Early Behavioral and Physiological Predictors of Autism in 12-month-old Siblings of Children with Autism

Author

Nicolas Poupore, B.S., Kayla Smith, B.S., Abigail L. Hogan, Ph.D., and Jane Roberts, Ph.D.

University of South Carolina

Abstract

Autism spectrum disorder (ASD) is a neurodevelopmental disorder defined by social-communicative deficits and repetitive behaviors. Complementing behavioral (i.e., The Autism Observation Scale for Infants (AOSI) scores) and physiological markers (i.e., respiratory sinus arrhythmia (RSA), heart rate (HR)) may improve early identification of ASD and provide insights into the neurobiological mechanisms contributing to early ASD symptoms. Younger siblings of a child with ASD (ASIBs) had significantly higher AOSI Total Scores and AOSI Marker Scores in comparison to Low Risk Controls (LRCs). RSA significantly decreased from a Baseline to AOSI for both groups, but only for the ASIB group was a negative correlation observed between RSA suppression and AOSI Marker Scores. Furthermore, the ASIB group was divided based on 24-month ASD outcomes (ASIB-ASD and ASIB-NonASD). At 12 months of age, these three groups differed significantly in both AOSI Total Score and AOSI Marker Score. Although a stepwise pattern emerged in regards to behavioral and physiological markers, no significant differences between groups were found. ASIBs have a different behavioral but not physiological profile than the LRCs. When the ASIB group was divided based on 24 month ASD outcomes, the ASIB-ASD exhibited the most divergent behavioral and physiological profile from the LRCs; however, the ASIB-NonASD also exhibited an a noticeably different profile from the LRCs.

Autism spectrum disorder (ASD) is a neurodevelopmental disorder defined by social communicative deficits and a pattern of repetitive interests and behaviors (American Psychiatric Association, 2013). The prevalence of ASD is high, impacting 1 in 68 children, with boys 4.5 times more likely to be diagnosed with ASD than girls (Christensen et al., 2016). Although the exact etiological mechanisms that cause ASD are unknown, studies suggest that both genetic and environmental factors increase the risk for ASD (Geschwind, 2011; Hallmayer et al., 2011).

While there is currently no cure for ASD, previous research has shown that early identification and intervention are crucial to improving the long-term outcomes of children diagnosed with ASD (Ben-Itzchak & Zachor, 2007; Dawson et al., 2010). Studies have shown that a diagnosis of ASD at two years of age is reliable and stable (Kleinman et al., 2008; Lord et al., 2006). However, the median age for a clinical diagnosis of ASD is almost four years of age (Christensen et al., 2016). As a result of this discrepancy between research and clinical practice, researchers have developed screening measures to detect young children at high risk for ASD. One such screening measure is the Autism Observation Scale for Infants (AOSI). The AOSI is a semi-structured play observation designed to identify early atypical behaviors related to ASD in children aged 6 to 18 months (Bryson et al., 2008). While the AOSI is successful at identifying early ASD markers in research settings, it appears to be less effective as a clinical screening measure, especially in highly verbal children (Bryson & Zwaigenbaum, 2014). Therefore, researchers are investigating other measures and screeners in conjunction with the AOSI to improve early identification.

There are several challenges associated with observational behavioral screening measures. First, the possibility of reduced objectivity for behaviors can impact sensitivity and specificity. Furthermore, due the short time constraint of the measure, rare and atypical behaviors that would aid the screening process may not be evoked. Another difficulty is that behavior is the most distal marker of a pathology or disorder. While atypical behaviors are currently the screening and diagnostic measure for ASD, potentially finding and using the physiological markers that influence these behavioral changes can increase accuracy and sensitivity in identifying children at higher risk for ASD.

Given that many early behaviors of ASD have been well-described, additional research has focused on atypical physiology that could be shaping these divergent behaviors. One development stems from the Polyvagal Theory that proposed respiratory sinus arrhythmia (RSA) as an important marker of physiological, emotional, cognitive, and behavioral regulation (Porges, 1986; Porges, Doussard-Roosevelt, Portales, & Greenspan, 1996). Further research on RSA has shown that it is an effective physiological biomarker for the function of the parasympathetic nervous system (PNS; Katona & Jih, 1975). Decreased baseline RSA in children has been theoretically linked to social-communicative problems (Neuhaus, Bernier, & Beauchaine, 2014). RSA suppression, the difference between RSA at a basal state and RSA at a reactive state, has also been studied. Decreased RSA suppression has been shown in some studies to be related to decreased physiological, emotional, and behavioral regulation (Porges et al., 1996). Another study has shown that poor RSA suppression is a significant predictor of an atypical developmental trajectory of stranger fear in infants (Brooker et al., 2013). These studies suggest that both static and reactive RSA can lend important information on the functioning of the PNS and may be influencing the atypical behaviors seen in infants with ASD.

