Frequency of occurrence in the input is a main factor determining the ease of acquisition in first language learners. However, little is known about the factors relevant for the acquisition of low-frequency items. We examine the use of aspectual markers in a longitudinal corpus of Chintang (Sino-Tibetan, Nepal) children (ages 2;1-4;5). Only 7.7% of all Chintang verbs are overtly marked for aspect. Chintang has three aspect markers, one of which is substantially more frequent than the others. One of the low-frequency markers is positionally and prosodically more salient, appearing at the word-boundary. Using a Bayesian beta-binomial model, we assess the distribution and flexibility of use of aspectual markers in the input and children's production. Our analysis shows that the most frequent marker was acquired earliest, as predicted. For the low-frequency markers, position, segmentability and uniformity are better predictors of ease of acquisition.

The initial phase of linguistic production by children is char-acterized by rote-learned, lexically restricted forms and con-structions. Only during later phases of language acquisitiondo they develop flexibility across a paradigm and mix lexicaland grammatical material more freely. In the development ofverb morphology, a correlation between the use of tense andaspect has been observed in many languages. It has been sug-gested that this leads to an intermediary state of paradigm cat-egorization based on temporal categories. So far the flexibilityof individual verbs occurring in different tense-aspect combi-nations has not been examined in detail. Here we evaluate theflexibility of verb use in a large longitudinal corpus of 4 Rus-sian children. We compute the Shannon entropy of verb stemsdistributed over individual grammatical forms. Results showthat children do not pass through a stage of paradigm cate-gorization based on aspecto-temporal categories. After a briefitem-specific phase of rote learned forms, they quickly becomeflexible users of verbs in both aspects.

We study repair in child-directed, child-surrounding and child speech in longitudinal corpora of 4 languages: English, Russian, Chintang and Indonesian (age range: 2;01-3;04). We distinguish open requests (e.g. 'huh?'), restricted requests (e.g. 'you saw what?'), and restricted offers (reformulation or recast, e.g. 'you saw a bird?'). Our results indicate that in the aggregated model, clarification requests develop in children independently from adult speech, pointing to early universal emergence. When we analyse repair types separately, only restricted offers in both CDS and CSS are significantly predictive factors for the number of reformulations in child speech. Since this repair type is used by caregivers to provide both positive and negative feedback to children, they follow a special path of acquisition dependent on input distributions. Therefore, we propose that early repair acquisition relies on individual cognitive development of children as well as language exposure to the recast frequency in the caregiver speech.

A key aspect in child language development involves inducing the rules that determine the relations of the arguments to their verbal predicate, i.e., semantic roles. Here, we investigate whether child-directed speech facilitates learning ‘who does what to whom' in English and Russian, two languages that strongly differ in their amount of case-marking and word order variation. We ask whether a contextual, distributional learner can more easily learn to assign semantic roles to arguments based on child-directed speech versus adult-directed speech. To this end, we represent the arguments of a verb with contextualised word embeddings extracted from neural language models. We compare the classification accuracy of semantic roles based on these representations between utterances extracted from corpora of child-directed speech and adult-directed speech. We further study to what extent semantic roles can be predicted based on arguments represented by different levels of information, such as non-contextualised representations, the position in the sentence, and case marking. We find that child-directed speech facilitates the learning of semantic roles, an important cornerstone for learning the morphosyntactic features of a language. However, the effect of child-directed speech is more pronounced in Russian than in English, indicating that child-directed speech may be optimised more strongly in a language where arguments are expressed in more varied forms and positions, as is the case in Russian.

Learning to understand and use agreement is an integral part of children's linguistic development. In Romance languages, this includes gender and number agreement between the controller and attributive or predicative adjectives or participles. We examine the development of this category in a case where children's task is complicated by syncretisms, multiple paradigms, and unequal input distributions. Romansh Tuatschin (Romance, Indo-European, Switzerland) presents children with two distinct paradigms for attributive (masculine and feminine only) and predicative (masculine, feminine, and neuter/unmarked) contexts of adjective and participle use. The masculine form in predicative use is the same as the neuter form in attributive usage. Thus the masculine form in these two paradigms differs. This could be challenging for the language learner. The distribution of these forms is heavily skewed towards the \emph{neuter} in predicative contexts but balanced in attributives. Examining production errors in children between 2;0 and 4;3, we evaluate the effects of frequency and syncretism and find that error-rate is affected by skewed distributions and less affected by syncretisms. This demonstrates the strong effect of input distributions on first language acquisition.

Learning to establish joint reference is an important milestone of communicative and linguistic development. Pointing is one of the first entry points into this process, since gestures often precede verbal communication. During early development, as well as later language use, pointing and linguistic utterances interact in many ways, complementing each other. However, little is known about the development of this relationship during development. In this paper, we focus on the development of the co-occurrence of finger pointing and accompanying utterances in two different cultures: Russia and Chintang (Sino-Tibetan, Eastern Nepal). We show that despite the differences in environment, the development of finger pointing and accompanying language use show substantial similarities. Early on, a larger proportion of points is not accompanied by language. As the children's linguistic abilities develop, children first use language to specify what is being pointed at, and later elaborate on some aspect of the referent.

Infant points are often accompanied by preverbal vocalisations but little is known about their acoustic properties. In this study we explore the role of point-accompanying preverbal infant vocalisations in expressing intention, declarative or imperative. We test whether the interpretation of these preverbal expressions is independent from the communicative and cultural context by assessing Swiss adult speakers’ ability to interpret preverbal point-accompanying vocalisations from Shipibo-Konibo (Peru) children. Results suggest that adults can easily distinguish declarative and imperative preverbal pointing acts by their accompanying vocalisations. We thus show that acoustic properties of these vocalisations contain a rich amount of information which is understood by members of a different culture. This hints towards a larger preverbal repertoire of communicative tools than previously assumed.

Despite considerable differences in the structures of the world's languages and child-rearing practices, children show remarkable cross-linguistic similarities in their early lexical development, including a preference for nouns. Here, we analyze children's early lexical production in naturalistic longitudinal corpora in a large-scale cross-linguistic comparison of 10 typologically highly diverse languages. We assess morphological complexity as a possible explanatory variable for children's higher noun-to-verb ratios and evaluate whether children's gradual increase in morphological productivity is correlated with their gradual decrease in noun-to-verb ratios towards the level found in their ambient language. We show that in languages with complex verb morphology, children exhibit a higher deviation in their noun-to-verb ratio compared to adults. This deviation gradually diminishes as they become more productive in the use of their target language. This effect holds across languages, despite their differences in morphological complexity.