Recently, Bowling, Purves, and Gill (2018a), found that individuals perceive chords with spectra resembling a harmonic series as more consonant. This is consistent with their vocal similarity hypothesis (VSH), the notion that the experience of consonance is based on an evolved preference for sounds that resemble human vocalizations. To rule out confounding between harmonicity and familiarity, we extended Bowling et al.’s (2018a) procedure to chords from the unconventional Bohlen-Pierce chromatic just (BPCJ) scale. We also assessed whether the association between harmonicity and consonance was moderated by timbre by presenting chords generated from either piano or clarinet samples. Results failed to straightforwardly replicate this association; however, evidence of a positive correlation between harmonicity and consonance did emerge across timbres following post hoc exclusion of chords containing intervals that were particularly similar to conventional equal-tempered dyads. Supplementary regression analyses using a more comprehensive measure of harmonicity confirmed its positive association with consonance ratings of BPCJ chords, yet also showed that spectral interference independently contributed to these ratings. In sum, our results are consistent with the VSH; however, they also suggest that a composite model, incorporating both harmonicity as well as spectral interference as predictors, would best account for variance in consonance judgments.

中文翻译：

---

非常规调音系统中的谐音偏好

最近，Bowling、Purves 和 Gill (2018a) 发现个人认为频谱类似于谐波系列的和弦更辅音。这与他们的声音相似性假设 (VSH) 一致，即和谐的体验是基于对类似于人类发声的声音的进化偏好。为了排除和声和熟悉度之间的混淆，我们将 Bowling 等人 (2018a) 的程序扩展到非常规 Bohlen-Pierce 半音阶 (BPCJ) 音阶的和弦。我们还通过呈现钢琴或单簧管样本产生的和弦来评估谐波和谐音之间的关联是否受到音色的调节。结果未能直接复制这种关联；然而，在事后排除了包含与传统等律二重音特别相似的音程的和弦后，在音色中确实出现了谐波和和音之间正相关的证据。使用更全面的谐波测量的补充回归分析证实了它与 BPCJ 和弦的协和评级正相关，但也表明频谱干扰独立地促成了这些评级。总之，我们的结果与VSH一致；然而，他们还建议，将谐波和光谱干扰作为预测因子的复合模型最能解释协和判断的差异。使用更全面的谐波测量的补充回归分析证实了它与 BPCJ 和弦的协和评级正相关，但也表明频谱干扰独立地促成了这些评级。总之，我们的结果与VSH一致；然而，他们还建议，将谐波和光谱干扰作为预测因子的复合模型最能解释协和判断的差异。使用更全面的谐波测量的补充回归分析证实了它与 BPCJ 和弦的协和评级正相关，但也表明频谱干扰独立地促成了这些评级。总之，我们的结果与VSH一致；然而，他们还建议，将谐波和光谱干扰作为预测因子的复合模型最能解释协和判断的差异。