Shanley E. M. Allen

How humans come to acquire language has been a persistent object of philosophical and scientific inquiry, in part because it bears on fundamental questions about the nature of human cognition.Classically, this field of inquiry has centered innateness as a critical component of language acquisition, traditionally characterizing innate capability in terms of richly specified, biologically endowed linguistic structure, such as Universal Grammar. Contemporary theories have evolved considerably from these earlier formulations and now present several differing candidates for innateness in language. Throughout this evolution, though, questions of innateness have remained central to theories of language acquisition, even as the nature of those questions has shifted in response to new sources of evidence. In this Perspective, we consider evidence from three complementary sources of evidence—linguistic diversity, human social interaction, and language models (LMs)—that together inform the question of what children must bring to the task of language learning. We conclude by identifying desiderata for theories of language acquisition that seek to specify the nature, scope, and explanatory role of innate contributions to language development.

Messages sent in digital communication today, through apps such as WhatsApp, can often be understood in different ways due to the lack of non-verbal cues such as tone and body language. This causes communication problems to be more prevalent in digital chats than in face-to-face communication. With the help of emojis, some of these aspects can be brought back, such as emotion. Because messages can contain more than one emoji, we aim to determine how the reduplication of emojis influences the perception of messages, and whether demographic aspects have an impact on this effect. We conducted a survey with 1,005 participants that analyzed how participants rate the sentiment of neutral sentences and how their perception changes when the sentences are enriched with increasing numbers of positive, negative, or neutral emojis – up to three. The results show that the perceived sentiment of the sentences got stronger when the emojis were reduplicated. Furthermore, this effect per emoji got weaker the more reduplications were in the sentence. This means that the more reduplications appear, the weaker the effect of each emoji is. Our data showed no evidence that demographic aspects influenced the observed effect.

Intra-individual language contact in bilinguals is considered a potential source for the emergence of structural innovations in a language, eventually leading to grammatical language change. This study investigates the psycholinguistic mechanisms involved in this process, focusing on cross-linguistic structural innovation priming. In a web-based self-paced reading experiment with production pre- and posttests, we tested Canadian French–English bilinguals on innovative French ditransitive and monotransitive structures primed by English sentences with the same structure or by control primes. No priming effect emerged for monotransitives. For ditransitives, however, reading times in the segment immediately following the innovation were significantly faster when primed by the corresponding English structure. In production, the proportion of innovative sentences did not significantly increase from pretest to posttest for either structure. Yet, production rates of innovative forms in both tasks were modulated by the individual degree of French contact. We discuss these differential outcomes with reference to theoretical accounts of the psycholinguistics of contact-induced change. Overall, these findings suggest that cross-linguistic priming can provide a pathway for structural innovations to enter bilingual grammars, potentially leading to language change. However, such processes are apparently constrained by the linguistic properties of the respective structure.

Left dislocation (LD) is described as a feature of informal, spoken discourse. However, most research on LD relies on a collection of examples or on datasets with a restricted mode/formality. To understand LD use across discourse types, we investigate the frequency of LDs in a corpus of 287 participants with four narrative types: informal spoken, formal spoken, informal written, formal written. We consider the speaker characteristics of age (adolescent and adult), gender (female and male), and speaker group (English monolinguals and German, Greek, Russian, and Turkish heritage speakers). We also analyze LD characteristics (e.g., noun phrase, function) based on formality. We found that the spoken mode contained more LDs than the written one and that informal narratives had more LDs than formal ones. However, the effect of formality was modulated by speaker group: only Turkish heritage speakers produced more LDs in the informal than formal narratives, while the other groups showed no evidence of a formality effect. No evidence of age or gender effects was discovered. Lastly, participants used LDs of the same characteristics in both formalities. Thus, the study confirms previous findings on LD use across discourse types and raises the importance of including bilingualism as a speaker characteristic.

