Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Frontiers in Neuroendocrinology journal homepage: www.elsevier.com/locate/yfrne

Review

On the causes of early life experience effects: Evaluating the role of mom Akaysha C. Tang a,b,c,⇑, Bethany C. Reeb-Sutherland d,e, Russell D. Romeo f, Bruce S. McEwen g a

Cognitive Neuroscience Program, BCS/SBE, National Science Foundation, Arlington, VA 22230, United States Department of Psychology, University of New Mexico, Albuquerque, NM 87131, United States c Department of Neurosciences, University of New Mexico, Albuquerque, NM 87131, United States d Department of Psychology, Florida International University, Miami, FL 33199, United States e Center for Children and Families, Florida International University, Miami, FL 33199, United States f Neuroscience and Behavior Program, Department of Psychology, Barnard College of Columbia University, New York, NY 10027, United States g Laboratory of Neuroendocrinology, Rockefeller University, New York, NY 10065, United States b

a r t i c l e

i n f o

Article history: Available online xxxx Keywords: Maternal care Stress CORT HPA Self-regulation Novelty Maternal mediation Maternal modulation Early experience Cognitive development

a b s t r a c t Early life experiences are thought to have long-lasting effects on cognitive, emotional, and social function during adulthood. Changes in neuroendocrine function, particularly the hypothalamic–pituitary–adrenal (HPA) axis, contribute to these systems-level behavioral effects. In searching for causal mechanisms underlying these early experience effects, pioneering research has demonstrated an important role for maternal care in offspring development, and this has led to two persistent ideas that permeate current research and thinking: first, environmental impact on the developing infant is mediated through maternal care behavior; second, the more care that a mother provides, the better off her offspring. While a good beginning, the reality is likely more complex. In this review, we critically examine these ideas and propose a computationally-motivated theoretical framework, and within this framework, we consider evidence supporting a hypothesis of maternal modulation. These findings may inform policy decisions in the context of child health and development. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Few findings regarding early life experience have generated more enduring interest among researchers and the general public than the large body of rodent literature on the relation between postnatal maternal care and the development of the hypothalamic–pituitary–adrenal (HPA) axis (Meaney, 2010; Zhang and Meaney, 2010). This long-lasting interest is in part due to its implied relevance to human child development as suggested by the title of the commentary ‘‘The Importance of a Well-Groomed Child’’ (Sapolsky, 1997) when this line of work was first published in Science (Liu et al., 1997) and by the title of a recent New York Times Sunday Review, ‘‘Cuddle Your Kid!’’ (Kristof, 2012). As the human mother plays a significant role in the shaping of her child’s stress response system and the development of later psychopathology (for review see Baram et al., 2012; McCrory et al., 2010), this line of animal research has been viewed as relevant to not only understanding child development (Kaffam and Meaney, 2007; Sanchez, 2006; Veenema, 2009) but also to early child development policymaking (Shonkoff and Garner, 2012). Thus, there is ⇑ Corresponding author at: Cognitive Neuroscience Program, BCS/SBE, National Science Foundation, Arlington, VA 22230, United States. Tel.: +1 703 292 7281. E-mail addresses: [email protected], [email protected] (A.C. Tang).

an urgent need to review recent conceptual progress in this line of research. The genesis of this influential literature was the initial demonstration that early stimulation through the ‘‘handling’’ procedure led to both an increase in maternal licking of the pups and longlasting differences in offspring development of the HPA axis (Liu et al., 1997). Follow-up studies showed further correlations between the naturally occurring variation in maternal licking of the pups and variations in many offspring developmental outcome measures (Meaney, 2010; Zhang and Meaney, 2010). Evidence for a causal relation between maternal licking and offspring development was strengthened by cross-fostering studies in which offspring outcome measures were found to correlate with the foster mother’s licking quantity (high and low licking) but not correlate with the birth mother’s licking quantity (Francis et al., 1999). This line of research has stimulated interest among human child development researchers to examine the influence of the amount of maternal contact on emotional and physiological outcomes (Sharp et al., 2012). Two general conclusions were reached through these studies. First, the so-called ‘‘handling’’ effects were mediated via maternal care behavior. An implication of this conclusion is that such early stimulation itself has no effect on offspring development; and maternal care, specifically, the licking of the pups, following the

0091-3022/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.yfrne.2013.11.002

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

2

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

stimulation of the pups from the ‘‘handling’’ manipulation is the cause of variation in offspring developmental outcomes. Second, the higher the quantity of licking, the better off the offspring will be. A practical and political implication of this conclusion may be that mothers need to more frequently engage in a specific type of maternal care behavior, for example, stroking of their babies as a presumed human analogue to the licking behavior of the rodent mother (Sharp et al., 2012). In this review, we hope to show that the role of maternal care in shaping the physiology and behavior of her offspring is more complex than that currently assumed by the majority of the rodent studies designed to assess the contributions of maternal variables on offspring development. Specifically, we will (1) review relevant literature that led to the maternal modulation hypothesis; and (2) present supporting evidence for this maternal modulation hypothesis from a line of research utilizing the neonatal novelty exposure paradigm. Given these specific goals, this review will not be comprehensive but highly focused on evidence that bears direct relevance to the discrimination of the two competing hypotheses.

