Detection of unexpected tones with short and long durations

How to define thresholds for level and interaural-level-difference discrimination: Insights from scedasticities and distributions

Sensitivity to changes in the stimulus level at one or at both ears and to changes in the interaural level difference (ILD) between the two ears has been studied widely. Several different definitions of threshold and, for one of them, two different ways of averaging single-listener thresholds have been used (i.e., arithmetically and geometrically), but it is unclear which definition and which way of averaging is most suitable. Here, we addressed this issue by examining which of the differently defined thresholds yielded the highest degree of homoscedasticity (homogeneity of the variance). We also examined how closely the differently defined thresholds followed the normal distribution. We measured thresholds from a large number of human listeners as a function of stimulus duration in six experimental conditions, using an adaptive two-alternative forced-choice paradigm. Thresholds defined as the logarithm of the ratio of the intensities or amplitudes of the target and the reference stimulus (i.e., as the difference in their levels or ILDs; the most commonly used definition) were clearly heteroscedastic. Log-transformation of these latter thresholds, as sometimes performed, did not result in homoscedasticity. Thresholds defined as the logarithm of the Weber fraction for stimulus intensity and thresholds defined as the logarithm of the Weber fraction for stimulus amplitude (the most rarely used definition) were consistent with homoscedasticity, but the latter were closer to the ideal case. Thresholds defined as the logarithm of the Weber fraction for stimulus amplitude also followed the normal distribution most closely. The discrimination thresholds should therefore be expressed as the logarithm of the Weber fraction for stimulus amplitude and be averaged arithmetically across listeners. Other implications are discussed, and the obtained differences between the thresholds in different conditions are compared to the literature.

Robust and Efficient Online Auditory Psychophysics

Most human auditory psychophysics research has historically been conducted in carefully controlled environments with calibrated audio equipment, and over potentially hours of repetitive testing with expert listeners. Here, we operationally define such conditions as having high ‘auditory hygiene’. From this perspective, conducting auditory psychophysical paradigms online presents a serious challenge, in that results may hinge on absolute sound presentation level, reliably estimated perceptual thresholds, low and controlled background noise levels, and sustained motivation and attention. We introduce a set of procedures that address these challenges and facilitate auditory hygiene for online auditory psychophysics. First, we establish a simple means of setting sound presentation levels. Across a set of four level-setting conditions conducted in person, we demonstrate the stability and robustness of this level setting procedure in open air and controlled settings. Second, we test participants’ tone-in-noise thresholds using widely adopted online experiment platforms and demonstrate that reliable threshold estimates can be derived online in approximately one minute of testing. Third, using these level and threshold setting procedures to establish participant-specific stimulus conditions, we show that an online implementation of the classic probe-signal paradigm can be used to demonstrate frequency-selective attention on an individual-participant basis, using a third of the trials used in recent in-lab experiments. Finally, we show how threshold and attentional measures relate to well-validated assays of online participants’ in-task motivation, fatigue, and confidence. This demonstrates the promise of online auditory psychophysics for addressing new auditory perception and neuroscience questions quickly, efficiently, and with more diverse samples. Code for the tests is publicly available through Pavlovia and Gorilla.

The Effects of Uncertainty in Level on Speech-on-Speech Masking

Identification of speech from a "target" talker was measured in a speech-on-speech masking task with two simultaneous "masker" talkers. The overall level of each talker was either fixed or randomized throughout each stimulus presentation to investigate the effectiveness of level as a cue for segregating competing talkers and attending to the target. Experimental manipulations included varying the level difference between talkers and imposing three types of target level uncertainty: 1) fixed target level across trials, 2) random target level across trials, or 3) random target levels on a word-by-word basis within a trial. When the target level was predictable performance was better than corresponding conditions when the target level was uncertain. Masker confusions were consistent with a high degree of informational masking (IM). Furthermore, evidence was found for "tuning" in level and a level "release" from IM. These findings suggest that conforming to listener expectation about relative level, in addition to cues signaling talker identity, facilitates segregation of, and maintaining focus of attention on, a specific talker in multiple-talker communication situations.

Towards a unifying basis of auditory thresholds: Thresholds for multicomponent stimuli

Sounds consisting of multiple simultaneous or consecutive components can be detected by listeners when the stimulus levels of the components are lower than those needed to detect the individual components alone. The mechanisms underlying such spectral, spectrotemporal, temporal, or across-ear integration are not completely understood. Here, we report threshold measurements from human subjects for multicomponent stimuli (tone complexes, tone sequences, diotic or dichotic tones) and for their individual sinusoidal components in quiet. We examine whether the data are compatible with the detection model developed by Heil, Matysiak, and Neubauer (HMN model) to account for temporal integration (Heil et al. 2017), and we compare its performance to that of the statistical summation model (Green 1958), the model commonly used to account for spectral and spectrotemporal integration. In addition, we compare the performance of both models with respect to previously published thresholds for sequences of identical tones and for diotic tones. The HMN model is similar to the statistical summation model but is based on the assumption that the decision variable is a number of sensory events generated by the components via independent Poisson point processes. The rate of events is low without stimulation and increases with stimulation. The increase is proportional to the time-varying amplitude envelope of the bandpass-filtered component(s) raised to an exponent of 3. For an ideal observer, the decision variable is the sum of the events from all channels carrying information, for as long as they carry information. We find that the HMN model provides a better account of the thresholds for multicomponent stimuli than the statistical summation model, and it offers a unifying account of spectral, spectrotemporal, temporal, and across-ear integration at threshold.

Humans attend to signal duration but not temporal structure for sound detection: Steady-state versus pulse-train signals

Most sounds fluctuate in amplitude, but do listeners attend to the temporal structure of those fluctuations when trying to detect the mere presence of those sounds? This question was addressed by leading listeners to expect a faint sound with a fixed temporal structure (pulse train or steady-state tone) and total duration (300 ms) and measuring their ability to detect equally faint sounds of unexpected temporal structure (pulse train when expecting steady state) and/or total duration (<300 ms). Detection was poorer for sounds with unexpected than with expected total durations, replicating previous outcomes, but was uninfluenced by the temporal structure of the expected sound. The results disagree with computational predictions of the multiple-look model, which posits that listeners attend to both the total duration and temporal structure of the signal, but agree with predictions of the matched-window energy-detector model, which posits that listeners attend to the total duration but not the temporal structure of the signal. Moreover, the matched-window energy-detector model could also account for previous results, including some that were originally interpreted as supporting the multiple-look model. Taken together, at least when detecting faint sounds, listeners appear to attend to the total duration of expected sounds but to ignore their detailed temporal structure.

