Each human language possesses a set of distinctive syntactic rules. Here, we show that balanced Welsh-English bilinguals reading in English unconsciously apply a morphosyntactic rule that only exists in Welsh. The Welsh soft mutation rule determines whether the initial consonant of a noun changes based on the grammatical context (e.g., the feminine noun cath-"cat" mutates into gath in the phrase y gath-"the cat"). Using event-related brain potentials, we establish that English nouns artificially mutated according to the Welsh mutation rule (e.g., "goncert" instead of "concert") require significantly less processing effort than the same nouns implicitly violating Welsh syntax. Crucially, this effect is found whether or not the mutation affects the same initial consonant in English and Welsh, showing that Welsh syntax is applied to English regardless of phonological overlap between the two languages. Overall, these results demonstrate for the first time that abstract syntactic rules transfer anomalously from one language to the other, even when such rules exist only in one language.

Here, we used event-related potentials to test the predictions of two prominent accounts of code-switching in bilinguals: The Matrix Language Framework (MLF; Myers-Scotton, 1993 ) and an application of the Minimalist Programme (MP; Cantone and MacSwan, 2009 ). We focused on the relative order of the noun with respect to the adjective in mixed Welsh–English nominal constructions given the clear contrast between pre- and post-nominal adjective position between Welsh and English. MP would predict that the language of the adjective should determine felicitous word order (i.e., English adjectives should appear pre-nominally and Welsh adjectives post-nominally). In contrast, MLF contends that it is the language of the finite verb inflexion rather than that of a particular word that governs felicitous word order. To assess the predictions of the two models, we constructed sentences featuring a code-switch between the adjective and the noun, that complied with either English or Welsh word-order. Highly proficient Welsh–English bilinguals made semantic acceptability judgements upon reading the last word of sentences which could violate MP assumptions, MLF assumptions, both assumptions, or neither. Behaviourally, MP violations had no significant effect, whereas MLF violations induced an average drop of 11% in acceptability judgements. Neurophysiologically, MP violations elicited a significant Left Anterior Negativity (LAN) modulation, whereas MLF violations modulated both P600 and LAN mean amplitudes. In addition, there was a significant interaction between MP and MLF status in the P600 range: When MP was violated, MLF status did not matter, and when MP criteria were met, MLF violations resulted in a P600 modulation. This interaction possibly reflects a general preference for noun over adjective insertions, and may provide support for MLF over MP at a global sentence processing level. Model predictions also manifested differently in each of the matrix languages (MLs): When the ML was Welsh, MP and MLF violations elicited greater P600 mean amplitudes than MP and MLF adherences, however, this pattern was not observed when the ML was English. We discuss methodological considerations relating to the neuroscientific study of code-switching, and the extent to which our results shed light on adjective-noun code-switching beyond findings from production and experimental-behavioural studies.

The power of poetry is universally acknowledged, but it is debatable whether its appreciation is reserved for experts. Here, we show that readers with no particular knowledge of a traditional form of Welsh poetry unconsciously distinguish phrases conforming to its complex poetic construction rules from those that violate them. We studied the brain response of native speakers of Welsh as they read meaningful sentences ending in a word that either complied with strict poetic construction rules, violated rules of consonantal repetition, violated stress pattern, or violated both these constraints. Upon reading the last word of each sentence, participants indicated sentence acceptability. As expected, our inexperienced participants did not explicitly distinguish between sentences that conformed to the poetic rules from those that violated them. However, in the case of orthodox sentences, the critical word elicited a distinctive brain response characteristic of target detection –the P3b– as compared to the other conditions, showing that speakers of Welsh with no expertise of this particular form of poetry implicitly detect poetic harmony. These results show for the first time that before we even consider literal meaning, the musical properties of poetry speak to the human mind in ways that escape consciousness.

Recent studies have revealed remarkable interactions between language and emotion. Here, we show that such interactions influence judgments made regarding cultural information. Balanced Welsh–English bilinguals categorized statements about their native Welsh culture as true or false. Whilst participants categorized positive statements as true when they were true, they were biased towards categorizing them as true also when they were false, irrespective of the language in which they read them. Surprisingly, participants were unbiased when categorizing negative statements presented in their native language Welsh, but showed a reverse bias - categorizing sentences as false, even when they were true - for negative statements when they read them in English. The locus of this behavior originated from online semantic evaluation of the statements, shown in corresponding modulations of the N400 peak of event-related brain potentials. These findings suggest that bilinguals perceive and react to cultural information in a language-dependent fashion.

The growing interdisciplinary research field of psycholinguistics is in constant need of new and up-to-date tools which will allow researchers to answer complex questions, but also expand on languages other than English, which dominates the field. One type of such tools are picture datasets which provide naming norms for everyday objects. However, existing databases tend to be small in terms of the number of items they include, and have also been normed in a limited number of languages, despite the recent boom in multilingualism research. In this paper we present the Multilingual Picture (Multipic) database, containing naming norms and familiarity scores for 500 coloured pictures, in thirty-two languages or language varieties from around the world. The data was validated with standard methods that have been used for existing picture datasets. This is the first dataset to provide naming norms, and translation equivalents, for such a variety of languages; as such, it will be of particular value to psycholinguists and other interested researchers. The dataset has been made freely available.