Does Talker Familiarity or Time of Testing Facilitate Sentence Recognition When Listening in Noise?

At the most elementary level, the speech signal is comprised of two parts: linguistic information and indexical information. The linguistic information is the phonetic information of the signal and indexical information is speaker specific and is the paralinguistic information of the signal. Part of this indexical information is talker specific characteristics; which have been shown to help people understand speech. The talker specific characteristic we looked at was talker familiarity. Talker familiarity has been shown to help babies segment speech and adults listen in noise and recall stories. We looked at talker familiarity to see if it would benefit typically developing adults listen in ecologically valid background noise. Our hypotheses were: two significant main effects and interaction. Our study had two independent variables; talker (familiar, novel) and time of testing (Time 1, Time 2) and the dependent variable was keyword accuracy. A total of 93 individuals participated in this study; 41 of which were familiar with the talker due to the talker being their university professor. Our results showed a main effect of talker and a main effect of time of testing but there was no interaction between talker and time of testing. Implications are discussed