Comparing and modeling absolute auditory thresholds in an alternative-forced-choice and a yes-no procedure

Detecting sounds in quiet is arguably the simplest task performed by an auditory system, but the underlying mechanisms are still a matter of debate. Threshold stimulus levels depend not only on the physical properties of the sounds to be detected but also on the experimental procedure used to measure them. Here, thresholds of human subjects were measured for sounds consisting of different numbers of bursts using both an alternative-forced-choice and a yes-no procedure in the same experimental sessions. Thresholds measured with the yes-no procedure were typically higher than thresholds measured with the alternative-forced choice procedure. The difference between the two thresholds decreased as stimulus duration increased. It also varied between subjects and varied with the probability of false alarms in the yes-no procedure. It is shown that a previously proposed model of detection (Heil et al., Hear Res 2017) can account for these findings better than other models. It can also account for the shapes of the psychometric functions. The model is consistent with basic concepts of signal detection theory but is based on a decision variable that follows Poisson statistics. It also differs from other models of detection with respect to the transformation of the stimulus into the decision variable. The findings in this study further support the model.

Auditory attentional filter in the absence of masking noise

Signals containing attended frequencies are facilitated while those with unexpected frequencies are suppressed by an auditory filtering process. The neurocognitive mechanism underlying the auditory attentional filter is, however, poorly understood. The olivocochlear bundle (OCB), a brainstem neural circuit that is part of the efferent system, has been suggested to be partly responsible for the filtering via its noise-dependent antimasking effect. The current study examined the role of the OCB in attentional filtering, particularly the validity of the antimasking hypothesis, by comparing attentional filters measured in quiet and in the presence of background noise in a group of normal-hearing listeners. Filters obtained in both conditions were comparable, suggesting that the presence of background noise is not crucial for attentional filter generation. In addition, comparison of frequency-specific changes of the cue-evoked enhancement component of filters in quiet and noise also did not reveal any major contribution of background noise to the cue effect. These findings argue against the involvement of an antimasking effect in the attentional process. Instead of the antimasking effect mediated via medial olivocochlear fibers, results from current and earlier studies can be explained by frequency-specific modulation of afferent spontaneous activity by lateral olivocochlear fibers. It is proposed that the activity of these lateral fibers could be driven by top-down cortical control via a noise-independent mechanism. SIGNIFICANCE: The neural basis for auditory attentional filter remains a fundamental but poorly understood area in auditory neuroscience. The efferent olivocochlear pathway that projects from the brainstem back to the cochlea has been suggested to mediate the attentional effect via its noise-dependent antimasking effect. The current study demonstrates that the filter generation is mostly independent of the background noise, and therefore is unlikely to be mediated by the olivocochlear brainstem reflex. It is proposed that the entire cortico-olivocochlear system might instead be used to alter the hearing sensitivity during focus attention via frequency-specific modulation of afferent spontaneous activity.

Contributions of Voice Expectations to Talker Selection in Younger and Older Adults With Normal Hearing

Focused attention on expected voice features, such as fundamental frequency (F0) and spectral envelope, may facilitate segregation and selection of a target talker in competing talker backgrounds. Age-related declines in attention may limit these abilities in older adults, resulting in poorer speech understanding in complex environments. To test this hypothesis, younger and older adults with normal hearing listened to sentences with a single competing talker. For most trials, listener attention was directed to the target by a cue phrase that matched the target talker's F0 and spectral envelope. For a small percentage of randomly occurring probe trials, the target's voice unexpectedly differed from the cue phrase in terms of F0 and spectral envelope. Overall, keyword recognition for the target talker was poorer for older adults than younger adults. Keyword recognition was poorer on probe trials than standard trials for both groups, and incorrect responses on probe trials contained keywords from the single-talker masker. No interaction was observed between age-group and the decline in keyword recognition on probe trials. Thus, reduced performance by older adults overall could not be attributed to declines in attention to an expected voice. Rather, other cognitive abilities, such as speed of processing and linguistic closure, were predictive of keyword recognition for younger and older adults. Moreover, the effects of age interacted with the sex of the target talker, such that older adults had greater difficulty understanding target keywords from female talkers than male talkers.

The Effects of Duration and Level on Spectral Modulation Perception

Purpose Spectral modulation detection is an increasingly common assay of suprathreshold auditory perception and has been correlated with speech perception performance. Here, the potential effects of stimulus duration and stimulus presentation level on spectral modulation detection were investigated. Method Spectral modulation detection thresholds were measured as a function of modulation frequency in young, normal-hearing listeners. The standard stimulus was a bandpass noise, and signal stimuli were created by superimposing sinusoidal spectral modulation on the bandpass noise carrier. The modulation was sinusoidal on a log₂ frequency axis and a log₁₀ (dB) amplitude scale with a random starting phase (0-2π radians). In 1 experiment, stimulus durations were 50, 100, 200, or 400 ms (at fixed level 81 dB SPL). In a 2nd experiment, stimuli were presented at sensation levels of 10, 20, 30, 40, and 60 dB SL (fixed at a duration of 400 ms). Results Spectral modulation detection thresholds were similarly low for the 400- and 200-ms durations, increased slightly for the 100-ms duration, and increased markedly for the 50-ms duration. Thresholds were lowest for 40 dB SL; increased slightly for 20, 30, and 60 dB SL; and markedly higher for the 10-dB SL condition. Conclusions The increase in thresholds for the shortest durations and lowest sensational levels is consistent with previous investigations of auditory spectral profile analysis. The effects of presentation level and stimulus duration are important considerations in the context of understanding potential relationships between the perception of spectral cues and speech perception, when designing investigations and interpreting data related to spectral envelope perception, and in the context of models of auditory perception. As examples, 2 simple models based on auditory nerve output that have been used to explain spectrotemporal modulation in previous investigations produced an output inconsistent with the present results. Plain language summary Intensity variations across audio frequency lead to spectral shapes that are essential and sometimes signature features of various sounds in the environment, including speech. Here, we show how laboratory measures of spectral shape perception depend on presentation level and stimulus duration.