The present study examined behavioral and physiological markers of atypical development to study their relationship in hopes of establishing a more effective screening measure for ASD. This study examined 12-month-old younger siblings of children with ASD (ASIBs) given that there is sufficient evidence indicating that there are noticeable behavioral changes just before 12 months in children later diagnosed with ASD. Examples of behavioral changes at 12 months include lack of babbling, showing objects to caregivers, gesturing, and following an adult’s pointed finger (Plauché Johnson, 2008; Watson & Crais, 2013; Zwaigenbaum et al., 2009). To our knowledge, the relationship between behavioral and physiological risk markers in 12-month-old ASIBs has not been studied.

This study included two primary objectives. First, we aimed to determine if ASIBs demonstrated atypical physiological regulation relative to low risk controls (LRCs) and if ASD behavioral risk markers were related to physiological regulation at 12 months of age. We hypothesized that the ASIBs would have atypical physiological control in both RSA and RSA suppression compared to the LRCs and that the behavioral ASD risk markers would be correlated to atypical physiological regulation. Second, we aimed to determine if physiological regulation differed between ASIBs later diagnosed with ASD compared to ASIBs without an ASD diagnosis and if there was a relationship between the ASD behavioral risk markers and physiological regulation in all three groups. We predicted that all three groups would differ on RSA and RSA suppression, with the ASIBs without an ASD diagnosis (ASIB-NonASD) showing an intermediate profile and the ASIBs diagnosed with ASD (ASIB-ASD) exhibiting the most divergent profile from the LRCs. These physiological profiles would parallel the behavioral risk markers between the three groups.

Method

Participants

For the first research question, 34 ASIBs and 33 LRCs were assessed at 12 months of age. All ASIBs were required to have an older sibling diagnosed with ASD. LRCs were required to be born full term, have no history or suspicion of developmental delay from parent report and research assessment results, and have no family history of ASD in first- or second-degree relatives. LRCs were also excluded from the study if they later received a diagnosis of ASD at 24 months. Descriptive statistics for participant demographics are shown in Table 1.

For the second research question, the ASIB group was divided into two groups based on ASD diagnostic outcomes at 24 months of age: 5 ASIBs-ASD and 24 ASIBs-NonASD. In order to ensure that our final sample only included children with diagnostic outcomes (ASD or non-ASD), we excluded children who had not completed their 24-month assessment (five ASIBs and seven LRCs). The 26 remaining LRCs were included for analysis. See Table 2 for the participant demographics.

Procedures

Participants were recruited as part of a larger longitudinal study at the University of South Carolina – Columbia (USC) that focused on the early emergence of ASD in high-risk infants. Infants were assessed at 12 and 24 months at either the USC laboratory or at their home, based on the needs of the family. The measures were administered by at least two trained examiners in a standard order as part of a larger standardized protocol assessing temperament and development over the course of one session. If possible, the assessments were administered at similar times of the day (mid-morning) to account for the effects of circadian rhythms on physiology and behavior. Parents of the infants were recruited using flyers, list-serves, and word of mouth. The families were compensated for their time and travel expenses. Parents provided informed written consent prior to data collection.

Measures

Cognitive Level. The Mullen Scales of Early Learning (MSEL; Mullen, 1995) was administered as an index of general cognitive abilities. The MSEL is designed for children 0 to 68 months of age and includes five subscales: Gross Motor, Visual Reception, Fine Motor, Receptive Language, and Expressive Language. An Early Learning Composite (ELC) score was derived from all subscales except Gross Motor and was calculated with a mean of 100 and a standard deviation of 15. Internal reliability was .75 to .83 for subscales and .91 for the ELC (Mullen, 1995).

