Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 1.
Published in final edited form as: Dev Psychobiol. 2020 Apr 20;63(2):237–246. doi: 10.1002/dev.21974

A Caretaker Acute Stress Paradigm: Effects on Behavior and Physiology of Caretaker and Infant

Ed Tronick 1,2, Isabelle Mueller 1, Jennifer DiCorcia 1, Richard Hunter 1, Nancy Snidman 1
PMCID: PMC7572457  NIHMSID: NIHMS1584158  PMID: 32311073

Abstract

While experimental stress paradigms of infants (arm restraint; the Still-Face) are powerful tools for infant research, no study has experimentally stressed mothers to observe its independent effects on infant stress regulation. Extant caretaker/maternal stress studies essentially are correlational and confounded by other conditions (e.g., depression). Here we present a standard procedure, the Caregiver Acute Stress Paradigm (CASP), for stressing mothers during en face interactions with their infants. We hypothesized that infants of the stressed mothers would be more distressed than infants of non-stressed mothers. 106 4-month- old infants and their mothers were randomly assigned to the experimental stress or non-stress manipulation. Confirming our hypothesis, infants of the stressed mothers were significantly more likely to become distressed and require terminating the procedure. While objective ratings of maternal behavior showed no difference between groups, mother in the stress-condition self-rated the episode following the caretaker-stress significantly lower than mothers in the non-stress group. The self-ratings in the maternal stress-group were reflected in infant cortisol. The findings indicate that CASP is an effective experimental paradigm for exploring the independent effects of an acute stress on caretakers, including effects of conditions, such as poverty or mental illness.

INTRODUCTION

A growing body of evidence suggests that early exposure to maternal stress can have long-lasting effects on a child’s socio-emotional, cognitive, and neuropsychological development (Beck, 1998; Cicchetti et al., 1998; Goodman et al 1999; Murray et al., 2001; Weinberg et al., 1998). Similarly, maternal stress has been reported to have long-term consequences on the child’s stress-response regulation (Kaufman et al., 1999) and latent outcomes, such as an increased vulnerability for mental health problems (Kessler et al., 2010; Carr et al., 2013; Reuben et al., 2016). Research investigating the underlying mechanisms connecting maternal stress with infant development hypothesizes that maternal stress may interfere with her caregiving capacity, such as a lowered sensitivity or inconsistency in co-regulation with the infant (Beeghly & Tronick, 2011; Burt et al, 2002; DiCorcia & Tronick, 2011; Feldman, 2007; Goodman et al., 1999; Serretti, 2006).

The quality of interactions between infants and their mothers is widely held as a primary process affecting infant development. Sensitive caregiving is crucial for the infant’s ability to cope with everyday stressors and regulate negative emotions during the first year of life before the infant’s self-regulatory skills become more robust. Along with other models (Feldman, 2007; Fogel, 1982, 1993; Trevarthen, 1979) the Mutual Regulation Model (MRM; Tronick, 2011) conceptualizes infants and their mothers as co-regulating their social interactions by responding to each other’s communicative displays of their needs and intentions on a moment-to-moment basis. A major tenet of these models is that stress of either interactant – infant or mother – will disrupt their regulatory capacity leading to interactive disorganization. Exposure to maternal stress during this important period may serve as an additional risk factor, disrupting the consolidation of early regulatory capacities that support the development of a well-adjusted stress response (Mueller et al., 2019; Mueller & Tronick, 2019).

One hypothesis on how maternal stress affects development can be derived from Buffer-Transducer Model (Tronick, 2018): Regulating stress depletes resources that typically would be utilized for purposes of growth and development (Hobfoll, 1989; Tronick, 2018). The caregiving task of the mother is to scaffold the infant’s endogenous regulatory resources and to buffer the infant from exposure to resource depleting factors. For example, sensitive caregiving buffers the infant’s hormonal stress response and therefore preserves resources for growth and also protects the developing brain from harmful effects of stress hormones (Gunnar and Donzella, 2001). Ongoing caregiver relationships that do not provide a sense of safety not only fail to buffer the infant they actually may transduce stress to the infant (Tronick, 2018). For example, a parent’s inability to co-regulate infant stress generates a hormonal stress response. The stress response is biologically costly and exposes the infants’ brain to adverse stress effects (Zeanah and Gleason, 2010). Also, the frontal lobe, central for the development of emotion regulatory skills, undergoes a period of rapid growth and synaptic excess around 6–18 months of age, making it a critical period for the infant to learn how to respond to emotions (Dawson, 1994; Nelson and Bosquet, 2004). Until self-regulation of emotion is well developed the caregivers are a source of external regulation for the child, hence the caregivers play a vital co-regulatory role in the development of infants’ emotion regulatory behaviors. Sensitive responding to the child’s regulatory needs by a caregiver helps the child to regulate stress and to better control stress hormones. This, in turn, helps the child to learn how to regulate herself more effectively. Furthermore, to complete this account of resource utilization and stress, one needs to recognize that the depletion of parental resource due to stress can compromise parent’s capacity for dyadic co-regulation. When the diminished capacity becomes chronic it may result in reducing resources for growth and undermine the quality of the emerging parent-infant interaction. In addition, calming cycle theory (Welch, 2016) has extended interpretations of dyadic co-regulation towards subconscious subcortical processes connected to Pavlovian learning mechanisms. The theory states that visceral/autonomic co-regulation between mother and infant is established in utero and after birth emotional communication between the dyad leads to autonomic co-regulation, as opposed to conventional attachment theory, emphasizing psychological regulation (Chambers, 2017). The theory links how increased sensitivity and emotional connection may facilitate a more optimal co-regulation and maturation of the infant’s autonomic regulation (Porges et al., 2019). In sum, these early experiences of sensitive regulation or dysregulation are critical for the child’s development be it good or ill (Sroufe and Rutter, 1984; Tronick, 2017).