Detection of tones of unexpected frequency in amplitude-modulated noise

Detection of a tonal signal in amplitude-modulated noise can improve with increases in noise bandwidth if the pattern of amplitude fluctuations is uniform across frequency, a phenomenon termed comodulation masking release (CMR). Most explanations for CMR rely on an assumption that listeners monitor frequency channels both at and remote from the signal frequency in conditions that yield the effect. To test this assumption, detectability was assessed for signals presented at expected and unexpected frequencies in wideband amplitude-modulated noise. Detection performance was high even for signals of unexpected frequency, suggesting that listeners were monitoring multiple frequency channels, as has been assumed.

Effect of echolocation behavior-related constant frequency-frequency modulation sound on the frequency tuning of inferior collicular neurons in Hipposideros armiger

In constant frequency-frequency modulation (CF-FM) bats, the CF-FM echolocation signals include both CF and FM components, yet the role of such complex acoustic signals in frequency resolution by bats remains unknown. Using CF and CF-FM echolocation signals as acoustic stimuli, the responses of inferior collicular (IC) neurons of Hipposideros armiger were obtained by extracellular recordings. We tested the effect of preceding CF or CF-FM sounds on the shape of the frequency tuning curves (FTCs) of IC neurons. Results showed that both CF-FM and CF sounds reduced the number of FTCs with tailed lower-frequency-side of IC neurons. However, more IC neurons experienced such conversion after adding CF-FM sound compared with CF sound. We also found that the Q 20 value of the FTC of IC neurons experienced the largest increase with the addition of CF-FM sound. Moreover, only CF-FM sound could cause an increase in the slope of the neurons' FTCs, and such increase occurred mainly in the lower-frequency edge. These results suggested that CF-FM sound could increase the accuracy of frequency analysis of echo and cut-off low-frequency elements from the habitat of bats more than CF sound.

Rapid estimation of high-parameter auditory-filter shapes

A Bayesian adaptive procedure, the quick-auditory-filter (qAF) procedure, was used to estimate auditory-filter shapes that were asymmetric about their peaks. In three experiments, listeners who were naive to psychoacoustic experiments detected a fixed-level, pure-tone target presented with a spectrally notched noise masker. The qAF procedure adaptively manipulated the masker spectrum level and the position of the masker notch, which was optimized for the efficient estimation of the five parameters of an auditory-filter model. Experiment I demonstrated that the qAF procedure provided a convergent estimate of the auditory-filter shape at 2 kHz within 150 to 200 trials (approximately 15 min to complete) and, for a majority of listeners, excellent test-retest reliability. In experiment II, asymmetric auditory filters were estimated for target frequencies of 1 and 4 kHz and target levels of 30 and 50 dB sound pressure level. The estimated filter shapes were generally consistent with published norms, especially at the low target level. It is known that the auditory-filter estimates are narrower for forward masking than simultaneous masking due to peripheral suppression, a result replicated in experiment III using fewer than 200 qAF trials.

Developmental hearing loss impairs signal detection in noise: putative central mechanisms

Listeners with hearing loss have difficulty processing sounds in noisy environments. This is most noticeable for speech perception, but is reflected in a basic auditory processing task: detecting a tonal signal in a noise background, i.e., simultaneous masking. It is unresolved whether the mechanisms underlying simultaneous masking arise from the auditory periphery or from the central auditory system. Poor detection in listeners with sensorineural hearing loss (SNHL) is attributed to cochlear hair cell damage. However, hearing loss alters neural processing in the central auditory system. Additionally, both psychophysical and neurophysiological data from normally hearing and impaired listeners suggest that there are additional contributions to simultaneous masking that arise centrally. With SNHL, it is difficult to separate peripheral from central contributions to signal detection deficits. We have thus excluded peripheral contributions by using an animal model of early conductive hearing loss (CHL) that provides auditory deprivation but does not induce cochlear damage. When tested as adults, animals raised with CHL had increased thresholds for detecting tones in simultaneous noise. Furthermore, intracellular in vivo recordings in control animals revealed a cortical correlate of simultaneous masking: local cortical processing reduced tone-evoked responses in the presence of noise. This raises the possibility that altered cortical responses which occur with early CHL can influence even simple signal detection in noise.

Objective and subjective psychophysical measures of auditory stream integration and segregation

The perceptual organization of sound sequences into auditory streams involves the integration of sounds into one stream and the segregation of sounds into separate streams. "Objective" psychophysical measures of auditory streaming can be obtained using behavioral tasks where performance is facilitated by segregation and hampered by integration, or vice versa. Traditionally, these two types of tasks have been tested in separate studies involving different listeners, procedures, and stimuli. Here, we tested subjects in two complementary temporal-gap discrimination tasks involving similar stimuli and procedures. One task was designed so that performance in it would be facilitated by perceptual integration; the other, so that performance would be facilitated by perceptual segregation. Thresholds were measured in both tasks under a wide range of conditions produced by varying three stimulus parameters known to influence stream formation: frequency separation, tone-presentation rate, and sequence length. In addition to these performance-based measures, subjective judgments of perceived segregation were collected in the same listeners under corresponding stimulus conditions. The patterns of results obtained in the two temporal-discrimination tasks, and the relationships between thresholds and perceived-segregation judgments, were mostly consistent with the hypothesis that stream segregation helped performance in one task and impaired performance in the other task. The tasks and stimuli described here may prove useful in future behavioral or neurophysiological experiments, which seek to manipulate and measure neural correlates of auditory streaming while minimizing differences between the physical stimuli.

Effects of temporal uncertainty and temporal expectancy on infants' auditory sensitivity

Adults are more sensitive to a sound if they know when the sound will occur. In the present experiment, the effects of temporal uncertainty and temporal expectancy on infants' and adults' detection of a 1 kHz tone in a broadband noise were examined. In one experiment, masked sensitivity was measured with an acoustic cue and without an acoustic cue to possible tone presentation times. Adults' sensitivity was greater for the cue than for the no-cue condition, while infants' sensitivity did not differ significantly between the cue and no-cue conditions. In a second experiment, the effect of temporal expectancy was investigated. The detection advantage for sounds occurring at an expected (most frequent) time, over sounds occurring at unexpected (less frequent) times, was examined. Both infants and adults detected a tone better when it occurred before or at an expected time following a cue than when it occurred at a later time. Thus, despite the fact that the auditory cue did not improve infants' sensitivity, it nonetheless provided the basis for temporal expectancies. Infants, like adults, are more sensitive to sounds that are consistent with temporal expectancy.

