normative sequence. It is inferred from that theory that these periods are accompanied by changing thresholds to stimulation and by affective changes, which are of crucial significance in the development of social relations. Because of this, we have paid particular attention to the behaviors of smiling and fussiness.
Our investigations have included detailed cross-sectional and longitudinal studies of smiling during the first three months (Emde and Koenig, 1969a), (1969b); (Emde, 1970). In the normal newborn, smiling is linked to the states of sleep REM and drowsy REM and occurs as one of many well-circumscribed state-related "spontaneous behaviors." Since it is determined not by external stimulation but by intrinsic physiological rhythms, we refer to it as endogenous smiling. It is found with increased frequency in prematures (Emde, McCartney, and Harmon, 1970). We have reason to believe endogenous smiling is mediated through brainstem mechanisms since it was present in a microcephalic infant with virtually no functioning cerebral tissue (Harmon and Emde, 1970). In the normal newborn, endogenous smiling occurs at a mean rate of 11 smiles per 100 minutes of REM period. It is evenly distributed across successive REM periods of an interfeeding interval; thus, it cannot be considered an expression of a "tensionless" or hunger-free condition.
Newborn frowning, on the other hand, appears to be of two types. Like smiling, it occurs as a spontaneous state-related behavior. In addition, during later REM periods of an interfeeding interval and also during wakefulness, it occurs as an expression of distress. Endogenous smiling and endogenous frowning along with other spontaneous REM-associated behaviors wane sharply in the period between eight and twelve weeks.
Exogenous smiling, on the other hand, is not present at birth. It begins as an irregular response which is first elicited during wakefulness at about three weeks of age. It may occur to a wide variety of nonspecific stimuli in auditory, kinesthetic, tactile, and visual modalities, but it is unpredictable, rare, and fleeting. With ensuing weeks it occurs more often, but it does not become predictable until the eight-to-twelve-week period at which time it is best elicited by the visual stimulus configuration which Spitz and Wolf (1946) termed the "essential sign Gestalt" (stimulus consisting of hairline, eyes, nose, and motion). This is commonly considered the time of onset of social smiling or of the "smiling response" (Ambrose, 1961); (Polak et al., 1964a); (Gewirtz, 1965). Within two weeks after its onset, however, the adequate stimulus for eliciting this response becomes more complex, as three dimensions are required (Polak et. al., 1964b), and very soon after that the face of the mothering person becomes the most potent elicitor. In other words, soon after the eight-to-twelve-week period, exogenous smiling becomes more specifically social and endowed with meaning. As exogenous smiling is becoming more specific and enriched with psychological meaning, endogenous smiling, which is physiologically determined, is decreasing.
In parallel with this countermovement is still another countermovement. In one of our longitudinal research projects (Tennes et al., 1970) we have studied in detail the fussiness which normally occurs in most infants between three and twelve weeks of age. Extreme fussiness at this age has often been characterized clinically as "colic"; although its precise etiology is unknown, several groups have documented the fussiness of this age period (Spitz, 1951); (Stewart et al., 1953); (Wessel et al., 1954); (Brazelton, 1962); (Paradise, 1966).
We found fussiness, unrelated to hunger, occurring in all of the infants studied during this age period. By twelve weeks the fussiness wanes and nothing else like it is observed through the first year of life. The time of waning of this fussiness is concomitant with the infant achieving capacities for being a more active regulator of contact with his environment, both in initiating stimulus contact and in terminating it. Again we are struck with the countermovement. As the capacity for mastery and active trial-and-error behavior is increasing, fussiness is decreasing.
In the foregoing we have presented some data from our studies on neonatal development. These form one part of our research program and we selected them because of the developmental parallels and connections they present. These studies are:
1. Neonatal sleep and REM states of infancy.

2. The development of smiling.

3. The development of differentiation of the "mature" form of deep sleep.

4. The effects of stress on neonatal sleep.
Our findings support a number of Freud's theories and the principles he elaborated on the origin, organization, and functioning of the psychic system.