Unfortunately, most research investigating the underlying mechanisms by that maternal stress affects infant stress reactivity has several major flaws that impair our ability to distinguish between primary causes and confounding influences. First, the concept of maternal stress is not well defined in the developmental psychobiological literature. The term has been conceptualized ever more broadly to include traumatic events, stressful life circumstances (such as poverty, loss of a loved one), as well as the psychological disorders emerging from these disparities (Sosnowski et al., 2018). Stress is currently used as an umbrella term for the presence of physiological and psychological stressors, psychopathologies, or even the occurrence of subclinical symptoms of mental problems (Doyle & Cicchetti, 2018). Especially the inclusion of psychopathologies such as anxiety and depressive symptoms has been criticized as the term “stress” appears to get increasingly applied to a wider range of experiences that exceedingly differ in their intensity, with some being closer to everyday challenges than the original conception of stress (Kagan, 2016; DiCorcia & Tronick, 2011). Moreover, the presence of psychological stressors is often documented through self-report measures, even though anxious and depressive symptoms often co-occur with negative attributions and a biased self-perception (Zbozinek et al, 2012).

Contributing to problems with a clear definition of maternal stress, studies often fail to address several methodological flaws that affect our ability to correctly estimate its effects. Studies on maternal stress often do not control for associated external conditions such as poverty, single parenthood, psychopathologies, and lack of social support (Apter-Levy et al., 2013). Our understanding is further limited because we are unable to evaluate the interplay of typical confounders such as postpartum depression, anxiety and psychiatric symptoms, parenting and life events, as well as chronic physiologic stress with the putative stressor. Moreover, most studies on maternal stress are inconsistent in their design, often retrospective or short-term. Prospective long-term studies would be needed to bring together the inconsistencies and variations of results reported in the literature on the effects of childhood exposure to maternal stress (Bremner & Vermettern, 2001; Gunnar & Donzella, 2002). Experimental manipulations to study infant stress exposure have been successfully developed and replicated, such as the Still-face Paradigm (Tronick et al., 1978) or arm restraint stress (Camras et al., 1992; Stifter & Fox, 1990). These paradigms are rightfully well-established and crucial tools for infancy research. They provide experimental control of the stress that enables causal interpretation of its effects. Thus, what has yet to be adequately explored in a controlled experimental fashion is how stress experienced by the mother affects her self-regulation and the resulting cascading behavioral and physiologic effect on her infant and their interaction.

The current study aims to introduce a new procedure, the Caretaker Acute Stress Paradigm (CASP), for stressing mothers during en face interactions in a controlled laboratory setting. We expect that CASP will combine the advantages of an experimental setting with a higher ecological validity, enabling us to eliminate the inconsistencies in the conception of maternal stress, and finally allow us to better control for external factors influencing results. In the CASP we introduce a paradigm with two stressors, first a maternal stressor, followed by a well-established infant stress paradigm (Still-Face) to see how dyadic regulation may get compromised when the mother is already under duress. The CASP paradigm begins with a baseline period to establish physiologic baselines on the mother and infant. It is followed by a play episode in which mother and infant engage en face as they would at home. Then depending upon a-priori group randomization, baseline play is followed by either a maternal stressor or a control condition. The maternal stressor consisted of infant cries, recorded during medical procedures presented through headphones. The cries were accompanied by pictures of infants in distress and a brief narrative that the infant is undergoing a medical procedure. Mother and infant are visually separated by a screen to ensure the mother is focused on the paradigm and of course that the infant is not affected by any reaction of the mother. Cries as a stressor were chosen for several reasons. While other methods are also known to likely induce stress (e.g. Kirschbaum et al., 1993), infant cries have relevance to parenting thereby adding to the ecological validity of the CASP. In previous research infant cries have been shown to produce a stress response (Krippl et al., 2010; LaGasse et al., 2005; Lester et al., 1992; Sheinkopf et al., 2006; Vecchio et al., 2009) and more distressing cries have been shown to recruit regions of the brain associated with arousal and attention (Montoya et al, 2012). In the control condition for mothers listen to a recording of infant giggles and laughter along with images of happy infants and a brief description that the infant is playing with the mother. Following the experimental episode all dyads are reunited for a play, allowing a comparison between the dyadic interaction before and after the maternal manipulations and between groups. To investigate the effects on infant regulation a classic Still-Face episode and a final en face play follow. The introduction of this infant stressor is important for understanding how mothers attempt to re-regulate the interaction when already under stress. By contrast and in emendation to previous studies of “stressed groups”, an experimental procedure with random subject assignment overcomes many methodologic issues and, when well-controlled, can potentially lead to causal interpretations as well as to an understanding of the interplay of mediators and moderators with stress. Measures of interest included maternal self-reported stress levels along with maternal and infant behavior and salivary cortisol.

We hypothesized that stressed mothers and their infants would be more distressed than non-stressed mothers and their infants. More specifically we hypothesized, for mothers in the stress condition their behavior following the stress condition compared to their play prior to the stress condition would be more intrusive or tense compared to the mothers’ in the non-stress condition. Further, we predicted the infants of the stressed mothers would express more negative affect and distress, and higher levels of cortisol.