Autism Observation Scale for Infants. At the 12-month assessment, the Autism Observation Scale for Infants (AOSI; Bryson, Zwaigenbaum, McDermott, Rombough, & Brian, 2008) was used to identify and measure early ASD behaviors. The AOSI is a semi-structured play observation designed to detect risk markers of ASD in infants between the ages of 6 and 18 months. The Total Score, the sum of all codes, ranges from 0-50 across 16 items and the Marker Score, the number of items endorsed (i.e., total number of items scored > 0), ranges from 0-16. A Total Score of 9 or higher and a Marker Score of 7 or higher are indicators of elevated ASD risk (Bryson et al., 2008; Zwaigenbaum et al., 2005). The AOSI has test-retest reliability of .61 for Total Score and .68 for Marker Score (Bryson et al., 2008) and strong sensitivity (84%) and specificity (98%) for ASD at 12 months (Zwaigenbaum et al., 2005). For the larger longitudinal study, item-level inter-rater agreement (81%) was checked for 20% of administrations.

Clinical Best Estimate Diagnosis. At 24 months of age, the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) was administered and scored by trained project staff. Adapted from standard procedures (Lord et al., 2006, 2012), a clinical best estimate (CBE) diagnosis of ASD was determined based on a review of the ADOS-2 and other clinical data (e.g., cognitive, adaptive behavior) collected at the 24-month visit. These diagnoses were determined by a multidisciplinary team that included a licensed psychologist and a licensed speech-language pathologist, both of whom were research reliable on the ADOS-2.

Heart Activity. Heart activity data were collected during a baseline condition and during the AOSI. During Baseline, participants watched a Baby Einstein video for three minutes. Electrocardiogram data were collected with an Alive Wireless Heart Monitor (Alive Technologies, Copyright 2005–2009) at a 300 Hz sampling rate. Data were edited to correct for false heart periods and artifacts with CardioEdit software (Brain-Body Center, University of Illinois at Chicago) by trained research assistants. Estimates for RSA and heart rate (HR; beats per minute) were then extracted. To compute RSA, sequential heart periods were sampled at 250 ms epochs and data were detrended with a 21-point moving polynomial algorithm (Porges & Bohrer, 1990). Data were then bandpass filtered to extract variance associated with spontaneous breathing parameters (0.24–1.04 Hz) and the variance was transformed to its natural logarithm to yield an estimate of RSA. The average RSA and HR for each condition, averaged over 30s periods, were used in analyses. RSA suppression was computed by subtracting AOSI RSA from Baseline RSA.

Statistical Analysis Plan

To investigate whether ASIBs exhibited atypical behavioral and physiological regulation, ASIBs and LRCs were compared using independent samples t-tests on the following variables: AOSI Total Score, AOSI Marker Score, Baseline RSA, Baseline HR, AOSI RSA, AOSI HR, and RSA suppression. Paired samples t-tests were used to examine whether RSA varied between the Baseline and AOSI conditions for both the ASIB and LRC groups. Pearson correlations were conducted to examine relationships between AOSI scores and physiological variables.

To determine if ASIBs exhibited atypical physiological regulation once divided by their ASD outcome, ASIBs-ASD, ASIBs-NonASD, and LRCs were compared using one-way analyses of variance (ANOVAs) for the variables described above. Paired samples t-tests were conducted to examine the differences in RSA during the Baseline and AOSI conditions. Pearson correlations were then used to evaluate the relationship between AOSI scores and physiological variables.

Results

Independent-samples t-tests were conducted to compare AOSI Total Scores and AOSI Marker Scores in LRCs and ASIBs. ASIBs had a significantly higher AOSI Marker Score in comparison to LRCs, t(65) = 2.29, p = 0.03, but only a marginally higher AOSI Total Score, t(65) = 1.70, p = 0.09. Groups did not differ on any physiological variables (see Table 3 for details).

To determine whether infants’ RSA values significantly differed between Baseline and AOSI contexts, paired-samples t-tests were conducted for each group. For the ASIB group, RSA decreased by an average of 0.39 (SD = 0.64) from Baseline to AOSI, t(14) = 2.33, p = 0.04. For the LRC group, RSA decreased by 0.37 (SD = 0.67), t(21) = 2.57, p = 0.03. Figure 1 depicts the suppression of RSA from Baseline to AOSI for both groups.