Our data impress us with the operation of two parallel lines acting and interacting in the progressive unfolding of the organism. These are the lines following maturation, the innate; and development, the experiential. They converge toward ever-increasing organization and regulation of the organism's functioning. The rates of change in the development of quiet sleep, the REM state, and smiling parallel each other in their progression. This progression is gradual from birth until the postnatal age of about six weeks, where we have found an important turning point in the nature of the interaction between maturation and development.
In the six weeks following this first turning point, the changes accelerate, with mounting evidence of incipient awareness and psychic functioning, to a second turning point in the third month. We have designated this turning point the emergence of the first organizer of the psyche. It is marked by clearly defined modifications in the behavioral pattern of the same three areas (quiet sleep, REM states, smiling). These modifications run parallel with a change in the EEG pattern, which has become organized and in which sleep spindles have appeared (see Table 1).
Until age six weeks the EEG pattern present when the above-mentioned behaviors were observed, was categorically different from the EEG patterns associated with a comparable behavior in the adult. After age six weeks, an increasing organization of the rather undifferentiated neonatal pattern begins to become rapidly evident. Sleep behavior becomes more specific; it approaches more and more the adult form. A good example of this is REM sleep, which represents 60 percent of total sleep in the neonate and only 20 percent in that of the adult. After the first turning point, the REM percentage decreases progressively. After the second turning point (age two and a half months) REM stages, which until then appeared indiscriminately during both sleep and apparently awake periods, disappear completely from the nonsleep periods.
0-6 weeks 6-12 weeks 10-12 weeks
REM STATE Occurs during behavioral sleep (eyes closed), drowsiness, nutritional sucking, fussing, and crying. Occurs decreasingly during drowsiness, sucking, fussiness, and crying. Occurs only during behavioral sleep and is more patterned from the physiological point of view.
Neonatal Pattern: sleep begins with a REM state. Continued neonatal pattern. Neonatal pattern disappears; "adult" pattern of sleep onset is now present.
QUIET SLEEP EEG rhythmic activity becomes organized at 4-6 weeks. No EEG spindles before 4 weeks. Spindle bursts first seen at 5-7 weeks. EEG rhythmic activity is poorly differentiated between "light" and "deep" sleep. Spindles become more defined. Stage III/IV EEG sleep ("deep" quiet sleep) begins to differentiate from stage II. Spindle maturation complete. Hypersynchronous drowsy pattern begins to emerge after 12 weeks.
SMILING Endogenous: Occurs during sleep REM and drowsy REM at a rate of approximately 11 per 100 minutes. Occurs at the same rate or at a slightly decreased rate. Endogenous smiling wanes.
Exogenous: Irregular response during wakefulness: elicited by a wide variety of stimulation in visual, auditory, tactile, and kinesthetic modalities. Occurs with increased frequency, but is still irregular and in response to nonspecific sensory stimulation. A regular response to the "essential sign Gestalt."
FUSSINESS Endogenous nonhunger fussiness appears during the latter part of this period (3-6 weeks). Intermittent bursts of prolonged nonhunger fussiness. Prolonged nonhunger fussiness disappears.
The development of smiling is another example of this unfolding. We found that smiling shows that the endogenous physiological origin of this behavior decreases progressively during the first six weeks; in the second six weeks it is progressively displaced by a psychologically determined behavior. As endogenous smiling decreases, there is a proportionate increase in exogenous smiling. The stimulus for exogenous smiling originates in the surround, whereas the endogenous smile occurs as a function of innate internal rhythmic processes. The exogenous smile begins as a response to a wide variety of stimuli and progressively is transformed into a specific response to the human face. After three months, the endogenous smile occurs mainly in REM sleep and is not a prominent behavior. In other terms, while the endogenous smiling had the characteristic of physiological rhythmicity, the exogenously determined behavior has the characteristic of awareness continuing into anticipation. Exogenous smiling represents a turning from inside perception (response to inner stimuli) to the perception of the surround. In terms of motivation, the progression goes from tension discharge to awareness with active exploration and anticipation. The latter is a specifically psychological process.
