Gautreau, A. , Hoen, M. & Meunier, F.
Centre de Recherche en Neurosciences de Lyon, CNRS – INSERM- Université Lyon
Our research aims at exploring psycholinguistic processes implicated in a situation favoring information competition: speech-in-speech comprehension. Our studies focus on the nature of interferences observed during speech-in-speech comprehension. Our goal is to identify information levels in which competitions, leading to interferences, can occur. In a first series of experiments, we used cocktail-party signals in different world languages: French, Breton, Irish and Italian. Cocktail-party signals were composed of 4 talkers. Participants had to identify French target-words inserted in a babble noise with a signal-to-noise ratio of 0 or -5dB. Results confirmed that it is more difficult to understand French words with French background noise than in a babble composed of languages unknown to listeners. Results obtained with unknown languages also showed differences in the observed performances depending on the language spoken in the background and demonstrated that some languages interfered more with French than some others. Overall these results suggested that the interference effect is not purely acoustic but also linguistic. To test it more directly, we used fluctuating noises matched to each cocktail-party signals in order to control for acoustic differences between the different language specific babbles such as speech rate and spectrum. Thus in a second experiment, the cocktail-party signals in the different world languages were used again as babble noise and served also to generate language specific fluctuating speech noises. Overall, results revealed a worse identification of the target-words with cocktail-party signals than with matched fluctuating noises of about 20%, showing that part of the interference is linguistic. Finally, a bigger difference between the two types of backgrounds (fluctuating noises - cocktail-party signals) was observed in French compared to all the other languages while no difference between the unknown languages is observed. These results suggested a strong effect of lexical competition in the French-in-French situation.