Evidence for causal top-down frontal contributions to predictive processes in speech perception.

Perception relies on the integration of sensory information and prior expectations. Here we show that selective neurodegeneration of human frontal speech regions results in delayed reconciliation of predictions in temporal cortex. These temporal regions were not atrophic, displayed normal evoked magnetic and electrical power, and preserved neural sensitivity to manipulations of sensory detail. Frontal neurodegeneration does not prevent the perceptual effects of contextual information; instead, prior expectations are applied inflexibly. The precision of predictions correlates with beta power, in line with theoretical models of the neural instantiation of predictive coding. Fronto-temporal interactions are enhanced while participants reconcile prior predictions with degraded sensory signals. Excessively precise predictions can explain several challenging phenomena in frontal aphasias, including agrammatism and subjective difficulties with speech perception. This work demonstrates that higher-level frontal mechanisms for cognitive and behavioural flexibility make a causal functional contribution to the hierarchical generative models underlying speech perception.

Subcortical neural synchrony and absolute thresholds predict frequency discrimination independently.

The neural mechanisms of pitch coding have been debated for more than a century. The two main mechanisms are coding based on the profiles of neural firing rates across auditory nerve fibers with different characteristic frequencies (place-rate coding), and coding based on the phase-locked temporal pattern of neural firing (temporal coding). Phase locking precision can be partly assessed by recording the frequency-following response (FFR), a scalp-recorded electrophysiological response that reflects synchronous activity in subcortical neurons. Although features of the FFR have been widely used as indices of pitch coding acuity, only a handful of studies have directly investigated the relation between the FFR and behavioral pitch judgments. Furthermore, the contribution of degraded neural synchrony (as indexed by the FFR) to the pitch perception impairments of older listeners and those with hearing loss is not well known. Here, the relation between the FFR and pure-tone frequency discrimination was investigated in listeners with a wide range of ages and absolute thresholds, to assess the respective contributions of subcortical neural synchrony and other age-related and hearing loss-related mechanisms to frequency discrimination performance. FFR measures of neural synchrony and absolute thresholds independently contributed to frequency discrimination performance. Age alone, i.e., once the effect of subcortical neural synchrony measures or absolute thresholds had been partialed out, did not contribute to frequency discrimination. Overall, the results suggest that frequency discrimination of pure tones may depend both on phase locking precision and on separate mechanisms affected in hearing loss.

Human auditory nerve compound action potentials and long latency responses.

CONCLUSION: The compound action potential (CAP) is followed by a long latency response (LLR), attributable to the post-auricular musculature. The LLR to one pulse may overlap with the CAP to a subsequent one, contributing to the clinically observed reduction in CAP at high pulse rates. OBJECTIVES: To measure refractory and other influences on CAPs in humans and guinea pigs. MATERIALS AND METHODS: CAPs were obtained from humans using trans-tympanic and extra-tympanic electrocochleography and from anaesthetized guinea pigs. Stimuli were single pulses presented at a slow rate, pairs of pulses, and 100 ms pulse trains where the inter-pulse interval alternated between 4 and 6 ms. RESULTS: For single pulses, the CAP shape was similar across species. For pairs of pulses, the CAP to the second pulse was smaller than that to the first, and decreased with increasing inter-pulse interval in a way that was similar across species. For pulse trains, CAPs were observed in response to each pulse in the train for the guinea pigs, but not for humans. For both filtered and unfiltered single pulses, there was a large LLR in humans, but not in guinea pigs, with peaks at latencies of 10-12 and 20-25 ms. Posture affected the LLR in a way consistent with the post-auricular response.

Dynamic aspects of the continuity illusion: perception of level and of the depth, rate, and phase of modulation.

