A tight connection between competence and performance is a central tenet of the usage-based model. Methodologically, however, corpus frequency is a poor predictor of acceptability – a phenomenon known as the “frequency/acceptability mismatch”. This article argues that the mismatch arises from a “methodological mismatch”, when simple frequency measures are mapped onto complex grammatical units. To illustrate, we discuss the results of acceptability judgments of go/come-v. The construction is subject to a formal constraint (Go see the doctor! vs. *He goes sees (...) the doctor), which results from its mandative semantics (directives, commissives). While a formal model makes no prediction with regard to gradient acceptability of bare (“grammatical”) go/come-v, the usage-based view assumes that acceptability is a function of compatibility with an abstract schema. The experimental ratings are compared with a number of corpus-derived measures: while acceptability is largely independent of (raw) frequency, it is not independent of frequency-related usage distribution. The results add to recent suggestions that the frequency/acceptability mismatch is substantially reduced if the syntactic complexity of a unit is appropriately captured in usage data. (shrink)

An important question within psycholinguistic research is whether grammatical features, such as number values on nouns, are probabilistic or discrete. Similarly, researchers have debated whether grammatical specifications are only set for individual lexical items, or whether certain types of noun phrases (NPs) also obtain number valuations at the phrasal level. Through a corpus analysis and an oral production task, we show that conjoined NPs can take both singular and plural verb agreement and that notional number (i.e., the numerosity of the (...) referent of the subject noun phrase) plays an important role in agreement with conjoined NPs. In two written production tasks, we show that participants who are exposed to plural (versus singular or unmarked) agreement with conjoined NPs in a biasing story are more likely to produce plural agreement with conjoined NPs on a subsequent production task. This suggests that, in addition to their sensitivity to notional information, conjoined NPs have probabilistic grammatical specifications that reflect their distributional properties in language. These results provide important evidence that grammatical number reflects language experience, and that this language experience impacts agreement at the phrasal level, and not just the lexical level. (shrink)

One of the central debates in the cognitive science of language has revolved around the nature of human linguistic competence. Whether syntactic competence should be characterized by abstract hierarchical structures or reduced to surface linear strings has been actively debated, but the nature of morphological competence has been insufficiently appreciated despite the parallel question in the cognitive science literature. In this paper, in order to investigate whether morphological competence should be characterized by abstract hierarchical structures, we conducted the crowdsourced acceptability (...) judgment experiment on morphologically complex words and evaluated five computational models of morphological competence against human acceptability judgments: Character Markov Models (Character), Syllable Markov Models (Syllable), Morpheme Markov Models (Morpheme), Hidden Markov Models (HMM), and Probabilistic Context-Free Grammars (PCFG). Our psycholinguistic experimentation and computational modeling demonstrated that "morphous" computational models with morpheme units outperformed "amorphous" computational models without morpheme units and, importantly, PCFG with hierarchical structures most accurately explained human acceptability judgments on several evaluation metrics, especially for morphologically complex words with nested morphological structures. Those results strongly suggest that human morphological competence should be characterized by abstract hierarchical structures internally generated by the grammar, not reduced to surface linear strings externally attested in large corpora. (shrink)

This paper contributes to the task of defining the relationship between the results of production and rating experiments in the context of language variation. We address the following research question: how may the grammatical options available to a single speaker be distributed in the two domains of production and perception? We argue that previous studies comparing acceptability judgments and frequencies of occurrence suffer from significant limitations. We approach the correspondence of production and perception data by adopting an experimental design different (...) from those used in previous research: (i) instead of using a corpus we use production data obtained experimentally from respondents who are later asked to make judgments, (ii) instead of pairwise phenomena we examine language variation, (iii) judgments are collected formally using the conditions and materials from the production experiment, (iv) we analyze the behavior of each participant across the production and acceptability judgment experiments. In particular, we examine three phenomena of variation in Russian: case variation in nominalizations, gender mismatch, and case variation in paucal constructions. Our results show that there is substantial alignment between acceptability ratings and frequency of occurrence. However, the distribution of frequencies and acceptability scores do not always correlate. Speakers are not consistent in choosing a single variant across the two types of experiment. Importantly, the types of inconsistency they display differ, which means that the variation can be characterized from this point of view. We conclude that the degree of coherence of the two experiments reflects the effects of the evolution of variation over time. Another result is that elicited production and acceptability judgments vary with respect to how they reveal variation in language. In the case of the development or disappearance of variants, production indicates this earlier than judgments, and the rating task has the effect of restricting the choices available to respondents. However, the production method should not thereby be considered more sensitive. We argue that only a combination of production and judgment data makes it possible to estimate the directionality of changes in variability and to see the full distribution of different variants. (shrink)

On the surface, bi- and multilingualism would seem to be an ideal context for exploring questions of typological proximity. The obvious intuition is that the more closely related two languages are, the easier it should be to implement the two languages in one mind. This is the starting point adopted here, but we immediately run into the difficulty that the overwhelming majority of cognitive, computational, and linguistic research on bi- and multilingualism exhibits a monolingual bias (i.e., where monolingual grammars are (...) used as the standard of comparison for outputs from bilingual grammars). The primary questions so far have focused on how bilinguals balance and switch between their two languages, but our perspective on typology leads us to consider the nature of bi- and multi-lingual systems as a whole. Following an initial proposal from Hsin ( 2014 ), we conjecture that bilingual grammars are neither isolated, nor (completely) conjoined with one another in the bilingual mind, but rather exist as integrated source grammars that are further mitigated by a common, combined grammar (Cook, 2016 ; Goldrick et al., 2016a, b ; Putnam and Klosinski, 2017 ). Here we conceive such a combined grammar in a parallel, distributed, and gradient architecture implemented in a shared vector-space model that employs compression through routinization and dimensionality reduction. We discuss the emergence of such representations and their function in the minds of bilinguals. This architecture aims to be consistent with empirical results on bilingual cognition and memory representations in computational cognitive architectures. (shrink)

A search for the terms ‘acceptability judgment tasks’ & ‘language’ and ‘grammaticality judgment tasks’ & ‘language’ produces results which report findings that are based on the exact same elicitation technique. Although certain scholars have argued that acceptability and grammaticality are two separable notions that refer to different concepts, there are contexts in which the two terms are used interchangeably. The present work shows that these two notions and their scales do not coincide: there are sentences that are acceptable, even though (...) they are ungrammatical, and sentences that are unacceptable, despite being grammatical. First we adduce a number of examples for both cases, including grammatical illusions, violations of Identity Avoidance, and sentences that involve a level of processing complexity that overloads the cognitive parser and tricks it into (un)acceptability. We then discuss whether the acceptability of grammatically ill-formed sentences entails that we actually assign a meaning to them. Last, it is shown that there are n ways of unacceptability, and two ways of ungrammaticality, in the absolute and the relative sense. Since the use of the terms ‘acceptable’ and ‘grammatical’ is often found in experiments that constitute the core of the evidential base of linguistics, disentangling their various uses is likely to aid the field reach a better level of terminological clarity. (shrink)