DEVELOPMENTAL PSYCHOBIOLOGY, Volume 9, Number 1: Pages 89-95,
January 1976.

Early Behavioral
Differences: Gender
or Circumcision?

Unit for Research on the
Medical Applications of Psychology

University of Cambridge
Cambridge, England

Department of Obstetrics and Gynecology
Georgetown University Medical School
Washington, D.C.

In human studies, the possible long-term effects of behavior of early physical insult or pharmacological agents have received little attention. We present both circumstantial and direct evidence that circumcision of male infants leads to behavioral changes. In some American studies using circumcised infants, reported gender differences may instead be the result of the altered behavior of circumcised males. We suggest that circumcision requires more study in its own right and that it requires description if not control in all neonatal and infancy studies.

From time to time a report appears in the literature on human development demonstrating that a "minor" stimulus event has consequences of such disproportionate magnitude and such temporal tenacity that it strains belief. An example of such a report was Whiting, Landauer, and Jones's (1968) suggestion that a single inoculation given before 2 yr of age produces taller adults than the same inoculation give after 2 yr. As adults and behavior scientists, we tend to view this and other reports with embarrassment if not disbelief.1 How can a single, brief, innocuous event have such a profound and lasting experience?

Given our license to do many things to infant animals that we may not do to human infants, the literature on animal development has a more extensive collection of such accounts which come from 2 sources. One source, inadvertently illuminating, is experimentation wherein treated animals are compared with sham-treated as well as untreated controls and where significant differences occur between the 2 control groups. The other source is the purposeful comparison of the size and duration of experimental treatments applied while the animal is biologically immature.

Both types of studies contain communalities, complexities, and subleties that are instructive in considering analogous experimentation with human beings. As an example, consider an animal study finding that a drug administered during infancy has a demonstrable effect on adult behavior, physiology, or anatomy. As adults, the experimental group is different from the 2 control groups (vehicle-injected group and untreated noninjected group) and, moreover, the 2 control groups are different from each other. To what processes are those differences attributable? In this prototypical case several possibilities exist and are typically confounded:

      (1) a pharmacological agent may have contributed to the result, e.g., the drug under study, the vehicle in which it was injected, and any anesthetic agent used at the time of injection.

      (2) a physical insult may have contributed to the result, e.g., the hypodermic injection by which the drug was administered.

      (3) a change in nonsocial sensory input may have been contributory, e.g., a sudden drop in ambient temperature upon removal from the nest.

      (4) a change in social stimulation may have been responsible for the effects, e.g., a change in maternal behavior toward those young who were removed and subsequently returned to the nest.

These possibilities are listed above in the order which, historically speaking, they have appeared obvious to animal experimenters. The first received earliest and most methodological attention whereas the last has attracted attention only relatively recently (Daly, 1973; Meier & Schutzman, 1968; Richards, 1966).

On the other hand, for experimenters concerned with human development, the historical order of obviousness has been reversed. Studies of the later effects of early mother-infant separation have been accumulating in large quantity for over a quarter of a century, whereas studies of the long-term effects of early pharmacological intervention are of recent introduction and few in number. Effects on infant development of analgesic drugs given to mothers during labor have only recently been established (e.g., Brackbill, Kane, Manniello, & Abramson, 1974). Some of these effects, including those on habituation measures, persist for at least 1 month - the oldest infants yet investigated (Aleksandrowicz & Aleksandrowicz, 1974; Conway and Brackbill, 1970), whereas changes in social interaction measures have been found up to 60 weeks (M.P.M. Richards & J.F.Bernal, unpublished data). Given the very rapid acceptance of concepts like deprivation and stimulation both by behavioral scientists and those responsible for making and administering social policies affecting children, that so little attention has been paid to what animal scientists are the more obvious sources of early influences on later behavior is indeed remarkable.

Our purpose here is to examine the possiblity that 1 instance of the 2nd category of significant events, i.e. physical insult in the form of male circumcision, has both behavioral and physiological consequences that may have been uniformly misinterpreted by developmental scientists.

The practice of circumcision varies widely throughout the world. In the United States, for example, it is so routine that no demographic statistics are kept on its incidence. Even the extensive data collection of the National Institute for Neurological Diseases and Stroke Collaborative Study did not include this statistic. The present authors had to inquire personally about the frequency of circumcision at 18 American teaching hospitals. Replies indicated that 83% of the 14,116 male infants born during 1973 had been circumcised as neonates in the hospital. In the United Kingdom and much of Europe, where medical opinion holds that circumcision is unnecessary and potentially dangerous, it is rarely performed. In 1972 the Hospital Inpatient Inquiry for England and Wales (Department of Health and Social Security, 1974) showed that for a population of 400,000 male children under 1 year of age, only .41% had been circumcised.2 We may assume, therefore, that when male infants had been used in neonatal behavioral research most American subjects will have been circumcised; most European subjects will not have been. We note that those studies reporting gender differences in neonatal behavior (e.g. Korner, 1973) are American rather than European and suggest that some of the so-called gender differences may be the result of altered behavior following circumcision. We will first consider evidence that circumcision does influence neonatal behavior and then return to the more general issue of gender differences in neonatal behavior.