The perception of modulation of a tone interrupted by a noise burst was investigated. The tone and its modulation were perceived as continuing through the noise. In experiment 1, subjects rated the similarity of an uninterrupted tone and a tone interrupted by noise, in terms of the perceived level and modulation depth of the sinusoidal carrier. The values of these parameters in the central portion of the uninterrupted tone were systematically varied. Both amplitude and frequency modulation (AM and FM) were used. The results indicated that the perceived level and modulation depth of the carrier did not change greatly during the noise burst. When the modulation rate differed before and after the noise burst, the modulation-rate transition was perceived to occur near the end of the noise burst for the FM stimuli. Hence, for these stimuli, the continuity illusion appears to be dominated by the portion of the tone before, rather than after, the interruption. Results for the AM stimuli showed a non-significant trend in the same direction. Experiment 2 used forced-choice tasks to evaluate the ability to detect a change in the ongoing phase of AM and FM following interruption by a noise burst. The results confirmed earlier findings for FM tones, and extended them to AM tones, showing that listeners lost track of the phase of the modulation, even though the modulation was perceived as continuous.

Detection, direction discrimination, and off-frequency interference of center-frequency modulations and glides for vowel formants.

Vowels are mainly classified by the positions of peaks in their frequency spectra, the formants. For normal-hearing subjects, change detection and direction discrimination were measured for linear glides in the center frequency (CF) of formantlike sounds. A CF rove was used to prevent subjects from using either the start or end points of the glides as cues. In addition, change detection and starting-phase (start-direction) discrimination were measured for similar stimuli with a sinusoidal 5-Hz formant-frequency modulation. The stimuli consisted of single formants generated using a number of different stimulus parameters including fundamental frequency, spectral slope, frequency region, and position of the formant relative to the harmonic spectrum. The change detection thresholds were in good agreement with the predictions of a model which analyzed and combined the effects of place-of-excitation and temporal cues. For most stimuli, thresholds were approximately equal for change detection and start-direction discrimination. Exceptions were found for stimuli that consisted of only one or two harmonics. In a separate experiment, it was shown that change detection and start-direction discrimination of linear and sinusoidal formant-frequency modulations were impaired by off-frequency frequency-modulated interferers. This frequency modulation detection interference was larger for formants with shallow than for those with steep spectral slopes.

The effects of real and illusory glides on pure-tone frequency discrimination.

Experiment 1 measured pure-tone frequency difference limens (DLs) at 1 and 4 kHz. The stimuli had two steady-state portions, which differed in frequency for the target. These portions were separated by a middle section of varying length, which consisted of a silent gap, a frequency glide, or a noise burst (conditions: gap, glide, and noise, respectively). The noise burst created an illusion of the tone continuing through the gap. In the first condition, the stimuli had an overall duration of 500 ms. In the second condition, stimuli had a fixed 50-ms middle section, and the overall duration was varied. DLs were lower for the glide than for the gap condition, consistent with the idea that the auditory system contains a mechanism specific for the detection of dynamic changes. DLs were generally lower for the noise than for the gap condition, suggesting that this mechanism extracts information from an illusory glide. In a second experiment, pure-tone frequency direction-discrimination thresholds were measured using similar stimuli as for the first experiment. For this task, the type of the middle section hardly affected the thresholds, suggesting that the frequency-change detection mechanism does not facilitate the identification of the direction of frequency changes.

Perceptual auditory stream segregation of sequences of complex sounds in subjects with normal and impaired hearing.

The influence of hearing loss and aging on the perceptual organization of sound sequences was investigated by comparing the ability of young normal-hearing subjects and elderly subjects having either impaired or normal hearing for their age to form perceptual auditory streams from sequences of harmonic complex tones as a function of differences in fundamental frequency (F0). The sequences consisted of repeating triplets of harmonic complex tones separated by a silence (ABA-). In conditions in which the F0s of the A and B tone were so low that the harmonics could not be individually resolved by the peripheral auditory system even in the young normal-hearing subjects, those subjects showed similar stream segregation performance to the elderly hearing-impaired subjects. In contrast, when the F0s of the tones were high enough for the harmonics to be largely resolved at the auditory periphery in normal-hearing subjects, but presumably unresolved in the elderly subjects, the former showed significantly more stream segregation than the latter. These results, which cannot be consistently explained in terms of age differences, suggest that auditory stream segregation is adversely affected by reduced peripheral frequency selectivity of elderly individuals. This finding has implications for the understanding of the listening difficulties experienced by elderly individuals in cocktail-party situations.

