Study title

Estonia, Finland, Poland

Individual

The three languages were tested on location, in nurseries in countries where the language is spoken by the majority. All tested children were reported to be typically developing, monolingual speakers of their respective languages. The ages ranged between 3;0–5;0, such that children would be sufficiently old to be able to complete the task, and sufficiently young to be expected to produce errors. We had a balanced sample size of 40 children per language group, with a span of just over two months between the mean ages of the groups. In all three languages, surface-form frequency and PND counts (both PND class and PND form) were obtained automatically from child-directed speech corpora, using a custom-written R script. The study employed an elicited-production method and within-subjects design. Due to the large number of cases in each of the languages, it was not possible for us to test children on every case. Instead, we selected the most frequently occurring singular cases, based on the child-directed speech corpora in each language. Each case was represented pictorially using a unique elicitation context involving a fox or hare character interacting with each target noun object. To prevent fatigue effects, each participant was tested on all nouns, but each in only three of the five (Polish) or six (Finnish and Estonian) case contexts. This resulted in 90 (Polish), 81 (Finnish) and 72 (Estonian) trials per participant. Each child was tested individually in a quiet setting at their nursery. The full set of trials for each participant was presented in two experimental sessions, lasting approximately 15 min each. There was a break between the experimental sessions; the second session was conducted either a few hours after the first session or on a subsequent visit during the next few days.In each experimental trial, the participant was presented with a picture of the object on the screen and told the name of the object in nominative form. S/he was then shown the stimulus picture (i.e., the object in a case elicitation context), and the experimenter produced the beginning of the context sentence, for example:1) Experimenter: ‘This is an apple-NOM.’ [screen shows an apple] 2) Experimenter: ‘The fox looks at…’ [screen shows picture of the fox looking at an apple] 3) Participant: ‘apple-ACC’