Publications

Abstract

Understanding the mechanisms that drive variation in children’s language acquisition requires large, population-representative datasets of children’s word learning across development. Parent report measures such as the MacArthur-Bates Communicative Development Inventories (CDI) are commonly used to collect such data, but the traditional paper-based forms make the curation of large datasets logistically challenging. Many CDI datasets are thus gathered using convenience samples, often recruited from communities in proximity to major research institutions. Here, we introduce Web-CDI, a web-based tool which allows researchers to collect CDI data online. Web-CDI contains functionality to collect and manage longitudinal data, share links to test administrations, and download vocabulary scores. To date, over 3,500 valid Web-CDI administrations have been completed. General trends found in past norming studies of the CDI are present in data collected from Web-CDI: scores of children’s productive vocabulary grow with age, female children show a slightly faster rate of vocabulary growth, and participants with higher levels of educational attainment report slightly higher vocabulary production scores than those with lower levels of education attainment. We also report results from an effort to oversample non-white, lower-education participants via online recruitment (N = 241). These data showed similar demographic trends to the full sample but this effort resulted in a high exclusion rate. We conclude by discussing implications and challenges for the collection of large, population-representative datasets.

Abstract

The discovery of the FOXP2 transcription factor, and its implication in a rare severe human speech and language disorder, has led to two decades of empirical studies focused on uncovering its roles in the brain using a range of in vitro and in vivo methods. Here, we discuss what we have learned about the regulation of FOXP2, its downstream effectors, and its modes of action as a transcription factor in brain development and function, providing an integrated overview of what is currently known about the critical molecular networks.

Abstract

Whereas large-scale statistical analyses can robustly identify disease-gene relationships, they do not accurately capture genotype-phenotype correlations or disease mechanisms. We use multiple lines of independent evidence to show that different variant types in a single gene, SATB1, cause clinically overlapping but distinct neurodevelopmental disorders. Clinical evaluation of 42 individuals carrying SATB1 variants identified overt genotype-phenotype relationships, associated with different pathophysiological mechanisms, established by functional assays. Missense variants in the CUT1 and CUT2 DNA-binding domains result in stronger chromatin binding, increased transcriptional repression and a severe phenotype. Contrastingly, variants predicted to result in haploinsufficiency are associated with a milder clinical presentation. A similarly mild phenotype is observed for individuals with premature protein truncating variants that escape nonsense-mediated decay and encode truncated proteins, which are transcriptionally active but mislocalized in the cell. Our results suggest that in-depth mutation-specific genotype-phenotype studies are essential to capture full disease complexity and to explain phenotypic variability.

Abstract

European cooperative effort seeks best practices architecture and procedures for international sites

Abstract

Sex differences in development and aging of human sulcal morphology have been understudied. We charted sex differences in trajectories and inter-individual variability of global sulcal depth, width, and length, pial surface area, exposed (hull) gyral surface area, unexposed sulcal surface area, cortical thickness, and cortex volume across the lifespan in a longitudinal sample (700 scans, 194 participants two scans, 104 three scans, age range: 16-70 years) of neurotypical males and females. After adjusting for brain volume, females had thicker cortex and steeper thickness decline until age 40 years; trajectories converged thereafter. Across sexes, sulcal shortening was faster before age 40, while sulcal shallowing and widening were faster thereafter. While hull area remained stable, sulcal surface area declined and was more strongly associated with sulcal shortening than with sulcal shallowing and widening. Males showed greater variability for cortex volume and thickness and lower variability for sulcal width. Across sexes, variability decreased with age for all measures except for cortical volume and thickness. Our findings highlight the association between loss of sulcal area, notably through sulcal shortening, with cortex volume loss. Studying sex differences in lifespan trajectories may improve knowledge of individual differences in brain development and the pathophysiology of neuropsychiatric conditions.

Abstract

This paper investigates when and how children figure out the force of modals: that possibility modals (e.g., can/might) express possibility, and necessity modals (e.g., must/have to) express necessity. Modals raise a classic subset problem: given that necessity entails possibility, what prevents learners from hypothesizing possibility meanings for necessity modals? Three solutions to such subset problems can be found in the literature: the first is for learners to rely on downward-entailing (DE) environments (Gualmini and Schwarz in J. Semant. 26(2):185–215, 2009); the second is a bias for strong (here, necessity) meanings; the third is for learners to rely on pragmatic cues stemming from the conversational context (Dieuleveut et al. in Proceedings of the 2019 Amsterdam colloqnium, pp. 111–122, 2019a; Rasin and Aravind in Nat. Lang. Semant. 29:339–375, 2020). This paper assesses the viability of each of these solutions by examining the modals used in speech to and by 2-year-old children, through a combination of corpus studies and experiments testing the guessability of modal force based on their context of use. Our results suggest that, given the way modals are used in speech to children, the first solution is not viable and the second is unnecessary. Instead, we argue that the conversational context in which modals occur is highly informative as to their force and sufficient, in principle, to sidestep the subset problem. Our child results further suggest an early mastery of possibility—but not necessity—modals and show no evidence for a necessity bias.

Abstract

To investigate how orthography and semantics interact during bilingual visual word recognition, Dutch–English bilinguals made lexical decisions in two masked priming experiments. Dutch primes and English targets were presented that were either neighbour cognates (boek – BOOK), noncognate translations (kooi – CAGE), orthographically related neighbours (neus – NEWS), or unrelated words (huid - COAT). Prime durations of 50 ms (Experiment 1) and 83 ms (Experiment 2) led to similar result patterns. Both experiments reported a large cognate facilitation effect, a smaller facilitatory noncognate translation effect, and the absence of inhibitory orthographic neighbour effects. These results indicate that cognate facilitation is in large part due to orthographic-semantic resonance. Priming results for each condition were simulated well (all r's >.50) by Multilink+, a recent computational model for word retrieval. Limitations to the role of lateral inhibition in bilingual word recognition are discussed.

Abstract

Age has a major effect on brain volume. However, the normative studies available are constrained by small sample sizes, restricted age coverage and significant methodological variability. These limitations introduce inconsistencies and may obscure or distort the lifespan trajectories of brain morphometry. In response, we capitalized on the resources of the Enhancing Neuroimaging Genetics through Meta‐Analysis (ENIGMA) Consortium to examine age‐related trajectories inferred from cross‐sectional measures of the ventricles, the basal ganglia (caudate, putamen, pallidum, and nucleus accumbens), the thalamus, hippocampus and amygdala using magnetic resonance imaging data obtained from 18,605 individuals aged 3–90 years. All subcortical structure volumes were at their maximum value early in life. The volume of the basal ganglia showed a monotonic negative association with age thereafter; there was no significant association between age and the volumes of the thalamus, amygdala and the hippocampus (with some degree of decline in thalamus) until the sixth decade of life after which they also showed a steep negative association with age. The lateral ventricles showed continuous enlargement throughout the lifespan. Age was positively associated with inter‐individual variability in the hippocampus and amygdala and the lateral ventricles. These results were robust to potential confounders and could be used to examine the functional significance of deviations from typical age‐related morphometric patterns.

Abstract

The article compares the integration of topic-related additive words at different stages of untutored L2 acquisition. Data stem from an ‘‘additive-elicitation task’’ that was designed in order to capture topic-related additive words in a context that is at the same time controlled for the underlying information structure and nondeviant from other kinds of narrative discourse. We relate the distinction between stressed and nonstressed forms of the German scope particles and adverbials auch ‘also’, noch ‘another’, wieder ‘again’, and immer noch ‘still’ to a uniform, information-structure-based principle: the stressed variants have scope over the topic information of the relevant utterances. It is then the common function of these additive words to express the additive link between the topic of the present utterance and some previous topic for which the same state of affairs is claimed to hold. This phenomenon has often been referred to as ‘‘contrastive topic,’’ but contrary to what this term suggests, these topic elements are by no means deviant from the default in coherent discourse. In the underlying information structure, the validity of some given state of affairs for the present topic must be under discussion. Topic-related additive words then express that the state of affairs indeed applies to this topic, their function therefore coming close to the function of assertion marking. While this functional correspondence goes along with the formal organization of the basic stages of untutored second-language acquisition, its expression brings linguistic constraints into conflict when the acquisition of finiteness pushes learners to reorganize their utterances according to target-language syntax.

Abstract

Continuers —words like mm, mmhm, uhum and the like— are among the most frequent types of responses in conversation. They play a key role in joint action coordination by showing positive evidence of understanding and scaffolding narrative delivery. Here we investigate the hypothesis that their functional importance along with their conversational ecology places selective pressures on their form and may lead to cross-linguistic similarities through convergent cultural evolution. We compare continuer tokens in linguistically diverse conversational corpora and find languages make available highly similar forms. We then approach the causal mechanism of convergent cultural evolution using exemplar modelling, simulating the process by which a combination of effort minimization and functional specialization may push continuers to a particular region of phonological possibility space. By combining comparative linguistics and computational modelling we shed new light on the question of how language structure is shaped by and for social interaction.

Abstract

Informal social interaction is the primordial home of human language. Linguistically diverse conversational corpora are an important and largely untapped resource for computational linguistics and language technology. Through the efforts of a worldwide language documentation movement, such corpora are increasingly becoming available. We show how interactional data from 63 languages (26 families) harbours insights about turn-taking, timing, sequential structure and social action, with implications for language technology, natural language understanding, and the design of conversational interfaces. Harnessing linguistically diverse conversational corpora will provide the empirical foundations for flexible, localizable, humane language technologies of the future.