Pearson correlation coefficients were computed to assess the relationship between the behavioral and physiological markers. In the ASIB group, there was a significant correlation between AOSI Marker Score and RSA suppression, r = -.52, n = 15, p = 0.02. Although not significant, the relationship between several physiological and behavioral markers trended towards significance. In the LRC group, there was a negative correlation between AOSI Total Score and AOSI RSA, r = -.37, n = 23, p = 0.09, as well as a negative correlation between AOSI Marker Score and AOSI RSA, r = -.36, n = 23, p = .10. An opposite relationship was seen in the ASIB group, where there was a positive correlation between AOSI Total Score and AOSI RSA, r = .44, n = 15, p = .10. Also in the ASIB group, a negative correlation was observed between AOSI Total Score and RSA suppression, r = -.46, n = 15, p = .09. There were no other significant correlations between the behavioral markers and the physiological markers (see Table 4 for details).

The ASIBs were then separated by ASD outcome and one-way ANOVAs were used to investigate group differences in AOSI scores and physiological regulation. For AOSI Total Score, a significant effect of group emerged, F(2, 52) = 5.74, p = 0.009. Bonferroni-corrected post-hoc comparisons indicated that the mean AOSI Total Score for the LRC group (M = 4.54, SD = 3.80) was significantly lower than that of the ASIB-ASD group (M = 10.60, SD = 3.98). The ASIB-NonASD group (M = 6.58, SD = 3.87) did not significantly differ from the LRC and ASIB-ASD groups. Furthermore, a one-way ANOVA was conducted to compare group effect on AOSI Marker Score, revealing a significant group effect on AOSI Marker Scores, F(2, 52) = 10.55, p < 0.001. Post-hoc comparisons indicated that the mean score for the LRCs (M = 2.96, SD = 2.12) was significantly lower than that of the ASIBs-NonASD (M = 4.38, SD = 1.81) and the ASIB-ASDs (M = 7.20, SD = 1.92). Results also indicated that the mean score for the ASIBs-NonASD was significantly lower than that of the ASIBs-ASD. Additionally, a one-way ANOVA was conducted to compare group effect on AOSI RSA. A group effect on AOSI RSA emerged, F(2, 30) = 3.30, p = .05, but was not significant. The groups did not differ on any other physiological variables (see Table 5 for details).

Next, paired-samples t-tests were conducted to investigate whether RSA values varied across Baseline and AOSI contexts for each of the three groups. For the ASIB-ASD group, RSA decreased by an average of 0.00 (SD = 0.31) from Baseline to AOSI, t(4) = 0.00, p = 1.00. For the ASIB-NonASD group, RSA decreased by an average of 0.54 (SD = 0.72) from Baseline to AOSI, t(8) = 2.27, p = .05. For the LRC group, RSA differed by 0.39 (SD = 0.65), t(18) = 2.62, p = 0.04.

Pearson correlation coefficients were computed to assess the relationship between the behavioral markers and the physiological markers for the three groups. No significant correlations were revealed (see Table 6 for details).

Discussion

ASD is a highly prevalent disorder, although the neurobiological mechanisms are still not fully understood. While screening measures for high-risk children have greatly improved, most are not sensitive enough to be used clinically. Therefore, examining behavioral and physiological markers together will help to identify biomarkers which can be used to screen and diagnosis children earlier in development as well as increase the understanding of the mechanisms of ASD which can lead to targeted treatments.

Both behavioral, measured by the AOSI, and physiological, measured by HR and RSA, markers were used to investigate whether significant differences could be detected at 12 months of age in ASIBs. While the current study is consistent with ongoing research that ASIBs exhibit more atypical behaviors compared to LRCs, the most important findings were noted when the ASIB group was divided based on 24-month outcome. In doing so, a stepwise pattern emerged in both behavior and physiology. The ASIB-NonASD showed an intermediate profile in AOSI scores and RSA suppression and the ASIB-ASD showed the most divergent profiles in both areas. However, significant correlations between behavioral and physiological markers did not emerge in the three groups. Regardless of 24-month outcomes, a significant negative correlation between AOSI Marker Score and RSA suppression was observed in ASIBs.

This study provides evidence that suggests that there are noticeable and quantifiable markers at 12 months of age in children later diagnosed with ASD at 24 months. The ability to identify children at 12 months would allow for therapies to be implemented earlier in development, which could drastically improve outcomes for these children. Given that the majority of neurodevelopment happens by three years of age (Gilmore et al., 2007; Nowakowski, 2006), implementing different evidence-based therapies early in development could greatly improve their social and communicative skills. More importantly, this study adds further evidence that there are subclinical differences in ASIBs compared to LRCs. Behavioral therapies may also be beneficial for ASIBs without a developmental disorder diagnosis.