METHODS

Participants

A total of 106 mothers and their 16-week old (+/− 3 days) infants (58 female) participated in the study. All infants were delivered full-term (37 weeks or greater) and were clinically healthy as determined by pediatric examination, with no chronic medical conditions or time in the neonatal intensive care unit. Infants also were clinically healthy at the time of testing. Mothers were between 20- and 42 years of age at time of birth, predominantly Caucasian (54.7 %, n= 58) or African-American (26.4%, n=28; other 9.6%, n=10; do not wish to answer 9.6%, n=10), with no serious chronic health conditions and at least a high school education. Dyads were recruited at the maternity ward of a large metropolitan Harvard Medical School-affiliated teaching hospital. Maternal and infant medical records were first reviewed by hospital nurses to exclude mothers who had a serious medical and/or mental health (e.g., psychotic disorders) condition and/or infants with serious health conditions. Twice a week a recruiter visited the rooms of mothers of full-term healthy infants in the well-child nursery and either talked with the mother about the study or, if the mother was unavailable, left written material and gathered contact information. All potential participants were contacted during the newborn period and later recruited by phone 3 weeks before the infants were 4-months of age. Mother-infant dyads who met the inclusion criteria were recruited regardless of race/ethnicity, mothers’ marital status, or socio-economic status. All mother-infant dyads were randomly assigned to stress or non-stress group before their arrival to the lab. A research assistant assigned each dyad a number between 0-9 using a web-based randomizer (randomizer.org), dyads with numbers between 0-4 were assigned to the non-stress group (n=54, 31 female) , dyads with numbers between 5-9 were assigned to the caretaker stress group (n=52, 27 female).

The Caretaker Acute Stress Paradigm

The Caretaker Acute Stress Paradigm (CASP) was developed to observe the influence of maternal stress on maternal and infant reactivity within an experimental setting. Overall, CASP has a standard 30-second physiology baseline and 5 episodes, each 2-minutes long. Mothers are seated in a standard chair, infants placed in a highchair facing the mother, close enough that the mother can touch and interact with her child. Following a resting baseline during which the mother sits quietly while the infant watches a video the dyad engages in a face-to-face natural play (episode E1). The caretaker acute stress experimental episode (E2) follows in which the caretaker is exposed to an auditory and matching visual stressor or control condition. Caretakers in the acute stress condition hear infant cries and hear a brief narrative over headphones that the infants were undergoing a medical procedure over while watching matching images of crying infants on a screen in front of them. In the control condition mothers hear infant positive vocalizations over the earphones and a narrative that the infants were playing with an adult while watching matched images on a screen. It is noteworthy that the narratives were suggested by several mothers who served as pilot participants. To ensure mothers would focus on the experimental condition and the infants would not be aware of the mothers’ reaction, a screen was set up between her and her infant. All infants stayed in their highchair and a research assistant entertained them with bubbles and finger puppets for the 2-minutes of the episode. Thus, infants had the exact same experience regardless of maternal experimental condition. Next, mother and infant were reunited for another face-to-face play episode (E3), followed by a classic still-face episode (E4) where the mother is asked to stop the interaction, sit back, and maintain a neutral “poker-face” (still-face). The final episode is another face-to-face play (E5; see figure 1 for entire paradigm). The CASP paradigm ended after E5 or was terminated early if an infant showed significant distress (e.g. crying) for more than 30 consecutive seconds. All infants who made it to E3 for 30 seconds were included in the study.

Figure 1:

Figure 1:

Open in a new tab

The Caretaker Acute Stress Paradigm, containing a standard 30-second physiology baseline and 5 episodes, each 2-minutes long. First the dyad engages in a face-to-face natural play (episode E1). The acute caretaker experimental episode (E2) follows: the caretaker is exposed to an auditory and matching visual stressor or control condition. Caretakers in the experimental condition hear infant cries, in the control condition mothers hear infant laughter and giggles over the earphones while watching matched images on a screen. Next, mother and infant were reunited for another face-to-face play episode (E3), followed by a classic still-face episode (E4) where the mother is asked to stop the interaction, sit back, and maintain a neutral “poker-face” (still-face). The final episode is another face-to-face play (E5).

Measures

Infant Behavior.

To evaluate infant stress reactivity in interaction with a stressed or non-stressed caretaker two behaviors were coded: the timepoint of first distress shown by the infant and the episode in which the infant had to be terminated because of distress, the drop-out episode. Two research assistants blind to condition coded the video recordings of the infant for the first time the infant showed distress during the paradigm following E2. Drop-out episode was recorded in real time on the study notes and then reviewed later on the video recording. Infants were considered to have dropped out when they met the 30 consecutive seconds distress criteria either in E3 (play episode after the caretaker stress), E4 (the still-face episode), or E5, the reunion play episode after the still-face.

Maternal Self-Rating.

To evaluate maternal perception of the stress of the paradigm mothers were instructed to rate their rate their emotional state on a Likert scale after each episode. To avoid suggestive questioning (e.g. “good”, “tense”, etc.) the self-report was labeled with faces, ranging from very positive affect (7) to very negative affect (1), with 4 being neutral.

Maternal Behavior.

Video recordings of each episode were divided into 30-second portions creating 4 30-second segments per episode. Two research assistants blind to condition rated the dominant maternal behavior for each segment on a 7-point rating scale in 3 different dimensions until they reached agreement. Maternal affect from negative (1) to positive (7), maternal activity from inactive (1) to active (7), and maternal tension from tense (1) to calm (7). In addition, every incident of mothers touching her infant’s face was recorded as a measure indicating maternal concern or stress that might differ between experimental and control mothers.

