We describe the methods for continuous monitoring of the autonomic nervous system under resting and challenge conditions with 18 month old children. Results revealed that this protocol can produce meaningful physiological responses in both branches of the autonomic nervous system and elicit significant individual variability in patterns of responses.
The autonomic nervous system (ANS) consists of two branches, the parasympathetic and sympathetic nervous systems, and controls the function of internal organs (e.g., heart rate, respiration, digestion) and responds to everyday and adverse experiences 1. ANS measures in children have been found to be related to behavior problems, emotion regulation, and health 2-7. Therefore, understanding the factors that affect ANS development during early childhood is important. Both branches of the ANS affect young children’s cardiovascular responses to stimuli and have been measured noninvasively, via external monitoring equipment, using valid and reliable measures of physiological change 8-11. However, there are few studies of very young children with simultaneous measures of the parasympathetic and sympathetic nervous systems, which limits understanding of the integrated functioning of the two systems. In addition, the majority of existing studies of young children report on infants’ resting ANS measures or their reactivity to commonly used mother-child interaction paradigms, and less is known about ANS reactivity to other challenging conditions. We present a study design and standardized protocol for a non-invasive and rapid assessment of cardiac autonomic control in 18 month old children. We describe methods for continuous monitoring of the parasympathetic and sympathetic branches of the ANS under resting and challenge conditions during a home or laboratory visit and provide descriptive findings from our sample of 140 ethnically diverse toddlers using validated equipment and scoring software. Results revealed that this protocol can produce a range of physiological responses to both resting and developmentally challenging conditions, as indicated by changes in heart rate and indices of parasympathetic and sympathetic activity. Individuals demonstrated variability in resting levels, responses to challenges, and challenge reactivity, which provides additional evidence that this protocol is useful for the examination of ANS individual differences for toddlers.
Individual differences in children’s autonomic nervous system (ANS) reactivity play an important role in the development and maintenance of physical and mental health problems 12-16. A growing body of evidence reveals that individual differences in ANS measures at rest and reactivity are predictive of variations in children’s internalizing and externalizing psychopathology 4,9,17-19 and physical health 7,20. In addition to these direct or main effects of ANS functioning, an accumulating body of studies have found that children’s reactivity can interact with contextual risk factors, such as adverse events or marital conflict, to moderate those effects on children’s wellbeing and health (see reference 21 for a review). Despite this growing evidence that children’s autonomic reactivity plays a role in health across the life course, and the accompanying interest in those processes, there is need for additional research examining the development of ANS functioning in children under three years of age.
The ANS consists of two branches, the parasympathetic and sympathetic nervous systems. It regulates the function of internal organs (e.g., heart rate, respiration, digestion, and sexual arousal), largely through unconscious mechanisms, in response to everyday and adverse experiences 1. The parasympathetic nervous system (PNS) is the ‘rest and restorative’ system that maintains a low resting heart rate and restorative state when sleeping or relaxing The sympathetic nervous system (SNS) is the ‘fight or flight’ system that responds to emergencies or threatening situations by accelerating one’s heart rate 23. Respiratory sinus arrhythmia (RSA) is a reliable index of the PNS influence on cardiac functioning, and pre-ejection period (PEP) is a reliable index of the SNS influence on cardiac functioning. Both RSA and PEP have been found to be valid measures through experimental pharmacological blockade in samples of adults 24,25. More recently, this work has been extended into child and adolescent samples, further establishing RSA and PEP as valid and reliable measures of ANS activity throughout development 8,26,27. Reliable impedance cardiography measurements (PEP) using band and spot electrodes have been used for considerable time in adult samples 28. More recently, parallel measurement and analytic techniques have also been shown to be reliable and valid in samples of children 11,29,30, although collection of PEP measures in young children is rare.
