Abstract
Proficient sensory processing affords adaptive responses required for daily functioning. Sensory processing differences are common across neurodevelopmental disabilities (NDDs), including autism and fragile X syndrome (FXS), and, in some cases, associated with challenging behaviors. Although sensory processing differences are common in both autism and FXS, little is known about possible variation as a function of genetic etiology, or autism classification. This study characterized sensory processing features and examined associations between these features and challenging behavior in 102 male participants—non-syndromic autism (n = 37), FXS-only (n = 15), FXS + ASD (n = 20), and neurotypical (NT) children (n = 30). Autistic children demonstrated the highest levels of sensory processing differences; however, nuanced group differences were also identified. The autistic groups (autism and FXS + ASD) were similar on hypo-responsivity and sensory seeking, and the FXS groups (FXS-only and FXS + ASD) were similar on hyper-responsivity and sensory seeking. Within-group associations between sensory differences and challenging behavior were relatively similar across groups. Findings have implications for potential etiological mechanisms of sensory processing differences and associated functional consequences. Implications for diagnostic specificity and intervention are discussed.
Lay Abstract
Sensory processing differences are common in neurodevelopmental disabilities such as autism and fragile X syndrome (FXS). Sensory differences include extreme sensitivity to sensory experiences (hyper-responsivity), decreased sensitivity to sensory experiences (hypo-responsivity), and seeking out sensory experiences to an unexpected degree (sensory seeking). While more is known about sensory differences and challenging behaviors in autism, less is known about these patterns in children with FXS. This study examined the profiles of sensory processing differences and patterns of association between types of sensory response and challenging behavior in males between 24 and 69 months of age with FXS with and without co-occurring autism compared with young autistic males. Participant’s parents filled out questionnaires about sensory responses and challenging behaviors. Results showed that sensory hyper-responsivity is likely part of the broader behavioral profile of FXS, whereas hypo-responsivity is more indicative of autism. In addition, results demonstrated comparable associations between types of atypical sensory processing and ratings of challenging behaviors across groups. These findings contribute to our understanding of the similarities and differences between autism and FXS and have important implications for targeted assessment and intervention practices.
Keywords
Introduction
Sensory processing, a primary function of the nervous system, involves the integration of input from the senses and proprioception (Ayres, 1972; Miller et al., 2007). Proficient sensory processing affords adaptive responses and environmental interactions that are necessary for daily functioning. Sensory processing differences, often characterized as hyper-responsivity—over-responsivity to sensory input, hypo-responsivity—lack of responsiveness to sensory input, and sensory seeking—seeking out stimuli to an excessive degree (Baranek, 1999), stem from difficulties with gating sensory input, challenges with proprioception, or difficulty discriminating across modalities of sensory input (Ayres & Robbins, 1979; Miller et al., 2007). Sensory processing differences can present across various modalities (e.g., auditory, tactile, visual), lead to challenges with adaptive functioning (Ayres & Robbins, 1979), and are linked to challenging behavior in some populations (Baker et al., 2008; Lane et al., 2010; Will et al., 2019). While a range of sensory processing reactivity is expected in neurotypical development (Baranek et al., 2014; He et al., 2023; Little et al., 2017), neurodevelopmental disabilities (NDDs), such as autism, are partially characterized by pervasive sensory processing differences that manifest as observable affective reactivity to sensory stimuli (Baranek, 2002; Ben-Sasson et al., 2007; He et al., 2023; Lane et al., 2012). The extent and nature of sensory processing difficulties in autism have been well established (Ashburner et al., 2008; Gourley et al., 2013; MacLennan et al., 2022; O’Donnell et al., 2012); however, less is known about patterns of sensory processing in other NDDs such as fragile X syndrome (FXS). Differentiating profiles of affective sensory reactivity and processing differences across NDDs can contribute to increased precision in distinguishing behavioral phenotypes within intellectual disability (ID) and elucidate potentially different mechanistic pathways to similar outcomes (Ausderau et al., 2016; He et al., 2023).
