Table 1.
Summary of participants’ musical training backgrounds.
Fig 1.
Distribution of semitone errors for each participant (S1-S6) and each testing session (Pre, Post, and Follow-Up) for the UCSF Piano Test (A), UCSF Sine Test (B), UCSD Test (C), and mean absolute deviation (MAD) across all participants for these tests (D). The dashed line in Panel D represents a MAD value that would be expected by random guessing (i.e., uniform distribution of errors ranging from 0 to 6 semitones removed from the correct note).
Table 2.
Performance measures across the Chicago Piano (C-P) and Multiple Timbre (C-MT) Blocks, the UCSF Piano (UCSF-P) and Sine (UCSF-S) Tests, as well as the UCSD Test.
Fig 2.
Distribution of semitone errors for each participant (S1-S6) and each testing session (Pre, Post, and Follow-Up) for the Chicago Test split by the Piano Block (A) and Multiple Timbre Block (B). Mean absolute deviation across all participants for this test is represented in panel C, whereas mean response time across all participants for this test is represented in panel D. The dashed line in Panel C represents a MAD value that would be expected by random guessing (i.e., uniform distribution of errors ranging from 0 to 6 semitones removed from the correct note).
Fig 3.
Averaged results from the weekly assessments during training.
The weekly test (WT) assessment is separated by mean absolute deviation (A) and response time (B). For Panels A and B, the shaded area to the left (i.e., Week 0) represents pretest performance, whereas the shaded area to the right (i.e., Week 8) represents immediate posttest performance. The name that key test (NTKT), which is represented in Panel C, is represented in terms of mean absolute deviation. Ribbons around the mean lines represent ± 1 standard error of the mean. Individual participant trajectories are represented by the dashed grey lines.
Table 3.
Analyses of improvements from pretest to posttest (top) and follow-up (bottom) for each participant for the Chicago Piano (C-P) and Multiple Timbre (C-MT) Blocks, as well as the UCSF Piano (UCSF-P) and Sine (UCSF-S) Tests.
Fig 4.
Comparison of the Chicago Test (combined performance on the Piano and Multiple Timbres Blocks) of the present experiment with Bermudez and Zatorre (2009), separated by testing session (Pre, Post, and Follow-Up).
The left column (“All”) represents performance across all trials, while the right column (“>8va”) represents performance on trials in which there was more than an octave separating the heard note from the previous note. An index consisting of mean absolute deviation and log response time is represented on the y-axis, while percent correct is represented on the x-axis. Prior to training, no participant was classified in the highest (“genuine”) AP group for either analysis. In the immediate test post-training, as well as in the follow-up test four months after training, participants S2 and S5 performed indistinguishably from the highest AP performers from Bermudez and Zatorre (2009), bounded by the dashed box (lower right hand corner) of each panel.