There is an extant large body of literature on PNS functioning in infants and young children, including vagal tone and respiratory sinus arrhythmia measures. There are fewer studies of the SNS using PEP, and very few simultaneously examining both the parasympathetic and sympathetic nervous systems, which is critical for building understanding of their highly integrated functioning. Of the few studies of young children’s PEP, most find that their protocols do not produce significant group level changes in PEP in response to challenges 8,31. Studies have demonstrated that there is stability in resting PNS measures starting in infancy and they have relations with temperament, behavior, and health,13,16,26,32,33 but there is a limited understanding of SNS responding and its stability over time, relations to development, and the factors that shape that its developmental trajectory. PNS responses have been attributed primarily to social engagement 16 and SNS responses proposed to reflect “fight or flight responses” 1 as well as reward sensitivity (see reference 34 for a review). In young children, it is challenging to simulate specific challenges in a laboratory and to engage them due to their short attention span. There are also measurement challenges for toddlers who may experience discomfort with the seven spot electrodes needed to measure PNS and SNS simultaneously. In addition, within the fields of developmental science, there few standards about how to elicit “stress reactivity” sufficiently while also respecting children’s developmental needs and the ethics in working with children. This article presents information on the administration and scoring of a standardized protocol comprised of resting and challenging conditions designed to elicit parasympathetic and sympathetic nervous system responses from 18 month old children.
ANS reactivity is typically conceptualized as the physiological response to a discrete external stimulus relative to a comparison or resting state, which varies across individual organisms 22. Examination of the early life etiology of mental and physical health problems has led scientists to be particularly interested in understanding children’s reactivity to situations that evoke adaptive responding or can be thought of as stressful challenges. The preponderance of literature refers to this phenomenon as “stress reactivity”, although encountering challenging stimuli has the potential to elicit responses across a broad range of domains. Thus, the protocol described below was designed to elicit ANS responses across multiple domains including two forms of rest (while listening to a soothing lullaby and while watching a calm, neutral video) and three developmentally-appropriate challenges for 18-month olds: anticipation/startle (jack-in-the-box), sensory (lemon juice), and socioemotional (listening to a sick infant cry). The protocol was adapted from our existing protocols for 12 month olds 8 and 3-5 year olds 11, in order to make it developmentally-challenging, engaging and tolerable for toddlers (18-21 months).
Here we present the Developmental Challenges Protocol (DCP) and ANS data from the 18 month old study visit of the Stress, Eating and Early Development (SEED) Study of pregnant women and their offspring. Maternal participants with healthy pregnancies were recruited during 2013-14 for a study of gestational weight gain and prenatal stress and were overweight, low income, and racially and ethnically diverse. Parents’ informed consent for the study of her offspring was obtained just after birth and again prior to the start of the data collection for the study reported here, when the offspring were 18-21 months of age during 2014-2015.
This study was approved by the Committee on Human Research of the University of California, San Francisco.
1. Pre-protocol Set-up (See Materials Spreadsheet for Complete List of Equipment)
Figure 1. BioLab Software Configuration settings. This figure displays a screenshot of the configuration settings to be used during ANS data collection as described in step 1.2.
2. Preparing the Mother and Child for the Developmental Challenges Protocol
3. Connecting the ANS Acquisition Equipment to the Child
Figure 2. Electrode Configuration for 18 month olds. Note: the 2 ECG electrodes are applied in a Lead II configuration. Also, the alternative electrode placement for the nape of the neck is shown here.
4. Administering the Tasks in the Developmental Challenges Protocol
5. Marking Protocol Segments During the Developmental Challenges Administration (CO)
Marker Key | Event | Duration after Start |
F1 | Lullaby 1 start | 60 sec |
F2 | Jack in the box start | 60 sec |
F3 | Lemon juice script start | At least 20 sec |
F4 | Lemon juice end (water drink offered) | At least 10 ses |
F5 | Infant cry start | 30 sec |
F6 | Lullaby 2 start | 60 sec |
F7 | Video start | 120 sec |
F10 | End of DCP (video end) | |
F11 | Serious irregular events |
Table 1: Marker-Keys for Protocol Event Marking During Challenge Tasks. Marker keys as described in step 5.1 are shown, used to indicate important protocol events and rare occurrences in the ANS data file.
6. Post Developmental Challenges Protocol
7. Respiratory Sinus Arrythmia Scoring
Figure 3. HRV Analysis Main Navigation Screen. Settings for scoring the toddler ANS data using the HRV Analysis scoring software are displayed. To score segments, drag events to the ‘Event Type” boxes.
8. Pre-ejection Period Scoring
NOTE: The Impedance (IMP) Analysis program allows one to clean the impedance data and obtain HR and PEP values.