Abstract

There is consensus that the adult lexicon exhibits lexical competition. In particular, substantial evidence demonstrates that words with more phonologically similar neighbors are recognized less efficiently than words with fewer neighbors. How and when these effects emerge in the child's lexicon is less clear. In the current paper, we build on previous research by testing whether phonological onset density slows lexical access in a large sample of 100 English-acquiring 30-month-olds. The children participated in a visual world looking-while-listening task, in which their attention was directed to one of two objects on a computer screen while their eye movements were recorded. We found moderate evidence of inhibitory effects of onset neighborhood density on lexical access and clear evidence for an interaction between onset neighborhood density and vocabulary, with larger effects of onset neighborhood density for children with larger vocabularies. Results suggest the lexicons of 30-month-olds exhibit lexical-level competition, with competition increasing with vocabulary size.

Abstract

How children learn grammar is one of the most fundamental questions in cognitive science. Two theoretical accounts, namely, the Early Abstraction and Usage-Based accounts, propose competing answers to this question. To compare the predictions of these accounts, we tested the comprehension of 92 24-month old children of transitive sentences with novel verbs (e.g., “The boy is gorping the girl!”) with the Intermodal Preferential Looking (IMPL) task. We found very little evidence that children looked to the target video at above-chance levels. Using mixed and mixture models, we tested the predictions the two accounts make about: (i) the structure of individual differences in the IMPL task and (ii) the relationship between vocabulary knowledge, lexical processing, and performance in the IMPL task. However, the results did not strongly support either of the two accounts. The implications for theories on language acquisition and for tasks developed for examining individual differences are discussed.

Abstract

Children acquire language embedded within the rich social context of interaction. This paper reports on a longitudinal study investigating the developmental relationship between conversational turn‐taking and vocabulary growth in English‐acquiring children (N = 122) followed between 9 and 24 months. Daylong audio recordings obtained every 3 months provided several indices of the language environment, including the number of adult words children heard in their environment and their number of conversational turns. Vocabulary was measured independently via parental report. Growth curve analyses revealed a bidirectional relationship between conversational turns and vocabulary growth, controlling for the amount of words in children’s environments. The results are consistent with theoretical approaches that identify social interaction as a core component of early language acquisition.

Abstract

Relationships among laurasiatherian clades represent one of the most highly disputed topics in mammalian phylogeny. In this study, we attempt to disentangle laurasiatherian interordinal relationships using two independent genome-level approaches: (1) quantifying retrotransposon presence/absence patterns, and (2) comparisons of exon datasets at the levels of nucleotides and amino acids. The two approaches revealed contradictory phylogenetic signals, possibly due to a high level of ancestral incomplete lineage sorting. The positions of Eulipotyphla and Chiroptera as the first and second earliest divergences were consistent across the approaches. However, the phylogenetic relationships of Perissodactyla, Cetartiodactyla, and Ferae, were contradictory. While retrotransposon insertion analyses suggest a clade with Cetartiodactyla and Ferae, the exon dataset favoured Cetartiodactyla and Perissodactyla. Future analyses of hitherto unsampled laurasiatherian lineages and synergistic analyses of retrotransposon insertions, exon and conserved intron/intergenic sequences might unravel the conflicting patterns of relationships in this major mammalian clade.

Abstract

People readily generalize knowledge to novel domains and stimuli. We present a theory, instantiated in a computational model, based on the idea that cross-domain generalization in humans is a case of analogical inference over structured (i.e., symbolic) relational representations. The model is an extension of the Learning and Inference with Schemas and Analogy (LISA; Hummel & Holyoak, 1997, 2003) and Discovery of Relations by Analogy (DORA; Doumas et al., 2008) models of relational inference and learning. The resulting model learns both the content and format (i.e., structure) of relational representations from nonrelational inputs without supervision, when augmented with the capacity for reinforcement learning it leverages these representations to learn about individual domains, and then generalizes to new domains on the first exposure (i.e., zero-shot learning) via analogical inference. We demonstrate the capacity of the model to learn structured relational representations from a variety of simple visual stimuli, and to perform cross-domain generalization between video games (Breakout and Pong) and between several psychological tasks. We demonstrate that the model’s trajectory closely mirrors the trajectory of children as they learn about relations, accounting for phenomena from the literature on the development of children’s reasoning and analogy making. The model’s ability to generalize between domains demonstrates the flexibility afforded by representing domains in terms of their underlying relational structure, rather than simply in terms of the statistical relations between their inputs and outputs.

Abstract

How a system represents information tightly constrains the kinds of problems it can solve. Humans routinely solve problems that appear to require abstract representations of stimulus properties and relations. How we acquire such representations has central importance in an account of human cognition. We briefly describe a theory of how a system can learn invariant responses to instances of similarity and relative magnitude, and how structured, relational representations can be learned from initially unstructured inputs. Two operations, comparing distributed representations and learning from the concomitant network dynamics in time, underpin the ability to learn these representations and to respond to invariance in the environment. Comparing analog representations of absolute magnitude produces invariant signals that carry information about similarity and relative magnitude. We describe how a system can then use this information to bootstrap learning structured (i.e., symbolic) concepts of relative magnitude from experience without assuming such representations a priori.

Abstract

Reading and writing are crucial life skills but roughly one in ten children are affected by dyslexia, which can persist into adulthood. Family studies of dyslexia suggest heritability up to 70%, yet few convincing genetic markers have been found. Here we performed a genome-wide association study of 51,800 adults self-reporting a dyslexia diagnosis and 1,087,070 controls and identified 42 independent genome-wide significant loci: 15 in genes linked to cognitive ability/educational attainment, and 27 new and potentially more specific to dyslexia. We validated 23 loci (13 new) in independent cohorts of Chinese and European ancestry. Genetic etiology of dyslexia was similar between sexes, and genetic covariance with many traits was found, including ambidexterity, but not neuroanatomical measures of language-related circuitry. Dyslexia polygenic scores explained up to 6% of variance in reading traits, and might in future contribute to earlier identification and remediation of dyslexia.

Abstract

Double verb constructions known as hendiadys have been studied primarily in literary texts and corpora of written language. Much less is known about their properties and usage in spoken language, where expressions such as ‘come and see’, ‘go and tell’, ‘sit and talk’ are particularly common, and where we can find an even richer diversity of other constructions. In this study, we investigate hendiadys in corpora of naturally occurring social interactions in four languages, Danish, English (US and UK), Finnish and Italian, with the objective of exploring whether hendiadys is used systematically in recurrent interactional and sequential circumstances, from which it is possible to identify the pragmatic function(s) that hendiadys may serve. Examining hendiadys in conversation also offers us a special window into its grammatical properties, for example when a speaker self-corrects from a non-hendiadic to a hendiadic expression, exposing the boundary between related grammatical forms and demonstrating the distinctiveness of hendiadys in context. More broadly, we demonstrate that hendiadys is systematically associated with talk about complainable matters, in environments characterised by a conflict, dissonance, or friction that is ongoing in the interaction or that is being reported by one participant to another. We also find that the utterance in which hendiadys is used is typically in a subsequent and possibly terminal position in the sequence, summarising or concluding it. Another key finding is that the complainable or conflictual element in these interactions is expressed primarily by the first conjunct of the hendiadic construction. Whilst the first conjunct is semantically subsidiary to the second, it is pragmatically the most important one. This analysis leads us to revisit a long-established asymmetry between the verbal components of hendiadys, and to bring to light the synergy of grammar and pragmatics in language usage.

Abstract

During communication in real-life settings, the brain integrates information from auditory and visual modalities to form a unified percept of our environment. In the current magnetoencephalography (MEG) study, we used rapid invisible frequency tagging (RIFT) to generate steady-state evoked fields and investigated the integration of audiovisual information in a semantic context. We presented participants with videos of an actress uttering action verbs (auditory; tagged at 61 Hz) accompanied by a gesture (visual; tagged at 68 Hz, using a projector with a 1440 Hz refresh rate). Integration ease was manipulated by auditory factors (clear/degraded speech) and visual factors (congruent/incongruent gesture). We identified MEG spectral peaks at the individual (61/68 Hz) tagging frequencies. We furthermore observed a peak at the intermodulation frequency of the auditory and visually tagged signals (fvisual – fauditory = 7 Hz), specifically when integration was easiest (i.e., when speech was clear and accompanied by a congruent gesture). This intermodulation peak is a signature of nonlinear audiovisual integration, and was strongest in left inferior frontal gyrus and left temporal regions; areas known to be involved in speech-gesture integration. The enhanced power at the intermodulation frequency thus reflects the ease of integration and demonstrates that speech-gesture information interacts in higher-order language areas. Furthermore, we provide a proof-of-principle of the use of RIFT to study the integration of audiovisual stimuli, in relation to, for instance, semantic context.

Abstract

We here demonstrate that face-to-face spatial orientation induces a special ‘social mode’ for neurocognitive processing during conversation, even in the absence of visibility. Participants conversed face-to-face, face-to-face but visually occluded, and back-to-back to tease apart effects caused by seeing visual communicative signals and by spatial orientation. Using dual-EEG, we found that 1) listeners’ brains engaged more strongly while conversing in face-to-face than back-to-back, irrespective of the visibility of communicative signals, 2) listeners attended to speech more strongly in a back-to-back compared to a face-to-face spatial orientation without visibility; visual signals further reduced the attention needed; 3) the brains of interlocutors were more in sync in a face-to-face compared to a back-to-back spatial orientation, even when they could not see each other; visual signals further enhanced this pattern. Communicating in face-to-face spatial orientation is thus sufficient to induce a special ‘social mode’ which fine-tunes the brain for neurocognitive processing in conversation.