2. Complexity in the role of the mother and a new theoretical framework In recent years, several lines of developmental research have brought forth complexities regarding the role of maternal care in animal models. First, certain levels of maternal stress may lead to an increase in maternal circulating corticosterone (CORT), which in turn can produce effects similar to those induced by early stimulation, as demonstrated via experiments in which CORT was added to the mother’s drinking water and presumably passed to her offspring via her milk (Catalani et al., 2011). This line of research clearly points to an additional source of early life experience from the mother that contributes to the HPA development of the offspring. Second, in experiments using non-human primates, in absence of differences in maternal care behaviors, intermittent stress of infant monkeys results in differences in HPA function that are similar to those produced by the ‘‘handling’’ treatment and to those associated with high versus low licking dams (Parker et al., 2006). These findings indicate that activation of the offspring’s HPA axis early in life contributes to the programming of later HPA function. Third, in research that manipulates maternal stress levels via the imposition of a variable foraging condition, it was found that higher quantity of maternal licking did not produce differences in HPA function of their offspring (Macrì et al., 2011) demonstrating that higher quantity of maternal licking does not necessarily correlate with adaptive changes in offspring HPA function. Finally, maternal presence or contact with her pups facilitates the recovery of the offspring’s stress response (Moriceau and Sullivan, 2006; Stanton and Levine, 1988, 1990; Stanton et al., 1987), thus setting the context for the modulation of the offspring’s HPA function. These pioneering lines of research demonstrate that the role of the mother is multi-dimensional, including not only the behavior of the mother, but the mother’s own HPA function. Furthermore, the mother is only one source of the developing infant’s environment and non-maternal aspects of the infant’s environment, particularly, change in this environment, represent another salient source of inputs to the developing HPA axis. Finally, maternal characteristics, whether behavioral or physiological, can interact with other environmental changes to jointly modify the offspring’s HPA axis. These have led us to develop a computationallymotivated conceptual framework for understanding early life experience effects on the development of HPA function. This framework has the following components: (1) change in the environment, including both maternal and non-maternal aspects, is the driving force that activates components of the HPA

axis; (2) key parameters of the HPA regulatory system are modifiable cumulatively by past experience; these parameters are related to the setting of the basal and peak circulating CORT levels and rates of initial rise and later recovery of a CORT response to environmental input; (3) through a process of learning, the same environmental input, or stressor, can lead to different HPA responses over time with relatively novel stimuli being more effective at activating the HPA axis and relatively familiar stimuli being less or not effective at all; and (4) the mother’s behavior prior to and after her infant’s experience of a stressor can change the key parameters of her infant’s regulatory system thus changing the effective experience of a given stressor by the infant. Furthermore, the mother’s circulating CORT and her own CORT surges in response to her own environment can also change these parameters via CORT levels transferred through her milk, thus also changing the effective experience of a given stressor by her infant. Within this framework, the HPA axis is viewed as a dynamic system consisting of multiple interacting variables, environmental novelty (or salient change) and various maternal characteristics. In such a dynamic system, no single variable can be the exclusive cause for changes in HPA function (Thelen and Smith, 1994). The task of understanding the mechanisms that underlie the effects of early life experiences on the HPA axis is not to isolate the influence of one variable while ruling out all other variables. Instead, the task is to define and describe the dynamic interactions among the multitude of equally important maternal and non-maternal variables. Moreover, in consideration of multiple maternal characteristics, it is important to recognize that maternal individual differences in her care-related behaviors are likely to be paralleled by individual differences in her stress physiology. Both are potential modulators of any environmental influence on the offspring. Even within the larger category of maternal care behavior, multiple types of maternal behaviors exist (e.g., anogenital and body licking of the pups, supine and arched-back nursing) and a mother may distribute her time differently across these different behavioral types (Moore, 1992; Stern, 1997). Within each type of behavior, a mother can further differ in average quantity and in the variability of the behavior over time. In the following sections, evidence from our laboratories gathered under the guidance of the above theoretical framework and utilizing the neonatal novelty exposure paradigm will be presented in support of a broader view of maternal contribution to early experience effects. 3. Programming Infant HPA axis by Novelty The neonatal novelty exposure paradigm is a split-litter design that consists of exposing rat pups from a given litter to a relatively novel non-home cage for 3 min/day for the first 3 weeks of life (Novel group) while the control pups from the same litter remain in the home cage (Home group) (Tang, 2001; Tang et al., 2006, 2012b). This neonatal novelty exposure paradigm is specifically designed to isolate environmental novelty, a salient component from the ‘‘neonatal handling’’ or ‘‘postnatal handling’’ paradigm (Denenberg, 1964; Levine, 1957, 1960) because novelty is known to activate the HPA axis (Dallman, 2000; Denenberg et al., 1967). This focus on novelty is motivated by a simple assumption that in order to shape or program the HPA axis, something has to be done to first activate at least some parts of the HPA axis. This notion directly challenges an assumption implied by the maternal mediation hypothesis that a mother’s highly familiar care-giving behavior, such as licking, can somehow serve to activate the HPA axis, thereby subsequently shaping HPA function. The maternal mediation hypothesis lacks a clearly articulated logical connection between the presumed cause, the licking, and the effect, some changes in HPA axis. Without such a clearly stated

