DEV (WILEJ) RIGHT INTERACTIVE Mark S. Blumberg Greta Sokoloff Department of Psychology Thermoregulatory Competence The University of Iowa Iowa City, Iowa 52242 and Behavioral Expression in the Young of Altricial SpeciesÐRevisited Received 24 July 1997; accepted 27 October 1997 ABSTRACT: The behavioral and physiological thermoregulatory capabilities of newborn and infant mammals have been studied for over half a century. Psychobiologists have noted that the infants of altricial species (e.g., rats) have physical and physiological limitations such that heat loss overwhelms heat production, thus forcing a reliance on behavioral thermoregulation for the maintenance of body temperature. Recent evidence, however, suggests that a modi®- cation of this view is justi®ed. Speci®cally, throughout a range of moderately cold air temper- atures, nonshivering thermogenesis by brown adipose tissue contributes signi®cantly to the infant rat's behavioral and physiological adaptations to cold challenge. Given the prominent use of altricial species for the study of infant behavior, increased understanding of the infant's physiological responses to cold and the effect of thermal factors on behavior is warranted. ᭧ 1998 John Wiley & Sons, Inc. Dev Psychobiol 33: 107±123, 1998 Keywords: rat; hamster; thermoregulation; brown adipose tissue; homeostasis; development INTRODUCTION protecting the brain from overheating, including the venous shunts of horned lizards (Heath, 1966) and the Like oxygen and gravity, temperature is a basic feature rete mirabile of many mammals (Baker, 1979). of the environment that shapes and constrains biolog- Threats of extreme cooling have led to the evolution ical activity. Temperature in¯uences cellular metabo- of similarly inventive solutions, such as a brain heater lism (Schmidt-Nielsen, 1990) and muscular contrac- tissue in sword®sh (Carey, 1982) and ªantifreezeº in tion (Bennett, 1984), it modulates many aspects of the blood of coldwater ®shes (DeVries, 1982). neuronal function from conduction velocity to refrac- Temperature is an especially important factor dur- tory period (e.g., Swadlow, Waxman, & Weyand, ing development. During embryonic development, and 1981), and it even guides sexual differentiation in regardless of whether we are considering a chick in- many reptiles (Deeming & Ferguson, 1989). The sen- side its egg or a human infant in utero, the thermal sitivity of neural tissue to high temperatures has made environment is maintained within narrow limits by the necessary the evolution of elaborate mechanisms for parent while the embryos are passive recipients of the heat supplied to them (Leon, 1986). With hatching or birth, however, young emerge from their thermally se- Correspondence to: M. S. Blumberg Contract grant sponsor: NIMH cure environments into a world where heat loss is a Contract grant number: MH 50701 constant threat to survival. Suddenly, they must now short ᭧ 1998 John Wiley & Sons, Inc. CCC 0012-1630/98/020107-17 standard long DEV (WILEJ) LEFT INTERACTIVE 108 Blumberg and Sokoloff become active participants in the regulation of their within which they were working. As was mentioned body temperature. earlier, there has been a tendency to assess thermoreg- The thermoregulatory capabilities of newborn and ulatory capabilities based on the maintenance of a sta- infant mammals have been the subject of experimental ble body temperature in the cold. But ªcoldº is a rel- research for over 50 years. As we will see, our under- ative concept, and any investigation of a regulatory standing and appreciation of those capabilities have system requires the choice of appropriate experimental been in¯uenced profoundly by our expectations as to stimuli. This is exactly what Conklin and Heggeness what constitutes ªsuccessfulº thermoregulation. In (1971) understood when, in discussing their experi- part, our expectations arise from the idea that there ments, they wrote that they exposed their infant rats exists a single core body temperature (e.g., rectal tem- to air temperatures ªthat would evoke, but not over- perature) that (a) is representative of all relevant body whelm, regulatory functionsº (p. 333). Therefore, test- temperatures and (b) will be maintained within a range ing infants at an extreme air temperature is not always of adultlike values if the infant is thermoregulating an effective way to assess thermoregulatory capabili- successfully. These expectations, however, have led us tiesÐif the mechanisms are overwhelmed, then one astray. Therefore, this article represents an attempt to cannot properly evaluate their control, ef®cacy, or de- reassess infant thermoregulation, especially that of al- velopment. tricial species such as the rat, based on recent physi- Taylor (1960) contributed