This study also suggests that physiological abnormalities may be present by one year of age in the ASIB group. The atypicality of RSA suppression in both the ASIB-NonASD and ASIB-ASD groups may stem from the development or functioning of the PNS. This atypical reactivity and control of the PNS to a social stressor situation may underlie the behavioral differences seen in both groups of ASIBs. This relationship is supported by the significant negative correlation results in the ASIB group between RSA suppression and AOSI Marker Scores. While there is evidence of an interaction between atypical RSA suppression and AOSI Marker Scores, more research is needed to define the exact nature of the relationship.

Two interesting trends were observed in the correlation data. Although nonsignificant, a positive correlation was observed between AOSI RSA and AOSI Total Score in the ASIB group. This trend is inconsistent with the hypothesized trend and contrasted with the negative correlation observed in the LRC group. Previous research indicates that lower RSA can lead to emotional, social, and communicative problems; therefore, future research should investigate whether RSA has a different effect in the ASIBs compared to LRCs. The other area of interest was that the three groups did not reveal any significant correlations between behavioral and physiological variables. The issue seems to be one of power as separating the ASIB group resulted in decreased sample sizes. Repeating this study with more participants could help answer this question.

There are a few limitations to the study. First is the small sample size of ASIB-ASD group. A larger sample size would allow for further conclusions to be drawn from this group. Another limitation is the absence of girls in the ASIB-ASD group. Although there are a higher proportion of boys with ASD than girls, including girls diagnosed with ASD would allow for a more representative sample. An additional concern is the use of the AOSI as a social or reactive stressor. While the AOSI is commonly viewed as a social, children with ASD may fail to properly recognize it as a social situation and as a result, may not be repressing their RSA. On the contrary, children with ASD may recognize the social stressor but may not be able to properly respond to it, potentially leading to atypical behaviors. Parsing out these differences could lead to a greater understanding of ASD as a social-communicative disorder.

Future research regarding ASD should focus on two areas: screening and understanding the physiological mechanisms of ASD. Research regarding screening for ASD should focus on evaluating whether studying behavior and RSA suppression in conjunction could be used clinically to identify children as high-risk earlier in development. A chief concern of future research includes investigating whether measuring RSA suppression would increase the sensitivity of the AOSI. To further understand the neurological mechanisms of ASD, the interaction of RSA suppression and behavioral markers should be studied to determine if this lack of RSA suppression is related to some or all of the atypical behavioral markers. On the other hand, since ASD has been shown to be mainly a polygenetic disease, characterization of mutations in the genome related to the development of the PNS may aid in defining the pathway from genetics to the behavioral phenotype.

Overall, this study suggests that noticeable step-wise physiological and behavioral profiles emerge between ASIBs-ASD, ASIBs-NonASD, and LRCs. These patterns can be used to improve screening and diagnostic measures and to increase the understanding of the neurobiological mechanisms of ASD in infants. This study helps fill the gap of research regarding how physiological and behavioral markers are present and interact in 12-month-old ASIBs.

About the Author

Nicolas Poupore

My name is Nicolas Poupore and I am from Greenville, South Carolina. I graduated with Honors in May 2017 with a Bachelor of Science in Biological Sciences. Currently, I am a second-year medical student at the University of South Carolina Greenville School of Medicine. Before medical school, I was working as a Research Associate in the Neurodevelopmental Disorders Lab. My main motivation for this project was to bring both behavioral and physiological markers of Autism Spectrum Disorder (ASD) together to both increase the understanding of the neurobiological mechanisms of ASD and improve the diagnostic measures so that interventions may be implemented earlier in life. Not only has my understanding of ASD increased, but I also have greatly increased my problem-solving and critical-thinking skills. I have learned how to ask the right questions and how to think about these problems that have never been answered before. I have also increased my ability to present scientific information clearly, concisely, and tailored for the audience. These skills will be vital in my career as a physician both in the clinic and in the lab. I would like to extend a special thanks to both of my mentors, Dr. Jane Roberts and Dr. Abigail L. Hogan for guiding me on this project and teaching me so much about ASD, neurodevelopment, and research in general. I would like to also thank everyone in the NDD Lab, for the entire lab was needed to gather, process, and analyze the data for this project. Lastly, I would like to thank the Magellan Travel Grant and South Carolina Honors College for partially supporting me financially to present this project on a poster at the Southeastern Psychological Association and National Conference on Undergraduate Research. I also gave an oral presentation at Discover USC that won first place for my section.