From the viewpoint of the psychic apparatus, we see at this age the emergence of a number of psychic structures (e.g., memory, anticipation, meaningful directed response). These psychic structures are modeled on neonatal physiological processes, which we have called "prototypes" (Spitz, 1958). Each of these physiological processes has its innate phylogenetically predetermined maturational trajectory, and develops independently from every other one. Under the influence of the first organizer of the psyche, situated at age three months, they tend at certain maturational levels to form a range of predictable relationships with each other. They become coherent. But this maturation does not proceed alone, uninfluenced by other factors. At all developmental levels maturationally guided processes are turned into developmental processes as a result of the adaptations enforced by exchanges with the surround and the organism's response to them.
In this sense, postnatally, the concept "maturation" is a useful construct, even though unrealizable: all is development. For development is the resultant of the constant interactions between environment (or experience) and the innate: these interactions operate at all levels, whether molecular, cellular, or organismic.
Maturation, as a functional reality, is that aspect of development which provides potentialities on the one hand, and limitations on the other. Organizers, particularly in the context of this discussion, were first described (Spitz, 1958) as "emergent, dominant centers of integration … [forming a] field of forces from which a dominant center of integration, the first organizer of the psyche, will emerge." Development is not blind. It is responsive to the surround in terms of the law of effect. In the ontogenesis of individual behavior the law of effect plays the same role as that played by natural selection (through survival) in phylogenesis.
Conversely, maturation is blind, for it is the product of phylogenesis over geological time spans. Development, through the impact of experience, is the means through which maturationally given potentialities are realized. It is one of the inducers of organizers. Organizers are formed out of species-specific, innate potentialities and directions, interacting with the species-unique demands of the surround.
Organizers introduce a new modus operandi into the psychic system. Indeed, we believe that the psychic organizer is equivalent to the development of a new modus operandi through adaptive exchanges. The organizer of the psyche is not a physical structure; it is not even a psychic structure. It is the emergence and establishment of a different way of processing the psychic givens. It introduces a better adapted way, which takes advantage of the opportunities offered by the surround. With this different way, the integration of these givens on a higher level of complexity becomes possible. From here on the new, more adapted modus operandi becomes predominant in the psychic processes. This will obtain until, at the next level, the growing complexity and number of elements achieved through the instrumentality of the organizer make a new step necessary. That step is the development of the next and more advanced modus operandi which will constitute the next organizer. As the term states, a more complex, more highly integrated reorganization of the psychic givens will now begin.
Postnatally the distinction between maturation and development becomes therefore increasingly constrained: by the same token a boundary line between psyche and soma becomes ultimately meaningless.
During the earliest postnatal days, approximately the first two to three weeks, physiological or somatic aspects of development are predominant as determinants of observable behavior. During this period, interactions between innate and surround become involved with an increasing variety of experience. The interactions increase in complexity and simultaneously become progressively more organized. These early patterned constellations of innate behavior and adaptive functioning, which culminate in volitional behavior and physiological adaptation, tempt one to speculate about their relationship to early ego or "pre-ego" development.
An outstanding example of the role of physiological prototypes for later psychic development and structure is breast-feeding behavior, and its "intaking" aspects in the neonate (Erikson, 1950); (Spitz, 1957).
Examining this earliest physiological process and its interplay with differential experience sheds light on its emergence as the physiological prototype of an ego nucleus. Psychic structures will encroach on the discharge processes which this nucleus provides. These psychic structures will in turn modify the functions of the ego nucleus to the point where their origin becomes unrecognizable, though they will continue to contain evidences of their somatic beginning.
In studying the prototype potentialities of neonatal sleep for subsequent development of the psychic system, of its structure, of its organization, of its activity, we have followed certain principles which have been outlined elsewhere. (Spitz, 1957), (1958), (1961); (Spitz and Cobliner, 1965). These principles are:
1. Psychic activity, function, structure, etc., are not innate.

2. What is innate is:

a. the variable capacity for learning and adaptation;

b. the capacity for making use of neurophysiological and morphological givens for coping with the environment.
