Method

Participants. Forty-two students from the University of Lyon participated in Experiment 3. Musical practice, as measured by years of instrumental instruction, varied from 0 to 7 years, with a mean of 1.7 (SD = 2.3) and a median of 0.

Material. The melody pairs of Experiment 1 were used with their last tone being repeated identically or lowered in pitch by 17 cents or 9 cents (a 1% or .5% deviation in frequency , respectively). Eighteen shorter melodies, which also ended on tonic or subdominant tones, were composed for participants to familiarize themselves with the task. These shorter melodies were one and a half 4-beat bars long, with the last 2 beats being the repeated target tones (identical repetition or mistuned repetition). Four example melodies were adapted from the 8 additional melodies of Experiment 2.

Procedure. The experiment contained a training phase and an experimental phase. In the training phase, participants were first familiarized with the two pitch deviations by listening to tone pairs, with the tone being repeated either identically or with one of the two pitch deviations. Participants were then trained for the task with the 18 short melodies: participants judged whether the last two tones were identical or not by using a four-point scale (1: sure different / 2: not sure different / 3: not sure same / 4: sure same). No time limit was imposed for responses. Finally, participants were trained with the 4 example melodies. In the experimental phase, participants judged 216 melodies (the 12 pairs of melodies with the target tone either in-tune or lowered in pitch by 9 or 17 cents, each presented 3 times in 3 successive blocks with different pseudorandom orders; 24*3*3=216). Participants received feedback on errors. Further details of the procedure were as described in Experiment 2.

Data Analysis. Discrimination performance between the in-tune condition and each of the two mistuned conditions was analyzed with areas under the ROC. Additionally, hits and false alarms (percentages of “in-tune” responses) were computed separately for each condition and participant for in-tune and out-of-tune stimuli, respectively.