Table 1.
List of musicians, age at onset of musical training, and primary instrument.
Figure 1.
Mean proportion of correctly identified words as a function of signal-to-noise ratio (SNR).
Musician (M; circles) and non-musician (NM; squares) data from continuous noise (solid lines) and gated noise (dashed lines) trials are included for each of the three speech types, shown in the three panels. No significant differences between the groups were observed. Error bars indicate +/−1 standard error (s.e.) of the mean.
Figure 2.
Average results for QuickSIN and Adaptive HINT measures of speech perception in noise.
Black bars denote the average performance of musicians, and grey bars denote the average performance of the non-musician data. The QuickSIN measure (left group) indicates dB of SNR loss, relative to an ideal level of speech understanding in 4-speaker background babble. The Adaptive HINT measure (right group) indicates the threshold SNR for speech understanding in a continuous speech-shaped noise. In both cases, lower scores denote better performance. Error bars represent +/−1 s.e. of the mean.
Figure 3.
Scatter plots of individual musician speech-in-noise data as a function of years of musical training and age of onset of musical training.
Individual non-musician data are shown in the far left column of each row. Distributions of musician and non-musician scores overlap, despite the significant within-group correlations with duration of musical training (but not age of onset).
Figure 4.
Fundamental-frequency difference limens.
Musicians (open circles) demonstrate significantly better F0 discrimination limens than non-musicians (grey filled squares) for lowpass-filtered harmonic complexes with F0s centered around 110 Hz and 210 Hz. Error bars represent +/−1 s.e. of the mean.