Publications

Abstract

Four probe latency experiments show that the ‘constituent boundary effect’ (transitions between constituents are more difficult than within constituents) is a retrieval and not a storage phenomenon. The experimental logic used is called paraphrastic reproduction: after verbatim memorization of some sentences, subjects were instructed to reproduce them both in their original wording and in the form of sentences that, whilst preserving the original meaning, embodied different syntactic constructions. Syntactic constructions are defined as pairs which consist of a pattern of conceptual information and a syntactic scheme, i.e. a sequence of syntactic word categories and function words. For example, the sequence noun + finite intransitive main verb (‘John runs’) expresses a conceptual actor-action relationship. It is proposed that for each overlearned and simple syntactic construction there exists a retrieval plan which does the following. It searches through the long-term memory information that has been designated as the conceptual content of the utterance(s) to be produced, looking for a token of its conceptual pattern. The retrieved information is then cast into the format of its syntactic scheme. The organization of such plans is held responsible for the constituent boundary effect.

Abstract

A series of experiments is reported in which subjects describe simple visual scenes by means of both sentential and non-sentential responses. The data support the following statements about the lexicalization (word finding) process. (1) Words used by speakers in overt naming or sentence production responses are selected by a sequence of two lexical retrieval processes, the first yielding abstract pre-phonological items (Ll -items), the second one adding their phonological shapes (L2-items). (2) The selection of several Ll-items for a multi-word utterance can take place simultaneously. (3) A monitoring process is watching the output of Ll-lexicalization to check if it is in keeping with prevailing constraints upon utterance format. (4) Retrieval of the L2-item which corresponds with a given LI-item waits until the Ld-item has been checked by the monitor, and all other Ll-items needed for the utterance under construction have become available. A coherent picture of the lexicalization process begins to emerge when these characteristics are brought together with other empirical results in the area of naming and sentence production, e.g., picture naming reaction times (Seymour, 1979), speech errors (Garrett, 1980), and word order preferences (Bock, 1982).

Abstract

Human communicative interaction is characterized by rapid and precise turn-taking. This is achieved by an intricate system that has been elucidated in the field of conversation analysis, based largely on the study of the auditory signal. This model suggests that transitions occur at points of possible completion identified in terms of linguistic units. Despite this, considerable evidence exists that visible bodily actions including gaze and gestures also play a role. To reconcile disparate models and observations in the literature, we combine qualitative and quantitative methods to analyse turn-taking in a corpus of multimodal interaction using eye-trackers and multiple cameras. We show that transitions seem to be inhibited when a speaker averts their gaze at a point of possible turn completion, or when a speaker produces gestures which are beginning or unfinished at such points. We further show that while the direction of a speaker's gaze does not affect the speed of transitions, the production of manual gestures does: turns with gestures have faster transitions. Our findings suggest that the coordination of transitions involves not only linguistic resources but also visual gestural ones and that the transition-relevance places in turns are multimodal in nature.

Abstract

The present study investigates the developmental trajectory of abstract representations for syntactic structures in children. In a structural priming experiment on the dative alternation in German, we primed children from three different age groups (3–4 years, 5–6 years, 7–8 years) and adults with double object datives (Dora sent Boots the rabbit) or prepositional object datives (Dora sent the rabbit to Boots). Importantly, the prepositional object structure in German is dispreferred and only rarely encountered by young children. While immediate as well as cumulative structural priming effects occurred across all age groups, these effects were strongest in the 3- to 4-year-old group and gradually decreased with increasing age. These results suggest that representations in young children are less stable than in adults and, therefore, more susceptible to adaptation both immediately and across time, presumably due to stronger surprisal. Lexical boost effects, in contrast, were not present in 3- to 4-year-olds but gradually emerged with increasing age, possibly due to limited working-memory capacity in the younger child groups.

Abstract

Statistical learning (SL)—sensitivity to statistical regularities in the environment—has been postulated to support language development. While even young infants are capable of using distributional statistics to learn in linguistic and non-linguistic domains, efforts to measure SL at the level of the individual and link it to language proficiency in individual differences designs have been mixed, which has at least in part been attributed to problems with task reliability. In the current study we present the first prospective longitudinal study of the relationship between both non-linguistic SL (measured with visual stimuli) and linguistic SL (measured with auditory stimuli) and language in a group of English-speaking children. One-hundred and twenty-one (N = 121) children in their first two years of formal schooling (Mage = 6;1 years, Range: 5;2 – 7;2) completed tests of visual SL (VSL) and auditory SL (ASL) and several control variables at time 1. Both forms of SL were then measured every 6 months for the next 18 months, and at the final testing session (time 4) their language proficiency was measured using a standardised test. The results showed that the reliability of the SL tasks increased across the course of the study. A series of path analyses showed that both VSL and ASL independently predicted individual differences in language proficiency at time 4. The evidence is consistent with the suggestion that, when measured reliably, an observable relationship between SL and language proficiency exists. Theoretical and methodological issues are discussed.

