Altered functional connectivity during speech perception in congenital amusia
Figures

Schematic of experimental paradigm.
(A) Example spectrograms of the early closure (top) and late closure (bottom) stimuli for the Both-Informative condition. Fundamental frequency contours are indicated with blue lines. The relative duration of the critical words are indicated with orange and green boxes. To the right, syntactic trees for the two sentences are shown to highlight the grammatical structure indicated by the phrase boundaries. (B) The time course of a single trial. Participants read a text version of the sentence from the screen, which was either early or late closure. This was followed by auditory presentation of the late and early closure versions. After both recordings were played, participants chose whether the first or second recording they heard matched the visually presentence sentence better. A single whole-brain volume was acquired after the button press, timed to capture the peak of the hemodynamic response roughly around presentation of the second sentence.

In-scanner performance.
Prosodic categorization performance measured in the scanner (proportion correct); each point represents the performance of a single participant.
-
Figure 2—source data 1
n-scanner performance.
- https://cdn.elifesciences.org/articles/53539/elife-53539-fig2-data1-v2.txt

Seed locations and group differences in seed-to-whole brain functional connectivity.
Inflated surfaces show the locations of False Discovery Rate-corrected group differences (Control > Amusia) in whole-brain connectivity (yellow crosses, minimum Z > 4.61), which were used as seeds in subsequent analyses (minimum Z > 3.57; warm colors indicate greater connectivity in the control than amusia participants). All four seed vertices were located in inferior frontal cortices (left inferior frontal gyrus, left DLPFC, right inferior frontal gyrus p. triangularis, and right inferior frontal gyrus p. orbitalis) (A) Significant group differences (Control > Amusia) in functional connectivity with left hemisphere seeds. The largest decreases in connectivity in the amusia group were located in right superior temporal plane and gyrus, the posterior middle temporal gyrus onto the inferior bank of the superior temporal sulcus, and anterior insula. (B) Significant group differences (control vs amusia) in functional connectivity with right hemisphere seeds. Prominent decreases in connectivity with the right inferior frontal gyrus in individuals with amusia were observed in the superior temporal plane and regions of occipital, frontal, and parietal cortex.

Connectivity between L DLPFC and insular (left) and auditory (right) cortex is modulated by normalized cue weights measured outside the scanner.
Correlation coefficients are Spearman rho.
-
Figure 4—source data 1
Brain-behaviour correlations.
- https://cdn.elifesciences.org/articles/53539/elife-53539-fig4-data1-v2.txt

Connectivity between L DLPFC and right insula (left) and between L DLPFC and right auditory cortex (right) were reduced in the amusia group during speech perception (Control >Amusia, p=0.0001 for both ROI pairs), but not during passive tone perception.
-
Figure 5—source data 1
LDLPFC-ROI connectivity.
- https://cdn.elifesciences.org/articles/53539/elife-53539-fig5-data1-v2.txt
Tables
Significant main effects of Group involving functional connectivity between seed areas and subcortical Structures.
All effects are Control > Amusia.
Seed | Region of interest | F(1,87) | p |
---|---|---|---|
L IFG | R Accumbens | 15.43 | 0.0002 |
L DLPFC | L Putamen | 15.78 | 0.0001459 |
R Putamen | 17.78 | 0.00006047 | |
L Caudate | 25.23 | 0.0000027 | |
R Caudate | 11.51 | 0.001044 | |
L Cerebellum | 24.47 | 0.00000364 | |
R Cerebellum | 16.23 | 0.0001194 | |
L Pallidum | 14.60 | 0.0002484 | |
R Pallidum | 12.44 | 0.0006739 | |
L Thalamus | 14.83 | 0.0002245 | |
R Thalamus | 15.72 | 0.0001501 | |
R IFG (orbit) | L Thalamus | 14.83 | 0.0002245 |
R IFG (triang.) | L Accumbens | 10.10 | 0.002054 |