Autism spectrum disorder (ASD) is a neurodevelopmental disorder in which individuals demonstrate deficits in social communication and repetitive or restricted behaviors or interests. About one-third never attain fluent expressive language and remain minimally or low verbal (ASD-MLV). We hypothesized that central auditory processing is particularly disrupted in ASD-MLV given the shared neural substrates of sound processing and language. To address this hypothesis, we conducted four empirical studies designed to capture neural and behavioral correlates of central auditory processing in children and adolescents who were either typically developing (TD), verbally fluent with ASD (ASD-V), or ASD-MLV. Our first study established that adult-like neural indices of sound organization as measured by mismatch responses (MMRs) during a passive stream segregation task were not observable until adolescence in TD participants, ages 3 to 21 (N=65). Findings led us to focus subsequent studies on adolescents. In our second study, we conducted a meta-analysis of experiments that had compared MMRs between TD and ASD samples (N=721). We identified that while there was some evidence suggesting that perceptual sound organization (as evidenced by MMR) was more dysfunctional in those with ASD, more research was needed to validate this pattern in adolescents and low verbal samples. In our third study (N=83), we determined that atypical auditory behaviors occurred most often in those with ASD with combined expressive and receptive language impairments; furthermore, the percentage of time exhibiting such behaviors was associated with weaker MMRs. In our final study (N=74), we quantified neural orienting responses to one’s own name in a multispeaker setting – a task that requires higher-order stream segregation and social auditory attention – in adolescents. We found that responses were weaker in ASD-MLV compared to ASD-V and TD participants. In addition, strength of response in those with ASD was negatively correlated with parent-reported signs of auditory-specific attentional deficits, as measured by the Short Sensory Profile Auditory Filtering Subscale. With this dissertation, we found atypical neural indices of auditory processing in ASD-MLV adolescents and discuss theoretical implications for why central auditory processing might be particularly pronounced in the ASD-MLV phenotype. / 2022-01-30T00:00:00Z

The prevalence of autism spectrum disorder (ASD) is higher in males than females. Previous research has uncovered a female protective effect in which females have a higher threshold for genetic abnormality before being affected by ASD. Because this also suggests that females with ASD will have more severe symptoms, sex differences in symptoms could provide support for this theory. The present study investigated differences in autistic symptoms and characteristics in 20 boys and 5 girls aged 5-21 who represent a unique subset of the autistic population: minimally verbal individuals. The participants, the majority of whom failed to develop language beyond phrase speech, completed a variety of assessments at Boston University’s Center for Autism Research Excellence (CARE). The present study was a secondary analysis of cross-sectional data from two previous studies conducted between May 2012 and July 2014 at CARE. Sex differences in overall symptoms and the domains of nonverbal cognitive ability, social functioning, repetitive and maladaptive behavior, and structural and pragmatic language from standardized tests and language samples were examined. Language sample videos and the associated transcriptions were coded based on whether the child directed his/her speech to other people and the intention behind their language: whether they used speech for the purpose of behavior regulation, joint attention, social interaction, or an unknown purpose. There were no significant sex differences found in any of the measures and no effect of sex when controlling for age and IQ. There was a significant effect of IQ found for maladaptive behavior, social functioning, language level, directed speech, and speech used for joint attention. There was also a significant effect of age on language ability and directed speech. These findings partially conflict with those of previous literature in verbal autistic individuals, suggesting that minimally verbal individuals have distinct symptomatic profiles. The finding of no sex differences in autism symptoms suggests that if a female protective effect exists, it does not manifest in more severe symptoms in minimally verbal females. Therefore, this study contributes to the knowledge of sex differences in autism by characterizing the symptomatic profiles of this subgroup.

2023 / Difficulties in communication during social interactions are a feature of the diagnosis of autism spectrum disorder (ASD). Little is known about the communication abilities of minimally-verbal (MV) individuals, which comprises one-third of autistic individuals. Natural language sampling (NLS) provides an optimal approach to capture communication abilities in this population. The goal of this project was to use NLS methods to examine multimodal communication: speech, gesture, gesture-speech combinations, and augmentative and alternative communication in MV autistic children and adolescents during social interactions.Study 1 examined how and why MV autistic children and adolescents (N=50; 6-21 years; 12 females) used their communication abilities during social interactions. Unexpectedly, adolescents produced more gesture-speech combinations and communicative functions but did not differ in gesture frequency. I explored their range of communicative functions across different communication modalities. As hypothesized, they produced more requests using gesture but used speech more often to disagree/agree/acknowledge and to respond to questions. I investigated the concurrent relation between gestural communication and spoken language. As hypothesized, fewer speech utterances were related to more gesturing after accounting for age, nonverbal IQ, and receptive communication ability. Study 2 explored how MV autistic children’s (N=47; 48-95 months; 10 females) modality and form of communication influenced parent responsivity. As expected, parents provided semantically-relevant (contingent) responses across children’s communication modalities. Unexpectedly, the precision of the child’s form (e.g., approximated gesture) did not influence parent contingent responding. I investigated the parent’s modality of response following the child’s modality of communication. Across children’s modalities of communication (e.g., gesture), parents responded most frequently using speech. Study 3 evaluated the predictive role of gestural communication in later spoken language ability in MV autistic children (N=50; 54-105 months; 10 females) receiving intervention. Unexpectedly, frequency of deictic gestures did not predict speech outcomes. As hypothesized, supplementary combinations predicted speech outcomes. Findings from this project underscore the importance of 1) examining multiple modalities and forms of communication to obtain a comprehensive understanding of communication skills; 2) inclusion of interactionist models to capture children’s input on parent responses in shaping language learning; and 3) the predictive role of gesture-speech combinations in facilitating spoken language outcomes.