Abstract

The previous literature has suggested that the hand movement in co-speech gestures and signs consists of a series of phases with qualitatively different dynamic characteristics. In this paper, we propose a syntagmatic rule system for movement phases that applies to both co-speech gestures and signs. Descriptive criteria for the rule system were developed for the analysis video-recorded continuous production of signs and gesture. It involves segmenting a stream of body movement into phases and identifying different phase types. Two human coders used the criteria to analyze signs and cospeech gestures that are produced in natural discourse. It was found that the criteria yielded good inter-coder reliability. These criteria can be used for the technology of automatic recognition of signs and co-speech gestures in order to segment continuous production and identify the potentially meaningbearing phase.

Abstract

During the last 25 years, second language acquisition (SLA) research hasmade considerable progress, but is still far from proving a solid basis for foreign language teaching, or from a general theory of SLA. In addition, its status within the linguistic disciplines is still very low. I argue this has not much to do with low empirical or theoretical standards in the field—in this regard, SLA research is fully competitive—but with a particular perspective on the acquisition process: SLA researches learners' utterances as deviations from a certain target, instead of genuine manifestations of underlying language capacity; it analyses them in terms of what they are not rather than what they are. For some purposes such a "target deviation perspective" makes sense, but it will not help SLA researchers to substantially and independently contribute to a deeper understanding of the structure and function of the human language faculty. Therefore, these findings will remain of limited interest to other scientists until SLA researchers consider learner varieties a normal, in fact typical, manifestation of this unique human capacity.

Abstract

Recently, the Max Planck Society founded a new Project group for Psycholinguistics. This article reviews some of the kernel issues of the group's research program. The central concern is with the context dependency of the speaker's linguistic behavior. The process of linguistically formulating depends not only on what the speaker wants to express, but also on what has been said previously (linguistic context), and on the physical and social situation (nonlinguistic context). Special attention is paid to two context-dependent phenomena.

Abstract

Rules of behavior are fundamental to human sociality. Whether on the road, at the dinner table, or during a game, people monitor one another’s behavior for conformity to rules and may take action to rectify violations. In this study, we examine two ways in which rules are enforced during games: instructions and reminders. Building on prior research, we identify instructions as actions produced to rectify violations based on another’s lack of knowledge of the relevant rule; knowledge that the instruction is designed to impart. In contrast to this, the actions we refer to as reminders are designed to enforce rules presupposing the transgressor’s competence and treating the violation as the result of forgetfulness or oversight. We show that instructing and reminding actions differ in turn design, sequential development, the epistemic stances taken by transgressors and enforcers, and in how the action affects the progressivity of the interaction. Data are in German and Italian from the Parallel European Corpus of Informal Interaction (PECII).

Abstract

During listening, brain activity tracks the rhythmic structures of speech signals. Here, we directly dissociated the contribution of neural envelope tracking in the processing of speech acoustic cues from that related to linguistic processing. We examined the neural changes associated with the comprehension of Noise-Vocoded (NV) speech using magnetoencephalography (MEG). Participants listened to NV sentences in a 3-phase training paradigm: (1) pre-training, where NV stimuli were barely comprehended, (2) training with exposure of the original clear version of speech stimulus, and (3) post-training, where the same stimuli gained intelligibility from the training phase. Using this paradigm, we tested if the neural responses of a speech signal was modulated by its intelligibility without any change in its acoustic structure. To test the influence of spectral degradation on neural envelope tracking independently of training, participants listened to two types of NV sentences (4-band and 2-band NV speech), but were only trained to understand 4-band NV speech. Significant changes in neural tracking were observed in the delta range in relation to the acoustic degradation of speech. However, we failed to find a direct effect of intelligibility on the neural tracking of speech envelope in both theta and delta ranges, in both auditory regions-of-interest and whole-brain sensor-space analyses. This suggests that acoustics greatly influence the neural tracking response to speech envelope, and that caution needs to be taken when choosing the control signals for speech-brain tracking analyses, considering that a slight change in acoustic parameters can have strong effects on the neural tracking response.