Effects of attention and unilateral neglect on auditory stream segregation.

Two pairs of experiments studied the effects of attention and of unilateral neglect on auditory streaming. The first pair showed that the build up of auditory streaming in normal participants is greatly reduced or absent when they attend to a competing task in the contralateral ear. It was concluded that the effective build up of streaming depends on attention. The second pair showed that patients with an attentional deficit toward the left side of space (unilateral neglect) show less stream segregation of tone sequences presented to their left than to their right ears. Streaming in their right ears was similar to that for stimuli presented to either ear of healthy and of brain-damaged controls, who showed no across-ear asymmetry. This result is consistent with an effect of attention on streaming, constrains the neural sites involved, and reveals a qualitative difference between the perception of left- and right-sided sounds by neglect patients.

Temporal pitch perception and the binaural system.

Two experiments examined the relationship between temporal pitch (and, more generally, rate) perception and auditory lateralization. Both used dichotic pulse trains that were filtered into the same high (3,900-5,400-Hz) frequency region in order to eliminate place-of-excitation cues. In experiment 1, a 1-s periodic pulse train of rate Fr was presented to one ear, and a pulse train of rate 2Fr was presented to the other. In the "synchronous" condition, every other pulse in the 2Fr train was simultaneous with a pulse in the opposite ear. In each trial, subjects concentrated on one of the two binaural images produced by this mixture: they matched its perceived location by adjusting the interaural level difference (ILD) of a bandpass noise, and its rate/pitch was then matched by adjusting the rate of a regular pulse train. The results showed that at low Fr (e.g., 2 Hz), subjects heard two pulse trains of rate Fr, one in the "higher rate" ear, and one in the middle of the head. At higher Fr (>25 Hz) subjects heard two pulse trains on opposite sides of the midline, with the image on the higher rate side having a higher pitch than that on the "lower rate" side. The results were compared to those in a control condition, in which the pulses in the two ears were asynchronous. This comparison revealed a duplex region at Fr > 25 Hz, where across-ear synchrony still affected the perceived locations of the pulse trains, but did not affect their pitches. Experiment 2 used a 1.4-s 200-Hz dichotic pulse train, whose first 0.7 s contained a constant interaural time difference (ITD), after which the sign of the ITD alternated between subsequent pulses. Subjects matched the location and then the pitch of the "new" sound that started halfway through the pulse train. The matched location became more lateralized with increasing ITD, but subjects always matched a pitch near 200 Hz, even though the rate of pulses sharing the new ITD was only 100 Hz. It is concluded from both experiments that temporal pitch perception is not driven by the output of binaural mechanisms.

Influence of rate of change of frequency on the overall pitch of frequency-modulated tones.