Sensory Processing in Non-Syndromic Autism
Sensory processing differences are common in autism, with prevalence estimates ranging from 69% to 95%, (Baranek et al., 2006; MacLennan et al., 2022; Schaaf & Nightlinger, 2007; Tomchek & Dunn, 2007) and are considered a core aspect of the autistic phenotype (DSM-5; American Psychiatric Association, 2013). Delineating sensory processing profiles in autism can identify points of phenotypic variation and help inform etiology, prognosis, assessment, and intervention (Ausderau et al., 2016; Baranek et al., 2006; Lane et al., 2014). Existing evidence suggests that autistic children display significantly higher rates of sensory-related reactivity and processing differences compared with children with other NDDs, particularly hypo-responsivity (Baranek et al., 2006; Ben-Sasson et al., 2008; Rogers et al., 2003; Watling et al., 2001). Although sensory hypo-responsivity is robustly associated with the core features of autism, patterns of hyper-responsivity and sensory seeking are also evident (Feldman et al., 2020; Sinclair et al., 2017). For example, a recent meta-analysis highlights hyper-responsivity as a particularly salient feature differentiating autism from other non-developmental disability clinical groups (Ben-Sasson et al., 2019). Sensory processing features are not mutually exclusive, and many children with autism display marked differences across multiple features, frequently in co-occurring hypo- and hyper-responsivity (Baker et al., 2008; Baranek et al., 2006; Hilton et al., 2007; Lane et al., 2010; Liss et al., 2006).
While sensory processing differences are common in autism regardless of cognitive ability, differences may impact autistic individuals with co-occurring ID to a greater extent. For example, Ausderau et al. (2016) found that 2- to 12-year-old autistic children with lower cognitive ability were significantly more likely to show combined hypo-responsivity and sensory seeking compared with autistic children who had higher cognitive ability. In contrast, other studies have concluded that cognitive ability or the presence of co-occurring ID has little impact on sensory processing in autism (Green et al., 2016; O’Donnell et al., 2012; Rogers et al., 2003). Thus, the role of cognitive ability and co-occurring ID in sensory processing in autism remains unclear.
Fragile X Syndrome
FXS, a rare X-linked condition caused by a mutation in the FMR1 gene, affects approximately 1:7,000 males and 1:11,000 females (Hunter et al., 2014). A notable degree of phenotypic overlap exists between FXS and autism (Abbeduto et al., 2014; Roberts et al., 2018; see also Hall et al., 2010) and estimates place rates of co-occurring autism spectrum disorder (ASD) in FXS between 20% and 75% (Abbeduto et al., 2019; Kaufmann et al., 2017; Klusek et al., 2016; Roberts et al., 2020). The co-occurrence of ASD in FXS unequivocally leads to compounded negative outcomes including lower cognitive and adaptive skills and lower quality of life relative to those with FXS without ASD (Abbeduto et al., 2014; Bailey et al., 2008; Roberts et al., 2020). Delineating the shared and unique phenotypic features within FXS versus FXS + ASD, as well as between FXS and non-syndromic autism, can help refine differential evaluation and diagnosis and identify optimal support strategies for mitigating compounded negative outcomes for individuals with FXS + ASD. Given their prominence and implications for behavioral outcomes, delineating sensory processing features may be particularly useful in achieving better phenotypic differentiation across non-syndromic autism, FXS, and FXS + ASD.
Sensory Processing in FXS
Sensory processing differences in FXS appear at rates similar to autism and higher than both neurotypical (NT) children and children with other NDDs (McIntosh et al., 1999; Rogers et al., 2003). Specifically, sensory differences affect an estimated 90% of children with FXS, emerge early in life, and remain prevalent into adulthood (Baranek et al., 2008; Hagerman & Hagerman, 2002). Sensory processing differences are well established across both caregiver reports (Baranek, 2002; Rogers et al., 2003) and direct observation measures (Baranek et al., 2008), in preschool-aged children with FXS, with hyper-responsivity of salience. In fact, sensory hyper-responsivity appears more common in FXS compared with other genetic conditions, including Angelman and Cornelia de Lange syndromes (Heald et al., 2020), suggesting that hyper-responsivity may be specific to the FXS phenotype.
Prior research has demonstrated a correspondence between the degree of sensory differences and degree of autistic feature presentation in children with FXS (Kolacz et al., 2018; Rogers et al., 2003), suggesting these differences are more prominent for children with FXS + ASD (Bailey et al., 1998). The extent to which the greater degree of ID observed in children with FXS + ASD affects the type and degree of sensory differences remains unclear, and findings are mixed (Baranek et al., 2005; Rogers et al., 2003). For example, some evidence suggests that lower cognitive ability exacerbates both sensory differences and autistic feature expression in FXS (Kolacz et al., 2018; Rogers et al., 2003), whereas other evidence indicates that links between sensory differences and autistic feature expression in FXS are independent of cognitive ability (Baranek et al., 2005). While sensory differences are clearly linked to autistic feature expression in FXS, variation in profiles of sensory processing and the associated function impact across non-syndromic autism, FXS, and FXS + ASD remains unclear.