References

American Psychiatric Association. (2013). DSM 5. American Psychiatric Association.

Ben-Itzchak, E., & Zachor, D. A. (2007). The effects of intellectual functioning and autism severity on outcome of early behavioral intervention for children with autism. Research in Developmental Disabilities, 28(3), 287–303. https://doi.org/10.1016/j.ridd.2006.03.002

Brooker, R. J., Buss, K. A., Lemery-Chalfant, K., Aksan, N., Davidson, R. J., & Goldsmith, H. H. (2013). The development of stranger fear in infancy and toddlerhood: normative development, individual differences, antecedents, and outcomes. Developmental Science, 16(6), n/a-n/a. https://doi.org/10.1111/desc.12058

Bryson, S. E., & Zwaigenbaum, L. (2014). Autism Observation Scale for Infants. In Comprehensive Guide to Autism (pp. 299–310). New York, NY: Springer New York. https://doi.org/10.1007/978-1-4614-4788-7_12

Bryson, S. E., Zwaigenbaum, L., McDermott, C., Rombough, V., & Brian, J. (2008). The autism observation scale for infants: Scale development and reliability data. Journal of Autism and Developmental Disorders, 38(4), 731–738. https://doi.org/10.1007/s10803-007-0440-y

Christensen, D. L., Baio, J., Braun, K. V. N., Bilder, D., Charles, J., Constantino, J. N., … Yeargin-Allsopp, M. (2016). Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years — Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2012. MMWR. Surveillance Summaries, 65(3), 1–23. https://doi.org/10.15585/mmwr.ss6503a1

Dawson, G., Rogers, S., Munson, J., Smith, M., Winter, J., Greenson, J., … Varley, J. (2010). Randomized, Controlled Trial of an Intervention for Toddlers With Autism: The Early Start Denver Model. Pediatrics, 125(1).

Geschwind, D. H. (2011). Genetics of autism spectrum disorders. Trends in Cognitive Sciences, 15, 409–416. https://doi.org/10.1016/j.tics.2011.07.003

Gilmore, J. H., Lin, W., Prastawa, M. W., Looney, C. B., Vetsa, Y. S. K., Knickmeyer, R. C., … Gerig, G. (2007). Regional gray matter growth, sexual dimorphism, and cerebral asymmetry in the neonatal brain. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 27(6), 1255–60. https://doi.org/10.1523/JNEUROSCI.3339-06.2007

Hall, S. S., Lightbody, A. A., Huffman, L. C., Lazzeroni, L. C., & Reiss, A. L. (2009). Physiological Correlates of Social Avoidance Behavior in Children and Adolescents With Fragile X Syndrome. Journal of the American Academy of Child & Adolescent Psychiatry, 48(3), 320–329. https://doi.org/10.1097/CHI.0b013e318195bd15

Hallmayer, J., Cleveland, S., Torres, A., Phillips, J., Cohen, B., Torigoe, T., … Risch, N. (2011). Genetic Heritability and Shared Environmental Factors Among Twin Pairs With Autism. Archives of General Psychiatry, 68(11), 1095. https://doi.org/10.1001/archgenpsychiatry.2011.76

Katona, P. G., & Jih, F. (1975). Respiratory sinus arrhythmia: noninvasive measure of parasympathetic cardiac control. JOURNALOF APPLIED PHYSIOLOGY, 39(5).

Kleinman, J. M., Ventola, P. E., Pandey, J., Verbalis, A. D., Barton, M., Hodgson, S., … Fein, D. (2008). Diagnostic Stability in Very Young Children with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 38(4), 606–615. https://doi.org/10.1007/s10803-007-0427-8

Lord, C., Petkova, E., Hus, V., Gan, W., Lu, F., Martin, D. M., … Risi, S. (2012). A Multisite Study of the Clinical Diagnosis of Different Autism Spectrum Disorders. Archives of General Psychiatry, 69(3), 306. https://doi.org/10.1001/archgenpsychiatry.2011.148

Lord, C., Risi, S., DiLavore, P. S., Shulman, C., Thurm, A., & Pickles, A. (2006). Autism From 2 to 9 Years of Age. Archives of General Psychiatry, 63(6), 694. https://doi.org/10.1001/archpsyc.63.6.694

