Schematic of a hypothetical neural circuit in the ventral cochlear nucleus to account for the captor effect, as proposed by Roberts and Holmes (2006a). BF refers to the best frequency of a cell. The captor tone is assumed to reduce the excitatory response of transient chopper (CT) cells, tuned to the leading portion of the tone, by increasing the wideband inhibitory input to these cells from onset chopper (OC) cells with the same BF. Adapted with permission from “Asynchrony and the grouping of vowel components: Captor tones revisited,” by B. Roberts and S. D. Holmes, 2006, J. Acoust. Soc. Am., 119, p. 2916. Copyright 2006 by the Acoustical Society of America.
Stimulus configurations used in this study for: (a) the incremented-4th condition; (b) the leading-4th condition; (c) an example captor condition, for which the captor was a pure tone; (d) the corresponding captor-control condition. Each part displays harmonics in the F1 region of a vowel (the continuation of the spectrum is shown by a vertical dashed line). In each case, a tone has been added to the vowel to increase the level of its 4th harmonic. The vowel and the additional tone were always presented in the left ear. The captor was presented either in the left ear (ipsilateral presentation) or in the right ear (contralateral presentation). Adapted with permission from “Asynchrony and the grouping of vowel components: Captor tones revisited,” by B. Roberts and S. D. Holmes, 2006, J. Acoust. Soc. Am., 119, p. 2907. Copyright 2006 by the Acoustical Society of America.
Results for experiment 1—effect of pure-tone captors under ipsilateral and contralateral presentation. Mean phoneme boundaries are expressed as nominal first-formant frequencies for 12 listeners (with inter-subject standard errors). The boundary for the vowel-alone condition is indicated by the open bar and dotted horizontal line. The boundary for the leading-4th condition is shown by the square symbol and solid horizontal line, and the boundary for the incremented-4th condition is shown by the narrow-hashed bar. The boundaries for the captor conditions and for their controls are shown by the solid bars and wide-hashed bars, respectively. The extent to which each captor condition restores the effect of the leading-4th component on vowel quality, relative to its own control, is shown above each pair of bars. Each restoration effect is quoted as an F1 boundary shift in Hz and as a percentage of the boundary difference between the leading-4th and incremented-4th conditions. Note that the captor is at least as effective under contralateral as under ipsilateral presentation.
Captor configurations used in experiment 2. Three different multi-component captors were used, each comprising harmonics 1–2 and 4–5 of F0. For the shared-F0 captor , the leading-4th component of the vowel in the opposite ear corresponds to the missing 3rd harmonic of the captor. For the different-F0 captors , the leading-4th component is mistuned from the frequency of the missing 3rd harmonic by and , respectively. These frequency relationships are illustrated by the dotted line in the spectral center of each captor.
Results for experiment 2—the influence of harmonic relations on the effect of multi-component captors under contralateral presentation. Phoneme boundaries and restoration effects are displayed as for Fig. 3. Note that all three captors are similarly effective, irrespective of whether they are harmonically related to the leading-4th component.
Article metrics loading...
Full text loading...