The mechanism(s) determining pitch may assign less weight to portions of a sound where the frequency is changing rapidly. The present experiments explored the possible effect of this on the overall pitch of frequency-modulated sounds. Pitch matches were obtained between an adjustable unmodulated sinusoid and a sinusoidal carrier that was frequency modulated using a highly asymmetric function with the form of a repeating U or inverted U shaped function. The amplitude was constant during the 400-ms presentation time of each stimulus, except for 10-ms raised-cosine onset and offset ramps. In experiment 1, the carrier level was 50 dB SPL and the geometric mean of the instantaneous frequency of the modulated carrier, fc, was either 0.5, 1, 2, or 8 kHz. The modulation rate (fm) was 5, 10, or 20 Hz. The overall depth (maximum to minimum) of the FM was 8% of fc. For all carrier frequencies, the matched frequency was shifted away from the mean carrier frequency, downwards for the U shaped function stimuli and upwards for the repeated inverted U shaped function stimuli. The shift was typically slightly greater than 1% of fc, and did not vary markedly with fc. The effect of fm was small, but there was a trend for the shifts to decrease with increasing fm for fc = 0.5 kHz and to increase with increasing fm for fc = 2 kHz. In experiment 2, the carrier level was reduced to 20 dB SL and matches were obtained only for fc = 2 kHz. Shifts in matched frequency of about 1% were still observed, but the trend for the shifts to increase with increasing fm no longer occurred. In experiment 3, matches were obtained for a 4-kHz carrier at 50 dB SPL. Shifts of about 1% again occurred, which did not vary markedly with fm. The shifts in matched frequency observed in all three experiments are not predicted by models based on the amplitude- or intensity-weighted average of instantaneous frequency (EWAIF or IWAIF). The shifts (and the pitch shifts observed earlier for two-tone complexes and for stimuli with simultaneous AM and FM) are consistent with a model based on the assumption that the overall pitch of a frequency-modulated sound is determined from a weighted average of period estimates, with the weight attached to a given estimate being inversely related to the short-term rate of change of period and directly related to a compressive function of the amplitude.

Frequency modulation detection interference produced by asynchronous and nonsimultaneous interferers.

The effect of asynchronous and nonsimultaneous interferers on detection of sinusoidal frequency modulation (FM) was compared with the effect of a synchronous interferer. In a two-interval, two-alternative forced-choice (2I-2AFC) adaptive procedure, listeners had to detect FM with a modulation frequency of 15 Hz, imposed on a 1-kHz sinusoidal carrier (the target). The 200-ms target was presented either alone (baseline condition), or with an interferer whose timing relative to the target was varied. The interferer was a 2.3-kHz sinusoidal carrier which was also frequency modulated at a rate of 15 Hz. Experiment one showed that thresholds for detection of FM increased significantly, both with a synchronous FM interferer, and also with asynchronous interferers (starting 200 ms before and stopping 200 ms after the target). Moreover, "gapped" interferers that were turned off during presentation of the target (presented for 200 ms before and for 200 ms after the target but not simultaneously) produced the same significant increase in thresholds as an asynchronous interferer that was not interrupted. In contrast, thresholds were not affected by the presence of a gapped unmodulated sinusoidal interferer. Experiment two showed that increasing the duration of the silent gap (centered on presentation of the target) between FM interferers from 200 to 600 ms did not abolish the interference. Thus nonsimultaneous FM interferers produced frequency modulation detection interference (FMDI) even when the silent gap between the interferers and target clearly led to the interferers and target being perceived as separate auditory objects. A possible explanation for the findings is the existence of an asymmetry in perception of steady and modulated sounds, as recently proposed by Cusack and Carlyon [Br. J. Audiol. 34.2, 112 (2000)]. Alternative explanations in terms of ringing in a hypothetical modulation filter bank and adaptation seem unlikely.

Binaural effects in center-frequency modulation detection interference for vowel formants.

The detection of slow (5 Hz) center-frequency modulations of formants (signals) can be impaired by the simultaneous presentation of off-frequency modulated formants (maskers) to the same ear [J. Lyzenga and R. P. Carlyon, J. Acoust. Soc. Am. 105, 2792-2806 (1999)]. In the present study we examine this "formant-frequency modulation detection interference (FMDI)" for various binaural masker presentation schemes. Signals and maskers were formantlike complex tones, centered around 1500 and 3000 Hz, respectively. Fundamentals of 80 and 240 Hz were used. The signals were presented to the right ear. The maskers were presented either to the right, the left, or to both ears, and they were either unmodulated or modulated at a slow rate (10 Hz). They had the same fundamental as the signals. Hardly any interference was found for the unmodulated maskers. For modulated maskers, the amount of FMDI depended strongly on the binaural masker presentation scheme. Substantial interference was found for the ipsilateral maskers. Interference was smaller for the contralateral maskers. In both cases the FMDI increased with increasing masker level. Substantial interference was also found for the binaural maskers. Imposing different interaural time and level differences (ITDs and ILDs) on maskers and signals did not affect FMDI. The same was true for the ITD condition when the maskers had different fundamentals than the signals, though FMDI was slightly smaller here. The amount of interference for the binaural maskers was roughly equal to that of the corresponding monaural masker with the largest effect. The data could not be described accurately using a model based on the loudness of the maskers. On the other hand, they were well described by a model in which the amount of FMDI was predicted from a "weighted combination" of the monaural masker levels.

