Effects of Fundamental Frequency Removal and Low-Pass Filtering on Pitch Comparisons

Previous research has established that pitch and timbre perceptually interact within a variety of tasks, but has not fully established the causes of this interaction. Using synthesized missing fundamental stimuli, the present study investigated the possibility that 2 acoustic characteristics commonl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Psychomusicology 2014-09, Vol.24 (3), p.195-206
Hauptverfasser: Becker, Christopher J, Hall, Michael D
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Previous research has established that pitch and timbre perceptually interact within a variety of tasks, but has not fully established the causes of this interaction. Using synthesized missing fundamental stimuli, the present study investigated the possibility that 2 acoustic characteristics commonly associated with timbre, spectral centroid and spectral envelope shape, affect pitch judgments. Listeners were asked to determine the pitch relationship (ascending/descending) between 2 sequential tones: a standard tone (A4) and a comparison tone (either A4, A#4, or F5). The comparison tone either included, or lacked, fundamental frequency (F0). For A4 and A#4, alternative versions of the Missing F0 comparison tones also were created that were low-pass filtered such that the spectral centroid matched the standard. When the fundamental of the comparison tone was higher than that of the standard A4 tone, removing its F0 (and thus, increasing centroid) led to increased reports of ascending pitch change. When shifts in centroid were eliminated through filtering (A4 and A#4 comparison tones), the impact of removing F0 was not observed, suggesting that spectral centroid influenced pitch comparison to a greater degree than envelope shape. In contrast, when the chroma for the standard and comparison tones were identical (i.e., both were A4), differences in spectral envelope shape led to increased reports of different pitches, while shifts in spectral centroid had no apparent effect. Furthermore, listeners with more years of musical training tended to be less affected by spectral manipulations in making pitch judgments. Thus, it appears that different aspects of timbre affect pitch judgments in different ways depending on whether or not the tones to be compared are based on the same fundamental frequency.
ISSN:0275-3987
2162-1535
DOI:10.1037/a0037933