Echo frequency selectivity of duration-tuned inferior collicular neurons of the big brown bat, Eptesicus fuscus, determined with pulse-echo pairs

During hunting, insectivorous bats such as Eptesicus fuscus progressively vary the repetition rate, duration, frequency and amplitude of emitted pulses such that analysis of an echo parameter by bats would be inevitably affected by other co-varying echo parameters. The present study is to determine the variation of echo frequency selectivity of duration-tuned inferior collicular neurons during different phases of hunting using pulse-echo (P-E) pairs as stimuli. All collicular neurons discharge maximally to a tone at a particular frequency which is defined as the best frequency (BF). Most collicular neurons also discharge maximally to a BF pulse at a particular duration which is defined as the best duration (BD). A family of echo iso-level frequency tuning curves (iso-level FTC) of these duration-tuned collicular neurons is measured with the number of impulses in response to the echo pulse at selected frequencies when the P-E pairs are presented at varied P-E duration and gap. Our data show that these duration-tuned collicular neurons have narrower echo iso-level FTC when measured with BD than with non-BD echo pulses. Also, IC neurons with low BF and short BD have narrower echo iso-level FTC than IC neurons with high BF and long BD have. The bandwidth of echo iso-level FTC significantly decreases with shortening of P-E duration and P-E gap. These data suggest that duration-tuned collicular neurons not only can facilitate bat's echo recognition but also can enhance echo frequency selectivity for prey feature analysis throughout a target approaching sequence during hunting. These data also support previous behavior studies showing that bats prepare their auditory system to analyze expected returning echoes within a time window to extract target features after pulse emission.

Tuning in the spatial dimension: evidence from a masked speech identification task

Spatial release from masking was studied in a three-talker soundfield listening experiment. The target talker was presented at 0 degrees azimuth and the maskers were either colocated or symmetrically positioned around the target, with a different masker talker on each side. The symmetric placement greatly reduced any "better ear" listening advantage. When the maskers were separated from the target by +/-15 degrees , the average spatial release from masking was 8 dB. Wider separations increased the release to more than 12 dB. This large effect was eliminated when binaural cues and perceived spatial separation were degraded by covering one ear with an earplug and earmuff. Increasing reverberation in the room increased the target-to-masker ratio (TM) for the separated, but not colocated, conditions reducing the release from masking, although a significant advantage of spatial separation remained. Time reversing the masker speech improved performance in both the colocated and spatially separated cases but lowered TM the most for the colocated condition, also resulting in a reduction in the spatial release from masking. Overall, the spatial tuning observed appears to depend on the presence of interaural differences that improve the perceptual segregation of sources and facilitate the focus of attention at a point in space.

Predicting the path of a changing sound: velocity tracking and auditory continuity

Three studies demonstrate listeners' ability to use the rate of a sound's frequency change (velocity) to predict how the spectral path of the sound is likely to evolve, even in the event of an occlusion. Experiments 1 and 2 use a modified probe-signal method to measure attentional filters and demonstrate increased detection to sounds falling along implied paths of constant-linear velocity. Experiment 3 shows listeners perceive a suprathreshold tone as falling along a trajectory of constant velocity when the frequency is near to the region of greatest detection as measured in Experiments 1 and 2. Further, results show greater accuracy and decreased bias in the use of velocity information with increased exposure to a constant-velocity sound. As the duration of occlusion lengthens, results also show a downward shift (relative to a trajectory of constant velocity) in the frequency at which listeners' detection and experience of a continuous trajectory are greatest. A preliminary model of velocity processing is proposed to account for this downward shift. Results show listeners' use of velocity in extrapolating sounds with dynamically changing spectral and temporal properties and provide evidence for its role in perceptual auditory continuity within a noisy acoustic environment.

Role of attention in overshoot: frequency certainty versus uncertainty

Overshoot, the elevation in the threshold for a brief signal that comes on close to masker onset, was measured with signal frequency certain (same frequency on every trial) or uncertain (randomized over trials). In broadband noise, thresholds were higher 2 ms after masker onset than 200 ms later, by 9 dB with frequency certainty, by 6-7 dB with uncertainty. In narrowband noise centered on the signal frequency, thresholds at 2 ms were not elevated with certainty, but were elevated 4-5 dB with uncertainty. Thus, frequency uncertainty leads to less overshoot in broadband noise, to more overshoot in narrowband noise. Reduced overshoot in broadband noise may come about because the masker, given its many frequencies, disrupts focusing at onset as much under certainty as uncertainty. Once the initial disruption dissipates, threshold is lower with certainty so overshoot is greater. In contrast, a narrowband noise with frequencies only near the signal does not disrupt focusing when the signal frequency is known beforehand, so overshoot is absent. When frequency is uncertain, the narrowband noise serves to focus attention on the signal frequency; as this requires time, detection near noise onset is poorer than later on, so overshoot is present.

Selective attention increases both gain and feature selectivity of the human auditory cortex

BACKGROUND

An experienced car mechanic can often deduce what's wrong with a car by carefully listening to the sound of the ailing engine, despite the presence of multiple sources of noise. Indeed, the ability to select task-relevant sounds for awareness, whilst ignoring irrelevant ones, constitutes one of the most fundamental of human faculties, but the underlying neural mechanisms have remained elusive. While most of the literature explains the neural basis of selective attention by means of an increase in neural gain, a number of papers propose enhancement in neural selectivity as an alternative or a complementary mechanism.

METHODOLOGY/PRINCIPAL FINDINGS

Here, to address the question whether pure gain increase alone can explain auditory selective attention in humans, we quantified the auditory cortex frequency selectivity in 20 healthy subjects by masking 1000-Hz tones by continuous noise masker with parametrically varying frequency notches around the tone frequency (i.e., a notched-noise masker). The task of the subjects was, in different conditions, to selectively attend to either occasionally occurring slight increments in tone frequency (1020 Hz), tones of slightly longer duration, or ignore the sounds. In line with previous studies, in the ignore condition, the global field power (GFP) of event-related brain responses at 100 ms from the stimulus onset to the 1000-Hz tones was suppressed as a function of the narrowing of the notch width. During the selective attention conditions, the suppressant effect of the noise notch width on GFP was decreased, but as a function significantly different from a multiplicative one expected on the basis of simple gain model of selective attention.

CONCLUSIONS/SIGNIFICANCE

Our results suggest that auditory selective attention in humans cannot be explained by a gain model, where only the neural activity level is increased, but rather that selective attention additionally enhances auditory cortex frequency selectivity.