Salivary Cortisol.

To evaluate the influence of the caretaker stress on the maternal and infant stress response, a total of 4 salivary samples were collected. One at baseline after formal consent and three post-stress and then at 10 (p10), 20 (p20), and 40 (p40) minutes after the still-face episode, considered the peak stress experience for the infant. Each sample was taken at the same time for mother and infant with a cotton swap. The mothers were handed the swap without touching it and asked to swirl it in their mouth for one minute. A research assistant used a longer cotton swap to get the infant’s salivary sample during the same time. All samples were immediately stored in a freezer in an adjunct room at −20 degrees Celsius. After study completion all stored samples of mother-infant dyads that met study criteria were analyzed using a standard commercial ELISA salivary cortisol kit (Enzo Life Sciences).

Experimental Procedure.

Mothers and their infants were randomly assigned to either experimental or control group before their arrival. After arrival in the laboratory oral and written informed consent was given. All mothers signed the consent form. As a first step, a salivary baseline sample was collected from mother and infant. Next seven electrodes (MindWare Technologies Ldt.) were placed on mother and infant to collect cardiac and respiratory data during the paradigm (Not reported here). Next, mother and infant were brought to the study room, where the infant was seated in a highchair, while the mother sat on a chair facing the infant, close enough to touch or interact with the infant. The electrode wires from mother and infant were connected and the research assistant made sure the wires were tucked away so that the infant would not be able to reach them. Two video cameras installed on the wall were used to record mother and infant. Research assistants were able to monitor the study room and physiology from an adjacent control room. Physiology and video were initialized simultaneously through E-Prime® software to ensure exact timing on both measures. The mother was given an earpiece connected to a walky-talky so that a research assistant could give instructions about the procedure without entering the room.

All mother-infant dyads then participated in CASP according to protocol. During E4, the still-face episode (and presumably peak stress-exposure), research assistants started 3 timers for the following post-stress salivary cortisol samples at 10-, 20-, and 40-minutes after E4. If the infant dropped out due to the 30-second distress rule the timer was started at the drop-out time. After CASP completion the dyads returned to the laboratory waiting room, cortisol samples were taken, and mothers were asked to complete questionnaires while a research assistant was attending the infant.

Results

Infant Dropout after CASP

Drop-out episodes were evaluated to see if there was a difference in drop-out rate between the experimental and control groups. Using a Chi-Square comparing the acute stress group with the non-stress group there was a significant difference in drop-out rate for the play episode and the still-face episode after the CASP paradigm (E3, E4). The paradigm had to be terminated for significantly more infants in the acute stress group compared to infants in the non-stress group (X2 = (1, N=106) = 4.662, p = .031; see figure 2a).

Figure 2:

Figure 2:

Figure 2:

Figure 2:

Figure 2:

Open in a new tab

Effects of CASP on Infant and Mother. A) Infant drop-out in the paradigm due to severe distress, during E3 and E4 (Play & SF after the Paradigm/Vocalizations). The paradigm had to be terminated for significantly more infants in the acute stress group compared to infants in the non-stress group (X2 = (1, N=106) = 4.662, p = .031). B) Differences between maternal group and infant first distress during either a play episode or a stress episode. Chi-Square analysis indicated that infants whose mothers were in the experimental condition showed their first distress significantly more often during a (usually positive) play episode than infants whose mothers were in the non-stress group (X2 = (1, N=31) = 4.288, p = .038). C) Maternal Self-Rating after the maternal manipulation episode (E2). There was a significant difference in rating the vocalizations between groups (F(1, 105) = 677.226, p < .001). D) Maternal Self-Rating of the Play episodes before and after the vocalizations (E1, E3). There was a significant interaction between time and group (F(1, 105) = 7.456, p = .007), indicating that mothers in the acute stress group rated E3, the play episode after the vocalizations lower than E1, and lower than the non-stress group.

Infant First Distress

To evaluate whether CASP has an effect on infant distress the first sign of distress (e.g., negative or cry vocalization) was coded, regardless how long the distress lasted. Overall first distress showed a moderate correlation with dropout episode (r(105) = .699, p < .000). Further, we dummy coded whether first distress appeared during a play episode (E3, E5) or during a stress episode (E4). Chi-Square analysis indicated that infants whose mothers were in the experimental condition showed their first distress significantly more often during a (usually positive) play episode than infants whose mothers were in the non-stress group (X2 = (1, N=31) = 4.288, p = .038; see figure 2b).

Maternal Self-Rating

A one-way ANOVA was conducted to compare the effect of the maternal manipulation (Vocalizations in E2) on maternal self-rating between experimental and control group. There was a significant difference in rating the vocalizations between groups (F(1, 105) = 677.226, p < .001, figure 2c) where acute stress group mothers rated themselves as significantly more negative (M=2, range = 1-4) than non-stress group mothers (M=6, range = 5-7). In addition, a repeated measure ANOVA for maternal self-rating of the Play episodes before and after the vocalizations was conducted (E1, E3). There was a significant interaction between time and group (F(1, 105) = 7.456, p = .007), indicating mothers in the acute stress group rated E3, the play episode after the cries lower than E1, and lower than the non-stress group (see figure 2d).

Maternal Behavior

Overall no differences in maternal behavior were found between groups (p’s>.05). Repeated measure ANOVA showed a significant change in maternal affect across time, from play episode E1 to play episode E3 (F(1, 102) = 13.387, p < .000). There was no effect of episode x group (p = .793) or main effect of group (p = .521). Repeated measure analysis for maternal tension before and after the experimental manipulation (E1, E3) revealed no main effect of group (p = .618) or interaction of group x time (p = .098). Similarly, repeated measure analysis for maternal level of activity before and after the caretaker stress (E1, E3) revealed no main effect of groups or interaction (p>.05).

