Table 1.
Demographic details of the participants.
Fig 1.
Results from the accent-rating task.
Within the response options (i.e., "rating" plotted on the y-axis), 1 indicates “very good pronunciation” and 6 “poor pronunciation”. Here, means are shown only for a range from 1 to 4. The left panel shows the result for the rating of others' voices only, depending on their quartile, showing that higher quartiles got better ratings. The right panel shows the effect of voice source, with better ratings for own-voice trials. Error bars show standard errors estimated with the R package emmeans [37].