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

connection, the programming of the offspring’s HPA axis by maternal licking is only assumed. The logically well-motivated connections provided by the recent literature were made only between different parts of the neural system (e.g., CORT to receptor, receptor to cell counts, Liu et al., 1997), but not between maternal behavior and the neural system. Using this paradigm, we explored the effects of brief systematic exposures to novelty on not only the function of the HPA axis but a wide range of cognitive, social, emotional, and neurophysiological functions at different time points across the lifespan. Importantly, we have also replicated our results across the majority of the behavioral and physiological functional end points (ReebSutherland and Tang, 2011; Tang, 2001; Tang et al., 2006, 2003a, 2011c, 2003b, 2008; Tang and Zou, 2002). Using this split-litter design, experimenter handling, maternal separation, maternal genetics, and maternal stress, are all matched or shared between the Novel and Home pups (Tang, 2001; Tang et al., 2006, 2012b). That is, pups’ experience of novelty is manipulated while the other confounding factors are carefully controlled. These controls are critical in ensuring that if differences are found between Novel and Home rats, such differences can be attributed to the pups’ differential exposure to novelty. Novel and Home rats were both handled by the experimenter during the procedure; therefore, any observed differences between the Novel and Home rats should not be referred to as ‘‘handling’’ effects. Novel and Home rats were separated from their mothers for the same amount of time so that the Novel–Home differences are not confounded by maternal separation. Finally, because the Novel and Home pups are from the same litter (within-litter design) and randomly assigned to the Novel and Home groups, i.e. sharing the same mother, the novelty treatment is not confounded by prenatal or genetic differences or the mother’s circulating stress hormone. Using this controlled experimental paradigm, it has been shown that in the absence of preferential maternal care (Reeb-Sutherland and Tang, 2011; Tang et al., 2006), pups that experienced brief 3min daily exposures to a non-home cage have lower basal CORT levels (Tang et al., 2003b), greater CORT novelty response (Tang et al., 2006), greater plasticity in social interaction induced CORT response (Akers et al., 2008), greater hippocampal synaptic plasticity (Akers et al., 2006; Tang and Zou, 2002; Tang et al., 2008; Zou et al., 2001), and greater modulation of such synaptic plasticity by stress-induced levels of CORT (Zou et al., 2001), than their littermates who experienced only the familiar home cage environment. These neural and physiological changes induced by brief neonatal novelty exposures are also paralleled by enhanced spatial (Reeb-Sutherland and Tang, 2011; Tang, 2001; Tang et al., 2006, 2011c; Yang and Tang, 2011) and social memory (Reeb-Sutherland and Tang, 2011; Tang and Reeb, 2004; Tang et al., 2003b), reduced behavioral inhibition to novelty (Tang et al., 2003a, 2012a; Zou et al., 2001), reduced aggression toward a non-familiar conspecific (Akers et al., 2008; Reeb-Sutherland and Tang, 2011), increased dominance in social competition (Akers et al., 2008), increased early growth (Tang et al., 2012b), and changes in several distinct expressions of functional brain asymmetry (Akers et al., 2004; Tang et al., 2006, 2011b, 2003b, 2008; Tang and Reeb, 2004; Tang and Verstynen, 2002; Verstynen et al., 2001). Furthermore, these novelty-induced functional enhancements are not transient but can be demonstrated repeatedly in the same animals throughout their lives, including senescence (Reeb-Sutherland and Tang, 2011; Tang et al., 2011b, 2012b). Together, these studies clearly demonstrate that even very brief (3 min/day) but systematic (daily for the first 2–3 postnatal weeks) exposures to a relatively novel non-home environment is sufficient and perhaps represent the optimal input needed to program the HPA axis and enhance offspring cognitive, social, emotional, and physical development. Such findings suggest that environmental

3

novelty is a salient factor in programming infant development at the system and behavioral levels. 4. Maternal modulation of novelty effects on infant HPA function: maternal care reliability Within the dynamic systems framework (Thelen and Smith, 1994), instead of pitting maternal characteristics against environmental novelty as a competing causal factor, we consider maternal care characteristics as a contextual and modulating variable that interacts with environmental novelty. Instead of asking whether a given maternal characteristic alone is the cause for any offspring differences, we ask the question whether a given effect of environmental novelty on offspring developmental outcomes varies according to the value of a given maternal characteristic. If the novelty effects observed across a sample of rat families vary with a certain characteristic of the dam, then such evidence supports the maternal modulation hypothesis. In this and the next sections, we provide evidence to support this maternal modulation of novelty by her care reliability and her ability to regulate her own HPA function. It should be noted that these two characteristics are by no means the only important maternal characteristics that can interact with non-maternal environmental changes. For example, abusive maternal behaviors (Raineki et al., 2010) and excessive maternal care behaviors (Egeland et al., 1993) are salient factors that are likely to also interact with non-maternal environmental factors. Their contributions remain to be investigated. Human developmental researchers have long recognized the complexity of maternal care influence on infants and have created constructs such as maternal sensitivity or responsiveness (Ainsworth et al., 1978). This construct highlights the dynamic nature of the mother–infant interaction and goes beyond the quantity of maternal care. Our notion of maternal care reliability captures some aspects of maternal sensitivity in that a reliable and responsive mother generates a predictable maternal environment for her developing infant. Operationally, we measure this maternal care reliability by day-to-day maternal care variability in maternal care immediately after the litter disturbance associated with the novelty exposure. The lower the day-to-day variability, the greater the reliability. Most importantly, this maternal care reliability measure is based on observations made immediately after a perturbation of the pups’ home environment. This perturbation may be viewed as analogous to a stressful situation for a human infant. Observing maternal care behavior immediately after such a stressor allows the measure to capture the sensitivity aspect of human maternal responsiveness. To distinguish this maternal care reliability measure from the frequently used maternal licking quantity measure, it is important that we present sufficient experimental and analysis details to enable replication in other laboratories. Maternal licking was observed for a brief 10-min interval immediately after the daily novelty exposure procedure during the first 10 postnatal days. This period immediately after the novelty exposure was chosen because it is likely to be the time of greatest distress for the pups. This parallels human studies in which maternal sensitivity (i.e., consistency) has the greatest predictive power when behavioral outcomes are observed during times of infant distress (Leerkes, 2011; Nachmias et al., 1996). Variability is measured as the standard deviation of the daily maternal care quantity after removing the initial linear trend observed across the postnatal days. This maternal care measure captures random day-to-day variations in maternal care immediately after each litter disturbance. It is thus fundamentally different from the construct of ‘‘fragmented maternal care’’ which was defined as the number of times the mother’s observed behavior changes from one type to another within a limited time window of 75 min (maternal behavior was sampled once

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

4

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

Fig. 1. Maternal care during brief 10-min windows immediately after repeated daily postnatal novelty exposure predicts the effect of novelty exposure on CORT habituation among aged offspring. (A) Greater average amount of maternal licking and grooming (LG) was associated with greater day-to-day variability in maternal LG (dots and bars indicate average and range, respectively, of LG across days for individual dams; N = 11 l). (B) Greater average amount of maternal LG was associated with negative novelty scores for CORT habituation (marginally significant). (C) Smaller day-to-day variability in maternal LG was significantly correlated with positive novelty scores for CORT habituation. From Akers et al., 2008.

