Results

For correct responses (Table 1), only a main effect of Target Timbre was observed (F(1,22) = 5.92; MSE = 156; p < .05), with more correct responses for Timbre B. For correct response times (Figure 2a), the interaction between Tonal Relatedness and Target Timbre was significant (F(1,22) = 9.32; MSE = 2604; p < .01): Response times were faster for related than for less-related targets played by timbre A (even if not significantly, p = .13), and the reversed pattern was observed for Timbre B (F(1,22) = 5.25; p < .05). The main effect of Target timbre was significant (F(1,22) = 12.89; MSE = 12437; p < .01) with faster response times for Timbre B than Timbre A. No other effects were significant. An ANOVA including Musical Expertise found that Musical Expertise tended to interact with Tonal Relatedness (F(1,21) = 3.09; p = .09): The faster response times for related targets were rather observed for less-experienced participants, and most particularly for Timbre A (F(1,21) = 8.40; p < .01), even if the three-way interaction between Musical Expertise, Target Timbre and Tonal Relatedness was not significant (p = .11).

Table 1. Percentages of correct responses in Experiments 1 and 2. The Related and Less Related conditions refer to mediant and leading tone. The Timbre A/Timbre B conditions refer to piano/harp (with a piano context) in Experiment 1a and oboe/clarinet (with a flute context) in Experiment 1b. In Experiment 2, the Out-of-tune condition refers to target tones whose pitch was augmented by 35 cents. Between-participants standard errors are shown in brackets.