The time required to focus on a cued signal frequency

How quickly can a listener focus on a single tonal cue that indicates the frequency of an upcoming signal? Initial measurements were made with frequency uncertainty (signal frequency varies randomly from trial to trial) and with certainty (same frequency on all trials). Measured by a yes-no procedure, thresholds for 40- and 20-ms signals presented in continuous broadband noise at 50 dB SPL were higher in uncertainty than in certainty; the difference decreased monotonically from 5 dB at frequencies below 500 Hz to under 3 dB above about 2500 Hz. This decrease in the detrimental effect from uncertainty, which comes about with increasing signal frequency, may result from preferential attention to higher frequencies. In a second experiment, frequency again varied randomly, but each trial now began with a cue at the signal frequency. The critical variable was the delay from cue onset to signal onset. A delay of 352 ms eliminated the detrimental effect of frequency uncertainty at all frequencies. At the shortest delays of 52 and 82 ms the detrimental effect was reduced primarily at lower frequencies. Our analysis suggests that shifting focus to a cued frequency region, under optimal stimulus conditions, requires less than 52 ms.

Duration selectivity organization in the inferior colliculus of the big brown bat, Eptesicus fuscus

Duration selectivity of auditory neurons plays an important role in sound recognition. Previous studies show that GABA-mediated duration selectivity of neurons in the central nucleus of the inferior colliculus (IC) of many animal species behave as band-, short-, long- and all-pass filters to sound duration. The present study examines the organization of duration selectivity of IC neurons of the big brown bat, Eptesicus fuscus, in relation to graded spatial distribution of GABA(A) receptors, which are mostly distributed in the dorsomedial region of the IC but are sparsely distributed in the ventrolateral region. Duration selectivity of IC neuron is studied before and during iontophoretic application of GABA and its antagonist, bicuculline. Bicuculline application decreases and GABA application increases duration selectivity of IC neurons. Bicuculline application produces more pronounced broadening of the duration tuning curves of neurons at upper IC than at deeper IC but the opposite is observed during GABA application. The best duration of IC neurons progressively lengthens and duration selectivity decreases with recording depth both before and during drug application. As such, low best frequency neurons at upper IC have shorter best duration and sharper duration selectivity than high best frequency neurons in the deeper IC have. These data suggest that duration selectivity of IC neurons systematically varies with GABA(A) receptor distribution gradient within the IC.

Temporal masking in electric hearing

Temporal masking can be defined as the detection threshold of a brief signal as a function of the signal delay in a relatively long masker. The temporal masking pattern in normal acoustic hearing reveals temporal edge enhancement in which the signal detection threshold is greater near the masker onset than in the steady-state portion. Both peripheral and central mechanisms appear to underlie temporal edge enhancement, but their relative contributions remain elusive. Cochlear implants bypass cochlear mechanical processing and stimulate the auditory nerve directly, thereby providing a unique opportunity to separate the peripheral mechanisms from the central mechanisms. Here, we systematically measured temporal masking in electric hearing by examining whether a brief signal was harder to detect at the onset than in the steady-state portion of a long masker (the "overshoot" effect). The signal and the masker were presented (1) either to the same electrode or to different electrodes, (2) at the same stimulation or different rates, and (3) in a simultaneous or an interleaved fashion. A consistent pattern of results was observed, depending on the stimulus configuration between the signal and the masker. Simultaneous stimulation at the same rate and with the same electrode produced no difference in sensitivity between the onset and the steady-state conditions, but interleaved stimulation at different rates or with different electrodes produced a significant difference. Unlike acoustic hearing, high masker levels produced an overshoot effect, and low masker levels produced an undershoot effect. Although the present results are consistent with the "on-frequency vs. off-frequency" hypothesis for the overshoot effect, results also suggest a central "same vs. different" mechanism underlying temporal masking. These results have practical implications for improving cochlear implant design.

The advantage of knowing where to listen

This study examined the role of focused attention along the spatial (azimuthal) dimension in a highly uncertain multitalker listening situation. The task of the listener was to identify key words from a target talker in the presence of two other talkers simultaneously uttering similar sentences. When the listener had no a priori knowledge about target location, or which of the three sentences was the target sentence, performance was relatively poor-near the value expected simply from choosing to focus attention on only one of the three locations. When the target sentence was cued before the trial, but location was uncertain, performance improved significantly relative to the uncued case. When spatial location information was provided before the trial, performance improved significantly for both cued and uncued conditions. If the location of the target was certain, proportion correct identification performance was higher than 0.9 independent of whether the target was cued beforehand. In contrast to studies in which known versus unknown spatial locations were compared for relatively simple stimuli and tasks, the results of the current experiments suggest that the focus of attention along the spatial dimension can play a very significant role in solving the "cocktail party" problem.

Concurrent measurement of the detectability of tone bursts and their effect on the excitability of the human blink reflex using a probe-signal method

The probe-signal method has shown that auditory signals that are either presented more often in a series of trials or that are immediately preceded by cues of the same frequency on a single trial are detected more readily than signals of other frequencies. The frequency range in which detection is favored defines an attentional band, which is thought to result from an effective attenuation of deviant frequencies in the cochlea, possibly by activation of the olivocochlear bundle. In a 2IFC procedure in which the first observation interval was preceded by a 1300-Hz cue, subjects detected cued probe tones (at 1300 Hz) but not uncued probe tones (at 1000 Hz or 1600 Hz) at better than chance levels. Concurrent elicitation of a blink reflex by presentation of an air puff in the first observation interval on a random half of the trials showed that cued probes, but not uncued probes, inhibited the size of the blink reflex. These data show that uncued probes do not enter into the low-level sensory processing in the brainstem which is responsible for reflex modification. This finding is consistent with the view that stimuli whose frequency falls outside an attentional band are excluded at the auditory periphery.

Combined representations for frequency and duration in detection templates for expected signals

When trying to detect a tonal signal in a continuous broadband noise, listeners attend selectively to both the frequency and the duration of the expected signal. However, it is not known whether they monitor separate or combined representations of these two attributes. To investigate this question, a probe-signal method was used to measure the detectability of signals of expected and unexpected durations at two expected frequencies. The four listeners expected only one of two signals to be presented at random: a brief tone at one frequency or a long tone at another frequency. For each signal frequency, the detectability of the signals of unexpected duration decreased to near chance as the difference between the expected and unexpected duration, at that frequency, increased. The frequency specificity of this duration tuning indicates that both the frequency and the duration of an expected stimulus are represented in a single template.