Abstract

In various applications of forensic phonetics the question arises as to how far aural-perceptual speaker recognition performance is reliable. Therefore, it is necessary to examine the relationship between speaker recognition results and human perception/production abilities like musicality or speech sensitivity. In this study, performance in a speaker recognition experiment and a speech sensitivity test are correlated. The results show a moderately significant positive correlation between the two tasks. Generally, performance in the speaker recognition task was better than in the speech sensitivity test. Professionals in speech and singing yielded a more homogeneous correlation than non-experts. Training in speech as well as choir-singing seems to have a positive effect on performance in speaker recognition. It may be concluded, firstly, that in cases where the reliability of voice line-up results or the credibility of a testimony have to be considered, the speech sensitivity test could be a useful indicator. Secondly, the speech sensitivity test might be integrated into the canon of possible procedures for the accreditation of forensic phoneticians. Both tests may also be used in combination.

Abstract

Developmental Language Disorder (DLD), present in 2 out of every 30 children, affects primarily oral language abilities and development in the absence of associated biomedical conditions. We report the first experimental study that examines relative clause (RC) comprehension accuracy and processing (via looking preference) in Cantonese-speaking children with and without DLD, testing the predictions from competing domain-specific versus domain-general theoretical accounts. We compared children with DLD (N = 22) with their age-matched typically-developing (TD) children (AM-TD, N = 23) aged 6;6–9;7 and language-matched (and younger) TD children (YTD, N = 21) aged 4;7–7;6, using a referent selection task. Within-subject factors were: RC type (subject-RCs (SRCs) versus object-RCs (ORCs); relativizer (classifier (CL) versus relative marker ge3 RCs). Accuracy measures and looking preference to the target were analyzed using generalized linear mixed effects models. Results indicated Cantonese children with DLD scored significantly lower than their AM-TD peers in accuracy and processed RCs significantly slower than AM-TDs, but did not differ from the YTDs on either measure. Overall, while the results revealed evidence of a SRC advantage in the accuracy data, there was no indication of additional difficulty associated with ORCs in the eye-tracking data. All children showed a processing advantage for the frequent CL relativizer over the less frequent ge3 relativizer. These findings pose challenges to domain-specific representational deficit accounts of DLD, which primarily explain the disorder as a syntactic deficit, and are better explained by domain-general accounts that explain acquisition and processing as emergent properties of multiple converging linguistic and non-linguistic processes.

Abstract

Interpreting the meaning of co-speech gesture often involves
identifying a gesture’s ‘lexical affiliate’, the word or phrase to
which it most closely relates (Schegloff 1984). Though there is
work within gesture studies that resists this simplex mapping of
meaning from speech to gesture (e.g. de Ruiter 2000; Kendon
2014; Parrill 2008), including an evolving body of literature on
recurrent gesture and gesture families (e.g. Fricke et al. 2014; Müller 2017), it is still the lexical affiliate model that is most ap-
parent in formal linguistic models of multimodal meaning(e.g.
Alahverdzhieva et al. 2017; Lascarides and Stone 2009; Puste-
jovsky and Krishnaswamy 2021; Schlenker 2020). In this work,
I argue that the lexical affiliate should be carefully reconsidered
in the further development of such models.
In place of the lexical affiliate, I suggest a further shift
toward a frame-based, action schematic approach to gestural
meaning in line with that proposed in, for example, Parrill and
Sweetser (2004) and Müller (2017). To demonstrate the utility
of this approach I present three types of compositional gesture
sequences which I call spatial contrast, spatial embedding, and
cooperative abstract deixis. All three rely on gestural context,
rather than gesture-speech alignment, to convey interactive (i.e.
pragmatic) meaning. The centrality of gestural context to ges-
ture meaning in these examples demonstrates the necessity of
developing a model of gestural meaning independent of its in-
tegration with speech.

Abstract

There is a wealth of evidence demonstrating that executive function (EF) abilities are positively associated with language development during the preschool years, such that children with good executive functions also have larger vocabularies. However, why this is the case remains to be discovered. In this study, we focused on the hypothesis that sentence processing abilities mediate the association between EF skills and receptive vocabulary knowledge, in that the speed of language acquisition is at least partially dependent on a child’s processing ability, which is itself dependent on executive control. We tested this hypothesis in longitudinal data from a cohort of 3- and 4-year-old children at three age points (37, 43, and 49 months). We found evidence, consistent with previous research, for a significant association between three EF skills (cognitive flexibility, working memory [as measured by the Backward Digit Span], and inhibition) and receptive vocabulary knowledge across this age range. However, only one of the tested sentence processing abilities (the ability to maintain multiple possible referents in mind) significantly mediated this relationship and only for one of the tested EFs (inhibition). The results suggest that children who are better able to inhibit incorrect responses are also better able to maintain multiple possible referents in mind while a sentence unfolds, a sophisticated sentence processing ability that may facilitate vocabulary learning from complex input.