every 3 min) and without particular reference to the infants’ stress situation (Ivy et al., 2008). In one cohort of rat families, we tested the maternal modulation hypothesis using offspring developmental outcomes assessed when the offspring from these families were two years old, i.e., during senescence (Akers et al., 2008). In this particular set of experiments, rats were trained to criterion to retrieve chocolate from the end of a short runway prior to testing. They were then tested in a social competition situation where a similarly-trained conspecific appeared unexpectedly in the same testing environment. Competition for the reward ensues because only one of the two rats can fit into the runway. Blood samples were collected shortly after the completion of social competition on two consecutive days of testing and post-social stress CORT levels were measured. We found that when the testing situation was novel on the first testing day, Novel rats won more than Home rats but did not differ significantly in their post-competition CORT measure on each of the two testing days. Instead, habituation of the CORT response to social competition over the two testing days (i.e., Day 1 CORT–Day 2 CORT) differed significantly between the Novel and Home rats, with Novel rats displaying greater habituation in their CORT response than Home rats. Because this novelty exposure treatment is a within-litter manipulation, the effect of novelty can be computed for each individual rat family as a difference score (NE score) between the Novel and Home pups of that litter. In the case of CORT response habituation, a positive NE score indicates a novelty-induced enhancement in CORT response habituation while a negative NE score indicates novelty-induced impairment for that given litter. To test the maternal modulation hypothesis, we examined whether the amount of novelty-induced increase in CORT habituation was associated with variations in maternal care variability and found a negative correlation. Specifically, the less variable a dam was in her day-to-day post-novelty exposure maternal licking, i.e., the more reliable the postnatal maternal care, the greater the novelty-induced CORT habituation observed in her offspring (Fig. 1C). This finding supports the hypothesis that maternal care reliability sets the favorable context for the novelty exposure to have a positive impact on the plasticity of the HPA axis during old age. In a separate cohort of rat families, we also tested this maternal modulation hypothesis in the context of other behaviorally-based functional measures during infancy and early adulthood (ReebSutherland and Tang, 2012; Tang et al., 2012b). We tested for an interaction effect between the variable of maternal care variability and the variable of novelty exposure (Novelty) using Analysis of Covariance (ANCOVA).1 The measure of maternal care reliability 1 This test is conceptually equivalent to the correlation measure in the previous paragraph but a more conservative, thus a stronger test statistically.

significantly interacted with the neonatal novelty exposure to affect both cognitive and physical development. Offspring of dams with more reliable maternal care showed a greater novelty-induced enhancement in spatial working memory (measured by one-trial learning in the Morris water task) (Reeb-Sutherland and Tang, 2012) and greater novelty-induced enhancement in early growth (measured by body weight) (Tang et al., 2012b). In summary, consistent with the maternal modulation hypothesis, we demonstrated that maternal care reliability modulates novelty-induced effects across multiple functional domains (i.e., cognitive, social, and physical). These results support the hypothesis that maternal care reliability sets the context to determine whether novelty exposure leads to beneficial (positive NE score) or detrimental (negative NE score) effects in her offspring. For litters with more reliable mothers, novelty exposure has beneficial effects; for litters with less reliable mothers, novelty exposure can have less beneficial effects, and occasionally, have detrimental effects. 5. Maternal modulation of novelty effects on infant HPA function: maternal self-stress regulation Mothers that differ in their maternal care behavior are likely to also differ in their underlying physiology, particularly their regulation of the HPA axis. As mentioned earlier, providing the nursing dams with CORT in the drinking water to simulate stress-induced variations in maternal CORT can result in changes in the neuroendocrine and behavioral functions of the offspring (Catalani et al., 2011). Exposing the nursing dams to stress associated with uncertainty in food availability also results in changes in offspring HPA function (Macrì et al., 2011). Finally, it is well known that exposing the pregnant dam to stress affects the development of her offspring’s HPA function (Richardson et al., 2006). Therefore, both the human and animal literatures make the ability of the mother to regulate her own HPA function a likely variable to modulate the effects of environmental novelty. Motivated by these findings and the theoretical framework outlined earlier, in our recent studies, we collected blood samples from dams at two points in time, one at the trough of the circadian cycle (CORTb) reflecting a true resting level of CORT when the animal was least disturbed by external stimulation, and the other at a 5-min delay from the onset of a 1-min swim stressor (CORTs) to measure the rate of rise in CORT response on a separate day also near the time of circadian trough. These maternal basal and evoked response measures were collected shortly after weaning and are intended to be used as trait instead of state variables. Using these two measures, we derived a CORT response measure defined by CORTe = (CORTs CORTb)/CORTb. This CORTe measure reflects the ability of the rat to mount a rapid response to the challenge of a