However, the physiological way the innate copes with its environment is not the way the psyche does it. The prototype is not a blueprint for a future psychic entity. It provides the available means for the later developing psychic structure and the limits within which they can operate. We realize that when we say "structure," we are taking liberties with this term. Though the subsequent development certainly is structured, it mostly is rather a modus operandi (e.g., defense mechanisms) and more rarely a coherent psychic structure (e.g., the superego). The modus operandi itself is not homologous with its physiological prototype. In view of the adaptive requirements of development, a direct continuation of the prototype is most unlikely. What will develop is an analogy that we must explore in the spirit of Robert Oppenheimer (1956): "One has to widen the framework a little and find the disanalogy which enables us to preserve what was right about the analogy." In this spirit, when examining neonatal sleep as a possible physiological prototype for later psychic functions, we shall stress more how it differs from sleep in the adult than dwell on the obvious similarities.
Furthermore, the neonate, during "nonsleep" periods, which could pass for "wakefulness" in the adult, shows undifferentiated REM states. The neonate begins behavioral sleep in a REM state; whereas the adult falls asleep in a NREM state. And while the adult shows a generalized diminution of motor activity during REM sleep, the neonate is behaviorally active despite paradoxical suppression of muscle tonus.
These differences suggest that REM sleep probably serves a different function in the neonate. In the adult, REM states are connected with dreaming. What, we may ask, would correspond to dreaming in the neonate? The very fact that the organism is compelled to adjust neonatal sleep to the adult pattern within the first three postnatal months suggests that the neonatal sleep pattern is probably not suitable for the function sleep will have to perform after the emergence of the first organizer. It is highly probable that an interference with this adjustment would encroach upon the development of a sharp demarcation of diurnal wakefulness and, concurrently, with the perception of reality.
The function of sleep and REM in the neonate is categorically different from their function in the adult. It is a good example of the uses of disanalogy, for it permits us to sense where analogies lie—not where we expect them. For instance, the neonate's REM activity is triple that of the adult. Yet what can correspond in the neonate to dream activity, that rich mental life of the adult during REM sleep? And, even more significantly, in studies of premature infants, the percentage of REM activity increases in inverse proportion to the gestational age of the subject.
We need hardly argue that the psychic content, the material of dreams, is as yet nonexistent in the neonate, and a fortiori in the premature. We cannot therefore consider neonatal sleep and its REM activity in the same line as sleep REM in the adult. In the neonate's first three weeks of life, REM states are present, both during apparent wakefulness and sleep. We therefore come to the conclusion that they probably represent the operation of maturational processes in the CNS of the neonate, which are primarily related to physiological processes and hardly influenced as yet by experience. In our opinion, these maturational processes are part of an unfolding genetic progression, phylogenetically preformed for an average expectable environment. The processes involve the establishment, by practice and channelization, of the necessary connections within the CNS as well as in CNS behavioral integration.
What then does neonatal sleep represent? In our opinion, a physiological prototype for later function. It is sleep behavior in transition from an exclusively physiological phenomenon to adult sleep, a psychophysiological phenomenon. Neonatal sleep occurs during an existence practically devoid of interference from the surround, let alone exchanges with it. Conversely, adult sleep includes, among its functions, the enormously important psychic function of discharging the tension originating during the previous day's exchanges with a rich, varied, and ever-changing surround, probably with the help of the dream.
It is then not surprising that, by comparison with adult sleep, prototype REM sleep is poorly organized. But in the neonate it only takes three months of development to organize it. Three months is that developmental level at which a number of other psychic phenomena also converge, forming on the one hand a rudimentary ego organization, initiating on the other the preobjectal stage in the formation of object relations. And it is precisely at this level that the physiological patterning of the REM state becomes consistent. Now the REM state becomes firmly linked with behavioral sleep and can no longer be confused with wakefulness. Sleep spindles become definitely established in the sleep EEG.
