Fig 1.
The top melody shows the in-key condition in which no note is out-of-key (all notes are in G-major). The middle melody shows the out-of-key condition in which only the tone coinciding with the stressed syllable of the relative clause verb (circled) is out-of-key. The bottom melody shows the auditory anomaly condition in which all notes are in G-major but the critical tone is 10dB louder (boxed). The lowest pitch used across all melodies was F#2 (92.5 Hz) and the highest was E4 (329.6 Hz). The Dutch sentence in the figure means: The athletes that the mistress noticed looked out of the window.
Fig 2.
A) The language main effect (OR > SR) found in the whole-brain analysis (p < .005 uncorrected, cluster size = 87 voxels). B) Left hemisphere structural ROI. The BOLD effect of the linguistic manipulation is shown (OR—SR) with the associated p-value of a paired t-test above the bar. The significance level of the interaction effect is denoted above the line. Bars represent the activity difference (OR-SR) to sequences in which the stressed syllable of the critical word was sung in-key, out-of-key or unusually loudly (auditory anomaly). C) Right hemisphere structural ROI. The BOLD effect (compared to implicit baseline) is shown for each music condition. The p-value of a dependent t-test comparing two music conditions can be seen above the respective bars. The significance level of the music main effect is denoted above the line. D) Left hemisphere functional ROI. fROIs were individually defined in the left structural ROI. The inter-subject overlap in fROI locations is shown in the top panel. See methods for details. The BOLD effect is shown for the three different music conditions separately. Error = SEM. All p-values in structural ROI analyses are Bonferroni adjusted.