Cortisol

Maternal baseline cortisol was correlated with all 3 of the post-stress samples (rp10 (87) = .302, p = .005; rp20 (74) = .477, p < .000; rp40 (75) = .582, p < .000). Further, maternal baseline cortisol was associated with infant 20 min post-stress cortisol sample (r (46) = .649, p < .000). In addition, the dyads p20 and p40 time samples showed a positive correlation (rp20 (48) = .269, p = .041; rp40 (50) = .540, p < .000). When looking only at the acute stress group, infant 20ps cortisol levels showed a significant negative correlation with drop-out episode (r (22) = -.422, p = .050), indicating that the earlier an infant in the experimental condition dropped-out during the paradigm, the higher was her 20ps-cortisol level. There was no significant association between any cortisol timepoint and drop-out episode in the non-stress group.

Correlations between Maternal Self-Rating and Cortisol

Overall, there was no correlation between maternal self-rating and cortisol. But when looking at the control and acute stress group separately there were significant correlations observed. In the acute stress group maternal post-stress cortisol showed a significant association with maternal self-rating of E4 (Still-Face episode), indicating that the higher mothers rated E4 the higher their post-stress cortisol level was (r maternal p20(34) = .479, p = .004). Moreover, infant post-stress cortisol showed a negative association to maternal rating of E5 (Play after Still-Face) indicating that the lower the mother rated E5, the higher infant cortisol was (r infant p20 (20) = −.457, p = .043). In the control condition maternal self-rating of E5 was related to maternal post-stress cortisol levels (r maternal p40 (40) = .391, p = .013), but not with infant cortisol.

DISCUSSION

The results suggest that CASP successfully alters the mother-infant interaction by changing something in the mother’s behavior that is detected by the infant. Oddly, while studies of the putative effects of stress proliferate (e.g. Bremner & Vermettern, 2001; Gunnar & Donzella, 2002) to the best of our knowledge this is the first study to experimentally and independently stress the mother to observe how it affects her regulatory capacity and subsequently her infant. This is an important extension of our knowledge and a crucial tool for future research. Thus, for example the tenet of models on maternal stress effects, such as the MRM (Tronick, 2011) has been tested and verified for the first time in an experimental fashion. Maternal stress is ubiquitous, and the effects of maternal stress on typical, as well as atypical mother-infant interactions has yet to be explored. The CASP paradigm will helping us gain a deeper and more profound understanding of the development of regulatory processes.

The mothers’ in the acute stress condition found the infant cries to be significantly more stressful than the mothers found the infant positive vocalizations. While this was expected it supports the CASP paradigm and its method of experimental stimulus manipulation. Unexpected was the interaction of the maternal ratings of stress over episodes for the acute stress group. Experiencing the cries not only affected their ratings of the cries, but the stress lowered their ratings of the play episode. Thus, there is a lingering effect of the stress even in positive context.

The infants of the acute stress mothers became negative or distressed significantly earlier and more often than infants of mothers in the control condition. The finding that for the acute stress group the first post stress infant cortisol sample was negatively correlated with the drop out episode indicates that the infants of the stressed mothers were physiologically stressed. Importantly, infant distress was not only evident during the infant stressor, but even during the play episodes. The play episode is usually a positive setting for both mother and infant. Distress during the episode suggests an ongoing disruption of the infant and of maternal regulation. Clearly maternal stress does not only affect the co-regulation of the dyad in an emotional challenging situation, but even affects subsequent playful interactions. Adding a maternal stressor to appears to lower the threshold for effective co-regulation even without the ongoing presence of a stressor.

These cortisol findings are in line with the results of a recent meta-analysis which found a robust cortisol response in less than half of all still-face studies and only for those using the double still-face paradigm (Provenzi et al., 2016). In that case for a cortisol response it is assumed that an extended stressor is needed to trigger a measurable response. In the CASP with the introduction of a maternal stressor, we were able to measure a more clearly stress-related cortisol reactivity in infants of the experimental condition. This could indicate a lowered threshold for stress tolerance due to targeting the co-regulation of the dyad. The findings here on cortisol also suggest that not only may intensity of exposure affect cortisol reactivity but there may be robust individual differences that affect cortisol reactivity to the still-face. We found that mothers’ baseline cortisol levels were related to their subsequent cortisol levels and that there were cortisol relations between mothers and infants within dyads. Moreover, the dyadic findings suggest a process of physiologic mutual influence of mothers and infants.

Quite surprisingly, despite clear findings of the effects of maternal stress on infant reactivity, we did not find differences in maternal behavior following the caretaker stress paradigm. After all, we have to assume that the infants were responding to some signal of their mothers’ state; there is no magic involved. One possibility for the lack of finding differences in the maternal behavior may be due to using a less detailed micro-temporal coding system instead of a more detailed second-by-second coding system for maternal behavior (Beebe et al., 2018; Weinberg et al., 1998). More likely, a lack of findings may also indicate that we have to look for more subtle differences or different variables than behavior. There may be a difference in maternal voice spectrum following the stress condition, or touch, as well as natural maternal olfactory cues following a stressor. In preliminary digital analyses of maternal vocalization during the stressful episode, stressed mothers’ vocalizations were found to have a higher frequency and less prosody than the vocalizations of the control mothers. It is intriguing that adult coders listening to maternal vocalizations did not hear these differences. Perhaps the infants are more sensitive to these vocal stress features than adults or it may be a function of the relationship between mother and infant.