If REM physiology in the adult belongs exclusively to behavioral sleep, then it has no place in the psychic life of directed, volitional action. After the three-month level, action, whether in the motor or mental sphere, will become increasingly incompatible with the REM state. Can this provide an explanation of the spectacular decrease of REM activity after the ego becomes established?
At the present state of our knowledge the linking of adult REM sleep with dreaming permits us only the assumption that during dreaming a CNS activity is going on. Some aspects of these linkages will become clearer when problems related to memory, object formation, and perhaps mental representation will have been adequately investigated.
One thing is certain: REM sleep is neither the cause nor a direct result of dreaming. That is evident from the difference of its role in the neonate and in the adult. REM sleep is the indication of specific ongoing processes in the CNS: in the neonate these processes are exclusively physiological. In the adult the vast realm of the psychological is added.
The ages at which REM and EEG pattern changes appear in the infant are quite suggestive through their convergence. The organization and structuration of sleep spindles in the EEG pattern and their maturation are coterminous with the appearance of the smiling response, the behavioral indicator of the first organizer. At the same period the REM states become limited to behavioral sleep. The adult "pattern" of NREM sleep onset also becomes established. Independently, as a result of certain theoretical considerations, we also situated the emergence of the preobject as well as of a rudimentary ego structure (Spitz, 1957) at this stage. Lastly, we may ask whether it is the progressive patterning of EEG and REM states which is one more indicator of the emergence and structuration of the first organizer of the psyche.
As happens so frequently, answering one question makes us aware of more unanswered ones. We concluded that prototype sleep in the neonate represents a transition from physiological sleep to sleeping as a psychophysiological function in the adult. But if REM sleep is an indication of a specific CNS activity in the neonate, then it may well correspond more to the adult's waking mental activity than to adult REM sleep.
What then would be the function of the neonate's NREM sleep in terms of prototypes? Should we assume that it corresponds to Engel's (1962) "conservation-withdrawal" to stress? Our finding that the neonate's response to circumcision consists in a spectacular increase in NREM sleep certainly supports this hypothesis. If so, of what is this the prototype?
Evidently there are quite a number of defense mechanisms of which one might think: denial, repression, withdrawal, regression, etc. One might even consider the neonatal NREM response as a possible prototype for defense as such.
However, in our opinion these initial observations are not sufficient for drawing such fundamental theoretical conclusions. Much more research will be needed to establish the further developmental steps which will follow this first one. We will have to observe the subsequent modifications of behavior in the course of the first year before drawing any conclusions.
The same can be said of the problem of the age of the onset of dreaming. Our work does not yet bear on this, inasmuch as we have not yet adequately investigated problems related to memory, object constancy, and mental representation. Furthermore, any useful speculations will require solution of the problems of dream reporting (or of reasonable inferences about dream experiences) during the first two years of life. We may remind the reader here that Freud's daughter dreamed the "strawberry dream" when she was nineteen months old (Freud, 1900).
Recent dream and sleep research (some of which has been noted here) again raises the question from a different direction. Is there any evidence from the physiological study of sleep in general and REM sleep (associated with adult dreaming) in particular that sheds light on the age of onset of dreaming?
We do not know of any studies of infant REM sleep which explore this question. We have already mentioned our EEG work on the ontogenesis of quiet NREM sleep (stages II, III, IV) culminating in the development of spontaneous K-complexes at about six months. We would suggest that the development of this CNS information-processing signal is a prerequisite for CNS readiness to support dreaming during sleep. It is known that there is occasional mental content also during NREM sleep. It may be that both kinds of sleep are necessary for dreaming. One possibility is that NREM sleep favors the more formal, secondary process aspects of day-residue processing, while REM sleep subserves the rich, associational, affect-guided psychic activity of dreams as generally understood.
