Virginia M. Richards

Monaural envelope correlation perception

The ability to discriminate between simultaneously presented 100-Hz-wide bands of noise with envelopes that were either similar or dissimilar was measured. The center frequencies of the noise bands, fL and fL + delta f Hz, were systematically varied. When the bands of noise were separated by an octave, delta f = fL, discriminations were at chance levels. For frequency separations less than an octave, delta f less than fL, discrimination was best for fL = 2500 and 4000 Hz, somewhat poorer for fL = 1000 Hz, and impossible for fL = 350 Hz. Listeners were also asked to discriminate between bands of noise with envelopes that were either perfectly or partially correlated, and bands with envelopes that were either uncorrelated or partially correlated. The data suggest that, when transformed to an equal-variance scale (Fishers z), equal changes in Fishers z lead to equal changes in detectability, regardless of the correlation of the envelopes of the reference signal.

Multiple bursts, multiple looks, and stream coherence in the release from informational masking

In the simultaneous multitone masking paradigm introduced by Neff and Green [Percept. Psychophys. 41, 409-415 (1987)] the masker typically is a small number of tones having frequencies and levels that are randomly drawn on every presentation. Large amounts of masking for a pure-tone signal often occur that are thought to reflect central, rather than peripheral, limitations on processing. Previous work from this laboratory has indicated that playing a rapid succession of randomly drawn multitone maskers in each observation interval dramatically reduces the amount of masking that is observed relative to a single burst (SB). In this multiple-bursts-different (MBD) procedure, the signal tone is the only constant frequency component during the sequence of bursts and tends to perceptually segregate from the masker. In this study, the number of masker bursts and the interburst interval (IBI) were varied. The goals were to determine how the release from masking relative to the SB condition depends on the number of bursts and to examine whether increasing the IBI would cause each burst to be processed independently. If the latter were true, it might disrupt the perception of signal stream coherence, thereby diminishing the MBD advantage. However, multiple independent looks could also lead to an improvement in performance. For those subjects showing large amounts of informational masking in the SB condition, substantial reduction in masked thresholds occurred as the number of masker bursts increased, while masking increased as IBI lengthened. The results were not consistent with a simple version of a multiple-look model in which the information from each burst was combined optimally, but instead appear to be attributable to mechanisms involved in the perceptual organization of sounds.

Cuing effects for informational masking

The detection of a tone added to a random-frequency, multitone masker can be very poor even when the maskers have little energy in the frequency region of the signal. This paper examines the effects of adding a pretrial cue to reduce uncertainty for the masker or the signal. The first two experiments examined the effect of cuing a fixed-frequency signal as the number of masker components and presentation methods were manipulated. Cue effectiveness varied across observers, but could reduce thresholds by as much as 20 dB. Procedural comparisons indicated observers benefited more from having two masker samples to compare, with or without a signal cue, than having a single interval with one masker sample and a signal cue. The third experiment used random-frequency signals and compared no-cue, signal-cue, and masker-cue conditions, and also systematically varied the time interval between cue offset and trial onset. Thresholds with a cued random-frequency signal remained higher than for a cued fixed-frequency signal. For time intervals between the cue and trial of 50 ms or longer, thresholds were approximately the same with a signal or a masker cue and lower than when there was no cue. Without a cue or with a masker cue, analyses of possible decision strategies suggested observers attended to the potential signal frequencies, particularly the highest signal frequency. With a signal cue, observers appeared to attend to the frequency of the subsequent signal.

Stimulus step size and heterogeneous stimulus conditions in adaptive psychophysics

This paper reports on simulations of repeated estimates of stimulus thresholds using a two‐down, one‐up adaptive procedure and forced‐choice procedures using 2, 3, and 4 response alternatives. In the first section of the paper, the selection of an initial and final step size for the stimulus is the main focus. For a 2‐alternative forced‐choice psychometric function, a step size of roughly one‐quarter the useful range (60%–90%) provides a reasonable initial step size. A final step size of roughly half that value is recommended. The increase in the variability of threshold estimates is, however, not very great if values near those recommended are selected instead. The change in the variability of the threshold estimates with the change in the number of alternatives used in the forced‐choice procedure can be predicted by considering the sweat factor associated with the psychometric function. In the second part of the paper, the focus is on how the variability in threshold estimates increases if the stimulus ...

Distributional theories of impression formation

This paper examines theories of impression formation which represent stimulus information as distributions on subjective dimensions rather than points along a continuum. These theories explain why unfavorable information often overrides favorable information. Subjects were asked to imagine hypothetical persons described by single adjectives or adjective combinations and to estimate the probability that each person would have various degrees of likeableness. Three models were considered to describe how the likeableness distribution for the adjective combination depends on the distributions of the single adjectives. All three models assume that the mean of the distribution for the adjective combination is described by the equal probability criterion which, for symmetric distributions, implies a weighted average of the single adjective means with weights inversely proportional to their standard deviations. The models can be distinguished on the basis of the standard deviations of the adjective combinations. Estimated standard deviations of the single unfavorable adjectives were smaller than those for the single favorable adjectives. Furthermore, the standard deviation associated with an adjective combination tended to fall between the standard deviations of the single adjectives, consistent with the horizontal averaging model.

Auditory profile analysis: potential pitch cues.