Behavioral Outcomes Associated With Sensory Processing Differences
Children with clinically elevated sensory processing differences demonstrate a wide array of behavioral challenges, including internalizing and externalizing behaviors (Mangeot et al., 2001; Marcham & Tavassoli, 2024; Rossow et al., 2021, 2023). Globally elevated sensory differences are strongly associated with a variety of challenging behavior outcomes, including higher levels of behavioral and emotional problems for autistic children (Baker et al., 2008; Lane et al., 2010; Marcham & Tavassoli, 2024; McCormick et al., 2016). For example, Baker et al. (2008) found significant associations between sensory processing differences and parent-reported child anxiety, social relating, communication disturbances, self-absorption, and antisocial behaviors in a group of young autistic children. In addition, prior research shows a strong predictive association between global sensory dysregulation and the presence of general challenging behaviors in autistic children (Rossow et al., 2021), with over 50% of the variance in challenging behavior explained by sensory processing differences (Lane et al., 2010). These trends evidently persist regardless of etiology. Specifically, Gourley and colleagues (2013) found that 59% of children receiving outpatient mental health services for emotional and behavioral disturbances showed higher rates of sensory processing differences than neurotypical norms. Furthermore, increased sensory processing differences significantly correlated with elevated scores on the Externalizing, Internalizing, and Total Behavior Problems domains of the Child Behavior Checklist (CBCL). These findings suggest that elevated sensory processing differences and challenging behavior co-occur, and that the functional consequences of sensory processing differences extend beyond autism.
The role of specific domains of affective sensory reactivity and processing differences in the manifestation of discrete challenging behaviors (e.g., internalizing and externalizing behavior) is poorly understood. However, some evidence indicates that specific features of sensory processing differences—hyper-, hypo-, or seeking—differentially impact challenging behaviors for autistic children (Green & Ben-Sasson, 2010; Rossow et al., 2021, 2023). For example, increased sensory hyper-responsivity significantly relates to internalizing behavior, particularly anxiety (Ben-Sasson et al., 2008; Feldman et al., 2020; Green et al., 2012; MacLennan et al., 2021; Mazurek et al., 2013; Rossow et al., 2021), whereas hypo-responsivity is linked to globally challenging behavior (Ausderau et al., 2016; Baker et al., 2008) in autistic children. While these findings provide emerging evidence for differential impact of specific sensory processing features and challenging behavior in autism, these associations remain unexamined in FXS. Furthermore, the added impact of co-occurring ID on the relationship between sensory processing differences and behavior outcomes in autism remains unclear. Given the extensive phenotypic overlap between autism and FXS, and the evidence suggesting increased sensory processing differences in the FXS + ASD phenotype (Bailey et al., 1998), understanding the differential effects of sensory processing features on behavior for children with FXS versus FXS + ASD is critically important.
Present Study
The present study aimed to determine whether profiles of sensory processing differences vary as a function of ASD diagnostic status or genetic etiology (i.e., FXS) in young males with non-syndromic autism or FXS and evaluate profiles of sensory processing differences relative to NT children. We applied the “Optimal Engagement Band” conceptual model of sensory processing (Baranek et al., 2001), thereby focusing on broad features of sensory processing (i.e., hypo-, hyper-, and seeking) rather than specific modalities (e.g., auditory, visual). We also investigated within-group patterns of association between sensory processing features and challenging behavior to determine variation across groups. Considering the elevated prevalence of sensory processing differences and challenging behavior in children with autism or FXS, examining the association between sensory processing differences and challenging behaviors may shed light on the functional impact of these challenges in children with FXS with and without ASD, and potentially inform targeted assessment and intervention.
Method
Participants
Participants were drawn from a larger prospective longitudinal study focused on early development in FXS at the University of South Carolina. Participants included males with non-syndromic autism (n = 37), FXS (n = 35), and NT development (n = 30) between 24 and 69 months of age. Given known sex differences in the rate of occurrence and degree of symptom expression for both FXS and autism (Rinehart et al., 2011), females were excluded. The FXS group included children with co-occurring autism (FXS + ASD; n = 20) and without co-occurring autism (FXS-only; n = 15). Autistic participants in the non-syndromic autism group had no known genetic condition or family history of FXS, and diagnostic status was confirmed at study enrollment via diagnostic report and clinical best estimate (CBE) procedures. Genetic report verified positive FXS diagnosis (i.e., >200 CGG repeats on the FMR1 gene) and presence/absence of co-occurring autism spectrum disorder (ASD) diagnosis was determined via CBE (see below). Participants in the NT group had no known genetic conditions, no family history of autism or developmental delays, were born full-term with no medical complications, and did not meet diagnostic criteria for ASD as confirmed through CBE. Overall, most participants were White (81%) and Not Hispanic/Latino (96%). Table 1 depicts participant characteristics and descriptive statistics.
Participant Characteristics.
