Representations of Pitch and Timbre Variation in Human Auditory Cortex

Abstract

Pitch and timbre are two primary dimensions of auditory perception, but how they are represented in the human brain remains a matter of contention. Some animal studies of auditory cortical processing have suggested modular processing, with different brain regions preferentially coding for pitch or timbre, whereas other studies have suggested a distributed code for different attributes across the same population of neurons. This study tested whether variations in pitch and timbre elicit activity in distinct regions of the human temporal lobes. Listeners were presented with sequences of sounds that varied in either fundamental frequency (eliciting changes in pitch) or spectral centroid (eliciting changes in brightness, an important attribute of timbre), with the degree of pitch or timbre variation in each sequence parametrically manipulated. The BOLD responses from auditory cortex increased with increasing sequence variance along each perceptual dimension. The spatial extent, region, and laterality of the cortical regions most responsive to variations in pitch or timbre at the univariate level of analysis were largely overlapping. However, patterns of activation in response to pitch or timbre variations were discriminable in most subjects at an individual level using multivoxel pattern analysis, suggesting a distributed coding of the two dimensions bilaterally in human auditory cortex.

Significance statement: Pitch and timbre are two crucial aspects of auditory perception. Pitch governs our perception of musical melodies and harmonies, and conveys both prosodic and (in tone languages) lexical information in speech. Brightness-an aspect of timbre or sound quality-allows us to distinguish different musical instruments and speech sounds. Frequency-mapping studies have revealed tonotopic organization in primary auditory cortex, but the use of pure tones or noise bands has precluded the possibility of dissociating pitch from brightness. Our results suggest a distributed code, with no clear anatomical distinctions between auditory cortical regions responsive to changes in either pitch or timbre, but also reveal a population code that can differentiate between changes in either dimension within the same cortical regions.

Keywords: Heschl's gyrus; auditory cortex; fMRI; perception; pitch; timbre.

Figure 1.

Schematic diagrams of the stimuli. A, Spectral representation of the stimuli used in this study (plotted on log–log axes). Changing the F₀ value results in changes in the frequencies of the harmonics (represented by the vertical lines). Changing the CF of the filter results in changes in the spectral centroid of the sound and hence in changes in the amplitudes (but not frequencies) of the harmonics. Lighter-colored arrows indicate that shifting in the rightward direction results in a sound with a higher pitch (increase in F₀) or a brighter timbre (increase in spectral centroid). B, Tone sequences with small and large step sizes. For the pitch sequences, the y-axis is F₀, centered around 200 Hz; for the timbre sequences, the y-axis is spectral centroid, centered around 900 Hz. C, Experimental block design layout. Thirty second pitch- and timbre-varying sequences are indicated in blue and green, respectively. Fifteen second silent gaps for a baseline measure are indicated in gray. The presentation order of step sizes, indicated in white text, was randomized. All possible step sizes across both dimensions were included in each scan.

Figure 2.

Group-level statistical maps of pitch (top) and timbre (bottom), pooled across all step sizes, both contrasted with silence. A cluster in each of right and left superior temporal gyri for pitch [center of mass: right (R), 56, −16, 8; left (L), −53, −22, 9) and timbre (center of mass: R, 56, −18, 9; L, −53, −24, 9) conditions, respectively. Color scale values range from −1 to 1, in units of percentage change relative to baseline. No voxels survive the contrast of pitch and timbre (pitch-timbre).

Figure 3.

Bar graphs showing mean β weights in the percentage change across all subjects' iROIs at each step size (1, 2, 5, and 10 DL) for pitch (top row) and timbre (bottom row) in each hemisphere (left and right). Error bars indicate ± 1 SEM across subjects.

Figure 4.

A, B, Group-level correlation coefficient maps. A, Heat maps of positive mean Fisher's z-transformed correlation coefficients (ZCOR) for pitch (top) and timbre (bottom), limited to voxels within a union of all subjects' iROI masks. No significant negative correlations were found. A cluster is shown in each hemisphere for pitch [peak: right (R), 52, −10, 6; left (L), −46, −24, 10] and timbre (peak: R, 48, −20, 12; L, −52, −18, 6) conditions, respectively. B, Maps indicating which voxels the maps in A have in common. The significant correlation coefficients within the pitch map (blue), the significant correlation coefficients within the timbre map (green), and the voxels these two maps have in common (red).

Figure 5.

Spatial distribution of the iROI masks in the auditory cortex in each hemisphere with respect to Heschl's gyrus. A, Individual subject's inflated brain (left) with iROI mask and a flattened patch (right) of the auditory cortex. Heschl's gyrus (black dashed line) and superior temporal gyrus (STG; white dashed line) are labeled for this subject. B, Summation of iROI masks across all subjects in the left hemisphere (left) and right hemisphere (right), color coded to indicate the number of subjects for which each surface vertex was inside their iROI.

Figure 6.

Spatial distribution of correlation coefficients for pitch and timbre. A, B, Left hemisphere (blues) and right hemisphere (reds) contrast maps within the sound mask (vertices inside the auditory ROI of at least five subjects), with darker colors indicating that pitch had a higher correlation coefficient in a given voxel. To the right and bottom are projections of the mean (SD) proportion of variance explained, parallel and perpendicular to Heschl's gyrus. C, D, Distribution of the contrast between variance explained by pitch and timbre step size across all voxels within the mask in each hemisphere. E, Variance-weighted COM for each subject for each dimension in each hemisphere. Black lines connect the center of mass for each condition within a hemisphere for each subject. Inset above demonstrates how small the spatial range is for the COMs. STG, Superior temporal gyrus.

Figure 7.

A, Excitation patterns for the highest and lowest steps of the largest step size (10× DL) for the pitch and timbre conditions, respectively. Lighter colors indicate the higher pitch and brighter timbre, respectively. B, Scatter plot showing mean β weight across all 10 subjects at each step size, averaged across hemispheres as a function of ΔE with a linear regression line for each dimension. Lighter colors indicate larger step sizes. Error bars indicate ±1 SEM across subjects.