Abstract

Normative ratings are a means to control for the effects of confounding variables in psycholinguistic experiments. This paper introduces a new dataset of normative ratings for Swedish encompassing 111 concrete nouns and the corresponding picture stimuli in the MultiPic database (Duñabeitia et al. 2017). The norms for name agreement, category typicality, age of acquisition and subjective frequency were collected using online surveys among native speakers of the Finland-Swedish variety of Swedish. The paper discusses the inter-correlations between these variables and compares them against available ratings for other languages. In doing so, the paper argues that ratings for age of acquisition and subjective frequency collected for other languages may be applied to psycholinguistic studies on Finland-Swedish, at least with respect to concrete and highly imageable nouns. In contrast, norms for name agreement should be collected from speakers of the same language variety as represented by the subjects in the actual experiments.

Abstract

Emotional words have consistently been shown to be processed differently than neutral words. However, few studies have examined individual variability in emotion word processing with longer, ecologically valid stimuli (beyond isolated words, sentences, or paragraphs). In the current study, we re-analysed eye-tracking data collected during story reading to reveal how individual differences in need for affect and narrative absorption impact the speed of emotion word reading. Word emotionality was indexed by affective-aesthetic potentials (AAP) calculated by a sentiment analysis tool. We found that individuals with higher levels of need for affect and narrative absorption read positive words more slowly. On the other hand, these individual differences did not influence the reading time of more negative words, suggesting that high need for affect and narrative absorption are characterised by a positivity bias only. In general, unlike most previous studies using more isolated emotion word stimuli, we observed a quadratic (U-shaped) effect of word emotionality on reading speed, such that both positive and negative words were processed more slowly than neutral words. Taken together, this study emphasises the importance of taking into account individual differences and task context when studying emotion word processing.

Abstract

The expansion of the cerebral cortex is one of the most distinctive changes in the evolution of the human brain. Cortical expansion and related increases in cortical folding may have contributed to emergence of our capacities for high-order cognitive abilities. Molecular analysis of humans, archaic hominins, and non-human primates has allowed identification of chromosomal regions showing evolutionary changes at different points of our phylogenetic history. In this study, we assessed the contributions of genomic annotations spanning 30 million years to human sulcal morphology measured via MRI in more than 18,000 participants from the UK Biobank. We found that variation within brain-expressed human gained enhancers, regulatory genetic elements that emerged since our last common ancestor with Old World monkeys, explained more trait heritability than expected for the left and right calloso-marginal posterior fissures and the right central sulcus. Intriguingly, these are sulci that have been previously linked to the evolution of locomotion in primates and later on bipedalism in our hominin ancestors.

Abstract

The purpose of this study was to relate a psycholinguistic processing model of picture naming to the dynamics of cortical activation during picture naming. The activation was recorded from eight Dutch subjects with a whole-head neuromagnetometer. The processing model, based on extensive naming latency studies, is a stage model. In preparing a picture's name, the speaker performs a chain of specific operations. They are, in this order, computing the visual percept, activating an appropriate lexical concept, selecting the target word from the mental lexicon, phonological encoding, phonetic encoding, and initiation of articulation. The time windows for each of these operations are reasonably well known and could be related to the peak activity of dipole sources in the individual magnetic response patterns. The analyses showed a clear progression over these time windows from early occipital activation, via parietal and temporal to frontal activation. The major specific findings were that (1) a region in the left posterior temporal lobe, agreeing with the location of Wernicke's area, showed prominent activation starting about 200 msec after picture onset and peaking at about 350 msec, (i.e., within the stage of phonological encoding), and (2) a consistent activation was found in the right parietal cortex, peaking at about 230 msec after picture onset, thus preceding and partly overlapping with the left temporal response. An interpretation in terms of the management of visual attention is proposed.

Abstract

Making a self-repair in speech typically proceeds in three phases. The first phase involves the monitoring of one’s own speech and the interruption of the flow of speech when trouble is detected. From an analysis of 959 spontaneous self-repairs it appears that interrupting follows detection promptly, with the exception that correct words tend to be completed. Another finding is that detection of trouble improves towards the end of constituents. The second phase is characterized by hesitation, pausing, but especially the use of so-called editing terms. Which editing term is used depends on the nature of the speech trouble in a rather regular fashion: Speech errors induce other editing terms than words that are merely inappropriate, and trouble which is detected quickly by the speaker is preferably signalled by the use of ‘uh’. The third phase consists of making the repair proper The linguistic well-formedness of a repair is not dependent on the speaker’s respecting the integriv of constituents, but on the structural relation between original utterance and repair. A bi-conditional well-formedness rule links this relation to a corresponding relation between the conjuncts of a coordination. It is suggested that a similar relation holds also between question and answer. In all three cases the speaker respects certain Istructural commitments derived from an original utterance. It was finally shown that the editing term plus the first word of the repair proper almost always contain sufficient information for the listener to decide how the repair should be related to the original utterance. Speakers almost never produce misleading information in this respect. It is argued that speakers have little or no access to their speech production process; self-monitoring is probably based on parsing one’s own inner or overt speech.