We know of only 3 studies that directly attempt to examine the behavioral consequences of circumcision. All found significant effects. Emde, Harmon, Metcalf, Koenig and Wagonfeld (1971) found that circumcision was followed by prolonged nonrapid eye movement (NREM) sleep. Not only was the amount of NREM increased, but the latency to onset on NREM sleep decreased and both the total number of NREM sleep periods and the number of very long NREM sleep periods increased. The authors conceptualized circumcision as a form of stress. More recently, Anders and Chalemian (1974) also examined sleep after circumcision, find prolonged wakefulness with fussing and crying during the hours immediately after circumcision. They suggested that later NREM sleep shifts may be secondary to changes in wakefulness and also pointed out that the differing techniques used for circumcision (either acute incision or incision and ligature causing ischemic necrosis of the foreskin for several days) may have differential effects on behavior.

Brackbill (in press) has investigated the effects of circumcision on responses to continuous auditory stimulation in 30 male neonates. Comparison groups were females and uncircumcised males. Arousal level, as indexed by heart rate (HR) and state, was measured under continuous auditory stimulation (white noise) during he experimental period and no added stimulation during the control period. No differences appeared between females and uncircumcised males during control or experimental periods for any of the state of HR measures used. However, when scores for females and uncircumcised males were compared to those for circumcised males, 7 of the 16 comparisions were significantly different.

From the results of all 3 studies we may conclude that circumcision does have effects on infant behavior and psychophysiology. The direction of those effects and their duration await clarification. In view of all the evidence showing long-term behavioral, physiological, anatomical, and even neurophamacological effects of "minor" events in early animal development (e.g., Levine, 1969), we would be unwise to assume without empirical demonstration that the circumcision effects are short-lived.

      Gender Differences in Neonatal Behavior

      As we remarked earlier, American studies have frequently reported newborn gender differences in, for example studies of sensory thresholds or sensitivity to sensory stimuli. The reports include findings of differential sensitivity to cutaneous stimuli (Bell & Costello, 1964; Wolff, 1969), electrotactual stimulation (Lipsitt & Levy, 1959), photic stimulation (Engel, Crowell, & Nishijima, 1968), and to gustatory stimulation (Nisbett & Gurwitz, 1970).

British and Dutch studies of the newborn (e.g., Bench, 1969; Bench, Collyer, Langford, & Toms, 1972; Hutt, Lenard, & Prechtl, 1969; Prechtl, 1972) did not find sex differences, provided that the sample had been screened to preclude babies from abnormal deliveries. (In unscreened samples, a greater proportion of males will have suffered from perinatal complications which may be associated with behavioral differences [see Pederson & Bell, 1970].) As yet no study has appeared from Europe using measures of newborn sensitivity to cutaneous, electrotactual, or photic stimulation that are directly comparable to the measures used in American studies which report gender differences. Hence the possible contribution of the stressful effects of circumcision to these particular differences in threshold or sensitivity remains to be tested directly.

Comparable evidence does not exist in studies of gestation. Nisbett and Gurwitz (1970) reported that female subjects had a greater preference for sweet substances than did male, presumably circumcised subjects. On the other hand no such gender differences were found in studies where uncircumcised males served as subjects (Desor, Maller, & Turner, 1973; Dubignon, Campbell, Curtis, & Parrington, 1969). (Information on circumcision was provided by D. Campbell & J.A. Desor, personal communication.)

      Long-Term Consequences

Even though short-lived neonatal gender differences (whether or not related to circumcision) could not affect the parent-child interaction and set up persistent styles of interchange differing for boys and girls. These, in turn, may have consequences for later development. Evidence exists from several American studies on differences in maternal handling of girls and boys in the early weeks of life (e.g., Lewis, 1972; Moss & Robson, 1968; Thoman, Liederman, & Olson, 1972), but comparable studies in England using similar categories of behavior yields very few gender differences (Bernal, 1972, 1973; Richards & Bernal, 1972). Richards and Bernal studied a sample of 80 infants from birth to school age. No male subject was circumcised. During the first 10 days of life, extensive observations were carried out in feeding situations. Information was also gathered on sleeping, crying, and feeding patterns. The results have been systematically searched for gender differences; no more have been found than would have occurred on the basis of chance alone. Similarly, gender differences at 60 weeks were minimal in number.