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

stressor. These two measures are interrelated but are likely to reflect different underlying parameters of the HPA regulatory system. Using these measurements from both the dams and her pups in adulthood, we tested the hypothesis that maternal self-stress regulation modulates the effects of neonatal novelty exposure on offspring self-stress regulation. Using ANCOVA, we found not only a significant novelty-induced increase in offspring CORTe, but also significant interaction effects between maternal CORT measures (i.e., CORTb, CORTe) and Novelty, with the offspring of dams who can maintain low levels of basal CORT and mount a more rapid surge in the CORT response showing greater novelty-induced enhancement in offspring CORTe. These results support the hypothesis that the effects of neonatal novelty on offspring HPA function are modulated by maternal self-stress regulation (S.M. Dinces et al., 2013). In studies assessing offspring behavioral inhibition, we found that, maternal CORTb and CORTe also showed significant interaction effects with Novelty on offspring behavioral inhibition (Tang et al., 2012a) and early growth (Tang et al., 2012b). Specifically, the offspring of dams who can maintain a low resting CORT and mount a more rapid CORT surge showed a greater novelty-induced reduction in behavioral inhibition and greater novelty-induced enhancement in early growth. In studies assessing offspring spatial memory function, only maternal CORTb was found to interact with novelty exposure for measures of spatial working and reference memory. Novelty-induced enhancements in memory were found for litters whose dams displayed lower basal circulating CORT (Tang et al., 2011c). In assessing offspring acoustic start response (ASR), again only maternal CORTb showed a significant interaction with novelty exposure on offspring’s latency and long-term plasticity of the acoustic startle response (Tang et al., 2011a). As CORTb and CORTe reflect related but different aspects of HPA function, their differential roles in different offspring behavioral functions are expected and need to be further characterized as more empirical data accumulate. In summary, consistent with the maternal modulation hypothesis, we demonstrated that the mother’s capacity to regulate her own HPA function modulates novelty-induced effects across multiple functional domains (neuroendocrine, cognitive, emotional, and physical). Thus, similar to our data on maternal care reliability, these results support the hypothesis that maternal self-stress regulation sets the context to determine whether novelty exposure leads to beneficial (positive NE score) or detrimental (negative NE score) effects in her offspring. 6. Quantity of maternal care: does more mean better? In the studies reviewed above, we clearly demonstrate that maternal care variability interacts with the neonatal novelty exposure treatment to jointly influence multiple developmental outcomes of the offspring. Within this set of experiments, we also used the same post-novelty exposure maternal care observations to measure maternal care quantity. In this case, the daily quantity of maternal licking was averaged and this mean quantity of maternal care was used in place of maternal care variability following identical data analysis procedures. This analysis using the maternal care quantity as a covariate allows the testing of the hypothesis that the average quantity of post-stress maternal care can serve to modulate the effects of novelty exposure. The distinction between the two hypotheses is conceptually important. If it is the variability but not quantity that modulates early stimulation-induced enhancements, then in order to obtain optimal results, consistency in care without reference to quantity should be maximized; if it is the quantity as well as the variability that modulates early stimulation-induced enhancements, then both consistency and quantity should be maximized.

5

We found that for the offspring measure of HPA plasticity in response to repeated social competition, average quantity of maternal care also varies with the novelty-induced enhancement in HPA plasticity but the direction of correlation is surprising: dams with lower average quantity of maternal care had pups who showed novelty-induced enhancement in HPA plasticity (Akers et al., 2008) (Fig. 1B). For the measure of working memory in the Morris Water Task, it is also the case that dams with lower average quantity of maternal care had pups that showed novelty-induced enhancement in working memory (marginally significant) (ReebSutherland and Tang, 2012). Finally, for the measure of early growth, it was dams with lower average quantity of maternal care that had pups that showed novelty-induced enhancement in early growth (Tang et al., 2012b). These findings suggest that the higher quantity of maternal care may not necessarily have a positive impact on offspring development. Empirically, such a conclusion is consistent with what is known from human child development studies. Negative effects have been observed when a mother interacts excessively with her infant (i.e., maternal intrusiveness) (Egeland et al., 1993). Conceptually, while low variation in day-to-day maternal care creates a predictable maternal environment for the developing infant, more but highly variable and thus less dependable maternal care, creates uncertainty in the maternal environment. Therefore, these seemingly surprising findings concerning maternal care quantity are consistent within a multidimensional dynamic systems framework (Thelen and Smith, 1994). We do not view these data as suggesting that less maternal care can somehow lead to better outcomes for offspring development because there is no conceptual link between a low quantity of maternal care and better offspring development. This is a good example illustrating that not all correlations imply causations.

7. Conclusions, implications for early childhood policy, and future directions In this review, we considered two persistent ideas originating from studies of rodents that permeate current research and thinking: first, environmental impact on the developing infant is mediated through maternal care behavior; second, the more care that a mother provides, the better off her offspring. By adopting a general dynamic systems framework, we presented empirical data to support a conceptual model in which (1) both the maternal and nonmaternal aspects of the early life environment jointly influence the development of the HPA axis, (2) environmental novelty, i.e., salient changes in the environment, plays a key role in programming the development of the HPA axis; and (3) multiple maternal characteristics can play a modulatory role in determining whether environmental novelty will produce a positive or negative effect on the development of the HPA axis, maturation of cognitive, social, and emotional function, and early growth (Fig. 2). We provided theoretical motivations for introducing the constructs of maternal care reliability and maternal self-stress regulation and validated the specific operational definitions for these two constructs by demonstrating their ability to predict the direction and magnitude of novelty effects across a range of offspring functional measures. We have reached the following key conclusions regarding the two persistent ideas: the idea that maternal care, particularly licking, mediates all early life experience effects and the idea that more such maternal care is better for offspring development are simply not true. While the mother clearly plays an important role in offspring development, her role is rather complex and does not preclude a direct impact of environmental novelty. Instead, maternal care reliability and the mother’s regulation of her own stress response system are two salient contextual

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

6

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx

Fig. 2. Maternal modulation of infant HPA response to environmental change via maternal physiological and behavioral characteristics. This model has the following elements: (1) HPA axis is activated by a change in the environment (DE); (2) infant HPA axis is minimally characterized by four parameters, resting level of CORT (BREST), time constant of the rise in CORT following DE (r1), time constant of CORT recovery (r2), and the peak CORT response (RMAX); (3) activation of the HPA axis is mounted from the resting level of CORT, BREST, which is jointly set by CORT generated by the offspring (b1: CORTO) and CORT generated by the mother through her milk (b2: CORTM), with increasing self-control by the infant and decreasing control of the mother during post-natal development; (4) maternal care reliability following DE; modulates r2, with reliable maternal behavior facilitating recovery (green line) and unreliable maternal care attenuating recovery (orange dashed line); (5) the cumulative history of infant HPA activation by DE, under the influence of maternal behavioral and physiological modulation, modifies the four parameters of the infant HPA axis during post-natal development, with a quick response recovery at the present time leading to a lower b1, shorter latency to peak (t1) and time to recovery (t2), and smaller BREST in the future; (6) these changes in HPA axis in turn give rise to changes in the dynamics of neural networks underlying offspring learning and memory, emotion, and social interaction.