Abstract

This commentary discusses whether abstract metrical frames are stored. For stress-assigning languages (e.g., Dutch and English), which have a dominant stress pattern, metrical frames are stored only for words that deviate from the default stress pattern. The majority of the words in these languages are produced without retrieving any independent syllabic or metrical frame.

Abstract

Spontaneous self-corrections in speech pose a communication problem; the speaker must make clear to the listener not only that the original Utterance was faulty, but where it was faulty and how the fault is to be corrected. Prosodic marking of corrections - making the prosody of the repair noticeably different from that of the original utterance - offers a resource which the speaker can exploit to provide the listener with such information. A corpus of more than 400 spontaneous speech repairs was analysed, and the prosodic characteristics compared with the syntactic and semantic characteristics of each repair. Prosodic marking showed no relationship at all with the syntactic characteristics of repairs. Instead, marking was associated with certain semantic factors: repairs were marked when the original utterance had been actually erroneous, rather than simply less appropriate than the repair; and repairs tended to be marked more often when the set of items encompassing the error and the repair was small rather than when it was large. These findings lend further weight to the characterization of accent as essentially semantic in function.

Abstract

The core issue in the 19-century sources of psycholinguistics was the question, "Where does language come from?'' This genetic perspective unified the study of the ontogenesis, the phylogenesis, the microgenesis, and to some extent the neurogenesis of language. This paper makes the point that this original perspective is still a valid and attractive one. It is exemplified by a discussion of the genesis of spoken words.

Abstract

Philosophers, psychologists, and linguists have argued that spatial conception is pivotal to cognition in general, providing a general, egocentric, and universal framework for cognition as well as metaphors for conceptualizing many other domains. But in an aboriginal community in Northern Queensland, a system of cardinal directions informs not only language, but also memory for arbitrary spatial arrays and directions. This work suggests that fundamental cognitive parameters, like the system of coding spatial locations, can vary cross-culturally, in line with the language spoken by a community. This opens up the prospect of a fruitful dialogue between anthropology and the cognitive sciences on the complex interaction between cultural and universal factors in the constitution of mind.

Abstract

Wearing the hat of a cognitive anthropologist rather than an historian, I will try to amplify the ideas of Renn’s cited above. I argue that a particular subclass of material objects, namely “cognitive artifacts,” involves a close coupling of mind and artifact that acts like a brain prosthesis. Simple cognitive artifacts are external objects that act as aids to internal
computation, and not all cultures have extended inventories of these. Cognitive artifacts in this sense (e.g., calculating or measuring devices) have clearly played a central role in the history of science. But the notion can be widened to take in less material externalizations of cognition, like writing and language itself. A critical question here is how and why this close coupling of internal computation and external device actually works, a rather neglected question to which I’ll suggest some answers.

Abstract

Human communication displays a striking contrast between the diversity of languages and the universality of the principles underlying their use in conversation. Despite the importance of this interactional base, it is not obvious that it heavily imprints the structure of languages. However, a deep-time perspective suggests that early hominin communication was gestural, in line with all the other Hominidae. This gestural phase of early language development seems to have left its traces in the way in which spatial concepts, implemented in the hippocampus, provide organizing principles at the heart of grammar.

Abstract

All living beings try to save effort, and humans are no exception. This groundbreaking book shows how we save time and energy during communication by unconsciously making efficient choices in grammar, lexicon and phonology. It presents a new theory of 'communicative efficiency', the idea that language is designed to be as efficient as possible, as a system of communication. The new framework accounts for the diverse manifestations of communicative efficiency across a typologically broad range of languages, using various corpus-based and statistical approaches to explain speakers' bias towards efficiency. The author's unique interdisciplinary expertise allows her to provide rich evidence from a broad range of language sciences. She integrates diverse insights from over a hundred years of research into this comprehensible new theory, which she presents step-by-step in clear and accessible language. It is essential reading for language scientists, cognitive scientists and anyone interested in language use and communication.