Abstract

This paper summarizes the results of a descriptive syntactic category analysis of child English no which reveals that young children use and represent no as a determiner and negatives like no pen as NPs, contra standard analyses.

Abstract

The number of individuals affected by traumatic brain injury (TBI) is growing globally. TBIs may cause a range of physical, cognitive, and psychiatric deficits that can negatively impact employment, academic attainment, community independence, and interpersonal relationships. Although there has been a significant decrease in the number of injury related deaths over the past several decades, there has been no corresponding reduction in injury related disability over the same time period. We propose that patient registries with large, representative samples and rich multidimensional and longitudinal data have tremendous value in advancing basic and translational research and in capturing, characterizing, and predicting individual differences in deficit profile and outcomes. Patient registries, together with recent theoretical and methodological advances in analytic approaches and neuroscience, provide powerful tools for brain injury research and for leveraging the heterogeneity that has traditionally been cited as a barrier inhibiting progress in treatment research and clinical practice. We report on our experiences, and challenges, in developing and maintaining our own patient registry. We conclude by pointing to some future opportunities for discovery that are afforded by a registry model.

Abstract

Motricity is the most commonly affected ability after a stroke. While many clinical studies attempt to predict motor symptoms at different chronic time points after a stroke, longitudinal acute-to-chronic studies remain scarce. Taking advantage of recent advances in mapping brain disconnections, we predict motor outcomes in 62 patients assessed longitudinally two weeks, three months, and one year after their stroke. Results indicate that brain disconnection patterns accurately predict motor impairments. However, disconnection patterns leading to impairment differ between the three-time points and between left and right motor impairments. These results were cross-validated using resampling techniques. In sum, we demonstrated that while some neuroplasticity mechanisms exist changing the structure–function relationship, disconnection patterns prevail when predicting motor impairment at different time points after stroke.

Abstract

Understanding cetacean distribution is essential for conservation planning and decision-making, particularly in regions subject to rapid environmental changes. Nevertheless, information on their spatiotemporal distribution is commonly limited, especially from remote areas. Species distribution models (SDMs) are powerful tools, relating species occurrences to environmental variables to predict the species' potential distribution. This study aims at using presence-only SDMs (MaxEnt) to identify suitable habitats for fin whales (Balaenoptera physalus) on their Nordic and Barents Seas feeding grounds. We used spatial-block cross-validation to tune MaxEnt parameters and evaluate model performance using spatially independent testing data. We considered spatial sampling bias correction using four methods. Important environmental variables were distance to shore and sea ice edge, variability of sea surface temperature and sea surface salinity, and depth. Suitable fin whale habitats were predicted along the west coast of Svalbard, between Svalbard and the eastern Norwegian Sea, coastal areas off Iceland and southern East Greenland, and along the Knipovich Ridge to Jan Mayen. Results support that presence-only SDMs are effective tools to predict cetacean habitat suitability, particularly in remote areas like the Arctic Ocean. SDMs constitute a cost-effective method for targeting future surveys and identifying top priority sites for conservation measures.

Abstract

This paper examines the Papuan languages of Island Melanesia, with a view to considering their typological similarities and differences. The East Papuan languages are thought to be the descendants of the languages spoken by the original inhabitants of Island Melanesia, who arrived in the area up to 50,000 years ago. The Oceanic Austronesian languages are thought to have come into the area with the Lapita peoples 3,500 years ago. With this historical backdrop in view, our paper seeks to investigate the linguistic relationships between the scattered Papuan languages of Island Melanesia. To do this, we survey various structural features, including syntactic patterns such as constituent order in clauses and noun phrases and other features of clause structure, paradigmatic structures of pronouns, and the structure of verbal morphology. In particular, we seek to discern similarities between the languages that might call for closer investigation, with a view to establishing genetic relatedness between some or all of the languages. In addition, in examining structural relationships between languages, we aim to discover whether it is possible to distinguish between original Papuan elements and diffused Austronesian elements of these languages. As this is a vast task, our paper aims merely to lay the groundwork for investigation into these and related questions.

Abstract

Rhythmic neural activity synchronizes with certain rhythmic behaviors, such as breathing, sniffing, saccades, and speech. The extent to which neural oscillations synchronize with higher-level and more complex behaviors is largely unknown. Here we investigated electrophysiological synchronization with keyboard typing, which is an omnipresent behavior daily engaged by an uncountably large number of people. Keyboard typing is rhythmic with frequency characteristics roughly the same as neural oscillatory dynamics associated with cognitive control, notably through midfrontal theta (4 -7 Hz) oscillations. We tested the hypothesis that synchronization occurs between typing and midfrontal theta, and breaks down when errors are committed. Thirty healthy participants typed words and sentences on a keyboard without visual feedback, while EEG was recorded. Typing rhythmicity was investigated by inter-keystroke interval analyses and by a kernel density estimation method. We used a multivariate spatial filtering technique to investigate frequency-specific synchronization between typing and neuronal oscillations. Our results demonstrate theta rhythmicity in typing (around 6.5 Hz) through the two different behavioral analyses. Synchronization between typing and neuronal oscillations occurred at frequencies ranging from 4 to 15 Hz, but to a larger extent for lower frequencies. However, peak synchronization frequency was idiosyncratic across subjects, therefore not specific to theta nor to midfrontal regions, and correlated somewhat with peak typing frequency. Errors and trials associated with stronger cognitive control were not associated with changes in synchronization at any frequency. As a whole, this study shows that brain-behavior synchronization does occur during keyboard typing but is not specific to midfrontal theta.

Abstract

By the end of their first year, infants can interpret many different types of complex dynamic visual events, such as caused-motion, chasing, and goal-directed action. Infants of this age are also in the early stages of vocabulary development, producing their first words at around 12 months. The present work examined whether there are meaningful individual differences in infants’ ability to represent dynamic causal events in visual scenes, and whether these differences influence vocabulary development. As part of the longitudinal Language 0–5 Project, 78 10-month-old infants were tested on their ability to interpret three dynamic motion events, involving (a) caused-motion, (b) chasing behaviour, and (c) goal-directed movement. Planned analyses found that infants showed evidence of understanding the first two event types, but not the third. Looking behaviour in each task was not meaningfully related to vocabulary development, nor were there any correlations between the tasks. The results of additional exploratory analyses and simulations suggested that the infants’ understanding of each event may not be predictive of their vocabulary development, and that looking times in these tasks may not be reliably capturing any meaningful individual differences in their knowledge. This raises questions about how to convert experimental group designs to individual differences measures, and how to interpret infant looking time behaviour.

Abstract

This article explores the relationship between low- and high-level aspects of reading by studying the interplay between word processing, as measured with eye tracking, and narrative absorption and liking, as measured with questionnaires. Specifically, we focused on how individual differences in sensitivity to lexical word characteristics—measured as the effect of these characteristics on gaze duration—were related to narrative absorption and liking. By reanalyzing a large data set consisting of three previous eye-tracking experiments in which subjects (N = 171) read literary short stories, we replicated the well-established finding that word length, lemma frequency, position in sentence, age of acquisition, and orthographic neighborhood size of words influenced gaze duration. More importantly, we found that individual differences in the degree of sensitivity to three of these word characteristics, i.e., word length, lemma frequency, and age of acquisition, were negatively related to print exposure and to a lesser degree to narrative absorption and liking. Even though the underlying mechanisms of this relationship are still unclear, we believe the current findings underline the need to map out the interplay between, on the one hand, the technical and, on the other hand, the subjective processes of reading by studying reading behavior in more natural settings.

Abstract

It is often argued that narratives improve social cognition, either by appealing to social-cognitive abilities as we engage with the story world and its characters, or by conveying social knowledge. Empirical studies have found support for both a correlational and a causal link between exposure to (literary, fictional) narratives and social cognition. However, a series of failed replications has cast doubt on the robustness of these claims. Here, we review the existing empirical literature and identify open questions and challenges. An important conclusion of the review is that previous research has given too little consideration to the diversity of narratives, readers, and social-cognitive processes involved in the social-cognitive potential of narratives. We therefore establish a research agenda, proposing that future research should focus on (1) the specific text characteristics that drive the social-cognitive potential of narratives, (2) the individual differences between readers with respect to their sensitivity to this potential, and (3) the various aspects of social cognition that are potentially affected by reading narratives. Our recommendations can guide the design of future studies that will help us understand how, for whom, and in what respect exposure to narratives can advantage social cognition.

Abstract

Although various studies have shown that narrative reading draws on social-cognitive abilities, not much is known about the precise aspects of narrative processing that engage these abilities. We hypothesized that the linguistic processing of narrative viewpoint—expressed by elements that provide access to the inner world of characters—might play an important role in engaging social-cognitive abilities. Using eye tracking, we studied the effect of lexical markers of perceptual, cognitive, and emotional viewpoint on eye movements during reading of a 5,000-word narrative. Next, we investigated how this relationship was modulated by individual differences in social-cognitive abilities. Our results show diverging patterns of eye movements for perceptual viewpoint markers on the one hand, and cognitive and emotional viewpoint markers on the other. Whereas the former are processed relatively fast compared to non-viewpoint markers, the latter are processed relatively slow. Moreover, we found that social-cognitive abilities impacted the processing of words in general, and of perceptual and cognitive viewpoint markers in particular, such that both perspective-taking abilities and self-reported perspective-taking traits facilitated the processing of these markers. All in all, our study extends earlier findings that social cognition is of importance for story reading, showing that individual differences in social-cognitive abilities are related to the linguistic processing of narrative viewpoint.