Note. M = mean, SD = standard deviation, Nonverbal Mental Age equivalent was derived from the Mullen Scales of Early Learning (MSEL); the Sensory Experiences Questionnaire, version 2.1 (SEQ 2.1) was used to measure Sensory Processing; the Child Behavior Checklist (CBCL 1.5-5) was used to measure Challenging Behavior.
Measures
Cognitive Ability
The Mullen Scales of Early Learning (MSEL; Mullen, 1995) is a standardized developmental assessment designed for children from birth through 68 months of age comprised of five subscales: Gross Motor, Fine Motor, Visual Reception, Receptive Language, and Expressive Language. Each subscale yields raw and standardized scores (M = 50, SD = 10), with a standardized Early Learning Composite (ELC) derived from all domains except Gross Motor, estimating overall developmental level. We derived a nonverbal mental age (NVMA) equivalent by averaging the Visual Reception and Fine Motor Age Equivalent scores.
Sensory Processing
The Sensory Experiences Questionnaire, version 2.1 (SEQ 2.1; Baranek, 1999) is a 43-item caregiver questionnaire for children with autism and other developmental disabilities ages 6 months through 6 years old. The SEQ examines sensory processing features across response patterns (hyper-responsiveness, hypo-responsiveness, and sensory seeking) and modalities (auditory, tactile, visual, vestibular/proprioceptive, and gustatory/olfactory). Features are rated on a 5-point scale of “almost never (1)” to “almost always (5).” Scores include a total raw score, and mean scores for response patterns, modalities, and (social/nonsocial) context. Mean scores for neurotypical norms for hyper-responsivity, hypo-responsivity, and sensory seeking are 1.72, 1.45, and 2.29 respectively, with higher mean scores indicating more sensory differences (Baranek, 1999). The overall internal consistency of the SEQ is 0.80 (Baranek et al., 2006) and the test–retest reliability is 0.92 (Little et al., 2011). We used mean response pattern (hyper-, hypo-, seeking) scores.
Challenging Behavior
The Child Behavior Checklist (CBCL/1.5–5; Achenbach & Rescorla, 2000), a widely used 99-item parent rating scale, assesses internalizing and externalizing symptoms in children ages 18 months through 5 years. Items are scored using a 3-point Likert-type scale of “not true (0),” “somewhat or sometimes true (1),” or “very true (2).” Internal consistency of the CBCL ranges from 0.72 to 0.91 and test–retest reliability ranges from 0.79 to 0.90 (Achenbach & Rescorla, 2001). The Internalizing domain includes somatic, anxious/depressed, withdrawn, and emotionally reactive behavior subdomains. The Externalizing domain includes aggression and attention problems subdomains. We used Internalizing and Externalizing raw scores in analyses.
ASD Diagnostic Status
The larger study included the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2; Lord et al., 2012), a semi-structured direct observation assessment of autistic features, and the Autism Diagnostic Interview–Revised (ADI-R; Rutter et al., 2003), a semi-structured caregiver interview assessing core features associated with autism, administered by highly trained research staff and included in CBE procedures (see below).
Procedures
Procedures were approved by the Institutional Review Board at the University of South Carolina and caregivers provided written consent prior to participation. Participants were recruited through local and national organizations, parent support groups, and the local community. Assessments were conducted in participants’ homes or at the University of South Carolina, based on family preference. For the larger study, trained research staff administered standardized developmental assessments, behavioral assessments of temperament, and gold-standard autism diagnostic measures. Caregivers completed survey questionnaires, including the SEQ and CBCL.
CBE
Two independently research reliable ADOS-2 examiners, and a licensed psychologist and certified ADOS-2 trainer with expertise in autism diagnostics completed full case reviews involving the evaluation of scores on gold-standard autism diagnostic measures (i.e., ADOS and ADI-R), review of (at minimum) 15 min of video from ADOS-2 administrations, and evaluation of developmental and functional abilities to determine DSM-5 ASD diagnostic status. For complex cases, the full ADOS-2 video was reviewed and all information discussed until the CBE team reached diagnostic consensus. Although we did not employ formal participatory research methods, our work was informed by ongoing, reciprocal relationships with families, who are routinely invited to share their experiences and perspectives on research priorities and gaps. In developing the aims and methods for this grant, we consulted informally with three parents of young children with FXS, who provided observations and recommendations regarding early indicators and later consequences of anxiety. These consultations helped shape the conceptualization of the research questions.
Data Analysis
Analyses were completed using R software (R Core Team, 2018). Descriptive statistics were computed for each group (see Table 1). Groups were not statistically similar on chronological age, so bivariate correlations tested whether age significantly associated with any primary variables. No significant correlations were identified, thereby excluding age as a potential covariate. Correlations between NVMA and sensory domain scores also tested NVMA as a potential covariate. Because NVMA was only significantly correlated with sensory seeking in the autistic (r = −.37, p = .024) and FXS + ASD groups (r = −.51, p = .21) and hypo-responsivity in the non-syndromic autism group (r = −.49, p = .002), regression models were tested with and without NVMA included and retained it in all final models.
