-
Behavioral Sciences (Basel, Switzerland) May 2024Despite the consensus that early identification leads to better outcomes for individuals with autism spectrum disorder (ASD), recent research reveals that the average...
Despite the consensus that early identification leads to better outcomes for individuals with autism spectrum disorder (ASD), recent research reveals that the average age of diagnosis in the Greek population is approximately six years. However, this age of diagnosis is delayed by an additional two years for families from lower-income or minority backgrounds. These disparities result in adverse impacts on intervention outcomes, which are further burdened by the often time-consuming and labor-intensive language assessments for children with ASD. There is a crucial need for tools that increase access to early assessment and diagnosis that will be rigorous and objective. The current study leverages the capabilities of artificial intelligence to develop a reliable and practical model for distinguishing children with ASD from typically-developing peers based on their narrative and vocabulary skills. We applied natural language processing-based extraction techniques to automatically acquire language features (narrative and vocabulary skills) from storytelling in 68 children with ASD and 52 typically-developing children, and then trained machine learning models on the children's combined narrative and expressive vocabulary data to generate behavioral targets that effectively differentiate ASD from typically-developing children. According to the findings, the model could distinguish ASD from typically-developing children, achieving an accuracy of 96%. Specifically, out of the models used, hist gradient boosting and XGBoost showed slightly superior performance compared to the decision trees and gradient boosting models, particularly regarding accuracy and F1 score. These results bode well for the deployment of machine learning technology for children with ASD, especially those with limited access to early identification services.
PubMed: 38920791
DOI: 10.3390/bs14060459 -
IEEE Transactions on Image Processing :... 2024The ability to detect and track the dynamic objects in different scenes is fundamental to real-world applications, e.g., autonomous driving and robot navigation....
The ability to detect and track the dynamic objects in different scenes is fundamental to real-world applications, e.g., autonomous driving and robot navigation. However, traditional Multi-Object Tracking (MOT) is limited to track objects belonging to the pre-defined closed-set categories. Recently, Generic MOT (GMOT) is proposed to track interested objects beyond pre-defined categories and it can be divided into Open-Vocabulary MOT (OVMOT) and Template-Image-based MOT (TIMOT). Taking the consideration that the expensive well pre-trained (vision-)language model and fine-grained category annotations are required to train OVMOT models, in this paper, we focus on TIMOT and propose a simple but effective method, Siamese-DETR. Only the commonly used detection datasets (e.g., COCO) are required for training. Different from existing TIMOT methods, which train a Single Object Tracking (SOT) based detector to detect interested objects and then apply a data association based MOT tracker to get the trajectories, we leverage the inherent object queries in DETR variants. Specifically: 1) The multi-scale object queries are designed based on the given template image, which are effective for detecting different scales of objects with the same category as the template image; 2) A dynamic matching training strategy is introduced to train Siamese-DETR on commonly used detection datasets, which takes full advantage of provided annotations; 3) The online tracking pipeline is simplified through a tracking-by-query manner by incorporating the tracked boxes in the previous frame as additional query boxes. The complex data association is replaced with the much simpler Non-Maximum Suppression (NMS). Extensive experimental results show that Siamese-DETR surpasses existing MOT methods on GMOT-40 dataset by a large margin.
PubMed: 38917291
DOI: 10.1109/TIP.2024.3416880 -
Proceedings of the National Academy of... Jul 2024Languages disfavor word forms containing sequences of similar or identical consonants, due to the biomechanical and cognitive difficulties posed by patterns of this...
Languages disfavor word forms containing sequences of similar or identical consonants, due to the biomechanical and cognitive difficulties posed by patterns of this sort. However, the specific evolutionary processes responsible for this phenomenon are not fully understood. Words containing sequences of identical consonants may be more likely to arise than those without; processes of word form mutation may be more likely to remove than create sequences of identical consonants in word forms; finally, words containing identical consonants may die out more frequently than those without. Phylogenetic analyses of the evolution of homologous word forms indicate that words with identical consonants arise less frequently than those without. However, words with identical consonants do not die out more frequently than those without. Further analyses reveal that forms with identical consonants are replaced in basic meaning functions more frequently than words without. Taken together, results suggest that the underrepresentation of sequences of identical consonants is overwhelmingly a by-product of constraints on word form coinage, though processes related to word usage also serve to ensure that such patterns are infrequent in more salient vocabulary items. These findings clarify aspects of processes of lexical evolution and competition that take place during language change, optimizing communicative systems.
Topics: Language; Humans; Phylogeny; Biological Evolution; Phonetics; Vocabulary
PubMed: 38917001
DOI: 10.1073/pnas.2316677121 -
BioRxiv : the Preprint Server For... Jun 2024An important and largely unsolved problem in synthetic biology is how to target gene expression to specific cell types. Here, we apply iterative deep learning to design...