Figure 4. IMP Analysis Impedance Calibration Settings Screen. Settings for scoring the toddler ANS data using the Impedance Analysis scoring software are displayed.
The SEED study enrolled 162 mother-child dyads (37% African American; 30% Latina; 16% white; 17% other or multiracial). For the 18 month visit, we completed the in-person assessment with 140 children (87% of the enrolled sample); 6 participants moved away, and the remaining participants were unreachable or unavailable for this visit. The refusal rate for ANS data collection component of the visit was 3/140 (2%) for mothers available for this visit. Three visits were conducted after 21 months (more than 3 months beyond the target age), resulting in 135 children with ANS data collected at the target time period. Poor signal quality prevented us from scoring two children’s ANS data. Therefore, ANS data were available for 133 children. The mean age of the children with ANS data presented here was 18.89 months (range 17.59-21.60; SD = .80) and 53% of the children are female. Mean, sex- and age-adjusted weight-for-length percentile was 68.40 (SD = 27.47).
Of the 133 target-aged children with scoreable ANS data, 100% tolerated the application of the electrodes and also began the developmental challenges protocol. The protocol was discontinued mid-way for three subjects and after the baby cry task for five subjects. Two subjects were not shown the video due to equipment problems. All available data from the 133 children were scored. Of these, three children had no scoreable PEP data due to noise/movement artifact. Four more children had no scorable first resting lullaby data for PEP. Seven other children had unscorable PEP data for one or more segments after the baseline lullaby.
Descriptive statistics for the HR, RSA, and PEP measures by task are presented in Table 2. Table 2 also presents mean changes between tasks calculated in two ways: challenge task reactivity calculations (relative to resting level during the first lullaby) and task-to-task change calculations (relative to preceding task) for each ANS measure. The mean changes that were significantly different from zero are noted. Figure 5 presents boxplots for the three domains of challenge reactivity for HR, RSA, and PEP. Although the mean reactivity values presented in Table 2 and Figure 5 are small, the vast majority are significantly different from zero, and the boxplots reveal consistent evidence across all tasks and all ANS indices that children demonstrated variability in their responses to the challenges. Figure 6 presents mean RSA and PEP responses to each of the tasks throughout the protocol, and indicates when they significantly changed from task-to-task. RSA significantly changed between most of the adjacent tasks, indicating that, on average, children’s parasympathetic nervous system responses changed in response to each new task. The number of significant mean changes between tasks for PEP were fewer than for RSA, although significant change was identified from the lullaby to both the first and second challenge tasks, and from the last lullaby to the neutral video, indicating that, on average, children’s SNS responses were different from resting conditions to subsequent tasks. Child age was not significantly correlated with any challenge reactivity scores. These patterns suggest that this series of tasks elicited sample-level variability in responding throughout the protocol and across both branches of the ANS, and that there were different patterns between the PNS and SNS branches for some tasks.
Figure 5. Boxplots Representing Heart Rate, Parasympathetic, and Sympathetic Activity by Developmental Challenge Task. (A) represents heart rate reactivity, (B) represents RSA reactivity, and (C) represents PEP reactivity. Note: the boxes represent the interquartile range containing the middle 50% of values, with the line across the box representing the median. The whiskers extend from the highest and lowest values, excluding outliers which lie at least 1.5 box lengths outside the box. Please click here to view a larger version of this figure.
Figure 6. Mean Levels of Sympathetic and Parasympathetic Activity Across the Developmental Challenge Protocol. Decreases in RSA (red line) and PEP (blue line) reflect mean-level autonomic reactivity via PNS withdrawal and SNS activation, respectively. * indicates that the sample average score for that ANS measure changed significantly between those adjacent tasks (p <.05).
Supplemental File 1: Script A, B, C Please click here to download this file.
Table 2: Descriptives for ANS Measures: Task-level Arousal and Two Difference Scores of ANS Responding. Table 2 presents descriptive statistics for HR, RSA, and PEP. The left column displays task specific statistics, the middle column displays results for challenge tasks relative to the resting measure (mean of the 2 lullabies), and the right column presents the change for each task from the previous task. Please click here to view a larger version of this figure.