A series of multiple regression models were estimated to test group differences in sensory response domains, controlling for NVMA. The NT group was coded as the reference group, and simple slopes recovered by switching reference groups and re-estimating each model. Bivariate correlations were tested to determine variation of within-group patterns of association between sensory features (i.e., hyper-, hypo-, and seeking) and challenging (i.e., internalizing, externalizing) behavior.
Results
Sensory Processing Profiles
Regression model results are shown in Table 2 and Figure 1.
Sensory Processing Profiles.
Note. Models are estimated with the neurotypical (NT) group as the reference group; NVMA = nonverbal mental age derived from the MSEL.
Sensory profiles.
Hyper-Responsivity
Group differences in hyper-responsivity were identified between the autistic, NT, FXS-only, and FXS + ASD groups, controlling for NVMA. Specifically, the autistic group was predicted to score 0.52 mean points higher than the NT (b = 0.52; p < .001), 0.62 mean points higher than the FXS-only, (b = 0.62; p < .001), and 0.32 mean points higher than the FXS + ASD (b = 0.32; p = .02) groups. In addition, the FXS + ASD group was predicted to score 0.43 points higher than the NT group and 0.36 points higher than the FXS-only group; however, these effects lost significance when accounting for NVMA (b = 0.19; p = .319; b = 0.29; p = .08, respectively).
Hypo-Responsivity
Results showed group differences in hypo-responsivity between the autistic, NT, and FXS-only groups, and the FXS + ASD and FXS-only groups, controlling for NVMA. Specifically, the autistic group showed higher predicted scores than the NT group (b = 0.52; p < .001) and the FXS-only group (b = 0.67; p < .001) covarying for NVMA. The FXS + ASD group showed higher predicted scores than the FXS-only group (b = 0.51; p < .001), covarying for NVMA. The FXS + ASD group initially showed significantly higher predicted scores than the NT group (b = 0.79; p < .05) until accounting for NVMA (b = 0.36; p = .10).
Sensory Seeking
Results revealed significant differences in sensory seeking between the autistic, FXS + ASD, NT, and FXS-only groups, controlling for NVMA. Specifically, the autistic group showed 0.61 higher predicted scores (b = 0.61; p < .001), and the FXS + ASD group showed 0.50 (b = 0.50; p = .040) higher predicted scores than the NT group, respectively. In addition, the autistic group showed significantly higher predicted scores than the FXS-only group (b = 0.50; p = .010) accounting for NVMA. The FXS + ASD group showed higher predicted scores than the FXS-only group, yet (despite moderate-large effect size d = 0.78) these were no longer significant when accounting for NVMA (b = 0.38; p = .06).
Correlates of Challenging Behavior
Results from within-group correlations between sensory processing features and challenging behavior are shown in Table 3 and Figure 2.
Sensory Correlates of Challenging Behavior.
p < .05; **p < .01.
Sensory profiles and challenging behavior.
Autism
Within the autistic group, hyper-responsivity was significantly correlated with both internalizing (r = .58, p < .001) and externalizing behavior (r = .45, p = .005), hypo-responsivity was significantly correlated with internalizing behavior (r = .51, p = .001), yet no significant associations between sensory seeking and challenging behavior domains emerged.
FXS + ASD
No significant associations between any sensory processing and challenging behavior domains emerged within the FXS + ASD group; however, associations between hyper-responsivity and internalizing behavior (r = .38, p = .094) and sensory seeking and internalizing behavior (r = .34, p = .148) were moderate.
FXS-Only
Within the FXS-only group, hyper-responsivity was significantly correlated with internalizing behavior (r = .66, p = .007), hypo-responsivity was significantly correlated with internalizing (r = .59, p = .021) and externalizing behavior (r = .75, p = .001), yet no significant associations between sensory seeking and challenging behavior emerged.
NT
Within the NT group, hyper-responsivity was significantly associated with internalizing (r = .44, p = .015) and externalizing behavior (r = .56, p = .001), and hypo-responsivity significantly correlated with externalizing behavior (r = .41, p = .026), as did sensory seeking (r = .38, p = .040).