The time course of attention in a simple auditory detection task

What is the time course of human attention in a simple auditory detection task? To investigate this question, we determined the detectability of a 20-msec, 1000-Hz tone presented at expected and unexpected times. Twelve listeners who expected the tone to occur at a specific time after a 300-msec narrowband noise rarely detected signals presented 150-375 msec before or 100-200 msec after that expected time. The shape of this temporal-attention window depended on the expected presentation time of the tone and the temporal markers available in the trials. Further, though expecting the signal to occur in silence, listeners often detected signals presented at unexpected times during the noise. Combined with previous data, these results further clarify the listening strategy humans use when trying to detect an expected sound: Humans seem to listen specifically for that sound, while ignoring the background in which it is presented, around the time when the sound is expected to occur.

Informational masking and musical training

The relationship between musical training and informational masking was studied for 24 young adult listeners with normal hearing. The listeners were divided into two groups based on musical training. In one group, the listeners had little or no musical training; the other group was comprised of highly trained, currently active musicians. The hypothesis was that musicians may be less susceptible to informational masking, which is thought to reflect central, rather than peripheral, limitations on the processing of sound. Masked thresholds were measured in two conditions, similar to those used by Kidd et al. [J. Acoust. Soc. Am. 95, 3475-3480 (1994)]. In both conditions the signal was comprised of a series of repeated tone bursts at 1 kHz. The masker was comprised of a series of multitone bursts, gated with the signal. In one condition the frequencies of the masker were selected randomly for each burst; in the other condition the masker frequencies were selected randomly for the first burst of each interval and then remained constant throughout the interval. The difference in thresholds between the two conditions was taken as a measure of informational masking. Frequency selectivity, using the notched-noise method, was also estimated in the two groups. The results showed no difference in frequency selectivity between the two groups, but showed a large and significant difference in the amount of informational masking between musically trained and untrained listeners. This informational masking task, which requires no knowledge specific to musical training (such as note or interval names) and is generally not susceptible to systematic short- or medium-term training effects, may provide a basis for further studies of analytic listening abilities in different populations.

Enhancement, extension, and reversal of the frequency selectivity effect

The influence of a frequency cue on judgments of whether or not a subsequent target incorporated a brief silent gap was examined. In Experiment 1, there was no predictive frequency relation and evidence of auditory inhibition of return was obtained with frequency repetitions, producing a facilitative effect at 175-msec stimulus onset asynchrony (SOA) and an inhibitory effect at 775-msec SOA. Relative to this baseline performance pattern, increasing the probability of a frequency match to .75 (Experiment 2) served to generate a beneficial effect of frequency repetitions at lengthy SOAs and to enlarge its magnitude at 175-msec SOA. In contrast, a reduction in the probability of a frequency match to .25 (Experiment 3) resulted in the elimination of any facilitative effect of repetition at 175-msec SOA and the development of an inhibitory effect at 475- and 1,075-msec SOA. These results establish that a frequency cue may engage both exogenous and endogenous attentional processes within 175 msec following its presentation.

Capture of attention in selective frequency listening

Four signal-detection experiments demonstrated robust stimulus-driven, or exogenous, attentional processes in selective frequency listening. Detection of just-above-threshold signal tones was consistently better when the, signal matched the frequency of an uninformative cue tone, even with relatively long cue-signal delays (Experiment 1) or when as few as 1 in 8 signals were at the cued frequency (Experiment 2). Experiments 3 and 4 compared performance with informative and uninformative cues. The involvement of intentional, or endogenous, processes was found to only slightly increase the size of the cuing effect beyond that evident with solely exogenous processes, although the attention band, a measure of how narrowly attention is focused, was found to be wider when cues were informative. The implications for models of auditory attention are discussed.

Expectancy, attention, and time

Seven experiments examine the influence of contextual timing manipulations on prospective time judgments. Subjects judged durations of standard vs comparison time intervals in the context of a preceding induction (context) sequence. In some experiments, the rate of the induction sequence was systematically manipulated relative to the range of to-be-judged standard time intervals; in others, the induction sequence was omitted. Time judgments were strongly influenced by the rate of an induction sequence with best performance occurring when the standard time interval ended as expected, given context rate. An expectancy profile, in the form of an inverted U, indicated that time estimation accuracy declined systematically as a standard interval differed from a context rate. A similar expectancy profile emerged when the context rate was based on a harmonic subdivision (one-half) of an expected standard interval. Results are discussed in terms of various stimulus-based models of prospective time judgments, including those which appeal to attentional periodicities and entrainment.

Pitch matches between unresolved complex tones differing by a single interpulse interval

The experiment compared the pitches of complex tones consisting of unresolved harmonics. The fundamental frequency (F0) of the tones was 250 Hz and the harmonics were bandpass filtered between 5500 and 7500 Hz. Two 20-ms complex-tone bursts were presented, separated by a brief gap. The gap was an integer number of periods of the waveform: 0, 4, or 8 ms. The envelope phase of the second tone burst was shifted, such that the interpulse interval (IPI) across the gap was reduced or increased by 0.25 or 0.75 periods (1 or 3 ms). A "no shift" control was also included, where the IPI was held at an integer number of periods. Pitch matches were obtained by varying the F0 of a comparison tone with the same temporal parameters as the standard but without the shift. Relative to the no-shift control, the variations in IPI produced substantial pitch shifts when there was no gap between the bursts, but little effect was seen for gaps of 4 or 8 ms. However, for some conditions with the same IPI in the shifted interval, an increase in the IPI of the comparison interval from 4 to 8 ms (gap increased from 0 to 4 ms) changed the pitch match. The presence of a pitch shift suggests that the pitch mechanism is integrating information across the two tone bursts. It is argued that the results are consistent with a pitch mechanism employing a long integration time for continuous stimuli that is reset in response to temporal discontinuities. For a 250-Hz F0, an 8-ms IPI may be sufficient for resetting. Pitch models based on a spectral analysis of the simulated neural spike train, on an autocorrelation of the spike train, and on the mean rate of pitch pulses, all failed to account for the observed pitch matches.