This study revealed that, within a sample of 18 month-old children, standardized resting and challenge tasks designed to elicit responses from a range of domains (startle, sensory, social, resting) led to a range of ANS responses. The mean ANS responses found under these resting and challenge conditions were similar to ANS values reported by others who have measured ANS responsiveness with young children using similar tasks 8,11,27,29. The mean levels of the ANS measures found in our sample are between those typically found in young infants and those found in older children, which is in line with evidence that heart rate decreases and RSA and PEP increase with age during childhood 10. We also found there was a range of reactivity scores (the difference in physiological response from resting state to the challenge conditions) across the three physiological indices, with several statistically significant meaningful differences between many of the challenge and resting scores for both RSA and PEP. Collectively, these findings show that the protocol was successful at capturing ANS change during transitions between calming activities and developmentally challenging tasks and eliciting a measurable stress response within both the sympathetic and parasympathetic branches of the ANS in an understudied age group. In addition to the evidence for sample-level mean changes in ANS, individuals demonstrated significant variability in resting ANS levels, task ANS responses, and challenge reactivity scores, which provides further evidence that this protocol is useful for the examination of ANS individual differences in toddlers 35.
The majority of ANS studies with infants and toddlers include only RSA, the parasympathetic branch of the ANS, due to the difficulty in assessing sympathetic measures in this very young age group. Thus, little is known about the development of sympathetic branch of the ANS at rest or in response to challenge within this period of development or what factors shape its development and related trajectories of health. Additionally, we can learn more about the dynamic or synchronous or asynchronous relationship between the SNS and PNS by measuring them simultaneously, so dual measures are needed. Results from this protocol suggest that measurement of both SNS and PNS activity and change is possible in this age group.
Although reactivity protocols with adults typically involve segments of time with “rest” during which the adults sit or lie still for 5 min without stimuli, resting measures for young children are challenging. Engaging young children in some minimal manner is necessary to limit movement artifact in ANS measures of rest 29,36. In previous studies, “resting or baseline’’ in young children has been assessed as infants listened to a lullaby 8, or as preschool-aged children were read a relaxing story 11, shown a neutral movie clip 4 or a video screen with changing shapes 37. The consensus is that these may constitute reasonable reference points for assessing reactivity depending on the comparison condition. Previous experiments from our lab within a sample of 5 year olds 38 suggest that the psychomotor activity elicited by the movement-inhibiting baseline condition (e.g., talking, social engagement, gesturing) is an important consideration; findings suggested that the use of social engagement to keep children calm and focused, such as reading a story, is an appropriate baseline comparison for social challenges whereas neutral videos may be appropriate as comparison values for reactivity to arousing videos for children. Attending to a video is associated with increased attention and it has a calming response in both branches of the ANS. In this study, for toddlers, the neutral video did not elicit differences in HR, RSA or PEP measures, relative to the first resting lullaby, suggesting that either could have been used as resting values from which to calculate reactivity, or that its inclusion in the protocol is unnecessary. In addition, this protocol does not include rest periods between tasks, which may lead to “carryover” reactivity effects from task to task. This choice was made because of the difficulty in maintaining toddler’s attention for long periods. However, the significant task-to-task change scores, most commonly in opposing directions suggest that carryover effects may have been limited in this protocol.
The setting for the assessment data described here presents a possible limitation; roughly half of assessments were completed in the participant homes and the others were completed in our laboratory. This potential limitation is balanced by the flexibility of completing assessments in participant homes, which allows us to maintain data collection with participants who were unable or unwilling to travel to our lab. Further, although all children had a 5 min waiting period between electrode placement and the start of the Developmental Challenges Protocol, another potential limitation is that some children became upset by the placement of the electrodes and may have been stressed at the start of the protocol 8. Thus, we may not have obtained a valid resting measure at the start of the protocol. This protocol includes a second lullaby in order to offset this possibility, although its occurrence immediately following challenges may also lead to carryover effects. Limited ability to assess “true” resting conditions is a long-standing challenge for ANS researchers working with young children 38. More generally, assessing toddlers is challenging at times and requires occasional modification of procedures (such as stopping to change a diaper or allowing a child to hold their small soft toy to prevent a tantrum that would prohibit continuation of the assessment). In this protocol, we provided details on engaging and distracting young children to reduce the likelihood of need for these modifications. Conducting research with toddlers always requires flexibility and troubleshooting by developmentally-sensitive staff. Overall, it is important to develop a collaborative relationship with participating mothers.
