Zhenghan Qi - US grants

This award was provided as part of NSF's Social, Behavioral and Economic Sciences Postdoctoral Research Fellowships (SPRF) program and the Established Program to Stimulate Competitive Research (EPSCoR). The goal of the SPRF program is to prepare promising, early career doctoral-level scientists for scientific careers in academia, industry or private sector, and government. SPRF awards involve two years of training under the sponsorship of established scientists and encourage Postdoctoral Fellows to perform independent research. NSF seeks to promote the participation of scientists from all segments of the scientific community, including those from underrepresented groups, in its research programs and activities; the postdoctoral period is considered to be an important level of professional development in attaining this goal. Each Postdoctoral Fellow must address important scientific questions that advance their respective disciplinary fields. Under the sponsorship of Dr. Zhenghan Qi at the University of Delaware, this postdoctoral fellowship award supports an early career scientist investigating whether superior statistical learning abilities serve as a protective factor against the detrimental effects coming from a low socioeconomic (SES) background can have upon language development. Coming from a low SES home negatively affects language and brain development. As children progress through school, this "achievement gap" between low and higher SES children is exacerbated. While this may be attributed to early environmental factors, such as poor quality and quantity of parental inputs early on, there is substantial heterogeneity in the vocabulary outcomes of children from low SES families. This heterogeneity in vocabulary outcomes suggests language success is not fully accounted for by quantity and quality of maternal input. In fact, in infants, both the child-directed input and the child's ability to process that input account for the majority of variance seen in vocabulary skills at two years of age. This ability to extract and process incoming input in the environment, also known as statistical learning (SL), is often considered a core supporting mechanism of first language development. The current proposal therefore seeks to clarify how both maternal input and the processing of this input via statistical learning (SL) at the neurological level, account for variability in vocabulary among low SES children. This research is driven by the critical need to identify protective factors for children from low SES families, given the pervasiveness of the vocabulary gap. Additionally, no research to date has addressed the substantial heterogeneity within a low SES population, which holds important insights for understanding why some children from low SES homes perform better than others on measures of vocabulary.

The proposed research utilizes a multimodal, cross-disciplinary approach to studying the heterogeneity of vocabulary knowledge among children from socioeconomically adverse environments. This research will recruit sixty children from low SES households to examine the contribution that SL ability and quality and quantity of mother-child conversations have upon vocabulary variability within a low SES sample. Within this same sample, the current study will compare patterns of neural engagement using fMRI during statistical learning between good and poor learners and identify whether these patterns predict vocabulary size. The current study launches an investigation into the contribution of environmental factors, cognition, and brain activation on vocabulary variability within a low SES sample to isolate which factors are most influential for language success in this socioeconomically at-risk population.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Abstract ! Children with Autism Spectrum Disorder (ASD) show enormous heterogeneity in core language abilities, such as phonology and grammar. Over 50% of verbal children with ASD have borderline or impaired language. Deciphering how children with ASD learn language is important for understanding the heterogeneity of language abilities in this population, as well as for the development of more effective interventions for children with ASD who also have language difficulties. Based on decades of behavioral research in typical language acquisition, statistical learning is essential for encoding rich regularities and variability for first language development. Whereas statistical learning has been linked to typical language development and Developmental Language Disorder (DLD), existing studies investigating statistical learning in ASD lack consensus regarding whether statistical learning is impaired in ASD. Both the enormous heterogeneity of ASD population and the recently reported variation in one?s ability to learn statistics across linguistic and non-linguistic domains, as well as across sensory modalities, might have contributed to such inconclusive results. The overall goal of this research is to understand whether the language heterogeneity in ASD can be attributed to variations of statistical learning in a specific domain (linguistic vs. nonlinguistic) or modality (auditory vs. visual). Aim 1 proposes to first characterize the relationship between children?s language abilities and statistical-learning profiles across domains and modalities in a large sample of children with ASD (6 to 8 years old) and typically developing (TD) controls. Both groups will be evaluated using a novel web-based testing platform. Aim 2 will focus on probing the neural mechanisms during statistical learning in each domain and modality in ASD, and how relative strengths and weaknesses in the groupings are related to language performance. Comparing the neural responses during statistical learning tasks among children with ASD in relation to high vs. low statistical learning abilities grouped by domain and modality, the study will test whether language performance in ASD is related to domain-specific, modality-specific, or domain/modality-general neural substrates of statistical learning. The proposed study will pave the way towards a deeper understanding of the mechanisms involved in language impairment in ASD. The results will provide crucial preliminary data for future investigations aiming to elucidate the causal relationship between statistical learning ability and language learning in a longitudinal intervention study and how statistical learning profile is related to different genetic risks in children with ASD.