Efficient across-frequency integration: evidence from psychometric functions

Across-frequency integration of complex signals was investigated by measuring psychometric functions [log (d') versus signal level in dB SPL] for detection of brief and long signals presented in broadband noise. The signals were tones at 630, 1600, and 4000 Hz, and a nine-tone complex with components spaced at one-third-octave frequencies between 630 and 4000 Hz. The phase relationship of the components in the complex was varied such that adjacent components were in phase (at 0 degrees), 90, or 180 degrees out of phase. Signal durations (defined in terms of the number of cycles between the half-amplitude points of the Gaussian envelopes) of 4.7 and 150 cycles were tested. Results for six normal-hearing listeners showed that the slopes of the psychometric functions were steeper for the brief than for the long signals, and steeper for the tone complexes than for the tones, particularly for the brief signals. This suggests that the transformation from signal intensity to decision variable may be different for brief complex signals than for tonal signals and long complex signals. Thresholds obtained from the psychometric functions were in excellent agreement with those obtained with an adaptive procedure that employed three interleaved tracks. For the long signals, the threshold improvement for the tone complexes relative to a single tone was well described by a 5* log (n) integration rule. However, the threshold improvement for brief signals obeyed a more efficient integration rule of 7 to 8* log (n). A portion of this effect could be accounted for by the phase relationship of the tone complexes; thresholds for brief signals were lowest when the components were in phase at the envelope peak of the signal. This finding indicates that temporal synchrony across auditory channels may enhance detection of brief multi-tone complexes.

Influence of spatial and temporal coding on auditory gap detection

This study investigated the effect on gap detection of perceptual channels, hypothesized to be tuned to spatial location or fundamental frequency (f0). Thresholds were measured for the detection of a silent temporal gap between two markers. In the first experiment, the markers were broadband noise, presented either binaurally or monaurally. In the binaural conditions, the markers were either diotic, or had a 640-micros interaural time difference (ITD) or a 12-dB interaural level difference (ILD). Reversing the ITD across the two markers had no effect on gap detection relative to the diotic condition. Reversing the ILD across the two markers produced a marked deterioration in performance. However, the same deterioration was observed in the monaural conditions when a 12-dB level difference was introduced between the two markers. The results provide no evidence for the role of spatially tuned neural channels in gap detection. In the second experiment, the markers were harmonic tone complexes, filtered to contain only high, unresolved harmonics. Using complexes with a fixed spectral envelope, where the f0 (of 140 or 350 Hz) was different for the two markers, produced a deterioration in performance, relative to conditions where the f0 remained the same. A larger deterioration was observed when the two markers occupied different spectral regions but had the same f0. This supports the idea that peripheral coding is dominant in determining gap-detection thresholds when the two markers differ along any physical dimension. Higher-order neural coding mechanisms of f0 and spatial location seem to play a smaller role and no role, respectively.

Temporal integration in the presence of off-frequency maskers

Temporal integration was measured at a relatively low and a relatively high signal frequency under conditions of off-frequency masking. The masker was typically gated for 300 ms, and the signal was presented 70 ms after masker onset. In experiment 1, the signal frequency was 500 or 2000 Hz. Temporal integration was measured in quiet and in the presence of a masker whose frequency was lower or higher than the signal frequency. In all listening situations, there was less integration at 2000 Hz than at 500 Hz. This effect of frequency was particularly dramatic in the presence of a lower frequency masker, where there was almost no integration at 2000 Hz. Experiment 2 showed that this dramatic effect of frequency cannot be understood in terms of the underlying psychometric functions. Experiment 3 measured temporal integration at 750 and 2000 Hz for a large number of masker-signal frequency separations for both a tonal and a noise masker, and in conditions where the masker was gated or continuous. The results with the gated tonal masker largely confirmed the results of experiment 1. The results with the continuous tonal masker and the gated or continuous noise masker, however, were quite different. In those cases, the amount of temporal integration at both signal frequencies was more or less independent of the masker-signal separation; the masked temporal integration was nearly equal to the integration in quiet. Thus based on the conditions evaluated here, off-frequency masked temporal integration differs substantially from integration in quiet only for gated tonal maskers located considerably lower in frequency than the signal. It is unclear how to account for this finding, although it may be related to attentional factors.

Detection of sinusoidal amplitude modulation at unexpected rates

The detectability of sinusoidal amplitude modulation at unexpected modulation rates was assessed using a probe-signal method. With this method, three listeners were led to expect a target modulation rate (4, 32, or 256 Hz) by presenting the signal most often at that rate, and sensitivity to modulation at six other unexpected rates between 4 and 256 Hz was measured via occasionally presented probe modulation rates. The modulation phase was random on each two-interval forced-choice trial and the overall level of the 500-ms broadband carrier was randomly varied between 55 and 75 dB SPL across intervals. The modulation depth at each rate was set so that the modulation was detected on about 90% of the trials when only that rate was presented. Performance at the unexpected rates depended upon the target rate. For the 4-Hz target, modulation at all rates was detected on about 80% of the trials. For the 32- and 256-Hz targets, unexpected modulation rates of 16 Hz and above were detected on 80%-90% of the trials, but modulation rates below 16 Hz were detected nearly at chance. The influence of expectation of modulation rate on the detection of sinusoidal amplitude modulation is not readily predicted by current models of modulation detection.

Variable-duration notched-noise experiments in a broadband noise context

A variable-duration notched-noise experiment was conducted in a noise context. Broadband noise preceded and followed a tone and notched noise of similar duration. Thresholds were measured at four durations (10, 30, 100, and 300 ms), two center frequencies (0.6, 2.0 kHz), and five relative notch widths (0.0, 0.1, 0.2, 0.4, 0.8). At 0.6 kHz, 10-ms thresholds decrease 6 dB across notch widths, while 300-ms thresholds decrease over 35 dB. These trends are similar but less pronounced at 2 kHz. In a second experiment, the short-duration notched noise was replaced with a flat noise which provided an equivalent amount of simultaneous masking and thresholds dropped by as much as 20 dB. A simple combination of simultaneous and nonsimultaneous masking is unable to predict these results. Instead, it appears that the elevated thresholds at short durations are dependent on the spectral shape of the simultaneous masker.

"Central" auditory gap detection: a spatial case

Normal listeners were tested for their temporal auditory gap detection thresholds using free-field presentation of white-noise stimuli delivered from the left (L) and right (R) poles of the interaural axis. The noise bursts serving as the leading and trailing markers for the silent period were presented in either the same (LL,RR) or different (LR,RL) auditory locations. The duration of the leading marker was a second independent variable. Gap thresholds for stimuli in which the markers had the same location were low, and usually were independent of the duration of the leading marker. Gap thresholds for the LR and RL conditions were longer. These gap thresholds were sensitive to the duration of the leading marker, and increased as the leading marker duration decreased. This finding is consistent with the hypothesis that a relative timing operation mediates gap detection when the markers activate different perceptual channels. The present data suggest that this timing process can operate on perceptual channels emerging from central nervous system processing.