Nonetheless, though they are not straightforward there were relations between maternal ratings of the stress and infant cortisol. First, though expected there was no overall relations between maternal ratings of stress and maternal cortisol levels. The lack of finding may reflect the mild nature of the cry stress, mothers being in close contact to their infants, most of whom were in a positive state, or analyzing experimental and control mothers together. Second, as expected experimental mothers’ self-rated stress level was positively related to their cortisol level following the stress. And even for control mothers the higher their rating of stress for the last episode following the still-face the higher their cortisol level. Further research on these stress ratings and cortisol is needed.

While many studies rely on a limited number of measures to quantify stress (e.g., salivary cortisol or questionnaires), their value is weakened because a stress response is complex and involves the interplay of multiple behavioral and physiologic reactions. CASP enables us to assess stress reactivity and regulation on multiple levels of measures including behavioral, psychological, and physiologic/hormonal responses in a controlled setting. These multi-domain measures will generate contextually rich data that captures a more complete picture of each dyad’s stress response and further allows for the evaluation of their relations. This study builds on the extensive research on the Still-Face, which has provided a crucial tool for infancy research. Knowledge gained from CASP in future studies will add to our understanding of stress and its regulatory effects and help in identifying mothers who are more vulnerable to stress that in the long run has the potential to continually disrupt dyadic interactions and subsequently compromise the infant’s capacity for healthy emotion regulation and development.

Based on the MRM (Tronick, 2011) the quality of infant social experience over the course of development affects most aspects of their functioning. While a host of processes and factors affect the quality of infants’ social experience, the most proximal organizer is the parent, typically the mother. Her capacity to organize her own behavior in order to scaffold her infant’s own self-organized regulatory capacities and behavior organizes their dyadic exchanges into a coherent whole. The infant is also constantly affected by an endless number of intrinsic and extrinsic factors that may or may not support the functioning of the mother, infant, and the dyad (e.g., genetic and epigenetic exposures, economic, culture, relations with others). One increasingly common way to conceptualize these factors is to regard some of them as perturbations or stressors that disrupt the regulatory capacity of the mother, the infant, and/or the organization of the dyad. In this conceptualization the mother-infant dyad is seen as a stress regulating system with two regulatory subcomponents, mother and infant. As a stress regulating system, the dyad is a primary, final common pathway for resolving stressful disruptions. Successful regulation of a stressor’s disruptive effects by the mother, infant, and/or their coordinated regulatory activities leads to a return to coherent organization whereas unsuccessful regulation results in continued disruption and costs to the behavioral, physiological, cognitive, affective, and regulatory functioning of the infant as well as the mother and to the organization of the dyad. Moreover, the reiterated chronic experience with success or failure becomes constitutive of each infant’s, and to a lesser extent each mother’s, and each dyad’s regulatory capacities resulting in developmental robustness or derailment.

A limitation of the study is that risk measures of maternal characteristics and history, such as early adverse experiences, trauma or depression were not examined. Examination of risk measures would have provided information on how maternal reactivity to stress interacts with other factors and would help us to better understand their interplay, including their independent effects. The dyads studied were healthy and typically developing. There is a need to study dyads that were more vulnerable to stress, such as dyads with preterm infants or neurobehaviorally dysregulated infants. A significant limitation of the study is that it lacked any sort of follow-up on dyadic functioning or infant development. Moreover, the study did not fully examine physiologic, behavioral, and genetic and epigenetic factors associated with a parent’s stress reactivity. The study examined stress reactivity at only one point in time leaving many developmental questions unanswered such as is stress reactivity stable, is it a trait or state, and what factors affect a parent’s reactivity. A factor of interest would be how the quality of the play interactions prior to the stress affect mothers’ stress reactivity to the stress and the relation of their stress reactivity to their, and their infants’ behavior and physiology during the following play and still-face episodes. Future research on the CASP should address these issues.

In this study we presented an experimental maternal stress paradigm, the CASP and found evidence that mildly stressing mothers during face-to-face interactions affects mothers’ and infants’ regulatory capacity. Our findings indicate that CASP is an effective experimental paradigm for exploring the independent effects of an acute stressor on caretakers, including effects on mother-infant dyads experiencing other compromising conditions, such as poverty or mental illness that are likely to be particularly vulnerable to the effects of stress. The CASP gives us traction on evaluating the independent effects of stress and the interplay of stress and other risk factors. For example, examining the effects of the CASP on samples that are seen as stress vulnerable such as mothers of preterm infants or mothers with anxiety disorders. It will enable us to investigate those subtle differences in caregiving under the influence of duress in a more objective way than before, allowing for new insights on the underlying mechanisms how maternal stress affects infant sensitivity and reactivity to stress.