Neuhaus, E., Bernier, R., & Beauchaine, T. P. (2014). Brief Report: Social Skills, Internalizing and Externalizing Symptoms, and Respiratory Sinus Arrhythmia in Autism. Journal of Autism and Developmental Disorders, 44(3), 730–737. https://doi.org/10.1007/s10803-013-1923-7

Nowakowski, R. S. (2006). Stable neuron numbers from cradle to grave. Proceedings of the National Academy of Sciences of the United States of America, 103(33), 12219–20. https://doi.org/10.1073/pnas.0605605103

Ozonoff, S., Young, G. S., Carter, A., Messinger, D., Yirmiya, N., Zwaigenbaum, L., … Stone, W. L. (2011). Recurrence Risk for Autism Spectrum Disorders: A Baby Siblings Research Consortium Study. PEDIATRICS, 128(3), e488-95. https://doi.org/10.1542/peds.2010-2825

Plauché Johnson, C. (2008). Recognition of Autism Before Age 2 Years. Pediatr. Rev. Boulevard, 29, 86–96. https://doi.org/10.1542/pir.29-3-86

Porges, S. W. (1986). Respiratory Sinus Arrhythmia: Physiological Basis, Quantitative Methods, and Clinical Implications. In Cardiorespiratory and Cardiosomatic Psychophysiology (pp. 101–115). Boston, MA: Springer US. https://doi.org/10.1007/978-1-4757-0360-3_7

Porges, S. W. (2001). The polyvagal theory: phylogenetic substrates of a social nervous system. International Journal of Psychophysiology, 42(2), 123–146. https://doi.org/10.1016/S0167-8760(01)00162-3

Porges, S. W., & Bohrer, R. E. (1990). Analyses of periodic processes in psychophysiological research. In J. T. Cacioppo & L. G. Tassinary (Eds.), Principles of psychophysiology: Physical, social, and inferential elements (pp. 708–753). New York: Cambridge University Press.

Porges, S. W., Doussard-Roosevelt, J. A., Portales, A. L., & Greenspan, S. I. (1996). Infant regulation of the vagal "brake" predicts child behavior problems: A psychobiological model of social behavior. Developmental Psychobiology, 29(8), 697–712. https://doi.org/10.1002/(SICI)1098-2302(199612)29:8<697::AID-DEV5>3.0.CO;2-O

Roberts, J. E., Boccia, M. L., Bailey, D. B., Hatton, D. D., & Skinner, M. (2001). Cardiovascular indices of physiological arousal in boys with fragile X syndrome. Developmental Psychobiology, 39(2), 107–123. https://doi.org/10.1002/dev.1035

Roberts, J. E., Miranda, M., Boccia, M., Janes, H., Tonnsen, B. L., & Hatton, D. D. (2011). Treatment effects of stimulant medication in young boys with fragile X syndrome. Journal of Neurodevelopmental Disorders, 3(3), 175–184. https://doi.org/10.1007/s11689-011-9085-4

Watson, L. R., & Crais, E. R. (2013). Translating Between Research and Practice in Serving Infants at Risk for ASD. American Speech-Language-Hearing Association: Perspectives on Language Learning and Education, 20(1), 4–13. https://doi.org/10.1044/lle20.1.4

Zwaigenbaum, L., Bryson, S., Lord, C., Rogers, S., Carter, A., Carver, L., … Yirmiya, N. (2009). Clinical Assessment and Management of Toddlers With Suspected Autism Spectrum Disorder: Insights From Studies of High-Risk Infants. Pediatrics, 123, 1383–1391. https://doi.org/10.1542/peds.2008-1606

Zwaigenbaum, L., Bryson, S., Rogers, T., Roberts, W., Brian, J., & Szatmari, P. (2005). Behavioral manifestations of autism in the first year of life. International Journal of Developmental Neuroscience, 23(2), 143–152. https://doi.org/10.1016/j.ijdevneu.2004.05.001

Table 1. Participant Demographics

	ASIB (n = 34)	LRC (n = 33)
Age (months) – M (SD)	12.70 (.80)	12.38 (.60)
Gender – n (%) males	26 (76%)	26 (79%)
Mullen Early Learning Composite – M (SD)	98.33 (14.53)	100.25 (12.87)

Table 2. Participant Demographics

	ASIB-ASD (n = 5)	ASIB-NonASD (n = 24)	LRC (n = 26)
Age (months) – M (SD)	12.31 (.41)	12.73 (.78)	12.33 (.47)
Gender – n (%) males	5 (100%)	18 (75%)	20 (77%)
Mullen Early Learning Composite – M (SD)	97.6 (15.45)	98.30 (15.87)	101.6 (12.74)

