and the hallucinations are often pleasurable rather than frightening or threatening. Some
subjects sing along with the music, and some report that they can switch to another tune by
consciously thinking about it. In some cases the pleasurable aspects of the music deteriorate
over time, becoming distorted or too loud, or degenerating into tinnitus.14,15 The
pathophysiological basis of such hallucinations is uncertain; plausibly, reduced sensory
input results in disinhibition of perception-bearing circuits, and perceptual traces are
‘released’.
Synesthesia is an involuntary perception produced by stimulation of another sense. Thus,
sounds produce the perception of colors. Geometric patters can occur, but more elaborate
formed hallucinations do not. In some cases particular pitches produce the same colour; for
example, C produced red, D green, and E blue in a boy with poor vision and musical talent
(but not absolute pitch).^16 In addition to deafferentation, synesthesia is associated with hal-
lucinogenic drug use (e.g. LSD) and, in one case, with a brain stem neoplasm. Many cases
are idiopathic and noted in early childhood; some are familial. Women and left-handers are
overrepresented, as are people who have difficulties with spatial relations.^17 Alexander
Scriabin and Nicholai Rimsky-Korsakoff are believed to have had coloured-hearing synes-
thesia. The disorder, if it can be called that, is an example of unusually strong cross-modal
association, but the physiological basis is otherwise unclear.Negative phenomena: amusia and aphasia
Amusia is an acquired impairment of musical processing. As with aphasia—a comparable
impairment of language—the problem can be expressive, receptive, or both, and the dis-
order cannot be explained by damage either to the articulatory apparatus or to primary
receptor mechanisms. In other words, tongue paralysis is not expressive amusia, and deaf-
ness is not receptive amusia. Amusia and aphasia can occur together or each can occur in
the absence of the other.
Since the nineteenth century, clinicians investigating the amusias have tended to adopt
either a localizationist/anatomical or a holistic/psychological approach. Among the local-
izers, Henschen classified amusia as either motor or sensory and then identified ‘centres’ in
the left cerebral hemisphere: singing resided in the pars triangularis of the third frontal con-
volution, musical reception in the temporal pole, note reading near the angular gyrus, and
instrument playing at the foot of the second frontal convolution (with a separate centre for
violinists).18,19Similarly, Kleist described motor and sensory amusia, each affecting either
tone or melody. His anatomical localization included separate centers for singing and for
whistling.^20
Opposed to such rigid diagramming was Feuchtwanger, who considered music too com-
plex to localize or even to restrict to one cerebral hemisphere.^21 Ustvedt noted the hetero-
geneity of amusic patients and the lack, in most reports, of premorbid baseline information
or of standardized tests.^22 A review of such reports reveals that heterogeneity is indeed the
rule and that the presence or absence of aphasia does not predict the type of amusia.
The first report of aphasia and relatively preserved musical ability was in 1745; the
patient, who had right hemiparesis and speech limited to the word ‘yes’, could sing hymns
if someone else sang along.^23 Similar cases include two children with nonfluent aphasia, 183Zat-Ch12.qxd 6/6/03 3:43 AM Page 183