Processing of musical sounds in primary auditory cortex of the awake monkey

Abstract

The studies described in this thesis aim at clarifying physiological bases of music perception. A first step towards this end requires an understanding of how perceptual attributes of complex sounds and sound sequences, characteristic of, but not unique to music, are represented by the auditory system. Accordingly, the present studies examine neural responses in auditory cortex to complex tones and to simple tone sequences and explore possible relationships between these responses and perceptual features of the sounds. All experiments are based on multi-contact electrode recordings of neuronal ensemble responses in primary auditory cortex (A1) of the awake macaque monkey. The studies utilize auditory evoked potential (AEP), multiunit activity (MUA), and current source density (CSD) techniques to examine stimulus-evoked synaptic events and action potential firing of neuronal populations simultaneously across multiple laminae within A1.;Human non-invasive studies suggest two models of auditory cortical organization relating to the representation of the pitch of complex sounds. Implications of these models are tested in Chapters 1 and 2. Chapter 3 examines neural representation of roughness, a perceptual correlate of amplitude fluctuations in the waveform envelope of complex sounds. Building upon the findings described in Chapter 3, Chapter 4 investigates neural responses contributing to the perception of consonance and dissonance of musical intervals. Finally, Chapter 5 examines neural correlates of auditory stream segregation, a well-known psychoacoustic phenomenon underlying the perceptual organization of auditory sequences and hence melodic patterns. Conclusions include: (1) No evidence was found in support of a topographic representation of pitch in A1, either in parallel with, or orthogonal to the representation of pure tone frequency. These findings urge a reevaluation of models of auditory cortical organization proposed on the basis of non-invasive studies and suggest alternative hypotheses regarding the physiological representation of pitch. (2) In general, complex tone response amplitudes reflect critical band filtering mechanisms, and thus correlate with the perceived loudness and resolvability of frequency components of complex sounds. (3) Phase-locking of neuronal ensemble responses in A1 to the amplitude modulation frequency (or difference frequency) of complex sounds correlates with psychoacoustic data relating to the perception of roughness, and may also underlie temporal encoding of pitch for complex tones composed of unresolved harmonics. (4) The magnitude of phase-locked activity in A1 correlates with the perceived dissonance of complex tone musical intervals. (5) Temporal response patterns evoked by alternating frequency tone sequences in A1 parallel the perceptual organization of auditory sequences as reflected in auditory stream segregation. Thus, on the whole, features of sounds relevant for the perception of melodic patterns, temporal pitch, and roughness/dissonance of musical chords appear to be represented in A1 via transient "On" responses and phase-locked activity. The dominant form of pitch which forms the basis of chroma in music appears not to be represented in A1 (at least on the basis of the response measures examined in the present studies), and may instead be derived via information processing within networks of multiple parallel and hierarchically organized primary and non-primary auditory cortical fields.