The ability to detect changes in spectral shape, or profile analysis, was measured for both complex and simple changes in the power spectrum of a complex equal-amplitude standard. In an effort to determine whether detectability was mediated by changes in pitch that are concomitant with changes in spectral shape, the pitch of the stimuli were altered on a trial by trial basis. For moderate-range pitch randomizations, thresholds were on average 3 dB poorer than when no pitch randomization was employed. For large-range pitch randomizations, threshold changes were larger, but performance levels remained above chance levels. The psychophysical data, coupled with computer simulations of Feths envelope-weighted average instantaneous frequency pitch model, indicate that changes in pitch contribute little to the discriminability of complex spectra.

Components of monaural envelope correlation perception

The ability of listeners to discriminate between simultaneously presented bands of noise whose envelopes were either the same or statistically independent was determined. Bands of 100-Hz wide noise were employed which had low and high center frequencies of (2500, 2750), (2500, 3000), (2500, 3500) and (4000, 4400) Hz. Average discriminations were above 90% correct except for the (2500, 3500) Hz condition, which yielded an average of 77% correct. Next, a factorial stimulus design was employed in order to determine the relative importance of envelope and power spectrum cues. The results indicate that in the absence of power spectrum cues, bands with the same envelopes could be discriminated from bands with statistically independent envelopes. When the envelopes were always the same, listeners were able to discriminate between power spectra that were either the same or different. In contrast, when the envelopes were always different, listeners were unable to discriminate between the same and different power spectra.

Effects of modulator phase for comodulation masking release and modulation detection interference

In an effort to evaluate the importance of across-frequency comparisons of envelope patterns in comodulation masking release (CMR) experiments and to compare joint effects of target-masker frequency separation for both CMR and modulation detection interference (MDI) tasks, thresholds were measured for three tasks. These tasks were: (a) the detection of sinusoidal amplitude modulation (SAM) of a tone, (b) the detection of a reduction in the modulation depth of a fully modulated SAM tone, and (c) the detection of a tone added to a narrow band of noise. Thresholds were obtained for the target alone and for the target presented with two maskers. For the detection of SAM, thresholds did not depend on whether the modulation patterns of the target and masker elements were the same or random. For the latter two tasks, modulator phase effects were apparent for target-masker frequency separations less than 1-2 oct. In contrast, past work has shown that observers can compare modulator envelope phases across frequency separations larger than 1-2 oct [Strickland et al., J. Acoust. Soc. Am. 86, 2160-2166 (1989); Yost and Sheft, J. Acoust. Soc. Am. 85, 848-857 (1989)]. In a second experiment, thresholds for the detection of SAM were obtained after prolonged exposure to a fully modulated SAM tone. For four of the five observers, modulation-rate specific adaptation was obtained for test/adapting carrier-frequency separations approaching 2 oct below and 1 oct above the adaptor.

Sensitivity to changes in overall level and spectral shape: An evaluation of a channel model

Two experiments involving level and spectral shape discrimination which test an optimal channel model developed by Durlach et al. [J. Acoust. Soc Am. 80, 63-72 (1986)] are described. The model specifies how the auditory system compares and/or combines intensity information in different frequency channels. In the first experiment, psychometric functions were obtained for the discrimination of changes in level and discrimination of changes in spectral shape for an eight-tone complex sound. A variety of different base spectral shapes were tested. In some conditions, level randomization was introduced to reduce the reliability of across-interval changes in level. Increasing the amount of level variation degraded performance for the level discrimination task but had no effect on the shape discrimination task. In all conditions, sensitivity to changes in spectral shape was superior to sensitivity to changes in level. Consequently, two models of central noise are evaluated in an attempt to explain these results; one in which central noise acts prior to the formation of the likelihood ratio and one in which central noise degrades the likelihood ratio. The former model is more successful in accounting for the data. In a second experiment, the detectability of a level increment to one component of a multitone complex was measured. The frequency content of the complex was varied by systematically removing six components from a 23-component complex. Thresholds were measured for increments at three different signal frequencies. A common trend in the data was that when there was a spectral gap directly above the signal frequency, thresholds were lowest. This result differs from the predictions of a simple channel model, and contrasts with results presented by Green and Berg [Q. J. Exp. Psychol. 43A, 449-458 (1991)].

The detection of a tone added to a narrow band of noise: The energy model revisited

In an effort to determine whether cues related to changes in energy contribute to the detection of a tone added to a narrow band of noise, we examined the effect of level variation on detection thresholds. In the first experiment, the level of each waveform was randomly varied on each presentation. Level variation had only marginal effects on performance. In addition, detection thresholds were obtained using bands of noise with equal energy across intervals. Neither increasing nor decreasing the variance of the noise-alone and tone-plus-noise energy difference distributions altered the detectability of a tone added to noise. Thus, the changes in energy that are concomitant with the addition of the tone are not the sole cue for the detection of the tone. In a second experiment, three psychometric functions were measured. One function was determined using no level variation, one was measured in the presence of level variation, and one was measured in the context of level variation, but for trials without level variation. The context of level variation slightly reduced detectability. In a third experiment, we compared detectability in three conditions: no level variation, across-trial level variation, and across-interval level variation. The thresholds obtained in the absence of level variation were superior to those measured in the presence of level variation, regardless of the manner in which the level variation was incorporated.