Experiment 3

Experiments 1 and 2 showed an influence of listeners’ tonal expectations on pitch judgments with the use of either a graduated scale or speeded response times. However, a potential problem with these two tasks is that listeners need to use internal reference pitches indicating the correct pitch of an ‘in-tune’ tone to judge the target. This need of internal reference pitches might have enhanced possible response biases (see the discussions of Experiments 1 and 2). Even if area scores in Experiment 1 revealed a bias-free influence of tonal relatedness on pitch judgments, and response times in Experiment 2 cannot be solely explained by response bias, we further investigated the influence of tonal relatedness on pitch processing in Experiment 3 by using a same/different task. This task does not require an internal reference pitch and should also avoid the implication of tonal relatedness in the response. For this discrimination task, the last tone of the melodies (i.e., functioning as tonic or subdominant) was repeated (Figure 1B). The last tone was repeated either identically or with a slight mistuning, and participants were asked to judge whether the last two tones were the same or different (see Johnston & Jones, 2006 for the use of a similar paradigm in non-tonal contexts). Consequently, no internal reference pitch was needed to compare the two tones. In order to calculate areas under the ROC as an unbiased estimate of listeners’ pitch discrimination, Experiment 3 was designed as a rating scale signal-detection theory experiment (see Dowling, Tillmann & Ayers, 2001) and aimed to assess the influence of tonal relatedness on the detection of fine mistunings.