Abstract

We present a dataset of behavioural and fMRI observations acquired in the context of humans involved in multimodal referential communication. The dataset contains audio/video and motion-tracking recordings of face-to-face, task-based communicative interactions in Dutch, as well as behavioural and neural correlates of participants’ representations of dialogue referents. Seventy-one pairs of unacquainted participants performed two interleaved interactional tasks in which they described and located 16 novel geometrical objects (i.e., Fribbles) yielding spontaneous interactions of about one hour. We share high-quality video (from three cameras), audio (from head-mounted microphones), and motion-tracking (Kinect) data, as well as speech transcripts of the interactions. Before and after engaging in the face-to-face communicative interactions, participants’ individual representations of the 16 Fribbles were estimated. Behaviourally, participants provided a written description (one to three words) for each Fribble and positioned them along 29 independent conceptual dimensions (e.g., rounded, human, audible). Neurally, fMRI signal evoked by each Fribble was measured during a one-back working-memory task. To enable functional hyperalignment across participants, the dataset also includes fMRI measurements obtained during visual presentation of eight animated movies (35 minutes total). We present analyses for the various types of data demonstrating their quality and consistency with earlier research. Besides high-resolution multimodal interactional data, this dataset includes different correlates of communicative referents, obtained before and after face-to-face dialogue, allowing for novel investigations into the relation between communicative behaviours and the representational space shared by communicators. This unique combination of data can be used for research in neuroscience, psychology, linguistics, and beyond.

Abstract

The use of spoken and written language is a fundamental human capacity. Individual differences in reading- and language-related skills are influenced by genetic variation, with twin-based heritability estimates of 30 to 80% depending on the trait. The genetic architecture is complex, heterogeneous, and multifactorial, but investigations of contributions of single-nucleotide polymorphisms (SNPs) were thus far underpowered. We present a multicohort genome-wide association study (GWAS) of five traits assessed individually using psychometric measures (word reading, nonword reading, spelling, phoneme awareness, and nonword repetition) in samples of 13,633 to 33,959 participants aged 5 to 26 y. We identified genome-wide significant association with word reading (rs11208009, P = 1.098 × 10−8) at a locus that has not been associated with intelligence or educational attainment. All five reading-/language-related traits showed robust SNP heritability, accounting for 13 to 26% of trait variability. Genomic structural equation modeling revealed a shared genetic factor explaining most of the variation in word/nonword reading, spelling, and phoneme awareness, which only partially overlapped with genetic variation contributing to nonword repetition, intelligence, and educational attainment. A multivariate GWAS of word/nonword reading, spelling, and phoneme awareness maximized power for follow-up investigation. Genetic correlation analysis with neuroimaging traits identified an association with the surface area of the banks of the left superior temporal sulcus, a brain region linked to the processing of spoken and written language. Heritability was enriched for genomic elements regulating gene expression in the fetal brain and in chromosomal regions that are depleted of Neanderthal variants. Together, these results provide avenues for deciphering the biological underpinnings of uniquely human traits.

Abstract

The most basic question for the study of morphology and the mental lexicon is whether or not words are _decomposed_: informally, this is the question of whether words are represented (and processed) in terms of some kind of smaller units; that is, broken down into constituent parts. Formally, what it means to represent or process a word as decomposed or not turns out to be quite complex. One of the basic lines of division in the field classifies approaches according to whether they decompose all “complex” words (“Full Decomposition”), or none (“Full Listing”), or some but not all, according to some criterion (typical of “Dual-Route” models). However, if we are correct, there are at least three senses in which an approach might be said to be decompositional or not, with the result that ongoing discussions of what appears to be a single large issue might not always be addressing the same distinction. Put slightly differently, there is no single question of decomposition. Instead, there are independent but related questions that define current research. Our goal here is to identify this finer-grained set of questions, as they are the ones that should assume a central place in the study of morphological and lexical representation.

Abstract

Language is a uniquely human trait likely to have been a prerequisite for the development of human culture. The ability to develop articulate speech relies on capabilities, such as fine control of the larynx and mouth, that are absent in chimpanzees and other great apes. FOXP2 is the first gene relevant to the human ability to develop language. A point mutation in FOXP2 co-segregates with a disorder in a family in which half of the members have severe articulation difficulties accompanied by linguistic and grammatical impairment. This gene is disrupted by translocation in an unrelated individual who has a similar disorder. Thus, two functional copies of FOXP2 seem to be required for acquisition of normal spoken language. We sequenced the complementary DNAs that encode the FOXP2 protein in the chimpanzee, gorilla, orang-utan, rhesus macaque and mouse, and compared them with the human cDNA. We also investigated intraspecific variation of the human FOXP2 gene. Here we show that human FOXP2 contains changes in amino-acid coding and a pattern of nucleotide polymorphism, which strongly suggest that this gene has been the target of selection during recent human evolution.

Abstract

Investigation of the emotions entails reference to words and expressions conventionally used for the description of emotion experience. Important methodological issues arise for emotion researchers, and the issues are of similarly central concern in linguistic semantics more generally. I argue that superficial and/or inconsistent description of linguistic meaning can have seriously misleading results. This paper is firstly a critique of standards in emotion research for its tendency to underrate and ill-understand linguistic semantics. It is secondly a critique of standards in some approaches to linguistic semantics itself. Two major problems occur. The first is failure to distinguish between conceptually distinct meanings of single words, neglecting the well-established fact that a single phonological string can signify more than one conceptual category (i.e., that words can be polysemous). The second error involves failure to distinguish between two kinds of secondary uses of words: (1) those which are truly active “online” extensions, and (2) those which are conventionalised secondary meanings and not active (qua “extensions”) at all. These semantic considerations are crucial to conclusions one may draw about cognition and conceptualisation based on linguistic evidence.

Abstract

This article argues that it is not possible to establish distinctions between 'Lao', 'Thai', and 'Isan' as seperate languages or dialects by appealing to objective criteria. 'Lao', 'Thai', and 'Isan' are conceived linguistics varieties, and the ground-level reality reveals a great deal of variation, much of it not coinciding with the geographical boundaries of the 'Laos', 'Isan', and 'non-Isan Thailand' areas. Those who promote 'Lao', 'Thai', and/or 'Isan' as distinct linguistic varieties have subjective (e.g. political and/or sentimental) reasons for doing so. Objective linguistic criteria are not sufficient

Abstract

Anthropologists and linguists have long been aware that the body is explicitly referred to in conventional description of emotion in languages around the world. There is abundant linguistic data showing expression of emotions in terms of their imagined ‘locus’ in the physical body. The most important methodological issue in the study of emotions is language, for the ways people talk give us access to ‘folk descriptions’ of the emotions. ‘Technical terminology’, whether based on English or otherwise, is not excluded from this ‘folk’ status. It may appear to be safely ‘scientific’ and thus culturally neutral, but in fact it is not: technical English is a variety of English and reflects, to some extent, culture-specific ways of thinking (and categorising) associated with the English language. People — as researchers studying other people, or as people in real-life social association — cannot directly access the emotional experience of others, and language is the usual mode of ‘packaging’ one’s experience so it may be accessible to others. Careful description of linguistic data from as broad as possible a cross-linguistic base is thus an important part of emotion research. All people experience biological events and processes associated with certain thoughts (or, as psychologists say, ‘appraisals’), but there is more to ‘emotion’ than just these physiological phenomena. Speakers of some languages talk about their emotional experiences as if they are located in some internal organ such as ‘the liver’, yet they cannot localise feeling in this physical organ. This phenomenon needs to be understood better, and one of the problems is finding a method of comparison that allows us to compare descriptions from different languages which show apparently great formal and semantic variation. Some simple concepts including feel and body are universal or near-universal, and as such are good candidates for terms of description which may help to eradicate confusion and exoticism from cross-linguistic comparison and semantic typology. Semantic analysis reveals great variation in concepts of emotion across languages and cultures — but such analysis requires a sound and well-founded methodology. While leaving room for different approaches to the task, we suggest that such a methodology can be based on empirically established linguistic universal (or near-universal) concepts, and on ‘cognitive scenarios’ articulated in terms of these concepts. Also, we warn against the danger of exoticism involved in taking all body part references ‘literally’. Above all, we argue that what is needed is a combination of empirical cross-linguistic investigations and a theoretical and methodological awareness, recognising the impossibility of exploring other people’s emotions without keeping language in focus: both as an object and as a tool of study.

Abstract

This article addresses the recognition of reduced word forms, which are frequent in casual speech. We describe two experiments on Dutch showing that listeners only recognize highly reduced forms well when these forms are presented in their full context and that the probability that a listener recognizes a word form in limited context is strongly correlated with the degree of reduction of the form. Moreover, we show that the effect of degree of reduction can only partly be interpreted as the effect of the intelligibility of the acoustic signal, which is negatively correlated with degree of reduction. We discuss the consequences of our findings for models of spoken word recognition and especially for the role that storage plays in these models.

Abstract

Anaphora, a ubiquitous feature of natural language, poses a particular challenge to young children as they first learn language due to its referential ambiguity. In spite of this, parents and caregivers use anaphora frequently in child-directed speech, potentially presenting a risk to effective communication if children do not yet have the linguistic capabilities of resolving anaphora successfully. Through an eye-tracking study in a naturalistic free-play context, we examine the strategies that parents employ to calibrate their use of anaphora to their child's linguistic development level. We show that, in this way, parents are able to intuitively scaffold the complexity of their speech such that greater referential ambiguity does not hurt overall communication success.

