Cortisol Stress Response Patterns and Social Behaviors in Adolescent Boys with Fragile X Syndrome and Autism

Friday, May 16, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
S. McGrath1, J. Klusek2, E. Schworer3, J. Gunther4, L. Abbeduto5 and J. E. Roberts6, (1)School Psychology, University of South Carolina, Columbia, SC, (2)Department of Psychology, University of South Carolina, Columbia, SC, (3)University of South Carolina, Columbia, SC, (4)University of California Davis M.I.N.D. Institute, Sacramento, CA, (5)University of California Davis M.I.N.D. Institute, Sacremento, CA, (6)Psychology, University of South Carolina, Columbia, SC
Background: Fragile X syndrome (FXS) is the most common known single-gene cause of autism spectrum disorder (ASD) and is characterized by problems modulating social and emotional behaviors, particularly during stress. The hypothalamus-pituitary-adrenal axis (HPAA) measures stress response systems through secretion of cortisol (Foley & Kirshcbaum, 2010). Elevated and suppressed HPAA responses are correlated to shyness, social withdrawal, social anxiety and ASD in FXS (Roberts et al., 2009).  These behaviors are also prevalent in individuals with idiopathic ASD; however, direct between-group comparisons of physiological patterns are needed to assess similarities and differences in symptomology. 

Objectives:  This study compared the relationship between cortisol indices of stress and avoidance, gaze and behavioral problems in boys with idiopathic ASD and boys with FXS which may inform targets for interventions. It was hypothesized that elevated cortisol would relate to increased social/emotional problems in both groups, with FXS displaying stronger trends. 

Methods: This sample represents preliminary data from a larger ongoing study. Participants included boys with idiopathic ASD (n=7) and FXS (n=14). The FXS group was comparable (M=18.9, SD=2.4) in age to the ASD group (M=18.5, SD=2.4). The Social Approach Scale, a direct assessment measure, evaluated approach and avoidance behaviors on a 4-point Likert scale at pre-assessment and during the last hour of assessments. The Child Behavior Checklist, a parent-report instrument, assessed anxiety, withdrawn behaviors, externalizing problems, and overall behavioral problems. Cortisol was measured in two conditions: baseline and reactivity post-assessment. Change scores were computed as reactivity minus baseline. Cortisol was analyzed by radioimmoassay and values were natural-log transformed for normality.

Results: Greater cortisol change was significantly associated with decreased internalizing behavior scores (r=.81) and moderately related to decreased problem (r=.72), anxiety (r=.72) and externalizing behavioral scores (r=.69) in the ASD group. In contrast, no relationships were found between change scores and these behaviors in the FXS group. Baseline cortisol was significantly associated with increased internalizing (r=.76) and problem behavior scores (r=.73) in the ASD group.  No other relationships between cortisol and behaviors were observed within FXS and ASD.

Conclusions: Contrary to our hypothesis, the ASD group demonstrated stronger relationships between cortisol change scores and overall behavioral problems and internalizing and externalizing behaviors. This tentative finding suggests that individuals with ASD are modulating stress better as demonstrated by greater cortisol change predicting decreased problem behaviors. Greater cortisol change may be an adaptive response thus having a positive effect on social emotional behaviors in the ASD group. However, these data are initial findings from a subsample which will be expanded (n=60) by the time of the conference to permit more extensive analyses taking into account developmental characteristics that differ between the two disorders (e.g., IQ) and including analyses with a three group comparison: idiopathic ASD, FXS without ASD, and FXS+ASD.


Foley, P., & Kirschbaum, C. (2010). Human hypothalamus-pituitary- adrenal axis responses to acute psychosocial stress in laboratory settings. Neuroscience and     Biobehavioral Reviews, 35, 91-96.