In sum, the results of this study suggest that the Developmental Challenges protocol and ANS data collection procedures presented here may provide researchers with a set of structured challenges that elicit a broad range of responses across both branches of the ANS for children aged 18-21 months. As the children’s ages were not correlated with reactivity scores, these tasks may be appropriate with slightly older children, although this was not tested here; if attempted, ANS scoring settings will need to be adjusted. This protocol provides one option for studying stress physiology and its development early in life within a difficult-to-assess age group. The reactivity calculated for each domain can be examined separately as has been done with older children 6, or can be combined to create an “overall reactivity” score as has been done with infants 8, depending upon the associations among the reactivity variables in a given sample and the aims of the research. Results from the data collected during this protocol can be compared to similar age-appropriate tasks completed by younger 8 and older children 10,38. Assessing ANS functioning across different ages allows for the examination of developmental trajectories across time and the pursuit of related questions about both their ontogeny and their ability to predict mental and physical health outcomes.
The authors have nothing to disclose.
This research was supported by NIH 1 U01 HL097973, NHLBI 5 R01 HL116511-02, UCSF-CTSI Grant Number UL1 TR000004, the Robert Wood Johnson Health and Society Scholars Program, and the Lisa and John Pritzker Family Foundation. The authors also wish to acknowledge Michelle Stephens for her assistance with scoring the ANS data as well as Vanessa Tearnan, Marialma Gonzales-Cruz, Yurivia Cervantes, Amy Engler, Stephanie Grover and Karen Jones-Mason for their assistance in collecting the ANS data, and Michael Coccia for his help with the data. We are also thankful to the families for their generous participation in this research and to the volunteers who helped us to test and refine this protocol, including the first author’s two children. We want to acknowledge our mentors who taught us about measuring ANS and why measuring ANS responsivity is meaningful in young children: W. Thomas Boyce, Gary Bernston, and Dave Lozano.
ANS equipment | |||
Laptop and power cord | Dell | 60-0107-3.0 | Mindware BioLab software requires a PC with 2 USB ports |
ANS data collection software | Mindware | contact company | BioLab acquisition program |
ANS data acquisition unit and cords | Mindware | 50-3711-08 | BioNex 8-slot Chassis |
Snap leads | Mindware | 93-0498-00 | 2 white, 2 red, 2 brown, 1 black |
Spot electrodes | Kendall | 31112496 | 130 model; 7 for each assessment + extras-to be sealed in a ziplock bag at all times) |
ANS scoring/editing software | Mindware | contact company | HRV Analysis, version 3.1.0F and IMP Analysis, version 3.1.0I |
Protocol equipment | |||
tablet computer | Apple | iPad; for playing audio and video clips | |
portable speaker | Polariod | PBT510 | |
Jack in the Box | Schylling | SMJB | Sock Monkey theme |
Lemon juice | ReaLemon | 98077 | 0.14oz shelf-stable packets |
Pasture pipettes | Grainger | 6FAV9 | 1 ml size, disposable |
Portion cups | Dart Solo Con | 250PC | 2.5 oz, disposable, plastic |
audio files of baby cry | Alkon lab | 30 second clip recorded in a neonatal intensive care unit | |
audio file of baseline lullaby | Rockabye Baby Music | 60 seconds of "Across the Universe" on "Rockabye Baby Lullaby Renditions of the Beatles" CD | |
audio file of final lullaby | Rockabye Baby Music | 60 seconds of "Here Comes the Sun" on "Rockabye Baby Lullaby Renditions of the Beatles" CD | |
video file of neutral video | 2 minutes of Baby Einstein video | ||
Retractable measuring tape | Rollfix | TM04 | |
Miscellaneous materials | |||
Surge protector | to plug in all electronics | ||
2-prong adapter for surge protector | |||
Video camera | Sony | Handycam HDR-CX440 | |
Video Encoder, power and input cords | Mindware | 50-8614-01 | optional; to sync video and ANS data files |
Baby wipes and gauze | for cleaning and drying oily skin | ||
Toys | to distract children during assessment | ||
Visual aids | pictures of child with electrodes on, description of protocol to show participants | ||
Script of protocol for research assistants | |||
Charging cords for tablet and speaker | |||
Doll and reinforcer | for demonstration if necessary |