Abstract

This article argues for a gradient approach to word order, which treats word order preferences, both within and across languages, as a continuous variable. Word order variability should be regarded as a basic assumption, rather than as something exceptional. Although this approach follows naturally from the emergentist usage-based view of language, we argue that it can be beneficial for all frameworks and linguistic domains, including language acquisition, processing, typology, language contact, language evolution and change, and formal approaches. Gradient approaches have been very fruitful in some domains, such as language processing, but their potential is not fully realized yet. This may be due to practical reasons. We discuss the most pressing methodological challenges in corpus-based and experimental research of word order and propose some practical solutions.

Abstract

Word classes play a central role in corpus linguistics under the name of parts of speech (POS). Many popular corpora are provided with POS tags. This chapter gives examples of popular tagsets and discusses the methods of automatic tagging. It also considers bottom-up approaches to POS induction, which are particularly important for the ‘poverty of stimulus’ debate in language acquisition research. The choice of optimal POS tagging involves many difficult decisions, which are related to the level of granularity, redundancy at different levels of corpus annotation, cross-linguistic applicability, language-specific descriptive adequacy, and dealing with fuzzy boundaries between POS. The chapter also discusses the problem of flexible word classes and demonstrates how corpus data with POS tags and syntactic dependencies can be used to quantify the level of flexibility in a language.

Abstract

There remains some debate about whether beta power effects observed during sentence comprehension reflect ongoing syntactic unification operations (beta-syntax hypothesis), or instead reflect maintenance or updating of the sentence-level representation (beta-maintenance hypothesis). In this study, we used magnetoencephalography to investigate beta power neural dynamics while participants read relative clause sentences that were initially ambiguous between a subject- or an object-relative reading. An additional condition included a grammatical violation at the disambiguation point in the relative clause sentences. The beta-maintenance hypothesis predicts a decrease in beta power at the disambiguation point for unexpected (and less preferred) object-relative clause sentences and grammatical violations, as both signal a need to update the sentence-level representation. While the beta-syntax hypothesis also predicts a beta power decrease for grammatical violations due to a disruption of syntactic unification operations, it instead predicts an increase in beta power for the object-relative clause condition because syntactic unification at the point of disambiguation becomes more demanding. We observed decreased beta power for both the agreement violation and object-relative clause conditions in typical left hemisphere language regions, which provides compelling support for the beta-maintenance hypothesis. Mid-frontal theta power effects were also present for grammatical violations and object-relative clause sentences, suggesting that violations and unexpected sentence interpretations are registered as conflicts by the brain's domain-general error detection system.

Abstract

Large language models that exhibit instruction-following behaviour represent one of the biggest recent upheavals in conversational interfaces, a trend in large part fuelled by the release of OpenAI's ChatGPT, a proprietary large language model for text generation fine-tuned through reinforcement learning from human feedback (LLM+RLHF). We review the risks of relying on proprietary software and survey the first crop of open-source projects of comparable architecture and functionality. The main contribution of this paper is to show that openness is differentiated, and to offer scientific documentation of degrees of openness in this fast-moving field. We evaluate projects in terms of openness of code, training data, model weights, RLHF data, licensing, scientific documentation, and access methods. We find that while there is a fast-growing list of projects billing themselves as 'open source', many inherit undocumented data of dubious legality, few share the all-important instruction-tuning (a key site where human labour is involved), and careful scientific documentation is exceedingly rare. Degrees of openness are relevant to fairness and accountability at all points, from data collection and curation to model architecture, and from training and fine-tuning to release and deployment.

Abstract

Speech recognition systems are a key intermediary in voice-driven human-computer interaction. Although speech recognition works well for pristine monologic audio, real-life use cases in open-ended interactive settings still present many challenges. We argue that timing is mission-critical for dialogue systems, and evaluate 5 major commercial ASR systems for their conversational and multilingual support. We find that word error rates for natural conversational data in 6 languages remain abysmal, and that overlap remains a key challenge (study 1). This impacts especially the recognition of conversational words (study 2), and in turn has dire consequences for downstream intent recognition (study 3). Our findings help to evaluate the current state of conversational ASR, contribute towards multidimensional error analysis and evaluation, and identify phenomena that need most attention on the way to build robust interactive speech technologies.

Abstract

For shared book reading to be effective for language development, the adult and child need to be highly engaged. The current paper adopted a mixed-methods approach to investigate caregiver’s language-boosting behaviours and children’s engagement during shared book reading. The results revealed there were more instances of joint attention and caregiver’s use of prompts during moments of higher engagement. However, instances of most language-boosting behaviours were similar across episodes of higher and lower engagement. Qualitative analysis assessing the link between children’s engagement and caregiver’s use of speech acts, revealed that speech acts do seem to contribute to high engagement, in combination with other aspects of the interaction.