Abstract

Some theorists posit the existence of a ‘core’ grammar that virtually all native speakers acquire, and a ‘peripheral’ grammar that many do not. We investigated the viability of such a categorical distinction in the Dutch language. We first consulted linguists’ intuitions as to the ‘core’ or ‘peripheral’ status of a wide range of grammatical structures. We then tested a selection of core- and peripheral-rated structures on naïve participants with varying levels of literacy experience, using grammaticality judgment as a proxy for receptive knowledge. Overall, participants demonstrated better knowledge of ‘core’ structures than ‘peripheral’ structures, but the considerable variability within these categories was strongly suggestive of a continuum rather than a categorical distinction between them. We also hypothesised that individual differences in the knowledge of core and peripheral structures would reflect participants’ literacy experience. This was supported only by a small trend in our data. The results fit best with the notion that more frequent syntactic structures are mastered by more people than infrequent ones and challenge the received sense of a categorical core-periphery distinction.

Abstract

Language comprehenders can use syntactic cues to generate predictions online about upcoming language. Previous research with reading-impaired adults and healthy, low-proficiency adult and child learners suggests that reading skills are related to prediction in spoken language comprehension. Here we investigated whether differences in literacy are also related to predictive spoken language processing in non-reading-impaired proficient adult readers with varying levels of literacy experience. Using the visual world paradigm enabled us to measure prediction based on syntactic cues in the spoken sentence, prior to the (predicted) target word. Literacy experience was found to be the strongest predictor of target anticipation, independent of general cognitive abilities. These findings suggest that a) experience with written language can enhance syntactic prediction of spoken language in normal adult language users, and b) processing skills can be transferred to related tasks (from reading to listening) if the domains involve similar processes (e.g., predictive dependencies) and representations (e.g., syntactic).

Abstract

‘Book language’ offers a richer linguistic experience than typical conversational speech in terms of its syntactic properties. Here, we investigated the role of long-term syntactic experience on syntactic knowledge and processing. In a pre-registered study with 161 adult native Dutch speakers with varying levels of literacy, we assessed the contribution of individual differences in written language experience to offline and online syntactic processes. Offline syntactic knowledge was assessed as accuracy in an auditory grammaticality judgment task in which we tested violations of four Dutch grammatical norms. Online syntactic processing was indexed by syntactic priming of the Dutch dative alternation, using a comprehension-to-production priming paradigm with auditory presentation. Controlling for the contribution of non-verbal IQ, verbal working memory, and processing speed, we observed a robust effect of literacy experience on the detection of grammatical norm violations in spoken sentences, suggesting that exposure to the syntactic complexity and diversity of written language has specific benefits for general (modality-independent) syntactic knowledge. We replicated previous results by finding robust comprehension-to-production structural priming, both with and without lexical overlap between prime and target. Although literacy experience affected the usage of syntactic alternates in our large sample, it did not modulate their priming. We conclude that amount of experience with written language increases explicit awareness of grammatical norm violations and changes the usage of (PO vs. DO) dative spoken sentences but has no detectable effect on their implicit syntactic priming in proficient language users. These findings constrain theories about the effect of long-term experience on syntactic processing.

Abstract

Perceptual learning of novel accents is a critical skill for second-language speech perception, but little is known about the mechanisms that facilitate perceptual learning in communicative contexts. To study perceptual learning in an interactive dialogue setting while maintaining experimental control of the phonetic input, we employed an innovative experimental method incorporating prerecorded speech into a naturalistic conversation. Using both computer-based and face-to-face dialogue settings, we investigated the effect of two types of learning mechanisms in interaction: explicit corrective feedback and implicit lexical guidance. Dutch participants played an information-gap game featuring minimal pairs with an accented English speaker whose /ε/ pronunciations were shifted to /ɪ/. Evidence for the vowel shift came either from corrective feedback about participants’ perceptual mistakes or from onscreen lexical information that constrained their interpretation of the interlocutor’s words. Corrective feedback explicitly contrasting the minimal pairs was more effective than generic feedback. Additionally, both receiving lexical guidance and exhibiting more uptake for the vowel shift improved listeners’ subsequent online processing of accented words. Comparable learning effects were found in both the computer-based and face-to-face interactions, showing that our results can be generalized to a more naturalistic learning context than traditional computer-based perception training programs.

Abstract

Learning a script with mirrored graphs (e.g., d ≠ b) requires overcoming the evolutionary-old perceptual tendency to process mirror images as equivalent. Thus, breaking mirror invariance offers an important tool for understanding cultural re-shaping of evolutionarily ancient cognitive mechanisms. Here we investigated the role of script (i.e., presence vs. absence of mirrored graphs: Latin alphabet vs. Tamil) by revisiting mirror-image processing by illiterate, Tamil monoliterate, and Tamil-Latin-alphabet bi-literate adults. Participants performed two same-different tasks (one orientation-based, another shape-based) on Latin-alphabet letters. Tamil monoliterate were significantly better than illiterate and showed good explicit mirror-image discrimination. However, only bi-literate adults fully broke mirror invariance: slower shape-based judgments for mirrored than identical pairs and reduced disadvantage in orientation-based over shape-based judgments of mirrored pairs. These findings suggest learning a script with mirrored graphs is the strongest force for breaking mirror invariance.

Abstract

To form a percept of the multisensory world, the brain needs to integrate signals from common sources weighted by their reliabilities and segregate those from independent sources. Previously, we have shown that anterior parietal cortices combine sensory signals into representations that take into account the signals’ causal structure (i.e., common versus independent sources) and their sensory reliabilities as predicted by Bayesian causal inference. The current study asks to what extent and how attentional mechanisms can actively control how sensory signals are combined for perceptual inference. In a pre- and postcueing paradigm, we presented observers with audiovisual signals at variable spatial disparities. Observers were precued to attend to auditory or visual modalities prior to stimulus presentation and postcued to report their perceived auditory or visual location. Combining psychophysics, functional magnetic resonance imaging (fMRI), and Bayesian modelling, we demonstrate that the brain moulds multisensory inference via two distinct mechanisms. Prestimulus attention to vision enhances the reliability and influence of visual inputs on spatial representations in visual and posterior parietal cortices. Poststimulus report determines how parietal cortices flexibly combine sensory estimates into spatial representations consistent with Bayesian causal inference. Our results show that distinct neural mechanisms control how signals are combined for perceptual inference at different levels of the cortical hierarchy.

Abstract

The brain has the extraordinary capacity to construct predictive models of the environment by internalizing statistical regularities in the sensory inputs. The resulting sensory expectations shape how we perceive and react to the world; at the neural level, this relates to decreased neural responses to expected than unexpected stimuli (‘expectation suppression’). Crucially, expectations may need revision as context changes. However, existing research has often neglected this issue. Further, it is unclear whether contextual revisions apply selectively to expectations relevant to the task at hand, hence serving adaptive behaviour. The present fMRI study examined how contextual visual expectations spread throughout the cortical hierarchy as participants update their beliefs. We created a volatile environment with two state spaces presented over separate contexts and controlled by an independent contextualizing signal. Participants attended a training session before scanning to learn contextual temporal associations among pairs of object images. The fMRI experiment then tested for the emergence of contextual expectation suppression in two separate tasks, respectively with task-relevant and task-irrelevant expectations. Behavioural and neural effects of contextual expectation emerged progressively across the cortical hierarchy as participants attuned themselves to the context: expectation suppression appeared first in the insula, inferior frontal gyrus and posterior parietal cortex, followed by the ventral visual stream, up to early visual cortex. This applied selectively to task-relevant expectations. Taken together, the present results suggest that an insular and frontoparietal executive control network may guide the flexible deployment of contextual sensory expectations for adaptive behaviour in our complex and dynamic world.

Abstract

Dance is ubiquitous among humans and has received attention from several disciplines. Ethnographic documentation suggests that dance has a signaling function in social interaction. It can influence mate preferences and facilitate social bonds. Research has provided insights into the proximate mechanisms of dance, individually or when dancing with partners or in groups. Here, we review dance research from an evolutionary perspective. We propose that human dance evolved from ordinary (non-communicative) movements to communicate socially relevant information accurately. The need for accurate social signaling may have accompanied increases in group size and population density. Because of its complexity in production and display, dance may have evolved as a vehicle for expressing social and cultural information. Mating-related qualities and motives may have been the predominant information derived from individual dance movements, whereas group dance offers the opportunity for the exchange of socially relevant content, for coordinating actions among group members, for signaling coalitional strength, and for stabilizing group structures. We conclude that, despite the cultural diversity in dance movements and contexts, the primary communicative functions of dance may be the same across societies.

Abstract

Determining when a partner’s spoken or musical turn will end requires well-honed predictive abilities. Evidence suggests that our motor systems are activated during perception of both speech and music, and it has been argued that motor simulation is used to predict turn-ends across domains. Here we used a dual-task interference paradigm to investigate whether motor simulation of our partner’s action underlies our ability to make accurate turn-end predictions in speech and in music. Furthermore, we explored how specific this simulation is to the action being predicted. We conducted two experiments, one investigating speech turn-ends, and one investigating music turn-ends. In each, 34 proficient pianists predicted turn-endings while (1) passively listening, (2) producing an effector-specific motor activity (mouth/hand movement), or (3) producing a task- and effector-specific motor activity (mouthing words/fingering a piano melody). In the speech experiment, any movement during speech perception disrupted predictions of spoken turn-ends, whether the movement was task-specific or not. In the music experiment, only task-specific movement (i.e., fingering a piano melody) disrupted predictions of musical turn-ends. These findings support the use of motor simulation to make turn-end predictions in both speech and music but suggest that the specificity of this simulation may differ between domains.