Acknowledgements

This research was supported by grants from the National Science Foundation 1457111 (Co-PI: E.Tronick, N. Snidman), NICHD 1 R01 HD083267-01,(PI: E. Tronick), The Bial Foundation (PI: E.Tronick), and NICHD 1 R01 HD083267-01 (PII: E.Tronick). We also want to thank the mothers and infants who participated in the studies.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Footnotes

Conflict of Interest

The authors declare that the paper was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Submitted for special issue: “Advances and New Approaches to the Study of Stress, Early Experiences, and Developmental (SEED) Science” in Developmental Psychobiology. Drs. Sarah Watamura and Dr. Erika Manczak, editors

REFERENCES

  1. Ainsworth MS (1979). Infant–mother attachment. American psychologist, 34(10), 932. [DOI] [PubMed] [Google Scholar]
  2. Apter-Levy Y, Feldman M, Vakart A, Ebstein RP, & Feldman R (2013). Impact of maternal depression across the first 6 years of life on the child’s mental health, social engagement, and empathy: the moderating role of oxytocin. American Journal of Psychiatry, 170(10), 1161–1168. [DOI] [PubMed] [Google Scholar]
  3. Beck CT (1998). The effects of postpartum depression on child development: a meta-analysis. Archives of psychiatric nursing, 12(1), 12–20. [DOI] [PubMed] [Google Scholar]
  4. Beebe B, Myers MM, Lee SH, Lange A, Ewing J, Rubinchik N, … & Hofer. (2018). Family nurture intervention for preterm infants facilitates positive mother–infant face-to-face engagement at 4 months. Developmental psychology, 54(11), 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beeghly M, & Tronick E (2011). Early resilience in the context of parent–infant relationships: A social developmental perspective. Current problems in pediatric and adolescent health care, 41(7), 197–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bremner JD, & Vermetten E (2001). Stress and development: behavioral and biological consequences. Development and psychopathology, 13(3), 473–489. [DOI] [PubMed] [Google Scholar]
  7. Burt VK, & Stein K (2002). Epidemiology of depression throughout the female life cycle. The Journal of clinical psychiatry, 63, 9–15. [PubMed] [Google Scholar]
  8. Camras LA, Oster H, Campos JJ, Miyake K, & Bradshaw D (1992). Japanese and American infants’ responses to arm restraint. Developmental psychology, 28(4), 578. [Google Scholar]
  9. Carr CP, Martins CMS, Stingel AM, Lemgruber VB, & Juruena MF (2013). The role of early life stress in adult psychiatric disorders: a systematic review according to childhood trauma subtypes. The Journal of nervous and mental disease, 201(12), 1007–1020. [DOI] [PubMed] [Google Scholar]
  10. Chambers J (2017). The neurobiology of attachment: From infancy to clinical outcomes. Psychodynamic psychiatry, 45(4), 542–563. [DOI] [PubMed] [Google Scholar]
  11. Cicchetti D, & Toth SL (1998). The development of depression in children and adolescents. American psychologist, 53(2), 221. [DOI] [PubMed] [Google Scholar]
  12. Dawson G (1994). Development of emotional expression and emotion regulation in infancy: Contributions of the frontal lobe. In Dawson G & Fischer KW (Eds.), Human behavior and the developing brain (pp. 346–379). New York, NY, US: The Guilford Press. [Google Scholar]
  13. Del Vecchio T, Walter A, & O’Leary SG (2009). Affective and physiological factors predicting maternal response to infant crying. Infant Behavior and Development, 32(1), 117–122. [DOI] [PubMed] [Google Scholar]
  14. DiCorcia JA, & Tronick ED (2011). Quotidian resilience: Exploring mechanisms that drive resilience from a perspective of everyday stress and coping. Neuroscience & Biobehavioral Reviews, 35(7), 1593–1602. [DOI] [PubMed] [Google Scholar]
  15. Doyle C, & Cicchetti D (2018). Future directions in prenatal stress research: Challenges and opportunities related to advancing our understanding of prenatal developmental origins of risk for psychopathology. Development and Psychopathology, 30(3), 721–724. [DOI] [PubMed] [Google Scholar]
  16. Feldman R (2007). Maternal versus child risk and the development of parent–child and family relationships in five high-risk populations. Development and psychopathology, 19(2), 293–312. [DOI] [PubMed] [Google Scholar]
  17. Feldman R (2007). Parent–infant synchrony: Biological foundations and developmental outcomes. Current directions in psychological science, 16(6), 340–345. [Google Scholar]
  18. Fogel TFA (1982). Emotion and early interaction. Psychology Press. [Google Scholar]
  19. Fogel A (1993). Developing through relationships. University of Chicago Press. [Google Scholar]
  20. Goodman SH, & Gotlib IH (1999). Risk for psychopathology in the children of depressed mothers: a developmental model for understanding mechanisms of transmission. Psychological review, 106(3), 458. [DOI] [PubMed] [Google Scholar]
  21. Gunnar MR, & Donzella B (2001). Social regulation of the cortisol levels in early human development. Psychoneuroendocrinology, 27(1–2), 199–220. [DOI] [PubMed] [Google Scholar]
  22. Hobfoll SE (1989). Conservation of resources: A new attempt at conceptualizing stress. American psychologist, 44(3), 513. [DOI] [PubMed] [Google Scholar]
  23. Kaufman J, & Charney D (2003). The neurobiology of child and adolescent depression. Neurodevelopmental mechanisms in psychopathology, 461–490. [Google Scholar]
  24. Kessler RC, McLaughlin KA, Green JG, Gruber MJ, Sampson NA, Zaslavsky AM, … & Benjet C. (2010). Childhood adversities and adult psychopathology in the WHO World Mental Health Surveys. The British Journal of Psychiatry, 197(5), 378–385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kirschbaum C, Pirke KM, & Hellhammer DH (1993). The ‘Trier Social Stress Test’–a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology, 28(1–2), 76–81. [DOI] [PubMed] [Google Scholar]