These findings lead us to search carefully for the relation between developing integration of the physiological system and the development of psychic structures of ever-increasing complexity, i.e., ego development. We ask ourselves whether the appearance of spontaneous K-complexes can be brought into meaningful statistical relationship with behavior at somewhat more advanced stages of development, when object constancy is achieved (Hartmann, 1952); (Spitz and Cobliner, 1965), or insight behavior becomes manifest in problem solving.
Many more problems confront us and the opportunities for research in the field of prototypes are numerous. We will not touch on them in our present report, reserving these problems for our future publications on the subject. But we will end with a brief discussion of the methodology of our research on prototypes. By this time, prototypes have a long history. After Freud, Hartmann (1939) was the first to take up the concept of prototypes as the explanatory principle for some ego defenses. As discussed elsewhere (Spitz, 1961), a number of authors approached the tempting problems of prototypes with very little actual definition of the term prototype. We therefore thought of establishing criteria for identifying a physiological entity as a prototype. Although it is easy to detect similarities in one respect or another between a psychic phenomenon and some physiological process, that does not make it a prototype.
We believe that in our present report we have demonstrated the usefulness of two such criteria, both genetic in character. One is the criterion of convergence of different developmental lines (Anna Freud, 1965). The process of psychological development of any physiological prototype or rather of its component elements, be they behavior, function, EEG, physiology, etc., will inevitably mesh with various other developmental processes, progressively converging to a point where they become coterminous and interact to form what we have called an organizer of the psyche. They are the constituents of the organizer and, at the same time, components of the subsequent changes in the development which arise from it. We believe that in a prototype such a convergence should be demonstrable as a developmental line—a line that leads from the prototype itself to the end result.
Our second criterion is equally genetic. For, to become a credible physiological prototype for a psychic entity emerging much later, the prototype must be the starting point of a developmental line. Accordingly, it can be observationally, mensurationally, and experimentally followed in its development. Decrease of physiological component elements can be quantitatively demonstrated. Their replacement through psychic devices, mechanisms, structures can be observed and the modification of the EEG verifiably be demonstrated.
This is what we have attempted to show in our present report. These are the principles we are applying to the other projects of our research program.
And this is indeed an application of Oppenheimer's recommendation to explore the disanalogies. For analogies can be found everywhere. But the genetically progressing series of disanalogies forms the meaningful coherent link for us, leading from physiological inception to psychological completion.
Accordingly, we would like to conclude by offering a whimsical recommendation for psychoanalytic developmental research: "Take care of the disanalogies; the analogies will take care of themselves!"
1 The concepts, light sleep and deep sleep, are valuable but imprecise and do not refer to sufficiently invariant physiological correlates. Stage IV sleep, the deepest stage of quiet sleep in the human, is associated with very high and steady thresholds to arousal by sensory stimulation. Stage I REM (REM sleep), associated with dreaming, is so unique as to be thought of as a separate state. It has, for this reason, been termed the "D State" by Ernest Hartmann (1967).
2 During REM sleep there is much motor activity, especially of face (smiles, frowns, etc.), hands, and feet as well as body activity (twisting, stretching, etc.). Muscle tonus reduction occurs concurrently and is most evident around the mouth and chin. REM-sleep body motility to this degree is not seen in the adult; the age when change from infant motility toward adult motility takes place is under study.