Discussion
Summary of Findings
Affective sensory reactivity and processing differences are common features of different NDDs, highly associated with autism, and are linked to challenging behavior for children with certain NDDs. Detailing variation in sensory processing features across distinct NDDs (e.g., autism and FXS), as well as within subgroups of specific etiologies (e.g., FXS + ASD versus FXS-only), and understanding associations between profile differences and challenging behaviors (or not) across groups provides critical evidence for improved differential assessment and diagnosis and targeted support. We aimed to characterize sensory processing profiles and examine the association between profile features and challenging behavior across four groups—non-syndromic autism, FXS-only, FXS + ASD, and NT children. Findings revealed consistent sensory processing differences in some groups, with nuanced differences in others, and identified linkages to challenging behavior that slightly varied across groups. Findings highlight the functional impact of sensory processing differences across neurodevelopmental and neurotypical contexts.
Sensory Processing Profiles
Overall, autistic children demonstrated the highest levels of sensory processing differences, whereas children with FXS-only and NT children demonstrated the lowest levels of sensory processing differences. These findings indicate that autistic children consistently experience sensory processing differences across multiple features to a greater extent than children with FXS and are aligned with prior evidence identifying elevated rates of sensory processing differences in autistic children compared with children with other NDDs (Baranek et al., 2006; Ben-Sasson et al., 2008; Rogers et al., 2003; Watling et al., 2001). In contrast to prior studies (McIntosh et al., 1999; Rogers et al., 2003), however, we did not detect statistically significant differences in sensory processing between children with FXS-only and NT children, which may have implications for understanding the FXS phenotype.
Hyper-Responsivity
We found similar levels of hyper-responsivity for children with FXS irrespective of co-occurring ASD and elevated levels for autistic children without an identified genetic etiology versus children with FXS with and without ASD. These findings suggest that, although hyper-responsivity emerges as a prominent feature in FXS, it may in fact be a key differentiating marker of unique underlying etiologies of autism (i.e., a syndromic etiology of autism such as FXS, vs. a non-syndromic etiology). Our findings further support existing evidence suggesting that hyper-responsivity is salient in autism (Ben-Sasson et al., 2019).
Hypo-Responsivity
Hypo-responsivity has long been associated with broader autistic feature expression, and prior research suggests that reduced responding to environmental stimuli may play a foundational role in the development of autism (Simon et al., 2017). We found that children with non-syndromic autism and those with FXS + ASD showed significantly elevated levels of hypo-responsivity compared with children with FXS-only. Similarities in hypo-responsivity between autistic groups (i.e., autism and FXS + ASD) and within-etiology differences (i.e., FXS + ASD vs. FXS-Only) highlight sensory hypo-responsivity as a potential autism-specific marker across distinct underlying etiologies (i.e., syndromic vs. non-syndromic) of autism. Because FXS is typically characterized by hyper-arousal and sensory hyper-sensitivity (Heald et al., 2020; Rogers et al., 2003), patterns of hypo-responsivity may prove especially useful in differential identification of ASD in children with FXS. Our findings expand previous work highlighting the high prevalence of hypo-responsivity in autism to children with FXS + ASD and provide evidence that hypo-responsivity may serve as an important hallmark of autism, regardless of etiology (i.e., syndromic and non-syndromic).
Sensory Seeking
The FXS + ASD and FXS-only groups did not significantly differ on sensory seeking, suggesting that, unlike hyper-responsivity and hypo-responsivity, it may not differentiate ASD in FXS. However, like hypo-responsivity, elevated seeking was identified in the autism and FXS + ASD groups relative to NT children, suggesting that sensory seeking corresponds with autism regardless of etiology. Similarities in sensory seeking across autistic groups may go hand-in-hand with hypo-responsivity. Foundational theoretical models of sensory processing depict sensory seeking behaviors as the manifestation of hypo-responsivity to sensory input (Dunn, 1999), indicating that elevated levels in one aspect of sensory processing would correspond with elevated levels in the other. Similar patterns of elevated differences in these specific sensory processing features in the autistic and FXS + ASD groups in the present study provide additional evidence for the co-occurrence of these sensory features in autistic children. Furthermore, hypo-responsivity may instead manifest as sensory seeking in the FXS-only group, explaining the lack of difference in sensory seeking behavior between FXS groups, despite identified differences in hypo-responsivity.
Overall Sensory Processing Implications
These novel and nuanced findings have critical implications for an increased understanding of differential diagnosis of ASD within FXS and may contribute to increased diagnostic precision. Diagnostic determination of ASD in FXS is challenging given the high degree of phenotypic overlap (Abbeduto et al., 2014, 2019; Hall et al., 2010). Because sensory processing differences are included in the diagnostic criteria for ASD (APA, 2013), nuanced sensory processing differences between children with FXS + ASD versus FXS-only inform diagnostic specificity. Our findings preliminarily support the utility of phenotyping sensory processing differences for increased diagnostic precision of ASD in FXS and provide evidence that children with FXS + ASD experience unique sensory processing differences, especially compared with children with FXS-only.
