Table 3. Behavioral Discrimination Accuracy in Active Tasks 1, 2, and 3 (The Follow-Up)

3(3)

Table 3. Behavioral discrimination accuracy in Active tasks 1, 2, and 3 (the follow-up). The percentage value (%) tells the response rate, that is, how many participants could proceed from easy level to the medium level, and from the medium level to the difficult level.

Musicians (n=20) / Nonmusicians (n=21)
Deviant type / Difficulty level reached / Active task 1 / Active task 2 / Active task 3 / Active task 1 / Active task 2 / Active task 3
Pitch / Easy-to-Medium / All: 100%
FB: 100%
No-FB: 100% / All: 100%
FB: 100%
No-FB: 100% / All: 100% b
FB: 100%
No-FB: 100% / All: 90% c
FB: 100%
No-FB: 80% / All: 100% b
FB: 100%
No-FB: 100% / All: 94% d
FB: 89%
No-FB: 100%
Medium-to Difficult / All: 100%
FB: 100%
No-FB: 100% / All: 100%
FB: 100%
No-FB: 100% / All: 100% b
FB: 100%
No-FB: 100% / All: 65% c
FB: 80%
No-FB: 50% / All: 84% b
FB: 90%
No-FB: 78% / All: 72% d
FB: 78%
No-FB: 67%
Duration / Easy-to-Medium / All: 100%a
FB: 100%
No-FB: 100% / All: 100% a
FB: 100%
No-FB: 100% / All: 100% b
FB: 100%
No-FB: 100% / All: 95% c
FB: 100%
No-FB: 90% / All: 100% b
FB: 100%
No-FB: 100% / All: 94% d
FB: 89%
No-FB: 100%
Medium-to Difficult / All: 90%
FB: 90%
No-FB: 90% / All: 95% a
FB: 100%
No-FB: 89% / All: 100% b
FB: 100%
No-FB: 100% / All: 60% c
FB: 80%
No-FB: 40% / All: 74% b
FB: 90%
No-FB: 56% / All: 72% d
FB: 78%
No-FB: 67%
Table 3 (continued)
Pitch + Duration / Easy-to-Medium / All: 100% a
FB: 100%
No-FB: 100% / All: 100% a
FB: 100%
No-FB: 100% / 100% b
FB: 100%
No-FB: 100% / All: 95% c
FB: 100%
No-FB: 85% / All: 97% b
FB: 100%
No-FB: 100% / All: 94% d
FB: 89%
No-FB: 100%
Medium-to Difficult / All: 92%
FB: 95%
No-FB: 95% / All: 100% a
FB: 100%
No-FB: 94% / All: 100% b
FB: 100%
No-FB: 100% / All: 62% c
FB: 80%
No-FB: 45% / All: 79% b
FB: 90%
No-FB: 67% / All: 72% d
FB: 78%
No-FB: 67%

Note. a n = 19, b n = 16, c n = 20, and d n = 18. In order to investigate the discrimination accuracy for deviant sounds between active discrimination tasks, button presses in active tasks were analyzed from a 200-800 msec time window after the onset of the sound. Percentages of hit rates and false alarms (missed answers) were used to compute a modified Grier’s sensitivity A' (Grier, J., 1971. Nonparametric indices for sensitivity and bias: Computing formulas. Psychol. Bull. 75, 424-429; Stanislaw, H., Todorov, N., 1999. Calculation of signal detection theory measures. Behav. Res. Methods, Instr., Comp. 31, 137-149, equation 3). The improvement in behavioral discrimination accuracy was assessed in the Active tasks 1 and 2, presented between the passive blocks in the first experimental day, and in the Active task 3 which was administered after ~1 week as a follow-up (seven participants were missing from the follow-up). Since the active task became more difficult on the individual basis, the number of responses to deviants, as well as the access to the more difficult levels, varied between participants. To avoid comparing serially dependent probabilities in hit rate, we computed a categorical indicator (yes or no) of whether the participant reached the medium or difficult level for hit

Table 3 (continued)

rates. Due to small dispersion, a χ2-square test was used to compare the discrimination accuracy between Active tasks 1, 2, and 3 (reached the medium, reached the hard difficulty level) between musicians and nonmusicians. The data for Active Task 3 (follow-up) consisted of only 34 participants due to missing data which was caused by the technical failure of the response box. Although there were too few participants for ERP analysis of feedback and no-feedback groupings, the behavioral results for these groupings are presented here (n varying between 7 and 10). Groups are coded as following: All = both feedback and no-feedback groups together, FB = Feedback, No-FB = No-Feedback. Combined values from pitch and duration deviant types (shown as bold text) were used in statistical analyses.