The Effects of Singer Head Position on Listener Preferences and Perceptions of Vocal Timbre

Amelia A. Rollings

Some vocal pedagogy textbooks encourage singers to keep the head level with the ground (e.g., McKinney, 1994; Miller, 2004). However, other vocal pedagogy texts and articles recommend singers employ a slightly lowered head position (Davids & LaTour, 2012) or a slightly elevated head position (e.g., Austin, 2013). The purpose of this study was to determine the effect, if any, of 3 extreme alterations in the head position (lowered, neutral, elevated) of a classical female singer on long-term average spectra (LTAS) data and listener (N = 30) preferences and perceptions of vocal timbre. A singer participant performed a portion of an aria 2 times in each of the 3 different focal point conditions that required her to adjust her head position. Listener participants compared 14 pairs of recordings and completed a questionnaire that asked them to select the amount of difference they perceived in vocal timbre between the two recordings in each set and to select the recording they most preferred.Primary results indicated that (a) when the singer performed in an elevated head position, LTAS relative mean signal amplitude increased, and when the singer performed in a lowered head position, LTAS relative mean signal amplitude decreased with individual harmonic amplitude differences ranging from 0.51-4.18 dB; (b) listeners most often heard “a little difference” when comparing recordings of the singer performing her aria in different head positions, but some listeners erroneously heard a difference in two sets that compared the same recording; (c) listeners most preferred the vocal timbre of the singer in the first performance of the neutral head position condition (neutral 1) followed by neutral 2 and lowered 2, elevated 2, lowered 1, and elevated 1; and (d) listener preferences appeared to follow LTAS relative mean signal amplitude as listeners preferred the recording with the midmost relative mean signal amplitude compared to those recordings with the highest and lowest relative mean signal amplitudes.