Abstract

Attention deficit/hyperactivity disorder (ADHD) is a common heritable disorder with a childhood onset. Molecular genetic studies of ADHD have previously focused on examining the roles of specific candidate genes, primarily those involved in dopaminergic pathways. We have performed the first systematic genomewide linkage scan for loci influencing ADHD in 126 affected sib pairs, using a ∼10-cM grid of microsatellite markers. Allele-sharing linkage methods enabled us to exclude any loci with a λs of ⩾3 from 96% of the genome and those with a λs of ⩾2.5 from 91%, indicating that there is unlikely to be a major gene involved in ADHD susceptibility in our sample. Under a strict diagnostic scheme we could exclude all screened regions of the X chromosome for a locus-specific λs of ⩾2 in brother-brother pairs, demonstrating that the excess of affected males with ADHD is probably not attributable to a major X-linked effect. Qualitative trait maximum LOD score analyses pointed to a number of chromosomal sites that may contain genetic risk factors of moderate effect. None exceeded genomewide significance thresholds, but LOD scores were >1.5 for regions on 5p12, 10q26, 12q23, and 16p13. Quantitative-trait analysis of ADHD symptom counts implicated a region on 12p13 (maximum LOD 2.6) that also yielded a LOD >1 when qualitative methods were used. A survey of regions containing 36 genes that have been proposed as candidates for ADHD indicated that 29 of these genes, including DRD4 and DAT1, could be excluded for a λs of 2. Only three of the candidates—DRD5, 5HTT, and CALCYON—coincided with sites of positive linkage identified by our screen. Two of the regions highlighted in the present study, 2q24 and 16p13, coincided with the top linkage peaks reported by a recent genome-scan study of autistic sib pairs.

Abstract

Developmental dyslexia, a specific impairment of reading ability despite adequate intelligence and educational opportunity, is one of the most frequent childhood disorders. Since the first documented cases at the beginning of the last century, it has become increasingly apparent that the reading problems of people with dyslexia form part of a heritable neurobiological syndrome. As for most cognitive and behavioural traits, phenotypic definition is fraught with difficulties and the genetic basis is complex, making the isolation of genetic risk factors a formidable challenge. Against such a background, it is notable that several recent studies have reported the localization of genes that influence dyslexia and other language-related traits. These investigations exploit novel research approaches that are relevant to many areas of human neurogenetics.

Abstract

Developmental dyslexia is defined as a specific and significant impairment in reading ability that cannot be explained by deficits in intelligence, learning opportunity, motivation or sensory acuity. It is one of the most frequently diagnosed disorders in childhood, representing a major educational and social problem. It is well established that dyslexia is a significantly heritable trait with a neurobiological basis. The etiological mechanisms remain elusive, however, despite being the focus of intensive multidisciplinary research. All attempts to map quantitative-trait loci (QTLs) influencing dyslexia susceptibility have targeted specific chromosomal regions, so that inferences regarding genetic etiology have been made on the basis of very limited information. Here we present the first two complete QTL-based genome-wide scans for this trait, in large samples of families from the United Kingdom and United States. Using single-point analysis, linkage to marker D18S53 was independently identified as being one of the most significant results of the genome in each scan (P< or =0.0004 for single word-reading ability in each family sample). Multipoint analysis gave increased evidence of 18p11.2 linkage for single-word reading, yielding top empirical P values of 0.00001 (UK) and 0.0004 (US). Measures related to phonological and orthographic processing also showed linkage at this locus. We replicated linkage to 18p11.2 in a third independent sample of families (from the UK), in which the strongest evidence came from a phoneme-awareness measure (most significant P value=0.00004). A combined analysis of all UK families confirmed that this newly discovered 18p QTL is probably a general risk factor for dyslexia, influencing several reading-related processes. This is the first report of QTL-based genome-wide scanning for a human cognitive trait.

Abstract

Between 2 and 5% of children who are otherwise unimpaired have significant difficulties in acquiring expressive and/or receptive language, despite adequate intelligence and opportunity. While twin studies indicate a significant role for genetic factors in developmental disorders of speech and language, the majority of families segregating such disorders show complex patterns of inheritance, and are thus not amenable for conventional linkage analysis. A rare exception is the KE family, a large three-generation pedigree in which approximately half of the members are affected with a severe speech and language disorder which appears to be transmitted as an autosomal dominant monogenic trait. This family has been widely publicised as suffering primarily from a defect in the use of grammatical suffixation rules, thus supposedly supporting the existence of genes specific to grammar. The phenotype, however, is broader in nature, with virtually every aspect of grammar and of language affected. In addition, affected members have a severe orofacial dyspraxia, and their speech is largely incomprehensible to the naive listener. We initiated a genome-wide search for linkage in the KE family and have identified a region on chromosome 7 which co-segregates with the speech and language disorder (maximum lod score = 6.62 at theta = 0.0), confirming autosomal dominant inheritance with full penetrance. Further analysis of microsatellites from within the region enabled us to fine map the locus responsible (designated SPCH1) to a 5.6-cM interval in 7q31, thus providing an important step towards its identification. Isolation of SPCH1 may offer the first insight into the molecular genetics of the developmental process that culminates in speech and language.

Abstract

This chapter highlights the research in isolating genetic factors underlying specific language impairment (SLI), or developmental dysphasia, which exploits newly developed genotyping technology, novel statistical methodology, and DNA sequence data generated by the Human Genome Project. The author begins with an overview of results from family, twin, and adoption studies supporting genetic involvement and then goes on to outline progress in a number of genetic mapping efforts that have been recently completed or are currently under way. It has been possible for genetic researchers to pinpoint the specific mutation responsible for some speech and language disorders, providing an example of how the availability of human genomic sequence data can greatly accelerate the pace of disease gene discovery. Finally, the author discusses future prospects on how molecular genetics may offer new insight into the etiology underlying speech and language disorders, leading to improvements in diagnosis and treatment.

Abstract

Embodied song practices involve the transformation of songs from the acoustic modality into an embodied-visual form, to increase meaningful access for d/Deaf audiences. This goes beyond the translation of lyrics, by combining poetic sign language with other bodily movements to embody the para-linguistic expressive and musical features that enhance the message of a song. To date, the limited research into this phenomenon has focussed on linguistic features and interactions with rhythm. The relationship between bodily actions and music has not been probed beyond an assumed implication of conformance. However, as the primary objective is to communicate equivalent meanings, the ways that the acoustic and embodied-visual signals relate to each other should reveal something about underlying conceptual agreement. This paper draws together a range of pertinent theories from within a grounded cognition framework including semiotics, analogy mapping and cross-modal correspondences. These theories are applied to embodiment strategies used by prominent d/Deaf and hearing Dutch practitioners, to unpack the relationship between acoustic songs, their embodied representations, and their broader conceptual and affective meanings. This leads to the proposition that meaning primarily arises through shared patterns of internal relations across a range of amodal and cross-modal features with an emphasis on dynamic qualities. These analogous patterns can inform metaphorical interpretations and trigger shared emotional responses. This exploratory survey offers insights into the nature of cross-modal and embodied meaning-making, as a jumping-off point for further research.

Abstract

Like many Indigenous Australian languages, Murrinhpatha has flexible word order with no apparent configurational syntax. We analyzed an experimental corpus of Murrinhpatha utterances for associations between different thematic role orders, intonational phrasing patterns and pitch downtrends. We found that initial constituents (Agents or Patients) tend to carry the highest pitch targets (HiF0), followed by patterns of downstep and declination. Sentence-final verbs always have lower Hif0 values than either initial or medial Agents or Patients. Thematic role order does not influence intonational
patterns, with the results suggesting that Murrinhpatha has positional prosody, although final nominals can disrupt global
pitch downtrends regardless of thematic role.

Abstract

Reaction time to detect a phoneme target in a sentence was found to be faster when the word in which the target occurred formed part of the semantic focus of the sentence. Focus was determined by asking a question before the sentence; that part of the sentence which comprised the answer to the sentence was assumed to be focussed. This procedure made it possible to vary position offocus within the sentence while holding all acoustic aspects of the sentence itself constant. It is argued that sentence understanding is facilitated by rapid identification of focussed information. Since focussed words are usually accented, it is further argued that the active search for accented words demonstrated in previous research should be interpreted as a search for semantic focus.

Abstract

Cognitive functional neuroimaging has been around for over 30 years and has shed light on the brain areas relevant for reading. However, new methodological developments enable mapping the interaction between functional imaging and the underlying white matter networks. In this study, we used such a novel method, called the disconnectome, to decode the reading circuitry in the brain. We used the resulting disconnection patterns to predict a typical lesion that would lead to reading deficits after brain damage. Our results suggest that white matter connections critical for reading include fronto-parietal U-shaped fibres and the vertical occipital fasciculus (VOF). The lesion most predictive of a reading deficit would impinge on the left temporal, occipital, and inferior parietal gyri. This novel framework can systematically be applied to bridge the gap between the neuropathology of language and cognitive neuroscience.

Abstract

Lesion symptom mapping has revolutionized our understanding of the functioning of the human brain. Associating damaged voxels in the brain with loss of function has created a map of the brain that identifies critical areas. While these methods have significantly advanced our understanding, recent improvements have identified the need for multivariate and multimodal methods to map hidden lesions and damage to white matter networks beyond the lesion voxels. This article reviews the evolution of lesion-symptom mapping from single case studies to the human disconnectome.