Inhibitory processes in auditory selective attention: evidence of location-based and frequency-based inhibition of return

The possibility that there is an inhibitory component to auditory covert orienting was addressed. Each trial consisted of a cue followed by a target, and listeners were required to detect, localize, or identify the frequency of the target. At 150-msec stimulus onset asynchrony (SOA), performance was best when stimuli sounded from the same location or were of the same frequency. However, at 750-msec SOA, performance was best when stimuli differed in location or were of different frequencies. These results document the existence of both location-based and frequency-based auditory inhibition of return.

Detection of signals having expected and unexpected temporal structures

Two experiments used a variant of the 'probe-signal' method [Greenberg and Larkin, J. Acoust. Soc. Am. 44 (1968) 1513-1523] to examine the effect of expectedness on the detection of signals having fixed frequencies but uncertain temporal structures. Expectedness was manipulated by presenting one signal on 75% of the trials and the other on only 25% of trials. Experiment 1 measured sensitivity (d') to 4000-Hz sinusoidal signals having durations of 10 ms and 295 ms. The results confirmed the finding by Wright and Dai [J. Acoust. Soc. Am. 95 (1994) 931-938] that sensitivity was poorer when the duration of the signal was unexpected. The second experiment used two signals having the same overall duration: six 10-ms 4000-Hz tone pulses with a 0-ms inter-pulse interval, and two 10-ms 4000-Hz pulses separated by 40 ms. Again, sensitivity was lower when the temporal structure of the signal was unexpected. It is argued that this finding is inconsistent with 'long time constant' models of temporal integration, but can be accounted for by assuming that subjects combine information from a number of short 'looks' at the signal, with the number and temporal location of these looks being influenced by its expected temporal structure.

Forward-masked intensity discrimination: duration effects and spectral effects

Three experiments were completed to examine the effect of masker duration and spectrum on forward-masked intensity discrimination. Four listeners participated in each experiment. Intensity discrimination was measured in quiet and in the presence of forward maskers using adaptive forced-choice procedures. The standard duration was either short (10 ms) or long (250 ms) in experiment 1 and short (10 ms) in experiment 2. The standard always occurred 100 ms after the offset of the masker. In the first experiment employing 1.0-kHz maskers and standards, a short duration masker (10 ms) produced more masking than a long duration masker (250 ms). A mid-level elevation of the Weber fraction was observed for all conditions. To ensure that the results of experiment 1 were not influenced by off-frequency listening, the second experiment employed a broadband noise masker. As before, a short duration (10 ms) masker produced more masking than a long duration masker (100 ms) and a mid-level elevation of Weber fractions was observed. This outcome is inconsistent with a peripheral sensory effect for which an increase in masker duration should result in a greater amount of adaptation, and, as a consequence more masking. A third experiment employing a broadband noise masker and standard showed the greatest amount of masking for low-level standards, but only when the duration of the masker and standard was short. This result is similar to one seen for a single listener in the first experiment for short duration tonal maskers and standards. For this listener, a second tone presented at 4.133 kHz presented simultaneously with the 1.0 kHz masker reduced significantly the amount of masking for low-level standards, but the mid-level elevation of the Weber fraction remained. Taken together, these results suggest that perceptual similarity plays a role in forward-masked intensity discrimination but does not account entirely for the mid-level elevation of the Weber fraction.

Detection of silent intervals between noises activating different perceptual channels: some properties of "central" auditory gap detection

This article describes four experiments on gap detection by normal listeners, with the general goal being to examine the consequences of using noises in different perceptual channels to delimit a silent temporal gap to be detected. In experiment 1, subjects were presented with pairs of narrow-band noise sequences. The leading element in each pair had a center frequency of 2 kHz and the trailing element's center frequency was parametrically varied. Gap detection thresholds became increasingly poor, sometimes by up to an order of magnitude, as the spectral disparity was increased between the noise bursts that marked the gap. These data suggested that gap-detection performance is impoverished when the underlying perceptual timing operation requires a comparison of activity in different perceptual channels rather than a discontinuity detection within a given channel. In experiment 2, we assessed the effect of leading-element duration in within-channel and between-channel gap detection tasks. Gap detection thresholds rose when the duration of the leading element was less than about 30 ms, but only in the between-channel case. In experiment 3, the gap-detection stimulus was redesigned so that we could probe the perceptual mechanisms that might be involved in stop consonant discrimination. The leading element was a wideband noise burst, and the trailing element was a 300-ms bandpassed noise centered on 1.0 kHz. The independent variable was the duration of the leading element, and the dependent variable was the smallest detectable gap between the elements. When the leading element was short in duration (5-10 ms), gap thresholds were close to 30 ms, which is close to the voice onset time that parses some voiced from unvoiced stop consonants. In experiment 4, the generality of the leading-element duration effect in between-channel gap detection was examined. Spectrally identical noises defining the leading and trailing edges of the gap were presented to the same or to different ears. There was a leading-element duration effect only for the between channel case. The mean gap threshold was again close to 30 ms for short leading-element durations. Taken together, the data suggest that gap detection requiring a temporal correlation of activity in different perceptual channels is a fundamentally different task to the discontinuity detection used to execute gap detection performance in the traditional, within-channel paradigm.

A psychoacoustic model for the noise masking of plosive bursts

A model for predicting the masked thresholds of the voiceless plosive bursts /k,t,p/ in background noise is proposed. Because plosive bursts are brief, are generated by a noise source, and have different spectral characteristics, the modeling approach accounts for duration, center frequency, and signal bandwidth. Noise-in-noise masking experiments are conducted using a broadband masker and bandpass noise signals of varying bandwidth (100-5483 Hz), duration (10-300 ms), and center frequency (0.4-4 kHz). Data from these experiments are used to parametrize an auditory filter model in which the effective bandwidth and the signal-to-noise ratio at threshold for each filter are duration dependent. The duration-dependent filter model is then used to predict the thresholds of synthetic and naturally spoken plosive bursts in background noise. Finally, results from pilot notched-noise experiments are presented which support duration-dependent frequency selectivity.