Adaptation in Mandarin tone production with pitch-shifted auditory feedback: influence of tonal contrast requirements

ABSTRACT

We investigated Mandarin speakers’ control of lexical tone production with F0-perturbed auditory feedback. Subjects produced high level (T1), mid rising (T2), low dipping (T3), and high falling (T4) tones in conditions with (a) no perturbation, (b) T1 shifted down, (c) T1 shifted down and T3 shifted up, or (d) T1 shifted down and T3 shifted up but without producing other tones. Speakers and new subjects also completed a tone identification task with unaltered and F0-perturbed productions. With only T1 perturbed down, speakers adapted by raising F0 relative to no-perturbation. With simultaneous T1 down and T3 up perturbations, no T1 adaptation occurred, and T3 adaptation occurred only if T2 was also produced. Identification accuracy with stimuli representing adapted productions was comparable to baseline, but with simulated non-adapted productions it was reduced for T2 and T3. Thus, Mandarin speakers’ adaptation to F0 perturbations is linguistically constrained and serves to maintain tone contrast.

KEYWORDS:

• Auditory feedback
• tone production
• sensorimotor adaptation
• tone identification

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Ludo Max http://orcid.org/0000-0002-3967-5545

Notes

1 There are four lexical tones in Mandarin: high level, mid rising, low dipping, and high falling tones, conventionally referred to as Tones 1, 2, 3, & 4, respectively. Tone 3 has the most intrinsic variation in that the F0 falls and then rises in citation form (i.e. produced in isolation) or in phrase-final positions, but it only falls (without the subsequent rise) in non-final positions in connected speech (Liu & Samuel, 2004; Xu, 1994). Although the low dipping tone in citation form has a dynamic contour, a large body of studies has shown that the initial fall and final rise are not important for identifying the tone (Liu & Samuel, 2004; Whalen & Xu, 1992). In addition, a low flat pitch is used as one of the variants in running speech, so that the tone is essentially a low level tone according to the definition of level tones (Maddieson, 1978). Some speakers use a creaky voice quality during the middle portion of low dipping tones and at the offset of high falling tones, & low dipping tones are sometimes even characterised by a loss of voicing at their lowest point which results in irregularity or discontinuity in the F0 contour (Belotel-Grenié & Grenié, 2004; Chao, 1956; Davison, 1991; Gårding et al., 1986; Liu & Samuel, 2004). The primary acoustic cue for these tones is F0, with the two dimensions height and contour. Secondary cues include amplitude, duration, & voice quality (Lee, 2009; Liu & Samuel, 2004; Whalen & Xu, 1992). Unlike other tonal languages such as Cantonese (Peng et al., 2012), Taiwanese (Lin & Repp, 1989) or Thai (Abramson, 1978), the dominant feature for tonal contrasts in Mandarin is the distinctive pitch contours and there are no tone pairs that differ only in pitch height. Nevertheless, a previous study revealed partial effect of register on the perception of Mandarin tone segments (Whalen & Xu, 1992), and a tone perception study suggested that both pitch height and pitch contour have a significant effect on distinguishing the high level and mid rising tones (Massaro, Cohen, & Tseng, 1985). Thus, F0 manipulation in the current study is shifting the pitch height of Mandarin tones.

2 Type 1 and 3 stimuli both simulated auditory feedback signals actually perceived in Experiment 1. For two reasons, we chose here in Experiment 2 to play back and alter the speakers’ productions rather than playing back recordings of the feedback signals themselves. First, our set of recordings of the actual auditory feedback signals from Experiment 1 was incomplete due to some technical issues with this aspect of the experiment. Second, Type 2 stimuli were not perceived as feedback signals in any conditions of Experiment 1. Thus, to ensure consistency across the three types of stimuli, all stimuli were generated by playing recordings of the original production through the same equipment circuitry.

Additional information

Funding

This work was supported by the Institute of Acoustics, Chinese Academy of Sciences [grant number Y154221431] and the National Institute on Deafness and Other Communication Disorders [grant number R01DC007603] and [grant number R01DC014510].