Figure 2
From: A tonal-language benefit for pitch in normally-hearing and cochlear-implanted children
d′ data across children for 300-ms sweeps of ±0.5 (most-left) ±1 (middle-left), ±32 (middle-right), and ±64 (most-right) semitones/sec, in the labelling (top) or discrimination (bottom) task. Means are on the right-hand side of each panel, and error bars represent one standard error. At the shallow rates, none of the correlations reached significance among NH children, and at steep rates, age effects were more consistent for implanted children in the US than in Taiwan.