  26. Krippl M, Ast-Scheitenberger S, Bovenschen I, & Spangler G (2010). Maternal perception of infants’ expressions of emotion. Journal of Psychophysiology. [Google Scholar]
  27. LaGasse LL, Neal AR, & Lester BM (2005). Assessment of infant cry: acoustic cry analysis and parental perception. Mental retardation and developmental disabilities research reviews, 11(1), 83–93. [DOI] [PubMed] [Google Scholar]
  28. Lester BM, Boukydis CZ, Garcia-Coll CT, Hole W, & Peucker M (1992). Infantile colic: Acoustic cry characteristics, maternal perception of cry, and temperament. Infant Behavior and Development, 15(1), 15–26. [Google Scholar]
  29. Montoya JL, Landi N, Kober H, Worhunsky PD, Rutherford HJ, Mencl WE, … & Potenza MN. (2012). Regional brain responses in nulliparous women to emotional infant stimuli. PloS one, 7(5), e36270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mueller I, Beeghly M, & Tronick E (2019). Depression Is Not Gender-Biased: Maternal and Paternal Depression and Early Parent-Infant Interactions. In Early Interaction and Developmental Psychopathology (pp. 151–164). Springer, Cham. [Google Scholar]
  31. Mueller I, & Tronick E (2019). Early life exposure to violence: developmental consequences on brain and behavior. Frontiers in behavioral neuroscience, 13, 156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Murray L, Hipwell A, Woolgar M, & Cooper P (2001). Cognitive vulnerability to depression in 5-year-old children of depressed mothers. The Journal of Child Psychology and Psychiatry and Allied Disciplines, 42(7), 891–899. [DOI] [PubMed] [Google Scholar]
  33. Nelson CA, & Bosquet M (2004). Neurobiology of fetal and infant development: Implications for infant mental health. In Zeanah CH Jr. (Ed.), Handbook of Infant Mental Health (pp. 37–59)., (2nd Ed.). NY: The Guilford Press. [Google Scholar]
  34. Porges SW, Davila MI, Lewis GF, Kolacz J, Okonmah‐Obazee S, Hane AA, … & Welch MG. (2019). Autonomic regulation of preterm infants is enhanced by Family Nurture Intervention. Developmental psychobiology, 61(6), 942–952. [DOI] [PubMed] [Google Scholar]
  35. Provenzi Livio, Giusti Lorenzo, and Montirosso Rosario. “Do infants exhibit significant cortisol reactivity to the Face-to-Face Still-Face paradigm? A narrative review and meta-analysis.” Developmental Review 42 (2016): 34–55. [Google Scholar]
  36. Reuben A, Moffitt TE, Caspi A, Belsky DW, Harrington H, Schroeder F, … & Danese A. (2016). Lest we forget: comparing retrospective and prospective assessments of adverse childhood experiences in the prediction of adult health. Journal of Child Psychology and Psychiatry, 57(10), 1103–1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Serretti A, Olgiati P, & Colombo C (2006). Influence of postpartum onset on the course of mood disorders. BMC psychiatry, 6(1), 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sheinkopf SJ, Lester BM, LaGasse LL, Seifer R, Bauer CR, Shankaran S, … & Wright LL. (2005). Interactions between maternal characteristics and neonatal behavior in the prediction of parenting stress and perception of infant temperament. Journal of Pediatric Psychology, 31(1), 27–40. [DOI] [PubMed] [Google Scholar]
  39. Sosnowski DW, Booth C, York TP, Amstadter AB, & Kliewer W (2018). Maternal prenatal stress and infant DNA methylation: A systematic review. Developmental psychobiology, 60(2), 127–139. [DOI] [PubMed] [Google Scholar]
  40. Sroufe LA, & Rutter M (1984). The domain of developmental psychopathology. Child development, 17–29. [PubMed] [Google Scholar]
  41. Stifter CA, & Fox NA (1990). Infant reactivity: Physiological correlates of newborn and 5-month temperament. Developmental Psychology, 26(4), 582. [Google Scholar]
  42. Trevarthen C (1979). Communication and cooperation in early infancy: A description of primary intersubjectivity. Before speech: The beginning of interpersonal communication, 1, 530–571. [Google Scholar]
  43. Tronick E (2018). The Caregiver-Infant Dyad as a Buffer or Transducer of Resource Enhancing or Depleting Factors that Shape Psychobiological. Development Journal: Australian and New Zealand Journal of Family Therapy. DOI: 10.1002 [Google Scholar]
  44. Tronick E, Als H, Adamson L, Wise S, & Brazelton TB (1978). The infant’s response to entrapment between contradictory messages in face-to-face interaction. Journal of the American Academy of Child psychiatry, 17(1), 1–13. [DOI] [PubMed] [Google Scholar]
  45. Tronick E, & Beeghly M (2011). Meaning making and infant mental health. American Psychologist, 107–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Weinberg MK, & Tronick EZ (1998). The impact of maternal psychiatric illness on infant development. The Journal of clinical psychiatry. [PubMed] [Google Scholar]
  47. Welch MG (2016). Calming cycle theory: the role of visceral/autonomic learning in early mother and infant/child behaviour and development. Acta Paediatrica, 105(11), 1266–1274. [DOI] [PubMed] [Google Scholar]
  48. Zbozinek TD, Rose RD, Wolitzky‐Taylor KB, Sherbourne C, Sullivan G, Stein MB, … & Craske MG. (2012). Diagnostic overlap of generalized anxiety disorder and major depressive disorder in a primary care sample. Depression and anxiety, 29(12), 1065–1071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zeanah CH, & Gleason MM (2010). Reactive attachment disorder: A review for DSM-V. Report presented to the American Psychiatric Association. [Google Scholar]

RESOURCES