The experiment reported here investigated the neural correlates of predictive processing of angry and neutral speech. Twenty-six participants listened to recordings of angry and neutral conversation segments, as well as to speech-shaped noise, while their EEG was recorded. Oscillatory power in the gamma band (30–80 Hz) and the N400 component of event-related potentials (ERP) to sentence-final words were analyzed. In comparison to neutral words, negative emotional valence significantly reduced the amplitude of the N400 elicited by sentence-final words. Furthermore, there was larger gamma power during exposure to angry speech in comparison to neutral speech. The results generally suggest increased prediction and facilitated semantic integration in negative as compared to neutral speech. To date, the predictability effects on gamma power have been reported with relation to the semantic-lexical content of words. The present findings demonstrate that gamma power is also modulated by the emotional content of speech.

The present study investigated online neural indices of statistical learning of silent speech. Adult participants were exposed to naturally mouthed, silent syllable streams in an artificial language in two conditions. In one condition, 12 syllables occurred randomly; in the other the syllables were structured into four syllable triplets, i.e. statistical words. In the recorded EEG signal, phase synchronisation in neural oscillations was assessed at the rate of syllables and words occurring in the exposure streams. Largest phase synchronisation was detected for the word rate during exposure to the structured stream. Moreover, the neural synchronisation to word rate increased throughout the exposure within the structured stream. In a behavioural post-test, however, no learning effects were detected. The EEG results demonstrate sensitivity to statistical regularities in viewed silent speech. These findings indicate that statistical learning in speech and language can be effectively measured online even in the absence of auditory cues.

Humans learn much about their language while still in thewomb. Prenatal exposure has been repeatedly shown to affectnewborn infants’ processing of the prosodic characteristics ofnative language speech. Little is known about whether and howprenatal exposure affects infants’ perception of speech soundsegments. Here we simulated prenatal learning of vowels intwo virtual fetuses whose mothers spoke (slightly) differentlanguages. The learners were two-layer neural networks andwere each exposed to vowel tokens sampled from an existentfive-vowel language (Spanish and Czech, respectively). Theinput acoustic properties approximated the speech signal thatcould possibly be heard in the intrauterine environment, andthe learners’ auditory system was relatively immature. Withoutsupervision, the virtual fetuses came to warp the continuousacoustic signal into “proto-categories” that were specific totheir linguistic environment. Both learners came to create twocategorization patterns and did so in language-specific ways,primarily on the basis of the vowels’ first-formantcharacteristics. Such prenatally formed proto-categories werenot adult-like in that they entirely collapsed some of the native-language contrasts. At the same time, the categories reflectedfeatures of the adult language in that they were language-specific. These results can inspire future work on speech andlanguage acquisition in real young humans.

At birth, the human brain is tuned to spoken language in general and to some extent also to native language in particular. In behavioral studies, infants also prefer to listen to infant-directed speech (IDS) to adults-directed speech (ADS), apparently most robustly in their native language. Recent studies demonstrated that this preference has correlates at the neural level as well. We test whether newborns show differential neural tracking of native over foreign, rhythmically different, language. We assess neural tracking of native and non-native speech in Czech-exposed newborns. Newborns' were played a children's story in two rhythmically different languages, Czech (lacking acoustic cues to word-level stress) and Russian (acoustically salient word-level stress), in IDS or ADS, while their EEG was recorded. We predicted stronger neural tracking of the native Czech, evident in larger inter-trial phase coherence (ITC), and total power. Preliminary data (n = 27 out of planned 60) suggest this language-specific effect is most prominent in the theta band corresponding to the syllable rate. We will further test whether this native-language effect would be more prominent in ADS or IDS. Data collection is underway and the results will be presented & discussed at the conference.

Tuning into a novel language is a particularly difficult task for many adults. While the rhythmic and melodic patterns, i.e. prosody, bootstrap language acquisition in infancy, they are considerably challenging to learn in adulthood. Is it because of an age-related decline of the language-learning ability or because of unfavourable learning conditions? We investigated whether adults can auditorily sensitise to the prosody of a novel language, and whether such sensitisation is affected by concurrent presentation of alphabetic transcription. After 5 minutes of exposure to Māori, Czech listeners could reliably recognize this language in a post-test using low-pass filtered clips of Māori and Malay recorded by new speakers. Recognition accuracy was lower for participants exposed to the novel-language speech along with deep-orthography transcriptions or shallow orthography with unfamiliar characters. Adults can thus attune to novel-language prosody, but orthography hampers this ability. This has implications for language acquisition theories and learning practice.