This study was run additionally to those originally proposed and in response to the Covid-19 pandemic lockdown. Since we could not conduct in-person testing, we trialled online testing via video conference calls and explored the extent to which learning via linguistic experiences occurs during remote conversations. We investigated whether children implicitly learn language during online interactions as they do during in-person interactions, and whether this is affected by the visual co-presence of a partner. During an online conference call, three- and five-year-olds alternated describing pictures with an experimenter who produced active and passive prime descriptions; half the participants had video+audio calls, and half had audio-only. Both age groups produced more passives after passive than active primes, both immediately and with accumulating input across trials; neither effect was influenced by call format. These results demonstrate that implicit grammar learning mechanisms operate during online interactions, and highlight the potential of online methodologies for developmental language production research.How we learn and use language is, not surprisingly, related to the language we experience around us: ultimately children who are exposed to English learn English, but more specifically, research shows that children exposed to varied language input (wider vocabularies, diverse sentences) come to develop more extensive language skills than those with narrower input. While we know that children's experience with language is important for shaping their learning of language, it remains unclear precisely how our experience with language influences our language development: what aspects of language experience are important, and how do children make use of them? Our project investigates how children learn from their language experiences, and the underlying learning mechanisms that they use to do so. Understanding the mechanisms that support language learning is theoretically important, because it informs our understanding of a uniquely human ability. But it also has substantial societal implications, because successful language development is critical to later educational and social attainment. Our project focuses on how language experience affects children's use of particular grammatical or syntactic structures - the way words are ordered in a sentence. We know that one way in which both children and adults make use of the language they experience is by re-using it. This re-use occurs not just for specific words (e.g., saying 'sofa' instead of 'settee' because that's what your friend just called it), but also at more abstract levels, including grammar. For instance, you are more likely to say "those pictures were drawn by Quentin Blake" after hearing someone say "this book was written by Roald Dahl". This grammatical repetition is known as syntactic priming: hearing a word order makes it easier for you to re-use that order, even with different words. When a speaker repeats a structure they have been exposed to, it indicates that they have a mental representation for that structure that they can use when understanding a sentence, and then re-use when planning a new sentence to say. So patterns of syntactic priming effects in children provide evidence about which grammatical structures they know. More recent research suggests that the influence of such priming lasts longer than the immediate context. In fact, recent models of language processing propose that priming can lead to long-term learning, so that when someone experiences a grammatical structure, it sets in train a persistent change in their language representations. For example, when a child hears "this book was written by Roald Dahl", it does not just help them to re-use that structure immediately, it also strengthens their knowledge of that structure. But it is unclear how this learning occurs: how does children's immediate experience with language translate into longer-term development of adult-like representations? In this research we will conduct psycholinguistic experiments with children at different stages of language development in which we vary which types of structure children hear and say, in different contexts and at different time points. We will examine how likely they are to re-use these structures and how long-lasting such effects are, in order to cast light on the way in which children can learn from different elements of the language they experience. We will address the following questions: When in language development is this type of experience important? To what extent is experience-based learning linked to specific contexts, whether linguistic (experience of particular words) or non-linguistic (experience within a particular conversation or task), and to what extent does it generalise? Does learning occur to the same extent through the act of hearing different word orders and the act of saying them? Answering these questions will ultimately help us understand the mechanisms, conditions and trajectories underlying successful language development.

Fifty-eight three-year-olds (28 females; Mage: 3;6; range: 3;2–3;11 years) and sixty-two five-year-olds (31 females; Mage: 5;7; range: 5;2–5;11 years) with no reported developmental or language delays took part in the experiment. All participants were monolingual British English speakers except three, who were simultaneously acquiring another language but still heard English from their primary caregiver at least 80% of the time. Participants were recruited online as a convenience sample from all over the UK via lab databases and social media. The experimenter and participant alternated in describing pictures of transitive events in a ‘Snap’ game task (Branigan et al., 2005) that were displayed side-by-side within a PsychoPy experiment and shown via screen-share on a Microsoft Teams video call. The experimenter described the first picture using a scripted prime sentence (either active or passive) and revealed the participant’s target picture, which the participant then described. The experimenter and participant took turns to describe 56 pictures (48 prime-target pairs and 8 snap pairs). Each session was recorded and participants’ utterances were later transcribed and coded for syntactic form according to strict (adult like) and lax coding schemes.