A central question for cognitive science is whether children’s linguistic productivity can be captured by item-based learning, or whether the learner must be guided by abstract, system-wide principles governed by innate constraints. Here, we present a computational model of early language acquisition which learns to discover and use lexically-based frames in a fully incremental, on-line fashion. The model is rooted in simple prediction- and recognition-based processes, subject to the same memory limitations as language learners. When exposed to English corpora of child-directed speech, the model is able learn developmentally plausible frames and use them to capture over 70% of the utterances produced by target children aged 2 to 5. Across a typologically diverse range of 29 languages, the model is able to capture over 68% of child utterances. Together, these findings suggest that much of children’s early linguistic productivity can be captured by item-based learning through computationally simple mechanisms.

There are considerable differences in language processing skill among the normal population. A key question for cognitive science is whether these differences can be ascribed to variations in domain-general cognitive abilities, hypothesized to play a role in language, such as working memory and statistical learning. In this paper, we present experimental evidence pointing to a fundamental memory skill—chunking—as an important predictor of crossindividual variation in complex language processing. Specifically, we demonstrate that chunking ability reflects experience with language, as measured by a standard serial recall task involving consonant combinations drawn from naturally occurring text. Our results reveal considerable individual differences in participants’ ability to use chunk frequency information to facilitate sequence recall. Strikingly, these differences predict variations across participants in the on-line processing of complex sentences involving relative clauses. Our study thus presents the first evidence tying the fundamental ability for chunking to sentence processing skill, providing empirical support for construction-based approaches to language.

Several recent empirical findings have reinforced the notion that a basic learning and memory skill—chunking—plays a fundamental role in language processing. Here, we provide evidence that chunking shapes sentence processing at multiple levels of linguistic abstraction, consistent with a recent theoretical proposal by Christiansen and Chater (2016). Individual differences in chunking ability at two different levels is shown to predict on-line sentence processing in separate ways: i) phonological chunking ability, as assessed by a variation on the non-word repetition task, predicts processing of complex sentences featuring phonological overlap; ii) multiword chunking ability, as assessed by a variation on the serial recall task, is shown to predict reading times for sentences featuring long-distance number agreement with locally distracting number-marked nouns. Together, our findings suggest that individual differences in chunking ability shape language processing at multiple levels of abstraction, consistent with the notion of language acquisition as learning to process.

Subject-auxiliary inversion in interrogatives has been a topic of great interest in language acquisition research, and has often been held up as evidence for the structure-dependence of grammar. Usage-based and nativist approaches posit different representations and processes underlying children’s question formation and therefore predict different causes for these errors. Here, we explore the question of whether input statistics predict children’s spontaneous non-inversion errors with wh- questions. In contrast to previous studies, we look at properties of the non-inverted, errorful forms of questions. Through a series of corpus analyses, we show that the frequency of uninverted subsequences (e.g., “she is going” in “what she is going to do?*”) is a good predictor of children’s errors, consistent with recent evidence for multiword units in children’s comprehension and production. This finding has implications for the types of mental representations and cognitive processes researchers ascribe to children acquiring a first language.

Attempts to connect individual differences in statisticallearning with broader aspects of cognition have receivedconsiderable attention, but have yielded mixed results. Apossible explanation is that statistical learning is typicallytested using the two-alternative forced choice (2AFC) task.As a meta-cognitive task relying on explicit familiarityjudgments, 2AFC may not accurately capture implicitlyformed statistical computations. In this paper, we adapt theclassic serial-recall memory paradigm to implicitly teststatistical learning in a statistically-induced chunking recall(SICR) task. We hypothesized that artificial languageexposure would lead subjects to chunk recurring statisticalpatterns, facilitating recall of words from the input.Experiment 1 demonstrates that SICR offers more fine-grained insights into individual differences in statisticallearning than 2AFC. Experiment 2 shows that SICR hashigher test-retest reliability than that reported for 2AFC. Thus,SICR offers a more sensitive measure of individualdifferences, suggesting that basic chunking abilities mayexplain statistical learning.