An important and largely unsolved problem in synthetic biology is how to target gene expression to specific cell types. Here, we apply iterative deep learning to design synthetic enhancers with strong differential activity between two human cell lines. We initially train models on published datasets of enhancer activity and chromatin accessibility and use them to guide the design of synthetic enhancers that maximize predicted specificity. We experimentally validate these sequences, use the measurements to re-optimize the predictor, and design a second generation of enhancers with improved specificity. Our design methods embed relevant transcription factor binding site (TFBS) motifs with higher frequencies than comparable endogenous enhancers while using a more selective motif vocabulary, and we show that enhancer activity is correlated with transcription factor expression at the single cell level. Finally, we characterize causal features of top enhancers via perturbation experiments and show enhancers as short as 50bp can maintain specificity.
PubMed: 38915713
DOI: 10.1101/2024.06.14.599076 -
BioRxiv : the Preprint Server For... Jun 2024Postnatal genomic regulation significantly influences tissue and organ maturation but is under-studied relative to existing genomic catalogs of adult tissues or prenatal...
Postnatal genomic regulation significantly influences tissue and organ maturation but is under-studied relative to existing genomic catalogs of adult tissues or prenatal development in mouse. The ENCODE4 consortium generated the first comprehensive single-nucleus resource of postnatal regulatory events across a diverse set of mouse tissues. The collection spans seven postnatal time points, mirroring human development from childhood to adulthood, and encompasses five core tissues. We identified 30 cell types, further subdivided into 69 subtypes and cell states across adrenal gland, left cerebral cortex, hippocampus, heart, and gastrocnemius muscle. Our annotations cover both known and novel cell differentiation dynamics ranging from early hippocampal neurogenesis to a new sex-specific adrenal gland population during puberty. We used an ensemble Latent Dirichlet Allocation strategy with a curated vocabulary of 2,701 regulatory genes to identify regulatory "topics," each of which is a gene vector, linked to cell type differentiation, subtype specialization, and transitions between cell states. We find recurrent regulatory topics in tissue-resident macrophages, neural cell types, endothelial cells across multiple tissues, and cycling cells of the adrenal gland and heart. Cell-type-specific topics are enriched in transcription factors and microRNA host genes, while chromatin regulators dominate mitosis topics. Corresponding chromatin accessibility data reveal dynamic and sex-specific regulatory elements, with enriched motifs matching transcription factors in regulatory topics. Together, these analyses identify both tissue-specific and common regulatory programs in postnatal development across multiple tissues through the lens of the factors regulating transcription.
PubMed: 38915583
DOI: 10.1101/2024.06.12.598567 -
Journal of Experimental Psychology.... Jun 2024Young children learn language from their caregivers, family members, and friends. However, with few exceptions, contemporary developmental scientists have studied...
Young children learn language from their caregivers, family members, and friends. However, with few exceptions, contemporary developmental scientists have studied language input and language learning through the lens of the primary caregiver and the nuclear family, rather than the infants' broader communities. In many communities-and increasingly in the United States-nonnuclear family structures are common, and extended kin, fictive kin, and intergenerational relationships are relied upon for child care. Understanding children's relationships within kinship networks can allow for more inclusive depictions of children's social interactions and their language experiences. We drew upon methods used by researchers studying social networks to assess U.S. infants' and toddlers' network composition. Results showed that young children with a greater number of close relationships (but not those with larger networks overall) had larger vocabularies, after controlling for age and socioeconomic status. These findings suggest that distributed models of child-rearing are an influential factor in early language growth and call for increased attention to social networks for understanding children's developmental trajectories. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
PubMed: 38913749
DOI: 10.1037/xge0001609 -
Psychology and Aging Jun 2024Older adults may experience certain forms of cognitive decline, but some forms of semantic memory remain intact in older age. To address how metaphor comprehension...
Older adults may experience certain forms of cognitive decline, but some forms of semantic memory remain intact in older age. To address how metaphor comprehension changes with age and whether metaphor comprehension relies more heavily on analogical reasoning (supported by fluid intelligence) or on conceptual combination (supported by crystalized intelligence), we compared performance of younger and older adults. In two experiments, healthy older adults (54-88 years) scored lower on a measure of fluid intelligence (Ravens Progressive Matrices) but higher on a measure of crystalized intelligence (Mill Hill Vocabulary Test) relative to younger adults (18-34 years). Groups were equally successful in comprehending relatively easy metaphors (Study 1), but older adults showed a striking advantage over younger adults for novel literary metaphors (Study 2). Mixed-effects modeling showed that measures of fluid and crystalized intelligence each made separable contributions to metaphor comprehension for both groups, but older adults relied more on crystalized intelligence than did younger adults. These age-related dissociations clarify cognitive effects of aging and highlight the importance of crystalized intelligence for metaphor comprehension in both younger and older adults. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
PubMed: 38913736
DOI: 10.1037/pag0000836 -
Journal of Experimental Psychology.... Jun 2024Infants' early words tend to be phonologically similar. This may reflect a systematic approach to early production, as they adapt newly acquired forms to fit familiar...