variables, capable of making the same environmental stimulation into either a positive or negative event. To conclude this review, we outline three important areas for future research. The first is to conduct carefully controlled studies to investigate how the precise patterns of maternal modulation of novelty exposure effects might differ between male and female offspring because several earlier studies showed that the effects of novelty exposure differed between males and females (Akers et al., 2004; Reeb and Tang, 2005; Tang et al., 2003a) and that major sex differences in the HPA axis exist (Bangasser and Valentino, 2012; Becker et al., 2007). All studies presented so far in evaluating the maternal modulation versus mediation hypothesis were based on observations of male offspring. Regarding maternal mediation hypothesis, most studies were conducted in either male offspring alone (e.g. Liu et al., 1997; Bagot et al., 2012) or in female offspring alone when their reproductive function and infant care behaviors were evaluated (Cameron et al., 2005, 2008; Francis et al., 1999). Therefore, at the moment, no systematic knowledge is available to draw conclusions regarding male and female differences across a wide range of offspring function in the context of testing the maternal modulation versus mediation hypothesis. Although the maternal modulation studies conducted so far have been carried out on male offspring only as well, we expect that the impact of environmental novelty on the female offspring will also be modulated by the maternal behavioral and physiological context, because the theoretical framework embodying the maternal modulation of environmental influence is formulated without specific reference to the sex of the offspring. The second is to expand the range of offspring measures to couple the already diverse range of behavioral, neurophysiological, and physical functional end points examined in the context of maternal modulation with a richer array of cellular, molecular, and genetic measures. These additional measures may provide the data necessary for setting parameters of computational models that can make quantitative predictions at multiple levels of

analysis. A tradition in investigating neural mechanisms has been to identify ‘‘circuits’’ or particular brain regions affected by an experimental manipulation and this approach has lead to discoveries of unique roles played by the hippocampus, frontal cortex, and amygdala (Eiland et al., 2012; Isgor et al., 2004; Kolb et al., 2012; Muhammad et al., 2012). Here we suspect that the joint impact of environmental novelty and maternal behavior and physiology will likely be global because differences in the HPA axis will result in global changes throughout the central and peripheral nervous systems via the wide distribution of its receptors (de Kloet et al., 2008, 1993; McEwen et al., 1968). Moreover, changes in receptor concentrations in one brain region can affect neuronal activity and organization in another region without these receptors, which in turn can change the neural computation carried out by these regions. Therefore, a more fruitful approach may be to target global patterns of brain activity and connectivity as end points in testing the maternal modulation hypothesis. The third is to translate the novel experimental design paradigm used here to human child development study whether the impact of environmental novelty can be made positive by providing the right kind of maternal context, in terms of consistency of behavior and self-stress regulation. We believe that when the present findings are translated into human terms, they would suggest that a good caregiver may have several characteristics. First, she couples her encouragement of exposure to new situations with context-dependent and reliable caregiving behaviors without being overwhelming and smothering. Second, she recognizes the importance of regulating her own emotional, and therefore hormonal, responses to stress and takes time and action to develop this capacity. By providing ‘‘smarter care’’ instead of ‘‘more care’’ and by investing not only in her infant’s, but her own well-being, a mom can foster in her child a greater ability to self-regulate his or her own stress response, greater cognitive and emotional capacity, and reduced health risk associated with low early growth. These advantages in turn may ultimately translate into a sense of self-confidence and good self-esteem in the child. References Ainsworth, M.D.S., Blehar, M.C., Waters, E., Wall, S., 1978. Patterns of Attachment: A Psychological Study of the Strange Situation. Lawrence Erlbaum, Hillsdale, NJ. Akers, K.G., Nakazawa, M., Romeo, R.D., Connor, J.A., McEwen, B.S., Tang, A.C., 2006. Early life modulators and predictors of adult synaptic plasticity. Eur. J. Neurosci. 24, 547–554. Akers, K.G., Reeb, B.C., Tang, A.C., 2004. Developmentally stable sex-dependent modulation of turning asymmetry by neonatal novelty exposure. Behav. Brain Res. 155, 257–263. Akers, K.G., Yang, Z., DelVecchio, D.P., Reeb, B.C., Romeo, R.D., McEwen, B.S., Tang, A.C., 2008. Social competitiveness and plasticity of neuroendocrine function in old age: influence of neonatal novelty exposure and maternal care reliability. PLoS ONE 7, e2840. Bagot, R.C., Zhang, T.Y., Wen, X., Nguyen, T.T.T., Nguyen, H.B., Diorio, J., Meaney, M.J., 2012. Variations in postnatal maternal care and the epigenetic regulation of metabotropic glutamate receptor 1 expression and hippocampal function in the rat. Proc. Natl. Acad. Sci. USA 109 (Suppl 2), 17200–17207. Bangasser, D.A., Valentino, R.J., 2012. Sex differences in molecular and cellular substrates of stress. Cell. Mol. Neurobiol. 32, 709–723. Baram, T.Z., Davis, E.P., Obenaus, A., Sandman, C.A., Small, S.L., Solodkin, A., Stern, H., 2012. Fragmentation and unpredictability of early-life experience in mental disorders. Am. J. Psychiatry 169, 907–915. Becker, J.B., Monteggia, L.M., Perrot-Sinal, T.S., Romeo, R.D., Taylor, J.R., Yehuda, R., Bale, T.L., 2007. Stress and disease: is being female a predisposing factor? J. Neurosci. 27, 11851–11855. Cameron, N.M., Champagne, F.A., Parent, C., Fish, E.W., Ozaki-Kuroda, K., Meaney, M.J., 2005. The programming of individual differences in defensive responses and reproductive strategies in the rat through variations in maternal care. Neurosci. Biobehav. Rev. 29, 843–865. Cameron, N.M., Fish, E.W., Meaney, M.J., 2008. Maternal influences on the sexual behavior and reproductive success of the female rat. Horm. Behav. 54, 178–184. Catalani, A., Alemà, G.S., Cinque, C., Zuena, A.R., Casolini, P., 2011. Maternal corticosterone effects on hypothalamus-pituitary-adrenal axis regulation and behavior of the offspring in rodents. Neurosci. Biobehav. Rev. 35, 1502–1517. Dallman, M.F., 2000. Moments in time – the neonatal rat hypothalamo-pituitaryadrenal axis. Endocrinology 141, 1590–1592.