The extent to which circumcision does contribute to gender differences in behavior during the neonatal period, or subsequently, obviously demands detailed and focused study. Most certainly, the published description of any sample using male neonates should indicate circumcision status. At present, with rare exception, this information does not appear in any account of subject characteristics.


The psychological literature is replete with reports of gender differences in many areas of behavior. In almost all cases where such differences have been described, nothing is known of their origin - the term remains descriptive. Many possible processes could be involved linking the particular pattern of behavior to the genotypic gender. If future research confirms that some neonatal "gender" differences are actually attributable to circumcision, we would, in these cases have an understanding of the outline of the process involved. In our culture, circumcision is selectively applied to infants of the male gender so that some differences in behavior stem from a cultural institution which results in differential physical treatment of the two genders. These gender differences would then belong to the general class of gender linked behavior that has its origin in culturally placed processes that produce particular developmental environments for infants of the 2 genders.

We have reported here some direct and indirect evidence indicating that circumcision has an influence of unknown duration on the behavior and psychophysiology of male infants and we suggest that so-called neonatal "gender" differences may instead be circumcision differences. The term "stress" has already been used in the literature to describe the psychophysiological impact of circumcision (Emde et al., 1971). The animal literature warns us of the dangers of proposing any simple hypothesis linking adrenal physiology, neonatal behavior, and external events arbitrarily labelled "stressful". Nevertheless, indices of physical stress have been found in a proportion of human male newborns (Forsyth, 1974).

With the rise in obstetric and pediatric technology, young infants are increasingly subject to surgical treatments and pharmacological treatments either directly or indirectly via the mother. Existing animal research suggests that these treatments may have long-term physiological and behavior consequences for the young child. Unfortunately, interventions carried out under medical aegis have provoked little research or concern - especially by comparison with the behavioral scientist's long-standing interest in the effects of social and psychological factors. This is particularly worrisome in view of the evidence showing that the human brain is particularly vulnerable to exogenous and endogenous influences during the period of its maximum growth rate, from the latter part of pregnancy until approximately 18 months postnatal (Dobbing & Sands, 1973).


      1 The Scientific Citation Index of 1973 reveals but a single reference to this article in the five years since its publication.
      2 The number of circumcisions in the United Kingdom has in the past been higher: In 1949 Gairdner (1949) estimated that in England and Wales 90,000 boys were circumcised. Gairdner's discussion of the medical side effects of the operation was probably instrumental in bringing about its decline in Britain.
      This investigation was supported in part by Research Scientist Award No. MH 5925 from the National Institute of Mental Health and in part by National Science Foundation Research Grant No. GB 15155 to Dr. Brackbill. The work of Drs. Bernal and Roberts is supported by a grant from the Nuffield Foundation, London.
      This article was prepared during Dr. Brackbill's tenure as Overseas Fellow, Churchill College, Cambridge.
Current address for J.F.Bernal: Medical Research Council Unit for the Development and Integration of Behavior, High Street, Madingley, Cambridge, England.
Requests for reprints should be sent to: Dr. Y. Brackbill, Department of Psychology, University of Florida, Gainesville, Florida 32611, U.S.A.


Aleksandrowicz, M. K., and Aleksandrowicz, D. R. (1974). Obstetrical pain-relieving drugs as predictors of infant behavior variability. Child Dev., 45: 935-945. [Abstract]

Anders, T. F. and Chalemian, R. J. (1974). The effects of circumcision on sleep-awake states in human neonates. Psychosom Med., 36: 174-179.

Bell, R. Q., and Costello, N. (1964). Three tests for sex differences in tactile sensitivity in the newborn. Biologica Neonatorum, 7: 335-347.

Bench, J. (1969). Some effects of audio-frequency stimulation on the crying baby. J. Auditory Res., 9: 122-128.

Bench, J., Collyer, Y. Langford, C. and Toms, R. (1972). A comparison between the neonatal sound-evoked startle response and the head-drop (moro) reflex. Dev. Med. Child Neurol., 14: 308-317.

Bernal, J. F. (1972). Crying during the first ten days of life, and maternal responses. Dev. Med Child Neurol., 14: 362-372.

Bernal, J. F. (1973). Night waking in infants during the first 14 months. Dev. Med. Child Neurol., 15: 760-769.

Brackbill, Y. (in press). Continuous stimulation and arousal level in infancy: Effects of stimulus intensity and stress. Child Dev. [Abstract]

Brackbill, Y., Kane, J., Manniello, R. L., and Abramson, D. (1974) Obstetrical premedication and infant outcome. Am. J. Obstet. Gynecol., 118: 377-384.