Infants' early words tend to be phonologically similar. This may reflect a systematic approach to early production, as they adapt newly acquired forms to fit familiar structures in the output. This "rich-get-richer" approach to phonological acquisition, known as preferential attachment in network science, proposes that new words cluster together with existing phonologically similar words in the lexicon (or network). This contrasts with recent work (e.g., Fourtassi et al., 2020) showing that the learning environment is the key predictor of learning (preferential acquisition). This study expands on previous analyses of vocabulary norm data to analyze naturalistic data, namely phonetic transcriptions of nine infants' word productions, from word onset to age 2;6. Network growth models test whether (a) acquisition is best modeled through preferential attachment or preferential acquisition, (b) the trajectory of network growth changes over time, and (c) there are any differences in network growth of adult target forms versus infants' actual productions. Results show that preferential attachment predicts acquisition of new words more convincingly than preferential acquisition: newly acquired words are phonologically similar to existing words in the network. Furthermore, systematicity becomes increasingly apparent over the course of acquisition, and infants produce their early words more systematically than we would expect from looking at target forms alone. (PsycInfo Database Record (c) 2024 APA, all rights reserved).
PubMed: 38913728
DOI: 10.1037/xlm0001368 -
Animal Cognition Jun 2024Due to their outstanding ability of vocal imitation, parrots are often kept as pets. Research has shown that they do not just repeat human words. They can use words... (Comparative Study)
Comparative Study
Due to their outstanding ability of vocal imitation, parrots are often kept as pets. Research has shown that they do not just repeat human words. They can use words purposefully to label objects, persons, and animals, and they can even use conversational phrases in appropriate contexts. So far, the structure of pet parrots' vocabularies and the difference between them and human vocabulary acquisition has been studied only in one individual. This study quantitatively analyses parrot and child vocabularies in a larger sample using a vocabulary coding method suitable for assessing the vocabulary structure in both species. We have explored the composition of word-like sounds produced by 21 grey parrots (Psittacus erithacus) kept as pets in Czech- or Slovak-speaking homes, and compared it to the composition of early productive vocabularies of 21 children acquiring Czech (aged 8-18 months), who were matched to the parrots by vocabulary size. The results show that the 'vocabularies' of talking grey parrots and children differ: children use significantly more object labels, activity and situation labels, and emotional expressions, while parrots produce significantly more conversational expressions, greetings, and multiword utterances in general. These differences could reflect a strong link between learning spoken words and understanding the underlying concepts, an ability seemingly unique to human children (and absent in parrots), but also different communicative goals of the two species.
Topics: Animals; Parrots; Female; Male; Vocabulary; Humans; Infant; Czech Republic; Language Development; Pets; Slovakia
PubMed: 38913161
DOI: 10.1007/s10071-024-01883-5 -
Research in Developmental Disabilities Jun 2024Narrative ability is crucial for social participation in everyday and school life but involves different language abilities such as vocabulary and morpho-syntax. This is...
BACKGROUND
Narrative ability is crucial for social participation in everyday and school life but involves different language abilities such as vocabulary and morpho-syntax. This is particularly difficult for individuals who display both language and cognitive impairments. Previous research has identified productive vocabulary as a possible key factor for narrative performance in individuals with Down syndrome. Considering a close connection between lexical and morpho-syntactic performance within language acquisition and the distinct impairments that individuals with Down syndrome display concerning their morpho-syntactic skills, the nature of a relation between vocabulary and narrative skills under the influence of grammatical deficits requires further investigation.
METHODS
Narrations were obtained from 28 children and adolescents with Down syndrome (aged 10;0-20;1) using a non-verbal picture book. Narrative abilities were rated using the Narrative Scoring Scheme across seven narrative aspects (including macro- and microstructure). Vocabulary analyses and morpho-lexical context analyses including verb and conjunction enumerations, evaluation of verb position and MLU were conducted. Findings from the transcript analysis have been supplemented with data from standardized language measures evaluating expressive lexical and morpho-syntactic development. A multiple regression analysis was conducted to identify significant predictors for narrative outcome in the participants with Down syndrome.
RESULTS
Lexical analyses revealed a high heterogeneity in production of subordinating conjunctions as a link between lexical and morpho-syntactic abilities. Comparisons of standardized and narrative data demonstrated differences in subordinate clause production depending on the elicitation setting. A multiple regression analysis identified the number of different verbs in the narrative task as the most significant predictor for narrative performance in individuals with Down syndrome.
DISCUSSION AND IMPLICATIONS
The findings of this study contribute to the knowledge regarding factors that influence narrative performance in individuals with language impairment. A differentiated verb lexicon can be identified as the key ability for reaching advanced narrative skills in participants with Down syndrome. These findings are of clinical relevance for therapeutic and educational support and contribute to an understanding of the relation between strengths in vocabulary and morpho-syntactic weaknesses in individuals with Down syndrome within communicative participation.
PubMed: 38908111
DOI: 10.1016/j.ridd.2024.104781