Hemispheres in conflict, when the left is mad, but the right is sad

Abstract

The two cerebral hemispheres have distinct processing strengths. However, almost any task calls on the skills of both hemispheres. In this thesis, I explore the integration of left- and right-hemisphere processes in speech perception. Previous research has demonstrated that the left hemisphere is specialized for processing the linguistic aspects of speech, and that the right hemisphere is specialized for processing prosody, or information that is carried in the tone of voice. The present series of experiments used an interference paradigm in which the linguistic content of the stimulus conflicted with the tone of voice in which it was spoken. Two theoretical viewpoints were considered. According to the shielding hypothesis, the fact that linguistic and prosodic processes are carried out in opposite hemispheres should minimize the interference between them. However, an alternative view is that there is a bilateral speech processing module with a specialized callosal relay channel to maximize integration (and therefore interference) between the two dimensions. These hypotheses were tested in a series of four experiments.

The first two experiments were designed to demonstrate that the stimuli met two criteria - linguistic and prosodic dimensions were processed in opposite hemispheres, and they produced interference. Experiments 3 and 4 used dichotic listening techniques to compare interference within a hemisphere to interference between hemispheres. Results from both experiments were incompatible with the shielding hypothesis, and the results from Experiment 4 were consistent with the specialized callosal relay hypothesis, in that interference was greater across hemispheres than within hemisphere.

In summary, the findings are consistent with the hypothesis that there is a specialized callosal relay channel between linguistic processing centres in the left hemisphere and prosodic processing centres in the right hemisphere. In the present study, this bilateral processing system maximized interference between linguistic and prosodic processes. However, in most speech processing situations, linguistic and prosodic information is congruent. The bilateral processing system would therefore lead to highly efficient integration of both dimensions.