Conclusions
The study of melody perception provides both a challenge and an opportunity for cognit-
ive neuroscience. Melodies, like sentences, are complex structures, and unraveling the
neural mechanisms by which they are understood is a project which is still in its infancy.
However, melodies do afford some attractive properties to the cognitive neuroscientist.
Their raw materials can be extremely simple (e.g. pure tones instead of phonemes), and
their complexity can be systematically varied using quantitative principles.^34 Thus they
provide a valuable tool with which to study basic issues in cognitive neuroscience, such as
the interaction of bottom-up and top-down processes in sequence perception. Indeed, it is
possible that neural principles learned from using melodies may prove useful in under-
standing brain processing of other types of complex sequences, such as sentences.
This chapter has described a new approach to the neuroscience of melody, based on a
method which monitors stimulus-related cortical activity over time during the perception
of individual tone sequences.44,mUsing the aSSR, we have demonstrated that it is possible
to extract a signal from the human cerebral cortex which reflects the pitch contour an
individual is hearing. The accuracy with which this signal reflects the pitch contour
improves as the pitch sequence becomes more predictable. Thus there may be top-down
influences of musical expectancy which influence this brain signal, suggesting that future
studies may be able to use aSSR dynamics to monitor how expectancy is structured in time.
The basis of aSSR phase tracking is temporalinformation in cortical activity. When the
amountof activity was examined, no relationship with pitch contour was observed. This
suggests that dynamic imaging techniques have an important role to play in the study of
music perception, complementing techniques sensitive to the amount of neural activity but
insensitive to the fine temporal structure of that activity (e.g. PET & fMRI).
Dynamic imaging techniques also offer the opportunity to study how brain areas inter-
act during perception. It is clear from decades of neural research that the brain is divided
into different regions, each of which has a special role to play in perception and cognition.
Yet it is also clear that these brain areas must interact to form coherent and unified per-
cepts. Complex patterns such as music and speech engage multiple brain regions, and
sequences with different perceptual properties may be distinguished by the pattern of brain
interactions they engender rather than by the particular brain regions which respond to
them. Using phase coherence, we examined brain interactions as a function of stimulus
structure and found that sequences with melody-like statistics were associated with the
greatest degree of neural interactions. In particular, we found evidence for strong func-
tional coupling between left posterior hemisphere and right hemisphere regions during the
perception of melody-like sequences. This may reflect the perceptual integration of local
and global pitch patterns, and suggests that one neural signature of melody is the dynamic
integration of brain areas which process structure at different time scales.
Two obvious directions to pursue with the dynamic aSSR approach are the examination
of tonality’s influence on brain dynamics and the dynamic neural correlates of rhythm per-
ception. The dynamic aSSR approach may also provide a way to examine brain interactions
341mFor a related approach based on the visual steady-state response, see Harris and Silberstein. 63