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002

A.C. Tang et al. / Frontiers in Neuroendocrinology xxx (2013) xxx–xxx de Kloet, E.R., Karst, H., Joels, M., 2008. Corticosteroid hormones in the central stress response: quick-and-slow. Front. Neuroendocrinol. 29, 268–272. de Kloet, E.R., Oitzl, M.S., Joels, M., 1993. Functional implications of brain corticosteroid receptor diversity. Cell. Mol. Neurobiol. 13, 433–455. Denenberg, V.H., 1964. Critical periods, stimulus input, and emotional reactivity: a theory of infantile stimulation. Psychol. Rev. 71, 335–351. Denenberg, V.H., Brumaghim, J.T., Haltmeyer, G.C., Zarrow, M.X., 1967. Increased adrenocortical activity in the neonatal rat following handling. Endocrinology 81, 1047–1052. Dinces, S.M., Romeo, R.D., McEwen, B.S., Tang, A.C., 2013. Glucocorticoids modulated the effects of novelty on adult offspring HPA function. Paper Presented at the Society of Biological Psychiatry, San Francisco, CA. Egeland, B., Pianta, R., O’Brien, M.A., 1993. Maternal intrusiveness in infancy and child maladaptation in early school years. Development and Psychopathology 5, 359–370. Eiland, L., Ramroop, J., Hill, M.N., Manley, J., McEwen, B.S., 2012. Chronic juvenile stress produces corticolimbic dendritic architectural remodeling and modulates emotional behavior in male and female rats. Psychoneuroendocrinology 37, 39– 47. Francis, D., Diorio, J., Liu, D., Meaney, M.J., 1999. Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science 286, 1155–1158. Isgor, C., Kabbaj, M., Akil, H., Watson, S.J., 2004. Delayed effects of chronic variable stress during peripubertal-juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 14, 636–648. Ivy, A.S., Brunson, K.L., Sandman, C., Baram, T.Z., 2008. Dysfunctional nurturing behavior in rat dams with limited access to nesting material: a clinically relevant model for early-life stress. Neuroscience 154, 1132–1142. Kaffam, A., Meaney, M.J., 2007. Neurodevelopmental sequelae of postnatal maternal care in rodents: clinical and research implications of molecular insights. J. Child Psychol. Psychiatry 48, 224–244. Kolb, B., Mychasiuk, R., Muhammad, A., Li, Y., Frost, D.O., Gibb, R., 2012. Experience and the developing prefrontal cortex. Proc. Natl. Acad. Sci. USA 109 (Suppl 2), 17186–17193. Kristof, N.D., 2012. Cuddle Your Kid! New York Times. Leerkes, E.M., 2011. Maternal sensitivity during distressing tasks: a unique predictor of attachment security. Infant Behav. Develp. 34, 443–446. Levine, S., 1957. Infantile experience and resistance to physiological stress. Science 126, 405. Levine, S., 1960. Stimulation in infancy. Sci. Am. 202, 81–86. Liu, D., Diorio, J., Tannenbaum, B., Caldji, C., Francis, D., Freedman, A., Meaney, M.J., 1997. Maternal care, hippocampal glucocorticoid receptor gene expression and hypothalamic-pituitary-adrenal responses to stress. Science 277, 1659–1662. Macrì, S., Zoratto, F., Laviola, G., 2011. Early-stress regulates resilience, vulnerability and experimental validity in laboratory rodents through mother-offspring hormonal transfer. Neurosci. Biobehav. Rev. 35, 1534–1543. McCrory, E., De Brito, S.A., Viding, E., 2010. Research review: the neurobiology and genetics of maltreatment and adversity. J. Child Psychol. Psychiatry 51, 1079– 1095. McEwen, B.S., Weiss, J.M., Schwartz, L.S., 1968. Selective retention of corticosterone by limbic structures in rat brain. Nature 220, 911–912. Meaney, M.J., 2010. Epigenetics and the biological definition of gene x environment interactions. Child Dev. 81, 41–79. Moore, C.L., 1992. The role of maternal stimulation in the development of sexual behavior and its neural basis. Ann. N. Y. Acad. Sci. 662, 160–177. Moriceau, S., Sullivan, R.M., 2006. Maternal presence serves as a switch between learning fear and attraction in infancy. Nat. Neurosci. 9, 1004–1006. Muhammad, A., Carroll, C., Kolb, B., 2012. Stress during development alters dendritic morphology in the nucleus accumbens and prefrontal cortex. Neuroscience 216, 103–109. Nachmias, M., Gunnar, M., Mangelsdorf, S., Parritz, R.H., Buss, K., 1996. Behavioral inhibition and stress reactivity: the moderating role of attachment security. Child Dev. 67, 508–522. Parker, K.J., Buckmaster, C.L., Sundlass, K., Schatzberg, A.F., Lyons, D.M., 2006. Maternal mediation, stress inoculation, and the development of neuroendocrine stress resistance in primates. Proc. Natl. Acad. Sci. USA 103, 3000–3005. Raineki, C., Moriceau, S., Sullivan, R.M., 2010. Developing a neurobehavioral animal model of infant attachment to an abusive caregiver. Biol. Psychiatry 67, 1137– 1145. Reeb-Sutherland, B.C., Tang, A.C., 2011. Dissociation between neonatal noveltyinduced preferential maternal care and enhancement in cognitive, social, and emotional functions. Behav. Brain Res. 224, 318–325. Reeb-Sutherland, B.C., Tang, A.C., 2012. Functional specificity in the modulation of novelty exposure effects by reliability of maternal care. Behav. Brain Res. 226, 345–350. Reeb, B.C., Tang, A.C., 2005. Sex difference in temporal patterns of social interaction and its dependence upon neonatal novelty exposure. Behav. Brain Res. 158, 359–365.