1 Ambrose, J. A. 1961 The Development of the Smiling Response in Early Infancy In:Determinants of Infant Behavior ed. B. M. Foss. New York: Wiley, Vol I pp. 179-201
2 Aserinsky, E. & Kleitman, N. 1953 Regularly Occurring Periods of Eye Motility and Concomitant Phenomena During Sleep Science 118 273-274
3 Aserinsky, E. & Kleitman, N. 1955 A Motility Cycle in Sleeping Infants as Manifested by Ocular and Gross Bodily Activity J. Appl. Physiol. 8 11-18
4 Barlow, J. S. 1964 Evoked Responses in Relation to Visual Perception and Oculomotor Reaction Times in Man Ann. N.Y. Acad. Sci. 112 432-467
5 Berger, H. 1929 ber das Elektroenkephalogramm des Menschen Arch. Psychiat. Nervenkr. 87 527-570
6 Brazelton, T. B. 1962 Crying in Infancy Pediatrics 29 579-588
7 Cobb, W. A. 1963 The Normal Adult EEG In:Electroencephalography ed. D. Hill & G. Pharr. New York: Macmillan, p. 246
8 Dawson, G. D. 1958 The Central Control of Sensory Inflow Proc. Roy. Soc. Med. 51 531-535
9 Dement, W. C. 1955 Dream Recall and Eye Movements During Sleep in Schizophrenics and Normals J. Nerv. Ment. Dis. 122 263-269
10 Dement, W. C. & Kleitman, N. 1957 Cyclic Variations in EEG During Sleep and Their Relation to Eye Movements, Body Motility, and Dreaming EEG Clin. Neurophysiol. 9 673-690
11 Ellingson, R. J. 1967 The Study of Brain Electrical Activity in Infants In:Advances in Child Development and Behavior ed. L. P. Lipsitt & C. C. Spiker. New York: Academic Press, Vol. 3 pp. 53-97
12 Emde, R. N. 1970 Endogenous and Exogenous Smiling in Early Infancy Read at the American Psychiatric Association, San Francisco, Calif., May 13, 1970
13 Emde, R. N. Harmon, R. J., Metcalf, D. R., Koenig, K. L., & Wagonfeld, S. 1970 Stress and Neonatal Sleep Submitted to Science.
14 Emde, R. N. & Koenig, K. L. 1969a Neonatal Smiling and Rapid Eye Movement States J. Amer. Acad. Child Psychiat. 8 57-67
15 Emde, R. N. & Koenig, K. L. 1969b Neonatal Smiling, Frowning, and Rapid Eye Movement States J. Amer. Acad. Child Psychiat. 8 637-656
16 Emde, R. N. McCartney, R. D., & Harmon, R. J. 1970 Neonatal Smiling in REM States: IV. Premature Study Child Develpm (in press).
17 Emde, R. N. & Metcalf, D. R. 1970 An Electroencephalographic Study of Behavioral Rapid Eye Movement States in the Human Newborn J. Nerv. Ment. Dis. 150 376-386
18 Engel, G. L. 1962 Anxiety and Depression-Withdrawal Int. J. Psychoanal. 43 :89-97 (IJP.043.0089A)
19 Erikson, E. H. 1950 Childhood and Society New York: Norton.
20 Fisher, C. 1965 Psychoanalytic Implications of Recent Research on Sleep and Dreaming J. Am. Psychoanal. Assoc. 13 :197-303 (APA.013.0197A)
21 Fisher, C. Gross, J., & Zuch, F. 1965 A Cycle of Penile Erection Synchronous with Dreaming (REM) Sleep: Preliminary Report A.M.A. Arch. Gen. Psychiat. 12 29-45
22 Freud, A. 1965 Normality and Pathology in Childhood New York: International Universities Press.
23 Freud, S. 1900 The Interpretation of Dreams Standard Edition 4 & 5 London: Hogarth Press, 1953
24 Freud, S. 1901 The Psychopathology of Everyday Life Standard Edition 6 London: Hogarth Press, 1957
25 Freud, S. 1905 Three Essays on the Theory of Sexuality Standard Edition 7 125-243 London: Hogarth Press, 1953
26 Freud, S. 1925 A Note upon the 'Mystic Writing-Pad.' Standard Edition 19 227-232 London: Hogarth Press, 1961
27 Gewirtz, J. L. 1965 The Course of Infant Smiling in Four Child-Rearing Environments in Israel In:Determinants of Infant Behavior ed. B. M. Foss. New York: Wiley, Vol. III pp. 205-248
