Three experimental studies have been completed during the 24 months of the fellowship. These studies explored the interaction between auditory and speech processing by combining two approaches: developmental psycholinguistics studying the acquisition of a language, and psychoacoustics, describing the auditory mechanisms linked to the perception of speech sounds.
A first study explored in normal-hearing adults, whether the native language influences the auditory processing of the speech features. The auditory brain responses involved in the processing of speech were measured using electroencephalography (EEG) in three groups of listeners: French, Mandarin and English. Consonants from these three languages show subtle acoustic differences in timing and voicing (vibrations of the vocal cords). Preliminary analyses suggest that the EEG responses for the consonants differed according to their acoustic properties, and also that the native language of the listeners might influence these responses.
A second study used a series of child-friendly behavioural tasks to assess the interaction between auditory and speech processing during childhood. This experimental study looked at the relationship between children's hearing abilities to detect changes in sounds over time (amplitude modulations) and their perception of speech in noisy backgrounds (like in a noisy classroom). Time is a very important aspect in hearing, since almost all sounds vary over time. More importantly, temporal variations in speech sounds conveyed meaningful information. Three groups of children participated in the study: 5-to-6-year olds, 7-to-8-year-olds and 10-to-11-year olds. Half of the children were English monolinguals; the other half were bilinguals. Three child-friendly games were implemented on a touch-screen tablet. In all the games, three animals appeared on the screen and produced a sound one after the other. Participants were asked to find the “one-odd out”. The results showed that the ability to detect temporal variations in sounds is well developed by 5 years of age. The three groups of children showed exquisite ability to detect really small acoustic variations. However, the ability to focus on these temporal variations varied according to the duration of the sounds. When the sounds were really short, better detection was observed after 7 years of age. Then, the ability to memorise the temporal variations within long sounds takes even longer to develop, suggesting that the memory for this type of acoustic information improves over childhood. Finally, the ability to correctly identify consonants in noise was linked to children’s ability to hear these temporal variations within long sounds. In other words, children who showed better memory for the temporal variations tended to be better at hearing speech in noise. Interestingly, no difference was observed in any games between monolingual and bilingual children.
Finally, a third study assessed how hearing impairment affects the interaction between auditory mechanisms and language acquisition. Hearing-impaired people usually struggle to hear speech in challenging noisy conditions. A group of children with mild-to-moderate sensori-neural hearing loss, aged from 5 to 7 years, participated in the same series of behavioural games. Children were recruited with the help of parents’ support groups, local education authorities and clinicians in London. Results showed that hearing-impaired children show rather good abilities to detect changes in sounds over time. However, they show poor speech in noise perception compared to normal-hearing children. Regarding the relationship between the auditory tasks, the same trend as for the normal-hearing children was observed, suggesting that the ability to detect and use efficiently temporal information may contribute to a better understanding of speech in noise in this population as well.