Abstract

The intergenerational stability of auditory symbolic systems, such as music, is thought to rely on brain processes that allow the faithful transmission of complex sounds. Little is known about the functional and structural aspects of the human brain which support this ability, with a few studies pointing to the bilateral organization of auditory networks as a putative neural substrate. Here, we further tested this hypothesis by examining the role of left–right neuroanatomical asymmetries between auditory cortices. We collected neuroanatomical images from a large sample of participants (nonmusicians) and analyzed them with Freesurfer’s surface-based morphometry method. Weeks after scanning, the same individuals participated in a laboratory experiment that simulated music transmission: the signaling games. We found that high accuracy in the intergenerational transmission of an artificial tone system was associated with reduced rightward asymmetry of cortical thickness in Heschl’s sulcus. Our study suggests that the high-fidelity copying of melodic material may rely on the extent to which computational neuronal resources are distributed across hemispheres. Our data further support the role of interhemispheric brain organization in the cultural transmission and evolution of auditory symbolic systems.

Abstract

Mental simulation is an important aspect of narrative reading. In a previous study, we found that gaze durations are differentially impacted by different kinds of mental simulation. Motor simulation, perceptual simulation, and mentalizing as elicited by literary short stories influenced eye movements in distinguishable ways (Mak & Willems, 2019). In the current study, we investigated the existence of a common neural locus for these different kinds of simulation. We additionally investigated whether individual differences during reading, as indexed by the eye movements, are reflected in domain-specific activations in the brain. We found a variety of brain areas activated by simulation-eliciting content, both modality-specific brain areas and a general simulation area. Individual variation in percent signal change in activated areas was related to measures of story appreciation as well as personal characteristics (i.e., transportability, perspective taking). Taken together, these findings suggest that mental simulation is supported by both domain-specific processes grounded in previous experiences, and by the neural mechanisms that underlie higher-order language processing (e.g., situation model building, event indexing, integration).

Abstract

The human experience is shaped by information from different perceptual channels, but it is still debated whether and how differential experience influences language use. To address this, we compared congenitally blind, blindfolded, and sighted people's descriptions of the same motion events experienced auditorily by all participants (i.e., via sound alone) and conveyed in speech and gesture. Comparison of blind and sighted participants to blindfolded participants helped us disentangle the effects of a lifetime experience of being blind versus the task-specific effects of experiencing a motion event by sound alone. Compared to sighted people, blind people's speech focused more on path and less on manner of motion, and encoded paths in a more segmented fashion using more landmarks and path verbs. Gestures followed the speech, such that blind people pointed to landmarks more and depicted manner less than sighted people. This suggests that visual experience affects how people express spatial events in the multimodal language and that blindness may enhance sensitivity to paths of motion due to changes in event construal. These findings have implications for the claims that language processes are deeply rooted in our sensory experiences.

Abstract

Each sensory modality has different affordances: vision has higher spatial acuity than audition, whereas audition has better temporal acuity. This may have consequences for the encoding of events and its subsequent multimodal language production—an issue that has received relatively little attention to date. In this study, we compared motion events presented as audio-only, visual-only, or multimodal (visual + audio) input and measured speech and co-speech gesture depicting path and manner of motion in Turkish. Input modality affected speech production. Speakers with audio-only input produced more path descriptions and fewer manner descriptions in speech compared to speakers who received visual input. In contrast, the type and frequency of gestures did not change across conditions. Path-only gestures dominated throughout. Our results suggest that while speech is more susceptible to auditory vs. visual input in encoding aspects of motion events, gesture is less sensitive to such differences.

Abstract

We investigated cross-modal influences between speech and sign in hearing bimodal bilinguals, proficient in a spoken and a sign language, and its consequences on visual attention during message preparation using eye-tracking. We focused on spatial expressions in which sign languages, unlike spoken languages, have a modality-driven preference to mention grounds (big objects) prior to figures (smaller objects). We compared hearing bimodal bilinguals’ spatial expressions and visual attention in Dutch and Dutch Sign Language (N = 18) to those of their hearing non-signing (N = 20) and deaf signing peers (N = 18). In speech, hearing bimodal bilinguals expressed more ground-first descriptions and fixated grounds more than hearing non-signers, showing influence from sign. In sign, they used as many ground-first descriptions as deaf signers and fixated grounds equally often, demonstrating no influence from speech. Cross-linguistic influence of word order preference and visual attention in hearing bimodal bilinguals appears to be one-directional modulated by modality-driven differences.

