Evidence for Cerebellar Contributions to Adaptive Plasticity in Speech Perception

Abstract

Human speech perception rapidly adapts to maintain comprehension under adverse listening conditions. For example, with exposure listeners can adapt to heavily accented speech produced by a non-native speaker. Outside the domain of speech perception, adaptive changes in sensory and motor processing have been attributed to cerebellar functions. The present functional magnetic resonance imaging study investigates whether adaptation in speech perception also involves the cerebellum. Acoustic stimuli were distorted using a vocoding plus spectral-shift manipulation and presented in a word recognition task. Regions in the cerebellum that showed differences before versus after adaptation were identified, and the relationship between activity during adaptation and subsequent behavioral improvements was examined. These analyses implicated the right Crus I region of the cerebellum in adaptive changes in speech perception. A functional correlation analysis with the right Crus I as a seed region probed for cerebral cortical regions with covarying hemodynamic responses during the adaptation period. The results provided evidence of a functional network between the cerebellum and language-related regions in the temporal and parietal lobes of the cerebral cortex. Consistent with known cerebellar contributions to sensorimotor adaptation, cerebro-cerebellar interactions may support supervised learning mechanisms that rely on sensory prediction error signals in speech perception.

Keywords: adaptation; cerebellum; fMRI; perceptual learning; supervised learning.

Figure 1.

Waveforms (Time × Amplitude, top) and Spectrograms (Time × Frequency, with amplitude in gray scale, bottom) of an example word “zone.” On the left is the undistorted stimulus, in the middle the stimulus at the moderate distortion (13.25 mm), and on the right, the stimulus at the severe distortion (15.25 mm).

Figure 2.

This figure illustrates the experimental and trial design. The main analysis compares Block 2 and Block 5. The trial design depicts a Written-Response trial. Two-thirds of the trials in each condition required a Written-Response. The other one-third of the trials did not require a Written-Response and had a fixation instead of the “question mark” after stimulus presentation.

Figure 3.

Partial word accuracy performance for the distorted speech conditions used in the fMRI analyses. Error bars represent standard errors of the mean. The partial word accuracy score was computed by multiplying the total number of correct (i.e., in-order matching) phonemes by the ratio of the number of phonemes in the target stimulus to the number in the elicited response, or vice versa. For example, the score for a stimulus “Yeast” where the Response is “Least” is 0.75. The partial word accuracy scores used also penalize extraneous and missing phonemes. More examples of accuracy scores are provided in Supplementary Materials.

Figure 4.

Significant regions in the whole-brain Pretest versus Posttest contrast at a corrected cluster size for a voxel-wise threshold P < 0.001, at an alpha level of 0.05 from Table 1. Significant voxel clusters were found in the left superior frontal gyrus and right middle frontal gyrus (top panel) and the left postcentral gyrus and right superior temporal gyrus (bottom panel).

Figure 5.

Sagittal and coronal slices (left is right and right is left) for significant regions in the cerebellum Pretest versus Posttest contrast at P < 0.001. (a) Right Crus I. On the left is a sagittal view at x = 38 and on right is a coronal view at y = −44. (b) Regions in the left and right Lobule V/VI. On the left is a sagittal view at x = −21 and on the right a coronal view at y = −36 showing 2 regions in the left Lobule V/VI regions. (c) Time course showing a percent signal change from baseline for Pre- and Posttest in the right Crus I (top panel) and right Lobule V/VI (bottom panel). Error bars represent standard errors of the mean. (d) Time course showing a percent signal change for Written-Response compared with No-Response trials in the right Crus I (top panel) and right Lobule V/VI (bottom panel). Error bars represent standard errors of the mean.

Figure 6.

Scatter plot showing the relationship between behavioral adaptive plasticity measured as a residual gain of partial word accuracy scores (y-axis) and a rank of % BOLD signal from the baseline (x-axis).

Figure 7.

Functional connectivity map for the right Crus I. Sagittal slice at x = −50 and x = −35 (top panel) and right Lobule V/VI (bottom panel). Voxel-wise threshold at P < 0.005 (cluster threshold = 52 voxels) showing t-values from t-tests conducted on correlations of the Fisher z-transformation of the square root of R-values. Positive t-values are colored in red and negative t-values are colored in blue.