28 Harmon, R. J. & Emde, R. N. 1970 Spontaneous REM Behaviors in a Microcephalic Infant (in preparation).
29 Hartmann, E. 1967 The Biology of Dreaming Springfield: Thomas.
30 Hartmann, H. 1939 Ego Psychology and the Problem of Adaptation New York: International Universities Press, 1958
31 Hartmann, H. 1952 The Mutual Influences in the Development of Ego and Id Psychoanal. Study Child 7 :9-30 (PSC.007.0009A)
32 Hess, W. R. 1924, 1925 ber die Wechselbeziehungen zwischen psychischen und vegetativen Funktionen Schweiz. Arch. Neurol. Psychiat. 15 260-277 16 36-55, 285-306
33 Kleitman, N. 1929 Sleep Physiol. Rev. 9 624-665
34 Metcalf, D. R. 1969 Effect of Extrauterine Experience on Ontogenesis of EEG Sleep Spindles Psychosom. Med. 31 393-399
35 Metcalf, D. R. & Emde, R. N. 1969 Ontogenesis of Sleep in Early Human Infancy Psychophysiology 6 264
36 Metcalf, D. R. Mondale, J., & Butler, F. K. 1970 Ontogenesis of Spontaneous K. Complexes Psychophysiology (in press).
37 Oppenheimer, R. 1956 Analogy in Science Amer. Psychol. 11 127-135
38 Oswald, I. 1962 Sleep and Waking New York: Elsevier.
39 Paradise, J. 1966 Maternal and Other Factors in the Etiology of Infantile Colic J.A.M.A. 197 191-199
40 Polak, P. R., Emde, R. N., & Spitz, R. A. 1964a The Smiling Response to the Human Face: I. Methodology, Quantification, and Natural History J. Nerv. Ment. Dis. 139 103-109
41 Polak, P. R., Emde, R. N., & Spitz, R. A. 1964b The Smiling Response to the Human Face: II. Visual Discrimination and the Onset of Depth Perception J. Nerv. Ment. Dis. 139 407-415
42 Silberer, H. 1912 Symbolik des Erwachens und Schwellensymbolik berhaupt Jb. Psychoanal. & Psychopath. Forsch. 3 621-660
43 Spitz, R. A. 1951 The Psychogenic Diseases in Infancy Psychoanal. Study Child 6 :255-275 (PSC.006.0255A)
44 Spitz, R. A. 1954 Genèse des premières relations objectales Rev. Franç. Psychanal. 28 479-575
45 Spitz, R. A. 1957 No and Yes New York: International Universities Press.
46 Spitz, R. A. 1958 On the Genesis of Superego Components Psychoanal. Study Child 13 :375-404 (PSC.013.0375A)
47 Spitz, R. A. 1959 A Genetic Field Theory of Ego Formation New York: International Universities Press.
48 Spitz, R. A. 1961 Early Prototypes of Ego Defenses J. Am. Psychoanal. Assoc. 9 :626-651 (APA.009.0626A)
49 Spitz, R. A. & Cobliner, W. G. 1965 The First Year of Life New York: International Universities Press.
50 Spitz, R. A. & Wolf, K. M. 1946 The Smiling Response Genet. Psychol. Monogr. 34 57-125
51 Stewart, A., Weiland, J., Leider, A., Mangham, C., & Ripley, H. 1953 Excessive Infant Crying in Relation to Parent Behavior Amer. J. Psychiat. 110 687-694
52 Tennes, K., Emde, R. N., Kisley, A. J., & Metcalf, D. R. 1970 The "Stimulus Barrier" in the First Three Months of Life (in preparation).
53 von Economo, C. 1929 Die Encephalitis Lethargica Berlin & Vienna: Urban & Schwarzenberg.
54 Wessel, M., Cobb, J., Jackson, E., Harris, S., & Detwiler, A. 1954 Paroxysmal Fussing in Infancy, Sometimes Called Colic Pediatrics 15 421-434

This publication is protected by US and international copyright lawsand its content may not be copied without the copyright holder's express written permission except for the print or download capabilities of the retrieval software used for access. This content is intended solely for the use of the individual user.
Psychoanalytic Study of the Child, 1970; v.25, p417 (25pp.)