How we identify and control different languages

Elly Brosig (University of Stuttgart, Germany) [BIO]

Email: brosig.isapl@gmx.de

 

ABSTRACT:

When we hear or read a sound or word sequence, how do we decide to which language this input belongs, so it can be stored into the proper language system? - It can be assumed that words of different languages are identified by certain phonetic, morphological, and semantic features specific of a particular language which activate neural connections in our brain.

To find out exactly which features are responsible for identification, the author developed an experiment in which 106 subjects had to decide whether 10 written and 10 orally presented words or non-words could be English.- Results revealed that in written language subjects rely primarily on orthographic and morphological cues. Spoken speech, however, seems to be largly identified by phonetic similarity to the target language to the point that English words pronounced with German pronunciation were rejected by 70% of the subjects. - An interesting new finding was that prosodic elements also see to be very important. A word with an un-English syllable stress was rejected by 80% of the subjects.

Another interesting problem is how bi- or multilingual speakers exercise control over their different languages. Ordinarily they can separate their languages quite well by consciously activating the one or the other language system. This is done unconsciously, if the verbal input and the speaker’s output occur in only one language. Mixed bilingual or ambiguous language input, however, causes inferences in most people.- The author has developed a bilingual processing model demonstrating normal bilingual processing and faulty processing in the case of ambiguous input.