Results

For each melody, the tonal contextuality was computed over the entire sequence, and the mean contextuality value, averaged over the time windows corresponding to the target tone, was then extracted and used as the dependent variable.

Piano tones. Simulation data (i.e., tonal contextuality values) for piano tones were analysed with a two-sided paired Student t-test with melodies as random variable. Higher tonal contextuality values were observed for related targets than for less-related targets (t 2 (11) = 4.78, p < .001).

Pure tones. Simulation data (i.e., tonal contextuality values) for pure tones were analysed with a two-sided paired Student t-test with melodies as random variable. No significant difference was observed (t 2 (11) = 1.32, p = .22).

Combined Analysis. Simulation data for piano tones and pure tones were analyzed in a 2x2 ANOVA with Tonal Relatedness (Related / Less-related) and Sound Type (Piano tones / Pure tones) as repeated measure and melodies as random variable. This analysis confirmed an interaction between Tonal Relatedness and Sound Type (F 2 (1,11) = 15.9; MSE < .00001: p < .01): Contextuality values were higher for related targets than for less-related targets when played with piano tones, but not when played with pure tones (Figure 4). In addition, main effects of Tonal Relatedness and Sound Type were significant (F 2 (1,11) = 7.47; MSE < .00001; p < .05 and F 2 (1,11) = 29.6; MSE < .001; p < .001, respectively).

Results for the different echo values. Varying the echo parameters confirmed the interactive pattern between tonal relatedness and sound type. For piano tones, the two-sided paired Student t-tests with melodies as random variable showed higher tonal contextuality values for related targets than for less-related targets for all pairs of echo parameters (ps < .014). For pure tones, the two-sided paired Student t-tests never showed a significant effect of tonal contextuality (ps > .31).

Figure 4. Tonal contextuality values of the target presented as a function of tonal relatedness when melodies are played by piano tones (A) or pure tones (B). Error bars represent between-melodies standard errors.
Figure 4. Tonal contextuality values of the target presented as a function of tonal relatedness when melodies are played by piano tones (A) or pure tones (B). Error bars represent between-melodies standard errors.

A.

B.