Abstract

Inter-individual differences can inform treatment procedures and—if accounted for—have the potential to significantly improve patient outcomes. However, when studying brain anatomy, these inter-individual variations are commonly unaccounted for, despite reports of differences in gross anatomical features, cross-sectional, and connectional anatomy. Brain connections are essential to facilitate functional organization and, when severed, cause impairments or complete loss of function. Hence, the study of cerebral white matter may be an ideal compromise to capture inter-individual variability in structure and function. We reviewed the wealth of studies that associate cognitive functions and clinical symptoms with individual tracts using diffusion tractography. Our systematic review indicates that tractography has proven to be a sensitive method in neurology, psychiatry, and healthy populations to identify variability and its functional correlates. However, the literature may be biased, as the most commonly studied tracts are not necessarily those with the highest sensitivity to cognitive functions and pathologies. Additionally, the hemisphere of the studied tract is often unreported, thus neglecting functional laterality and asymmetries. Finally, we demonstrate that tracts, as we define them, are not correlated with one, but multiple cognitive domains or pathologies. While our systematic review identified some methodological caveats, it also suggests that tract–function correlations might still be a promising tool in identifying biomarkers for precision medicine. They can characterize variations in brain anatomy, differences in functional organization, and predicts resilience and recovery in patients.

Abstract

Progress in genome sequencing now enables the large-scale generation of reference genomes. Various international initiatives aim to generate reference genomes representing global biodiversity. These genomes provide unique insights into genomic diversity and architecture, thereby enabling comprehensive analyses of population and functional genomics, and are expected to revolutionize conservation genomics.

Abstract

Language perception studies on bilinguals often show that words that share form and meaning across languages (cognates) are easier to process than words that share only meaning. This facilitatory phenomenon is known as the cognate effect. Most previous studies have shown this effect visually, whereas the auditory modality as well as the interplay between type of similarity and modality remain largely unexplored. In this study, highly proficient late Spanish–English bilinguals carried out a lexical decision task in their second language, both visually and auditorily. Words had high or low phonological and orthographic similarity, fully crossed. We also included orthographically identical words (perfect cognates). Our results suggest that similarity in the same modality (i.e., orthographic similarity in the visual modality and phonological similarity in the auditory modality) leads to improved signal detection, whereas similarity across modalities hinders it. We provide support for the idea that perfect cognates are a special category within cognates. Results suggest a need for a conceptual and practical separation between types of similarity in cognate studies. The theoretical implication is that the representations of items are active in both modalities of the non-target language during language processing, which needs to be incorporated to our current processing models.

Abstract

Language perception studies on bilinguals often show that words that share form and meaning across
languages (cognates) are easier to process than words that share only meaning. This facilitatory
phenomenon is known as the cognate effect. Most previous studies have shown this effect visually,
whereas the auditory modality as well as the interplay between type of similarity and modality
remain largely unexplored. In this study, highly proficient late Spanish–English bilinguals carried out
a lexical decision task in their second language, both visually and auditorily. Words had high or low
phonological and orthographic similarity, fully crossed. We also included orthographically identical
words (perfect cognates). Our results suggest that similarity in the same modality (i.e., orthographic
similarity in the visual modality and phonological similarity in the auditory modality) leads to
improved signal detection, whereas similarity across modalities hinders it. We provide support for
the idea that perfect cognates are a special category within cognates. Results suggest a need for a
conceptual and practical separation between types of similarity in cognate studies. The theoretical
implication is that the representations of items are active in both modalities of the non‑target
language during language processing, which needs to be incorporated to our current processing
models.

Abstract

As knowing a foreign language becomes a necessity in the modern world, a large portion of
the population is faced with the challenge of learning a language in a classroom. This, in turn,
presents a unique set of difficulties. Acquiring a language with limited and artificial exposure makes
learning new information and vocabulary particularly difficult. The purpose of this thesis is to help us
understand how we can compensate—at least partially—for these difficulties by presenting
information in a way that aids learning. In particular, I focused on variables that affect semantic
richness—meaning the amount and variability of information associated with a word. Some factors
that affect semantic richness are intrinsic to the word and others pertain to that word’s relationship
with other items and information. This latter group depends on the context around the to-be-
learned items rather than the words themselves. These variables are easier to manipulate than
intrinsic qualities, making them more accessible tools for teaching and understanding learning. I
focused on two factors: emotionality of the surrounding semantic context and contextual diversity.
Publication 1 (Frances, de Bruin, et al., 2020b) focused on content learning in a foreign
language and whether the emotionality—positive or neutral—of the semantic context surrounding
key information aided its learning. This built on prior research that showed a reduction in
emotionality in a foreign language. Participants were taught information embedded in either
positive or neutral semantic contexts in either their native or foreign language. When they were
then tested on these embedded facts, participants’ performance decreased in the foreign language.
But, more importantly, they remembered better the information from the positive than the neutral
semantic contexts.
In Publication 2 (Frances, de Bruin, et al., 2020a), I focused on how emotionality affected
vocabulary learning. I taught participants the names of novel items described either in positive or
neutral terms in either their native or foreign language. Participants were then asked to recall and
recognize the object's name—when cued with its image. The effects of language varied with the
difficulty of the task—appearing in recall but not recognition tasks. Most importantly, learning the
words in a positive context improved learning, particularly of the association between the image of
the object and its name.
In Publication 3 (Frances, Martin, et al., 2020), I explored the effects of contextual
diversity—namely, the number of texts a word appears in—on native and foreign language word
learning. Participants read several texts that had novel pseudowords. The total number of
encounters with the novel words was held constant, but they appeared in 1, 2, 4, or 8 texts in either
their native or foreign language. Increasing contextual diversity—i.e., the number of texts a word
appeared in—improved recall and recognition, as well as the ability to match the word with its
meaning. Using a foreign language only affected performance when participants had to quickly
identify the meaning of the word.
Overall, I found that the tested contextual factors related to semantic richness—i.e.,
emotionality of the semantic context and contextual diversity—can be manipulated to improve
learning in a foreign language. Using positive emotionality not only improved learning in the foreign
language, but it did so to the same extent as in the native language. On a theoretical level, this
suggests that the reduction in emotionality in a foreign language is not ubiquitous and might relate
to the way in which that language as learned.
The third article shows an experimental manipulation of contextual diversity and how this
can affect learning of a lexical item, even if the amount of information known about the item is kept
constant. As in the case of emotionality, the effects of contextual diversity were also the same
between languages. Although deducing words from context is dependent on vocabulary size, this
does not seem to hinder the benefits of contextual diversity in the foreign language.
Finally, as a whole, the articles contained in this compendium provide evidence that some
aspects of semantic richness can be manipulated contextually to improve learning and memory. In
addition, the effects of these factors seem to be independent of language status—meaning, native
or foreign—when learning new content. This suggests that learning in a foreign and a native
language is not as different as I initially hypothesized, allowing us to take advantage of native
language learning tools in the foreign language, as well.

Abstract

The cognate effect refers to translation equivalents with similar form between languages—i.e., cognates, such as “band” (English) and “banda” (Spanish)—being processed faster than words with dissimilar forms—such as, “cloud” and “nube.” Substantive literature supports this claim, but is mostly based on orthographic similarity and tested in the visual modality. In a previous study, we found an inhibitory orthographic similarity effect in the auditory modality—i.e., greater orthographic similarity led to slower response times and reduced accuracy. The aim of the present study is to explain this effect. In doing so, we explore the role of the speaker's accent in auditory word recognition and whether native accents lead to a mismatch between the participants' phonological representation and the stimulus. Participants carried out a lexical decision task and a typing task in which they spelled out the word they heard. Words were produced by two speakers: one with a native English accent (Standard American) and the other with a non-native accent matching that of the participants (native Spanish speaker from Spain). We manipulated orthographic and phonological similarity orthogonally and found that accent did have some effect on both response time and accuracy as well as modulating the effects of similarity. Overall, the non-native accent improved performance, but it did not fully explain why high orthographic similarity items show an inhibitory effect in the auditory modality. Theoretical implications and future directions are discussed.

Abstract

Genomewide quantitative-trait locus (QTL) linkage analysis was performed using a continuous measure of relative hand skill (PegQ) in a sample of 195 reading-disabled sibling pairs from the United Kingdom. This was the first genomewide screen for any measure related to handedness. The mean PegQ in the sample was equivalent to that of normative data, and PegQ was not correlated with tests of reading ability (correlations between −0.13 and 0.05). Relative hand skill could therefore be considered normal within the sample. A QTL on chromosome 2p11.2-12 yielded strong evidence for linkage to PegQ (empirical P=.00007), and another suggestive QTL on 17p11-q23 was also identified (empirical P=.002). The 2p11.2-12 locus was further analyzed in an independent sample of 143 reading-disabled sibling pairs, and this analysis yielded an empirical P=.13. Relative hand skill therefore is probably a complex multifactorial phenotype with a heterogeneous background, but nevertheless is amenable to QTL-based gene-mapping approaches.