Interestingly, we found similarities between children with FXS-only and NT children, somewhat contrary to prior evidence suggesting that children with FXS experience greater sensory processing differences than neurotypical children. Our systematic differentiation of within-group heterogeneity in FXS by discerning FXS + ASD from FXS-only likely accounts for discrepancies with previous findings. The lack of significant differences between these groups within our study indicates that the presence of co-occurring ASD may drive sensory processing differences in FXS identified in prior research. This explanation is consistent with evidence that the severity of sensory processing differences in FXS corresponds with the degree of autistic features (Kolacz et al., 2018). Our findings extend this evidence by identifying important nuanced differences across NDDs, underscoring key areas of phenotypic distinction in the context of autism versus ID alone.
Effects on Challenging Behavior
Despite varying levels of sensory processing differences across groups, we substantiated associations between multiple sensory processing features and aspects of challenging behavior across both neurodevelopmental and neurotypical contexts. Although our findings revealed relatively comparable patterns of association between sensory features and internalizing and externalizing behaviors across groups, nuanced differences also emerged.
Autism
In the autistic group, sensory hyper-responsivity was associated with internalizing and externalizing behavior, and sensory hypo-responsivity was associated with internalizing behavior. These associations are largely consistent with prior research in autism showing links between hyper-responsivity and internalizing behavior (Ben-Sasson et al., 2008; Feldman et al., 2020; Green & Ben-Sasson, 2010; MacLennan et al., 2021; Mazurek et al., 2013; Rossow et al., 2021). Present study findings extend our current understanding of associations between hyper-responsivity and challenging behavior by substantiating the association between this sensory processing feature and externalizing behavior in autistic children. Similarly, our findings linking hypo-responsivity and challenging behavior are consistent with existing evidence (Ausderau et al., 2016; Baker et al., 2008; Marcham et al., 2023), while also providing preliminary evidence that hypo-responsivity may also relate to internalizing behavior in autistic children. Together, results suggest that autistic children demonstrate high levels of sensory differences that have implications for behavioral outcomes.
FXS + ASD
Interestingly, no significant associations between sensory processing features and challenging behavior were identified for children with FXS + ASD. Despite some similarities in the profiles of sensory processing differences, these features did not correspond with challenging behavior as they did in autistic children. The lack of significant associations in the FXS + ASD group suggests that sensory processing differences may impact phenotypic features other than challenging behavior, such as the expression of restricted and repetitive behaviors (RRBs). In addition, the lack of associations identified, coupled with the patterns of association found in the autistic group, suggests that these patterns may be specific to non-syndromic etiologies of autism.
FXS-Only
Although children with FXS-only showed lower levels of sensory processing differences than the other NDD groups, significant associations between hyper-responsivity and were identified. Similar patterns emerged between hypo-responsivity and internalizing and externalizing behavior between the FXS-only and NT groups. These findings suggest that, in addition to somewhat similar features of sensory processing differences, children with FXS-only and NT children are similar in how these features affect other aspects of behavior. Although children in these groups were reported to have fewer sensory processing differences than autistic children or children with FXS + ASD, findings indicate that they still experience related functional consequences that may require additional supports. These findings may have implications for within-syndrome specific targeted supports for children with FXS. That is, certain sensory processing differences may prove useful in targeting challenging behaviors for children with FXS + ASD, whereas other specific targets may warrant consideration for addressing these behaviors in children with FXS-only. Our findings on these associations for children with FXS-only provide new insight into within-syndrome variation and behavioral effects of sensory processing differences in children with FXS.
NT
Interestingly, hyper-responsivity was associated with externalizing behavior for the NT group, and a unique association between sensory seeking and externalizing behavior emerged for this group alone. Along with associations between hyper-responsivity, hypo-responsivity, and internalizing and externalizing behavior, these associations suggest that, as previously stated, features of sensory processing evidently impact these behavioral outcomes even for NT children. This supports the notion of a continuum of sensory processing features in all children and underscores the potential impact of sensory processing differences, even in the absence of a neurodevelopmental disability.
