Action–perception mismatch in tone-deafness

The source of conscious experience has fueled scientific and philosophical debates for centuries. In the auditory and motor domains, it is not yet known how consciously and unconsciously obtained information combine to enable the production and perception of speaking and singing. Both forms of vocalization rely upon the interaction of brain networks responsible for perception and action. While perceptual experience and executed actions are usually well coupled, dissociations between perception and action can be informative. Here we report such a dissociation: tone-deaf individuals, who cannot consciously perceive pitch differences, can paradoxically reproduce pitch intervals in correct directions. Our results suggest that multiple neural pathways have evolved for sound perception and production, so that pitch information sufficient for intact speech can be obtained separately from pathways necessary for conscious perception. Pitch perception is a central function of the human and animal auditory system [1,2]. Humans are generally able not only to consciously perceive pitch differences, but also to produce pitch intervals accurately via the interaction of perceptual and motor neural systems [3], thus enabling the communication of musical and linguistic information. However, this ability is selectively impaired in one special population: people affected by tone-deafness, also known as congenital amusia, report musical difficulties or have been told that they sing out of tune, but have normal audiometry and no obvious language problems [4,5]. Psychophysically, one hallmark of tone-deaf individuals is their inability to consciously discriminate pitches less than one semitone apart [5,6]. This dichotomy offers a unique model to test the relationship between conscious perception and unconscious actions, and between brain mechanisms responsible for action and perception. We tested the hypothesis that tone-deaf individuals might have dissociated abilities in pitch perception and production. Using psychophysics combined with sound analysis of singing in tone-deaf and control samples, we provide evidence for intact but imprecise abilities in pitch production despite impaired perception. Tone-deaf listeners were identified based on their performance on the Montreal Battery of Evaluation of Amusia (MBEA) (see Supplemental Data). In a first experiment assessing perception and production, pairs of pure tones, forming different small intervals, were presented to tonedeaf and normal control individuals. Listeners reproduced the tone intervals by humming (production task), and then indicated verbally whether the second tone was higher or lower than the first (perception task). Tone-deaf listeners performed at chance for the perception task, and were significantly worse than controls. In the production task, however, tone-deaf individuals were above chance at producing the correct pitch direction, with performance being indistinguishable from controls (Figure 1). This dichotomy between perception and production points to a dissociation between action and perception pathways in the auditory system [7], possibly analogous to action-blindsight in the visual system [8]. Pitch extraction of sung intervals showed that, while directions of interval production were intact, variability in produced pitches compared to target pitches was significantly higher in tone-deaf listeners than in controls (Figure 2A,B). Thus, although tone-deaf listeners could produce pitch intervals in target directions, the pitches they produce are imprecise and highly variable. This leads to the common observation that tone-deaf individuals are unable to sing in tune. To further characterize the dichotomy between production and perception, we conducted adaptive staircase procedures on a subset of subjects (tone-deaf group: N = 3; control group: N = 3) to assess the psychophysical thresholds of production and perception (see Supplemental Data for details). For tone-deaf individuals, production thresholds (mean = 12.3 Hz, SD = 2.5 Hz) were much smaller than perception thresholds (mean = 36.2 Hz, SD = 2.0 Hz). In contrast, control subjects showed nearly identical thresholds for production and perception, with a slightly larger production threshold (mean = 2.5 Hz, SD = 1.0 Hz) than perception threshold (mean = 2.0 Hz, SD = 0.8 Hz). Paired t-tests comparing thresholds for perception and production were conducted for each individual subject using confidence intervals obtained from reversal points in the adaptive staircase procedure (in Hz). The results show significantly different perception and production thresholds for every tone-deaf subject (subject 1: t(5) = 5.9, p = 0.002; subject 2: t(5) = 2.6, p = 0.047; s3: t(5) = 3.6, p = 0.02) but no such dissociation between perception and production thresholds for every control subject (subject 1: t(5) = 1.3, p = 0.3; subject 2: t(5) = 2.0, p = 0.1; s3: t(5) = 1.0, p = 0.4; see Figure S1 in the Supplemental Data). These individual statistics confirm that perception and production are mismatched in tonedeaf individuals but not in controls. Perception Production P er ce nt c or re ct ( % )