Table 3. Behavioral and Physiological Differences between Groups

Variables	ASIBs M (SD)	LRCs M (SD)	Test Statistic	p-value
AOSI Total Score	6.56 (4.24)	4.85 (4.00)	t(65) = -1.70	.09
AOSI Marker Score	4.41 (2.27)	3.15 (2.22)	t(65) = -2.30	.03
Baseline RSA	4.48 (1.00)	4.49 (1.07)	t(48) = .02	.98
Baseline HR	124.41 (12.77)	126.50 (10.98)	t(48) = .62	.54
AOSI RSA	4.00 (.69)	4.20 (.65)	t(36) = .92	.37
AOSI HR	128.51 (8.67)	128.72 (11.25)	t(36) = .06	.95
RSA Suppression	.39 (.64)	.37 (.67)	t(35) = -.07	.94

Note. Groups did not have any significant differences on any of the physiological variables.

Table 4. Correlations between Behavioral and Physiological Markers

		Baseline HR	Baseline RSA	AOSI HR	AOSI RSA	RSA Suppression
ASIB	AOSI Total Score	.19	-.08	.04	.44	-.46
	AOSI Marker Score	.22	-.18	.01	.39	-.52*
LRC	AOSI Total Score	.00	-.04	.10	-.37	-.06
	AOSI Marker Score	-.02	-.08	.05	-.36	-.09

Note.  p < .10; * p < .05. Although not significant, in the LRC group, there was a negative correlation between AOSI Total Score and AOSI RSA, r = -.37, n = 23, p = .09, and a negative correlation between AOSI Marker Score and AOSI RSA, r = -.36, n = 23, p = .10. In the ASIB group, there was a non-significant positive correlation between AOSI Total Score and AOSI RSA, r = .44, n = 15, p = .10, a non-significant negative correlation was observed between AOSI Total Score and RSA suppression, r = -.46, n = 15, p = .09, and a negative correlation was observed between AOSI Marker Score and RSA suppression, r = -.52, n = 15, p = .02.

Table 5. Behavioral and Physiological Differences between Groups

Variables	LRCs M (SD)	ASIB-NonASDs M (SD)	ASIB-ASD M (SD)	Test Statistic	p-value
AOSI Total Score	4.54 (3.80)	6.58 (3.87)	10.60 (3.98)	F(2,52) = 5.74	.01
AOSI Marker Score	2.96 (2.13)	4.38 (1.81)	7.20 (1.92)	F(2,52) = 10.55	.00
Baseline RSA	4.61 (1.04)	4.27 (1.02)	4.55 (.50)	F(2,40) = .58	.56
Baseline HR	126.33 (11.24)	124.94 (13.19)	129.13 (11.81)	F(2,40) = .24	.79
AOSI RSA	4.28 (.67)	3.72 (.64)	4.55 (.54)	F(2,30) = 3.30	.05
AOSI HR	127.84 (11.78)	128.92 (8.91)	128.16 (12.14)	F(2,30) = .03	.97
RSA Suppression	.39 (.65)	.54 (.72)	.00 (.31)	F(2,30) = 1.19	.32

Note. Groups did not have any significant differences in any physiological measures.

Table 6. Correlations between Behavioral and Physiological Markers

		Baseline HR	Baseline RSA	AOSI HR	AOSI RSA	RSA Suppression
LRC	AOSI Total Score	.08	-.17	-.03	-.34	-.28
	AOSI Marker Score	.05	-.22	-.07	-.31	-.28
ASIB-NonASD	AOSI Total Score	-.03	.14	-.47	.47	-.26
	AOSI Marker Score	.03	-.05	-.22	.10	-.38
ASIB-ASD	AOSI Total Score	.16	-.56	.53	-.45	-.17
	AOSI Marker Score	-.08	-.62	.28	-.48	-.17

Note. ^p < .10; *p < .05. Groups did not have any significant correlations.

Figure 1. RSA Suppression from Baseline to AOSI

Note. For the ASIB group, RSA significantly decreased by an average of 0.39 (SD = 0.64), t(14) = 2.33, p = .04. For the LRC group, RSA significantly differed by 0.37 (SD = 0.67), t(21) = 2.57, p = .03.

Office of the Vice President for Research