Abstract

A locus on chromosome 2p12-16 has been implicated in dyslexia susceptibility by two independent linkage studies, including our own study of 119 nuclear twin-based families, each with at least one reading-disabled child. Nonetheless, no variant of any gene has been reported to show association with dyslexia, and no consistent clinical evidence exists to identify candidate genes with any strong a priori logic. We used 21 microsatellite markers spanning 2p12-16 to refine our 1-LOD unit linkage support interval to 12cM between D2S337 and D2S286. Then, in quantitative association analysis, two microsatellites yielded P values<0.05 across a range of reading-related measures (D2S2378 and D2S2114). The exon/intron borders of two positional candidate genes within the region were characterized, and the exons were screened for polymorphisms. The genes were Semaphorin4F (SEMA4F), which encodes a protein involved in axonal growth cone guidance, and OTX1, encoding a homeodomain transcription factor involved in forebrain development. Two non-synonymous single nucleotide polymorphisms were found in SEMA4F, each with a heterozygosity of 0.03. One intronic single nucleotide polymorphism between exons 12 and 13 of SEMA4F was tested for quantitative association, but no significant association was found. Only one single nucleotide polymorphism was found in OTX1, which was exonic but silent. Our data therefore suggest that linkage with reading disability at 2p12-16 is not caused by coding variants of SEMA4F or OTX1. Our study outlines the approach necessary for the identification of genetic variants causing dyslexia susceptibility in an epidemiological population of dyslexics.

Abstract

Dyslexia, a disorder of reading and spelling, is a heterogeneous neurological syndrome with a complex genetic and environmental aetiology. People with dyslexia differ in their individual profiles across a range of cognitive, physiological, and behavioural measures related to reading disability. Some or all of the subtypes of dyslexia might have partly or wholly distinct genetic causes. An understanding of the role of genetics in dyslexia could help to diagnose and treat susceptible children more effectively and rapidly than is currently possible and in ways that account for their individual disabilities. This knowledge will also give new insights into the neurobiology of reading and language cognition. Genetic linkage analysis has identified regions of the genome that might harbour inherited variants that cause reading disability. In particular, loci on chromosomes 6 and 18 have shown strong and replicable effects on reading abilities. These genomic regions contain tens or hundreds of candidate genes, and studies aimed at the identification of the specific causal genetic variants are underway.

Abstract

Delineating the association of age and cortical thickness in healthy individuals is critical given the association of cortical thickness with cognition and behavior. Previous research has shown that robust estimates of the association between age and brain morphometry require large‐scale studies. In response, we used cross‐sectional data from 17,075 individuals aged 3–90 years from the Enhancing Neuroimaging Genetics through Meta‐Analysis (ENIGMA) Consortium to infer age‐related changes in cortical thickness. We used fractional polynomial (FP) regression to quantify the association between age and cortical thickness, and we computed normalized growth centiles using the parametric Lambda, Mu, and Sigma method. Interindividual variability was estimated using meta‐analysis and one‐way analysis of variance. For most regions, their highest cortical thickness value was observed in childhood. Age and cortical thickness showed a negative association; the slope was steeper up to the third decade of life and more gradual thereafter; notable exceptions to this general pattern were entorhinal, temporopolar, and anterior cingulate cortices. Interindividual variability was largest in temporal and frontal regions across the lifespan. Age and its FP combinations explained up to 59% variance in cortical thickness. These results may form the basis of further investigation on normative deviation in cortical thickness and its significance for behavioral and cognitive outcomes.

Abstract

Theeffects of spatial filtering in functional magnetic resonance imaging were investigated by reevaluating the data of a previous study of episodic memory encoding at 2 × 2 × 4-mm3 resolution with use of a SPM99 analysis involving a Gaussian kernel of 8-mm full width at half maximum. In addition, a multisubject analysis of activated regions was performed by normalizing the functional images to an approximate Talairach brain atlas. In individual subjects, spatial filtering merged activations in anatomically separated brain regions. Moreover, small foci of activated pixels which originated from veins became blurred and hence indistinguishable from parenchymal responses. The multisubject analysis resulted in activation of the hippocampus proper, a finding which could not be confirmed by the activation maps obtained at high resolution. It is concluded that the validity of multisubject fMRI analyses can be considerably improved by first analyzing individual data sets at optimum resolution to assess the effects of spatial filtering and minimize the risk of signal contamination by macroscopically visible vessels.

Abstract

The current study investigates whether self-talk phrases can influence behavior in Ultimatum Games. In our three self-talk treatments, participants were instructed to tell themselves (i) to keep their own interests in mind, (ii) to also think of the other person, or (iii) to take some time to contemplate their decision. We investigate how such so-called experimenter-determined strategic self-talk phrases affect behavior and emotions in comparison to a control treatment without instructed self-talk. The results demonstrate that other-focused self-talk can nudge proposers towards fair behavior, as offers were higher in this group than in the other conditions. For responders, self-talk tended to increase acceptance rates of unfair offers as compared to the condition without self-talk. This effect is significant for both other-focused and contemplation-inducing self-talk but not for self-focused self-talk. In the self-focused condition, responders were most dissatisfied with unfair offers. These findings suggest that use of self-talk can increase acceptance rates in responders, and that focusing on personal interests can undermine this effect as it negatively impacts the responders’ emotional experience. In sum, our study shows that strategic self-talk interventions can be used to affect behavior in bargaining situations.

Abstract

Evolution, as we currently understand it, strikes a delicate balance between animals' ancestral history and adaptations to their current niche. Similarities between species are generally considered inherited from a common ancestor whereas observed differences are considered as more recent evolution. Hence comparing species can provide insights into the evolutionary history. Comparative neuroimaging has recently emerged as a novel subdiscipline, which uses magnetic resonance imaging (MRI) to identify similarities and differences in brain structure and function across species. Whereas invasive histological and molecular techniques are superior in spatial resolution, they are laborious, post-mortem, and oftentimes limited to specific species. Neuroimaging, by comparison, has the advantages of being applicable across species and allows for fast, whole-brain, repeatable, and multi-modal measurements of the structure and function in living brains and post-mortem tissue. In this review, we summarise the current state of the art in comparative anatomy and function of the brain and gather together the main scientific questions to be explored in the future of the fascinating new field of brain evolution derived from comparative neuroimaging.

Abstract

Language acquisition is complex. However, one thing that has been suggested to help learning is the way that information is distributed throughout language; co-occurrences among particular items (e.g., syllables and words) have been shown to help learners discover the words that a language contains and figure out how those words are used. Humans’ ability to draw on this information—“statistical learning”—has been demonstrated across a broad range of studies. However, evidence from non-WEIRD (Western, Educated, Industrialized, Rich, and Democratic) societies is critically lacking, which limits theorizing on the universality of this skill. We extended work on statistical language learning to a new, non-WEIRD linguistic population: speakers of Yélî Dnye, who live on a remote island off mainland Papua New Guinea (Rossel Island). We performed a replication of an existing statistical learning study, training adults on an artificial language with statistically defined words, then examining what they had learnt using a two-alternative forced-choice test. Crucially, we implemented several key amendments to the original study to ensure the replication was suitable for remote field-site testing with speakers of Yélî Dnye. We made critical changes to the stimuli and materials (to test speakers of Yélî Dnye, rather than English), the instructions (we re-worked these significantly, and added practice tasks to optimize participants’ understanding), and the study format (shifting from a lab-based to a portable tablet-based setup). We discuss the requirement for acute sensitivity to linguistic, cultural, and environmental factors when adapting studies to test new populations.

Abstract

Graph neural networks (GNNs) have emerged as the standard method for numerous tasks on graph-structured data such as node classification. However, real-world graphs are often evolving over time and even new classes may arise. We model these challenges as an instance of lifelong learning, in which a learner faces a sequence of tasks and may take over knowledge acquired in past tasks. Such knowledge may be stored explicitly as historic data or implicitly within model parameters. In this work, we systematically analyze the influence of implicit and explicit knowledge. Therefore, we present an incremental training method for lifelong learning on graphs and introduce a new measure based on k-neighborhood time differences to address variances in the historic data. We apply our training method to five representative GNN architectures and evaluate them on three new lifelong node classification datasets. Our results show that no more than 50% of the GNN's receptive field is necessary to retain at least 95% accuracy compared to training over the complete history of the graph data. Furthermore, our experiments confirm that implicit knowledge becomes more important when fewer explicit knowledge is available.

Abstract

COVID-19 research datasets are crucial for analyzing research dynamics. Most collections of COVID-19 research items do not to include cited works and do not have annotations
from a controlled vocabulary. Starting with ZB MED KE data on COVID-19, which comprises CORD-19, we assemble a new dataset that includes cited work and MeSH annotations for all records. Furthermore, we conduct experiments on the analysis of research dynamics, in which we investigate predicting links in a co-annotation graph created on the basis of the new dataset. Surprisingly, we find that simple heuristic methods are better at
predicting future links than more sophisticated approaches such as graph neural networks.

Abstract

Large pretrained language models (PreLMs) are rev-olutionizing natural language processing across all benchmarks. However, their sheer size is prohibitive for small laboratories or for deployment on mobile devices. Approaches like pruning and distillation reduce the model size but typically retain the same model architecture. In contrast, we explore distilling PreLMs into a different, more efficient architecture, Continual Multiplication of Words (CMOW), which embeds each word as a matrix and uses matrix multiplication to encode sequences. We extend the CMOW architecture and its CMOW/CBOW-Hybrid variant with a bidirectional component for more expressive power, per-token representations for a general (task-agnostic) distillation during pretraining, and a two-sequence encoding scheme that facilitates downstream tasks on sentence pairs, such as sentence similarity and natural language inference. Our matrix-based bidirectional CMOW/CBOW-Hybrid model is competitive to DistilBERT on question similarity and recognizing textual entailment, but uses only half of the number of parameters and is three times faster in terms of inference speed. We match or exceed the scores of ELMo for all tasks of the GLUE benchmark except for the sentiment analysis task SST-2 and the linguistic acceptability task CoLA. However, compared to previous cross-architecture distillation approaches, we demonstrate a doubling of the scores on detecting linguistic acceptability. This shows that matrix-based embeddings can be used to distill large PreLM into competitive models and motivates further research in this direction.