Implications for Sensory Seeking in Children with NDDs
The lack of association between sensory seeking and challenging behavior across the neurodevelopmental groups was somewhat surprising given existing evidence. One important consideration that may explain this inconsistent finding is that children in each of the NDD groups—autism, FXS + ASD, and FXS-only—also had co-occurring ID. Children with co-occurring ID or high support needs are not well represented in existing autism research; accordingly, the lack of functional consequences identified for unexpected sensory seeking behavior may reflect cognitive differences in children included in our study compared with prior studies. Prior evidence shows that children who experience elevated sensory seeking behavior require greater support with their communication skills and exhibit more RRBs (Lane et al., 2010; Tomchek et al., 2015; Watson et al., 2011); thus, sensory seeking may indeed impact functional outcomes other than challenging behavior for children with NDDs (Boyd et al., 2010; Wolff et al., 2019). Children with patterns of under-responsive or sensory seeking behaviors are thought to require greater stimulation to regulate and effective process information, which could manifest as RRBs (Baker et al., 2008). Thus, the sensory seeking differences specifically identified for autistic children and children with FXS + ASD relative to NT children may alternatively contribute to elevated RRBs, characteristic to both ASD and FXS, and a core diagnostic feature of ASD (APA, 2013). While RRBs can indeed manifest as challenging behavior, broadband behavior indices reflecting internalizing and externalizing behavior specifically may not effectively capture or adequately categorize the RRBs characterizing autism and FXS. Future research is needed to better elucidate the nature of sensory seeking behavior and its functional consequences for children with NDDs, especially those with non-syndromic autism and FXS + ASD.
Collectively, present study findings suggest that elevated sensory processing differences have a considerable impact on various aspects of challenging behavior across both neurodevelopmental and neurotypical developmental contexts, and, consequently, also have critical implications for both individual and family functioning. For example, caregiver stress is one potential area that may be negatively impacted by sensory processing differences and associated challenging behavior. The negative effects of child challenging behavior on parent stress are well established, and parents of children with NDDs unequivocally experience elevated stress relative to parents of neurotypical children (Nieto et al., 2017). Thus, sensory processing differences experienced by children with NDDs may have downstream effects on parent stress and well-being within the broader family system via increased rates of challenging behavior. Targeted interventions addressing challenging behavior, and in turn, parental stress, could benefit from incorporating strategies to mitigate challenges with sensory processing in children with and without NDDs. Mothers of children with FXS may particularly benefit from such efforts given not only their increased vulnerability to parenting stress but also their added genetic vulnerability as carriers to challenges with their own mental health (Bailey et al., 2008; Fielding-Gebhardt et al., 2020).
Limitations and Future Directions
Our study is among the first to examine the associations between sensory profiles and challenging behavior and how they may differ between young children with non-syndromic autism, FXS-only, FXS + ASD, and NT development. Given the exclusion of females and the relatively small sample size, particularly for the two FXS groups, the generalizability of findings must be considered. Examining additional sources of within-syndrome variability (i.e., sex effects) is an important next step. In addition, the measures of sensory processing and challenging behavior are based on parent-report, reflecting their own perceptions of their child’s behavior, which may be affected by additional vulnerabilities such as stress and mental health. Accordingly, an even stronger association between sensory processing differences and challenging behavior may exist but require alternative measures, such as direct observation of both sensory processing and behavioral symptoms. Similarly, future work should consider exploring how other RRBs (e.g., instance on sameness; repetitive motor movements) contribute to the links between features of sensory processing and challenging behaviors. Findings are also somewhat limited by the cross-sectional nature of the study. While our results provide useful insight on concurrent associations between sensory processing differences and challenging behavior, longitudinal research could offer greater insight into the progression of the association between sensory processing and challenging behavior in children with autism, FXS-only, and FXS + ASD compared with NT children throughout early development. Finally, examining isolated sensory modalities (e.g., auditory, visual, tactile, gustatory/olfactory, vestibular/proprioceptive), contexts (e.g., social vs. nonsocial), or specific aspects of internalizing/externalizing behavior (e.g., anxiety, aggression) was beyond the scope of our study, but considering these fine-grained details in the association between sensory processing differences and challenging behavior may yield additional insight into specific features of sensory processing differences and their associated consequences for children with NDDs.
Summary and Conclusions
To our knowledge, the present study is the first to characterize features of sensory processing differences and identify their functional impact on domains of challenging behavior across young children with autism, FXS-only, and FXS + ASD compared with NT children. Our findings provide additional insight into nuanced differences in sensory processing features and their links to challenging behavior across both neurodevelopmental and neurotypical contexts. Results underscore the utility of considering specific features of affective sensory reactivity rather than global sensory processing differences and provide evidence for considering targeted assessment and intervention approaches, particularly in the differential diagnosis and targeted support of co-occurring ASD in FXS. Collectively, our findings contribute evidence on the similarities and differences between autism and FXS, which is critical for advancing our understanding of the phenotypic boundaries between non-syndromic autism and FXS.
Footnotes
Acknowledgements
We would like to thank all the families that participated in this research.
Author Contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institutes of Health through the following awards: K99HD105980, R00HD105980 (PI: Will); R01MH90194, R01MH107573 (PI: Roberts).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.