Conway, E., and Brackbill, Y. (1970) Delivery medication and infant outcome: an empirical study. Mongr. Soc. Res. Child Dev., 35:24-34.

Daly, M. (1973). Early stimulation of rodents: a critical review of present interpretations. Br. J. Psychol., 64: 435-460.

Department of Health and Social Security. (1974). Hospital Inpatient Inquiry for England and Wales (1972). London: Her Majesty's Stationery Office.

Desor, J. A. Maller, O., and Turner, R. A. (1973). Taste in acceptance; of sugars in human infants. J. Comp. Physiol. Psychol., 84:496-501.

Dobbing, J., and Sands, J. (1973). Quantitative growth and development of human brain. Arch. Dis. Child., 48: 757-767.

Dubignon, J. Campbell, D., Curtis, M., and Parrington, M. (1969). The relation between laboratory measures of sucking, food intake, and perinatal factors during the newborn period. Child Dev., 40: 1107-1120. [Abstract]

Emde, R. N., Harmon, R. J. Metcalf, D., Keonig, K. L., and Wagonfeld, S. (1971). Stress and neonatal sleep. Psychosom Med., 33: 491-497.

Engel, R., Crowel, D., and Nishijima, S. (1968). Visual and auditory response latencies in neonates. In Felicitation Volume in Honour of C.C. de Silva. Ceylon: Kulartne & Co.

Forsyth, C. C. (1974) Growth and development of the endocrine glands - adrenal cortex. In J. A. Davis and J. Dobbing (Eds.) Scientific Foundations of Paediatrics. London: Heineman.

Gairdner, D. (1949). The fate of the foreskin. Br. Med. J., 2: 1433-1437.

Hutt, S. J., Lenard, H. G., and Prechtl, H. F. R. (1969). Psychophysiological studies in newborn infants. In L. P. Lipsitt and H. W. Reese (Eds.), Advances in Child Development and Behavior, Vol. 4. New York: Academic Pp. 127-172.

Korner, A. F. (1973). Sex differences in newborns with special reference to the organization of oral behavior. J. Child. Psychol. Psychiatry, 14: 19-29.

Levine, S. (1969). An endocrine theory of infantile stimulation. In J. A. Ambrose (Ed.), Stimulation in Early Infancy. New York: Academic. Pp. 45-55.

Lewis, M. (1972). State as an infant-environment interaction: An analysis of mother-infant interaction as a function of sex. Merrill Palmer Q. Behavior. Dev., 18: 95-121. [Abstract]

Lipsitt, L. P., and Levy, N. (1959). Electrotactual threshold in the human neonate. Child Dev., 30: 547-554.

Meier, G. W., and Schutzman. L. H. (1968). Mother-infant interactions and experimental manipulations: Confounding or misidentification? Dev. Psychobiol., 1: 141-145.

Moss, H. A., and Robson, K. (1968). The role of protest behavior in the development of mother-infant attachment. Paper presented at the Meeting of the American Psychological Association, San Francisco, California, U.S.A.

Nisbett, R., and Gurwitz, S. (1970). Weight, sex, and the eating behavior of human newborns. J. Comp. Physiol. Psychol., 73: 245-253.

Pederson, F. A., and Bell, R. Q. (1970). Sex differences in preschool children without histories of complications of pregnancy and delivery. Dev. Psychol., 3: 10-15. [Abstract]

Prechtl, H. F. R. (1972). Patterns of reflex behavior related to sleep in the human infant. In C. D. Clemente, D. P. Purpura, and F. Mayer (Eds.), Sleep and the Maturing Nervous System. New York: Academic. Pp. 297-301.

Richards, M. P. M. (1966). Infantile handling in rodents: A reassessment in the light of recent studies of maternal behavior. Anim. Behav., 14: 582.

Thoman, E. B. Leiderman, P. H., and Olson, J. P. (1972). Neonate-mother interaction during breast-feeding. Dev. Psycho. 6: 110-118. [Abstract]

Whiting, J. W. M., Landauer, T. K. and Jones, T. M. (1968). Infantile immunization and adult stature. Child Dev., 39:59-67.

Wolff, P. H. (1969). The natural history of crying and other vocalizations in early infancy. In B. M. Foss (Ed.), Determinants of Infant Behaviour, Vol. IV. London: Metheun, Pp. 81-109.

Received for publication 14 January 1975
Revised for publication 10 April 1975
Developmental Psychobiology, 9(1): 89-95 (1976)
© 1976 by John Wiley & Sons, Inc.

(File revised 20 October 2006)

Return to CIRP library