We investigate timing and eye-movement behavior during semantic prediction in L1 and L2 speakers of English using the Visual World Paradigm, additionally exploring speech rate. We differentiate first-stage predictions, considered to be automatic and relatively cost-free, from second-stage predictions, which are non-automatic and more cognitively demanding, with differences between L1 and L2 speakers believed to arise in second-stage predictions. We found no differences in the divergence of looks to the target in first- or second-stage predictions across groups. However, speech rate played an important role. Both L1 and L2 speakers showed similar first-stage predictions at slower speech rates, but L1 speakers showed earlier predictions as the speech rate increased. L2 speakers showed reduced and more variable second-stage predictions, suggesting they were impacted during the more demanding second-stage prediction. This may indicate a wait-and-see strategy to help reduce costs associated with second-stage prediction.

Demonstratives and locative adverbs cross-linguistically are typically acquired relatively late, with children initially overusing proximal forms. However, these findings are largely based on research in languages with only two or three demonstratives. It is unclear whether the findings extend to languages with more complex systems. The present study examines data from Inuktitut, a language of the Inuit-Yupik-Unangan family, which has 20 demonstrative roots and 10 locative adverb roots representing six spatial distinctions. It uses data from 18 Inuktitut speakers (8–60 years) to investigate the target-like use of demonstratives/locatives and data from eight Inuktitut-speaking children (1–4 years) and their mothers to determine the acquisition trajectories of these structures. Children initially used only the proximal demonstratives/locatives, which aligns with prior research. The proportion of proximal forms out of all others decreased significantly with mean length of utterance in morphemes (MLUm), and by MLUm 2.50, children were using the full demonstrative/locative paradigm in a target-like manner. This differs from prior research and highlights the importance of language diversity in acquisition research.

Research has shown that first (L1) and second language (L2) speakers actively make predictions about upcoming linguistic information, though L2 speakers are less efficient. While prediction mechanisms are assumed to be qualitatively the same, quantitative prediction-driven processing differences may be modulated by individual differences We tested whether L2 proficiency and quality of lexical representation (QLR) impact the capacity of L2 readers to extract parafoveal information while reading, leading to quantitative differences in prediction. Using the same items as Slattery and Yates, we investigated the impact of predictability and length of a critical word on bottom-up parafoveal processing, measured by skipping rates, and top-down predictability processing, measured by reading times. Comparing our L2 English to their L1 English data, we found that L2 speakers skipped less and had longer gaze duration. However, both groups showed increased skipping rate and decreased gaze duration for predictable relative to unpredictable words and for shorter relative to longer words. We argue that L1 and L2 predictability mechanisms are qualitatively the same and quantitative differences stem from L2 speakers’ Reduced Capacity for Parafoveal Processing, the ReCaPP hypothesis.

Bilinguals have been shown to adapt to syntactic innovations (i.e., structures that deviate from the standard grammar) either by producing such structures more or by processing them faster after repeated exposure. However, research on whether they adapt by increasing their acceptability ratings for innovations is limited. We consider this to be a crucial gap in the literature, since it could provide insights into how speakers adapt their perception for innovations that they might otherwise not adapt to in their production and/or processing. On this basis, the present study investigates overall acceptability and trial-by-trial acceptability (adaptation) for different types of innovations in Canadian French with grammatical structural equivalents in English. Structure type and individual differences in language experience (dominance, proficiency, exposure, etc.) are considered as factors that influence these processes, as previous research has shown that they play a role in the acceptability of innovations in bilinguals. For this purpose, we employed a timed acceptability judgment task (TAJT), where adult bilingual speakers of French and English in Canada were asked to rate innovative sentences in French and their standard (grammatical) counterparts as fast and spontaneously as possible. Both acceptability ratings (offline measure) and response times (RTs) (online measure) across trials were measured to test whether speakers show adaptation on both levels. Results revealed that innovations were rated lower and for most structure types slower than their standard counterparts, with the different types of innovations showing differences. Crucially, adaptation on a group level was reflected only in response times and not in acceptability ratings. On an individual level, though, some participants adapted their ratings, but not consistently across all innovation types. Moreover, ratings and RTs were influenced by individual language experience, with participants with a higher contact with French (higher French Score) being faster and less accepting of innovative sentences compared to participants with a lower contact with French.