activation. The finding of an activation within the right STG while listening to melodies fits in
well with the lesion data, however, and likely reflects the specialization of neuronal networks
within the right secondary auditory cortices for perceptual analysis of tonal information.
Zatorre et al.^21 also examined working memory for pitch by subtracting the passive con-
dition from the two-note condition, which resulted in significant activation within the right
frontal lobe (both inferiorly and superiorly). The first/last minus passive melodies subtrac-
tion yielded a large number of cortical and subcortical activation sites in both hemispheres,
including a focus in the right inferior frontal cortex identical to that observed in the two-
note condition. A comparison not reported in the original study pits the first/last condition
against the two-note condition. The advantage of this contrast is that both involve pitch
comparisons, the main difference being the degree of working memory load. The interest-
ing finding here (Figure 16.4) is once again an area of increased blood flow in the right infe-
rior frontal region, though not in the same location as in the other subtractions. The pattern
of results from these conditions therefore implicates frontal-lobe mechanisms in effecting
pitch comparisons, with a particularly important contribution from right-frontal regions. It
should be recalled, however, that in the first/last minus passive melody comparison, a large
number of separate foci of CBF change were seen over a wide swath of cortical and subcor-
tical territory; this finding perhaps reflects the complexity and increased cognitive demands
of the task, which was also manifested in increased error rate and slower reaction times.
One may speculate that the numerous frontal lobe sites observed might be associated
with successful performance of distinct aspects of the task. The findings of Perry et al.^23 are
directly relevant to this issue; they observed CBF increases with a rightward asymmetry in dor-
solateral frontal areas while subjects were engaged in an auditory tonal working memory task
that required them to monitor a sequence of tones. This finding suggests a functional dissoci-
ation between the cognitive act of monitoring information as opposed to holding sensory
239Figure 16.5Cartoon depiction of putative differences in anatomical structure between left and right auditory
cortical areas.Trianglesand linesrepresent individual neurons with differing degrees of myelination (thicker on
the left). Neurons on the left are larger than those on the right, have wider spacing across tonotopically organized
cortex, and are more widely interconnected. These structural features would lead to relatively fast conduction
but poor spectral resolution on the left, and to poorer temporal resolution but greater sensitivity to fine spectral
differences on the right.