Influence of peripheral resolvability on the perceptual segregation of harmonic complex tones differing in fundamental frequency.

Two experiments investigated the influence of resolvability on the perceptual organization of sequential harmonic complexes differing in fundamental frequency (F0). Using a constant-stimuli method, streaming scores for ABA-... sequences of harmonic complexes were measured as a function of the F0 difference between the A and B tones. In the first experiment, streaming scores were measured for harmonic complexes having two different nominal F0s (88 and 250 Hz) and filtered in three frequency regions (a LOW, a MID, and a HIGH region with corner frequencies of 125-625 Hz, 1375-1875 Hz, and 3900-5400 Hz, respectively). Some streaming was observed in the HIGH region (in which the harmonics were always unresolved) but streaming scores remained generally lower than in the LOW and MID regions. The second experiment verified that the streaming observed in the HIGH region was not due to the use of distortion products. Overall, the results indicated that although streaming can occur in the absence of spectral cues, the degree of resolvability of the harmonics has a significant influence.

The effect of modulation rate on the detection of frequency modulation and mistuning of complex tones.

Experiment 1 measured frequency modulation detection thresholds (FMTs) for harmonic complex tones as a function of modulation rate. Six complexes were used, with fundamental frequencies (F0s) of either 88 or 250 Hz, bandpass filtered into a LOW (125-625 Hz), MID (1375-1875 Hz) or HIGH (3900-5400 Hz) frequency region. The FMTs were about an order of magnitude greater for the three complexes whose harmonics were unresolved by the peripheral auditory system (F0 = 88 Hz in the MID region and both F0s in the HIGH region) than for the other three complexes, which contained some resolved harmonics. Thresholds increased with increases in FM rate above 2 Hz for all conditions. The increase was larger when the F0 was 88 Hz than when it was 250 Hz, and was also larger in the LOW than in the MID and HIGH regions. Experiment 2 measured thresholds for detecting mistuning produced by modulating the F0s of two simultaneously presented complexes out of phase by 180 degrees. The size of the resulting mistuning oscillates at a rate equal to the rate of FM applied to the two carriers. At low FM rates, thresholds were lowest when the harmonics were either resolved for both complexes or unresolved for both complexes, and highest when resolvability differed across complexes. For pairs of complexes with resolved harmonics, mistuning thresholds increased dramatically as the FM rate was increased above 2-5 Hz, in a way which could not be accounted for by the effect of modulation rate on the FMTs for the individual complexes. A third experiment, in which listeners detected constant ("static") mistuning between pairs of frequency-modulated complexes, provided evidence that this deterioration was due the harmonics in one of the two "resolved" complexes becoming unresolved at high FM rates, when analyzed over some finite time window. It is concluded that the detection of time-varying mistuning between groups of harmonics is limited by factors that are not apparent in FM detection data.

Detection of small across-channel timing differences by cochlear implantees.

Five post-lingually deafened users of the LAURA cochlear implant were presented with two trains of biphasic pulses applied concurrently to two widely separated channels. They could all discriminate between stimuli where pulses on the two channels were nearly synchronous (inter-channel delay=0.1 ms) and those where there was a longer delay applied to one channel. All showed an asymmetry, being more sensitive when the longer delay was on either the more basal or, depending on the listener, the more apical channel. For four out of the five listeners this asymmetry could be at least partly attributed to one stimulus, with a 0.1-ms delay in either the apical (three listeners) or basal (one listener) channel, sounding markedly different from all other stimuli used in the experiment. