Tipultech logo

MR's Spared Speech Prosody

Dawn Senathi-Raja

Peretz's Taxonomic Model

Recent cases have shown comparable deficits for music perception and prosody perception (Nicholson et al., 2003; Patel et al., 1998). These parallel deficits have led Peretz and others to postulate the existence of a common cognitive module or neural mechanism at some level that co-processes melodic and rhythmic aspects of both speech and music. MR, however, provides the first example of dissociated prosodic and musical recognition of analogical tasks. MR has an abnormally large threshold for detecting pitch changes between tones (Beckwith, 2003); he also has difficulty in judging the direction of pitch change between tones and difficulty discriminating short tones sequences on the basis of pitch contour (Beckwith, 2003). Therefore, it is reasonable to suggest that the dissociation between his linguistic and non-linguistic performance may be due to the fact that pitch contrasts in sentences are coarser than in music and consequently, large enough to overcome his deficits in pitch change detection. However, the non-linguistic analogues exactly reproduced the pitch movements and temporal patterns of speech intonation. Thus, the discrepancy between Peretz's proposition and MR's results do not appear to be task-related. The dissociation between MR's recognition of music and prosodic sounds suggests that these two processes must be analysed separately at some level, and that it is at this level that MR's lesion causes disproportionate impairment.

Pitch Perception Account

The perception of speech prosody relies on an ability to detect slow changes in pitch, intensity, and timing of the melody and rhythm of language (Griffiths, 2002). MR retains the ability to detect slow modulations in continuous sound. Coupled with the finding that MR is not aprosodic, this encourages the view that slow pitch pattern processing, rather than fine temporal processing of pitch for which MR was impaired, may be necessary for perceiving the prosody of language. Thus, MR's case does not support the notion that the perception of prosodic speech and music rely on similar changes in the acoustic signal (Nicholson et al., 2003; Patel et al., 1998).

Pitch pattern perception of discrete sounds at a higher-order temporal level is necessary for processing rapidly changing energy peaks that are characteristic of many speech consonants (Griffiths, 2001; Griffiths et al., 1998; Griffiths et al., 2001; Nicholls, 1996). MR showed impairment for tasks that were, in all respects, identical to the speech prosody tasks, except that they were removed from a speech context. Thus, while MR's ability to perceive consonants may have contributed to his performance in the prosody domain, this ability did not benefit his performance in the musical domain.

Auditory Scene Analysis Account

Pitch-based sequential streaming enables the auditory system to keep track of sounds from one source (Bregman, 2002). Based on MR's sequential streaming deficit, it would be reasonable to expect an inability to keep track of a series of spoken inflections. On the contrary, MR was able to track inflections in order to perceive the prosody of speech. As the general or specific nature of MR's sequential streaming deficit is inconclusive, it is unknown whether he is capable of keeping track of a stream of slow changing inflections. If his streaming deficit is specific to sounds of short temporal durations, MR's preserved prosodic perception is completely accounted for. If, however, his streaming deficit occurs for all sequences of sound irrespective of temporal structure, MR's preserved prosodic perception may be explained by way of a buffering of his high-level language system, which may compensate for his lower-level deficits.

<< >>