to the development of a ological and behavioral data. revised perspective by showing that newborn rats in- crease oxygen consumption signi®cantly when ex- posed to an air temperature of 29ЊC. Taylor was A BRIEF HISTORY unique in cautioning against super®cial comparisons between the thermogenic capabilities of young rats of A common strategy in developmental studies of infant different agesÐdifferences in size and insulation are thermoregulation entails exposing pups to a very low intimately connected with age and cannot be ignored. air temperature and, by measuring rectal temperature, In effect, Taylor argued against the relatively simplis- monitoring the ªdevelopment of homeothermyº (e.g., tic approach of assessing thermoregulatory compe- Brody, 1943; Hahn, 1956). When tested in this way, tence by exposing pups to a single, arbitrary air tem- developing rat pups are increasingly able to maintain perature and measuring rectal temperature. rectal temperature in the cold from birth to 3 weeks of The determinants of heat loss in infants and adults age. This general experimental approach to studying were well known by the early 1960s: Surface-to-vol- the development of homeothermy has been popular ume ratio, insulation, and vasomotor activity all con- over the years, despite its considerable methodological tribute, passively or actively, to the promotion or re- and conceptual pitfalls. tardation of heat loss. The mechanisms of heat gain, In the experiment of Brody (1943), 1- to 22-day- however, were somewhat more obscure at that time. old pups were placed in a container maintained at an Speci®cally, while the importance of shivering ther- air temperature of 15ЊC and rectal temperature was mogenesis for adults was not in dispute, the ability of monitored. He concluded that thermoregulatory com- curarized animals to produce heat in the cold argued petence developed between the 2nd and 3rd week for a second form of heat production that was referred postpartum. Similarly, Hahn (1956) exposed young to as chemical or nonshivering thermogenesis (Hsieh, rats to an air temperature of 10ЊC and demonstrated Carlson, & Gray, 1957). This form of thermogenesis the thermoregulatory importance of the development was known to be stimulated by noradrenaline, but the of fur and the control of piloerection. Based on Hahn's site of action had not yet been identi®ed. Then, in the results, it was concluded that ªphysical thermoregu- early 1960s, Smith hypothesized that brown adipose latory mechanisms develop between the 14th and 18th tissue (BAT) is one source of nonshivering thermo- dayº postpartum (p. 430). In both of these experi- genesis in adult mammals (Smith, 1964). The role of ments, the particular choice of air temperature is not BAT as a thermogenic organ was quickly extended to discussed, and one must wonder to what extent these apply to heat production in newborn rabbits and other choices in¯uenced the conclusions of the experiments. mammalian species (e.g., Dawkins & Hull, 1964; Hull For example, would Brody's 22-day-old pups have ex- & Segall, 1965). By the end of the decade, the ther- hibited homeothermy had they been challenged at an mogenic function of BAT seemed irrefutable (Smith air temperature of 0ЊC? And would the 1-day-old pups & Horwitz, 1969). have exhibited homeothermy had they been chal- In reviewing the work of comparative physiologists lenged at an air temperature of only 30ЊC? following the discovery of the thermogenic function short That earlier investigators did not ask these ques- of BAT, BruÈck and Hinckel (1996) suggest that neo- standard tions provides insight into the conceptual framework nates can be divided into three categories. First, there long DEV (WILEJ) RIGHT INTERACTIVE Infant Thermoregulation 109 are the precocial newborns (e.g., guinea pig, lamb, afforded by the favorable changes in surface-to-vol- full-term human) that respond vigorously to cold chal- ume relations provides a signi®cant metabolic savings lenge and are able to maintain body temperature at a to the individual pups. He concluded that ªthrough stable level in the cold, albeit only within a narrow individual competitive adjustments the huddle be- range of air temperatures. Second, there are the altri- haves as a self-regulating unit which provides warmth cial newborns (e.g., rabbit, dog, rat) whose thermo- and insulation to all its membersº (Alberts, 1978, p. genic responses, although present shortly after birth, 231). are insuf®cient to retard heat loss so that their ªbody The elegance and power of Alberts' (1978) analy- temperature drops on exposure to environmental tem- ses crystallized for psychobiologists the signi®cance peratures slightly less than thermal neutralityº (BruÈck of behavior for infants facing the problem of