A5046843056- Language Development and Disorders
- Reading and Literacy Development
- Speech and dialogue systems
- Phonetics and Phonology Research
- Child and Animal Learning Development
- Second Language Acquisition and Learning
- Neurobiology of Language and Bilingualism
- Biomedical Text Mining and Ontologies
- Text Readability and Simplification
- Cognitive and developmental aspects of mathematical skills
- Natural Language Processing Techniques
- Topic Modeling
Cardiff University
2020-2023
Nottingham Trent University
2020-2021
Child-directed speech has long been known to influence children’s vocabulary learning. However, while we know that caregiver utterances differ from those directed at adults in various ways, little is about any differences the lexical properties of child-directed and adult-directed utterances. We compare over half a million word tokens adult children (from caregiver–child transcriptions) same quantity adults. show contains greater numbers words are lower phonemic length, higher frequency,...
Research shows that as toddlers grow and their vocabulary expands, ability to recognize a referent after hearing its name worsens when they encounter similar-sounding words (e.g., dog-door) or from the same category dog-chicken) compared unrelated dog-boat). This study investigated impact of phonological semantic similarities between on spoken word recognition in toddlers. We presented 21-month-old English monolinguals with Preferential Looking Task adapted priming paradigm while eye...
Abstract Word segmentation is a crucial step in children's vocabulary learning. While computational models of word can capture infants’ performance small‐scale artificial tasks, the examination early naturalistic settings has been limited by lack measures that relate models’ to developmental data. Here, we extended CLASSIC (Chunking Lexical and Sublexical Sequences Children; Jones et al., 2021), corpus‐trained chunking model simulate several memory phonological learning phenomena allow it...
Understanding how children process ambiguous words is a challenge because sense disambiguation depends on sentence context bottom-up and top-down aspects. Here, we seek in- sight into this phenomenon by investigating such com- petence might arise in large distributional learners (Transform- ers) that purport to acquire representations from lan- guage input largely unsupervised fashion. We investigated be achieved using model rep- resentations derived naturalistic child-directed speech....
Word segmentation is a crucial step in children’s vocabulary learning. While computational models of word can capture infants’ performance small-scale artificial tasks, the examination early naturalistic settings has been limited by lack measures that relate models’ to developmental data. Here, we extended CLASSIC (Jones et al., 2021) - corpus-trained chunking model simulate several memory, phonological and learning phenomena allow it perform using utterance boundary information (henceforth...