7

Richardson, H.N., Zorrilla, E.P., Mandyam, C.D., Rivier, C.L., 2006. Exposure to repetitive versus varied stress during prenatal development generates two distinct anxiogenic and neuroendocrine profiles in adulthood. Endocrinology 147, 2506–2517. Sanchez, M.M., 2006. The impact of early adverse care on HPA axis development: nonhuman primate models. Horm. Behav. 50, 623–631. Sapolsky, R.M., 1997. The importance of a well-groomed child. Science 277, 1620– 1621. Sharp, H., Pickles, A., Meaney, M.J., Marshall, K., Tibu, F., Hill, J., 2012. Frequency of infant stroking reported by mothers moderates the effect of prenatal depression on infant behavioural and physiological outcomes. PLoS ONE 7, e45446. Shonkoff, J.P., Garner, A.S., 2012. The lifelong effects of early childhood adversity and toxic stress. Pediatrics 129, e232–e246. Stanton, M.E., Levine, S., 1988. Maternal modulation of infant glucocorticoid stress response: role of age and maternal deprivation. Psychobiology 16, 223– 228. Stanton, M.E., Levine, S., 1990. Inhibition of infant glucocorticoid stress response: specific role of maternal cues. Dev. Psychobiol. 23, 411–426. Stanton, M.E., Wallstrom, J., Levine, S., 1987. Maternal contact inhibits pituitaryadrenal stress responses in preweanling rats. Dev. Psychobiol. 20, 131–145. Stern, J.M., 1997. Offspring-induced nurturance: animal-human parallels. Dev. Psychobiol. 31, 19–37. Tang, A.C., 2001. Neonatal exposure to novel environment enhances hippocampaldependent memory function during infancy and adulthood. Learn Mem. 8, 257– 264. Tang, A.C., Akers, K.G., Reeb, B.C., Romeo, R.D., McEwen, B.S., 2006. Programming social, cognitive, and neuroendocrine development by early exposure to novelty. Proc. Natl. Acad. Sci. USA 103, 15716–15721. Tang, A.C., Jiang, H., Yang, Z., Zhang, Y., Romeo, R.D., McEwen, B.S., 2011a. Converging influence of neonatal novelty experience and maternal self-stress regulation on the plasticity of offspring acoustic startle response latency. Behav. Brain Res. 221, 253–260. Tang, A.C., Nakazawa, M., Reeb, B.C., 2003a. Neonatal novelty exposure affects sex difference in open field disinhibition. NeuroReport 14 (12), 1553– 1556. Tang, A.C., Reeb-Sutherland, B.C., Romeo, R.D., McEwen, B.S., 2012a. Reducing behavioral inhibition to novelty via systematic neonatal novelty exposure: the influence of maternal hypothalamic-pituitary-adrenal regulation. Biol. Psychiatry 72, 150–156. Tang, A.C., Reeb-Sutherland, B.C., Yang, Z., 2011b. Functional brain asymmetry in adult novelty response: On fluidity of neonatal novelty exposure effects. Behav. Brain Res. 221, 91–97. Tang, A.C., Reeb-Sutherland, B.C., Yang, Z., Romeo, R.D., McEwen, B.S., 2011c. Neonatal novelty-induced persistent enhancement in offspring spatial memory and the modulatory role of maternal self-stress regulation. J. Neurosci. 31, 5348–5352. Tang, A.C., Reeb, B.C., 2004. Neonatal novelty exposure, dynamics of brain asymmetry, and social recognition memory. Dev. Psychobiol. 44, 84–93. Tang, A.C., Reeb, B.C., Romeo, R.D., McEwen, B.S., 2003b. Modification of social memory, hypothalamic-pituitary-adrenal axis, and brain asymmetry by neonatal novelty exposure. J. Neurosci. 23, 8254–8260. Tang, A.C., Verstynen, T., 2002. Early life environment modulates ‘handedness’ in rats. Behav. Brain Res. 131, 1–7. Tang, A.C., Yang, Z., Reeb-Sutherland, B.C., Romeo, R.D., McEwen, B.S., 2012b. Maternal modulation of novelty effects on physical development: the importance of self-stress regulation and care reliability. Proc. Natl. Acad. Sci. USA 106, 2120–2125. Tang, A.C., Zou, B., 2002. Neonatal exposure to novelty enhances long-term potentiation in CA1 of the rat hippocampus. Hippocampus 12, 398–404. Tang, A.C., Zou, B., Reeb, B.C., Connor, J.A., 2008. An epigenetic induction of a rightshift in hippocampal asymmetry: selectivity for short- and long-term potentiation but not post-tetanic potentiation. Hippocampus 18, 5–10. Thelen, E., Smith, L.B., 1994. A Dynamic Systems Approach to the Development of Cognition and Action. MIT Press, Cambridge, MA. Veenema, A.H., 2009. Early life stress, the development of aggression and neuroendocrine and neurobiological correlates: what can we learn from animal models? Front. Neuroendocrinol. 30, 497–518. Verstynen, T., Tierney, R., Urbanski, T., Tang, A., 2001. Neonatal novelty exposure modulates hippocampal volumetric asymmetry in the rat. NeuroReport 12, 3019–3022. Yang, Z., Tang, A.C., 2011. Novelty-induced enhancement in spatial memory: is infancy a critical period? Behav. Brain Res. 219, 47–54. Zhang, T.Y., Meaney, M.J., 2010. Epigenetics and the environmental regulation of the genome and its function. Annu. Rev. Psychol. 61, 439–466. Zou, B., Golarai, G., Connor, J.A., Tang, A.C., 2001. Neonatal exposure to a novel environment enhances the effects of corticosterone on neuronal excitability and plasticity in adult hippocampus. Brain Res. Dev. Brain Res. 130, 1–7.

Please cite this article in press as: Tang, A.C., et al. On the causes of early life experience effects: Evaluating the role of mom. Front. Neuroendocrinol. (2013), http://dx.doi.org/10.1016/j.yfrne.2013.11.002