Abstract

Gene duplication events can drive evolution by providing genetic material for new gene functions, and they create opportunities for diverse developmental strategies to emerge between species. To study the contribution of duplicated genes to human early development, we examined the evolution and function of NANOGP1, a tandem duplicate of the transcription factor NANOG. We found that NANOGP1 and NANOG have overlapping but distinct expression profiles, with high NANOGP1 expression restricted to early epiblast cells and naïve-state pluripotent stem cells. Sequence analysis and epitope-tagging revealed that NANOGP1 is protein coding with an intact homeobox domain. The duplication that created NANOGP1 occurred earlier in primate evolution than previously thought and has been retained only in great apes, whereas Old World monkeys have disabled the gene in different ways, including homeodomain point mutations. NANOGP1 is a strong inducer of naïve pluripotency; however, unlike NANOG, it is not required to maintain the undifferentiated status of human naïve pluripotent cells. By retaining expression, sequence and partial functional conservation with its ancestral copy, NANOGP1 exemplifies how gene duplication and subfunctionalisation can contribute to transcription factor activity in human pluripotency and development.

Abstract

We present a protocol to study naturalistic human communication using dual-EEG and audio-visual recordings. We describe preparatory steps for data collection including setup preparation, experiment design, and piloting. We then describe the data collection process in detail which consists of participant recruitment, experiment room preparation, and data collection. We also outline the kinds of research questions that can be addressed with the current protocol, including several analysis possibilities, from conversational to advanced time-frequency analyses.
For complete details on the use and execution of this protocol, please refer to Drijvers and Holler (2022).

Abstract

Individual differences in cognitive abilities are ubiquitous across the spectrum of proficient language users. Although speakers differ with regard to their memory capacity, ability for inhibiting distraction, and ability to shift between different processing levels, comprehension is generally successful. However, this does not mean it is identical across individuals; listeners and readers may rely on different processing strategies to exploit distributional information in the service of efficient understanding. In the following psycholinguistic reading experiment, we investigate potential sources of individual differences in the processing of co-occurring words. Participants read modifier-noun bigrams like absolute silence in a self-paced reading task. Backward transition probability (BTP) between the two lexemes was used to quantify the prominence of the bigram as a whole in comparison to the frequency of its parts. Of five individual difference measures (processing speed, verbal working memory, cognitive inhibition, global-local scope shifting, and personality), two proved to be significantly associated with the effect of BTP on reading times. Participants who could inhibit a distracting global environment in order to more efficiently retrieve a single part and those that preferred the local level in the shifting task showed greater effects of the co-occurrence probability of the parts. We conclude that some participants are more likely to retrieve bigrams via their parts and their co-occurrence statistics whereas others more readily retrieve the two words together as a single chunked unit.

Abstract

Iconicity in language is receiving increased attention from many fields, but our understanding of iconicity is only as good as the measures we use to quantify it. We collected iconicity measures for 304 Japanese words from English-speaking participants, using rating and guessing tasks. The words included ideophones (structurally marked depictive words) along with regular lexical items from similar semantic domains (e.g., fuwafuwa ‘fluffy’, jawarakai ‘soft’). The two measures correlated, speaking to their validity. However, ideophones received consistently higher iconicity ratings than other items, even when guessed at the same accuracies, suggesting the rating task is more sensitive to cues like structural markedness that frame words as iconic. These cues did not always guide participants to the meanings of ideophones in the guessing task, but they did make them more confident in their guesses, even when they were wrong. Consistently poor guessing results reflect the role different experiences play in shaping construals of iconicity. Using multiple measures in tandem allows us to explore the interplay between iconicity and these external factors. To facilitate this, we introduce a reproducible workflow for creating rating and guessing tasks from standardised wordlists, while also making improvements to the robustness, sensitivity and discriminability of previous approaches.

Abstract

The results of a word-spotting experiment are presented in which Dutch listeners tried to spot different types of bisyllabic Dutch words embedded in different types of nonsense contexts. Embedded verbs were not reliably harder to spot than embedded nouns; this suggests that nouns and verbs are recognised via the same basic processes. Iambic words were no harder to spot than trochaic words, suggesting that trochaic words are not in principle easier to recognise than iambic words. Words were harder to spot in consonantal contexts (i.e., contexts which themselves could not be words) than in longer contexts which contained at least one vowel (i.e., contexts which, though not words, were possible words of Dutch). A control experiment showed that this difference was not due to acoustic differences between the words in each context. The results support the claim that spoken-word recognition is sensitive to the viability of sound sequences as possible words.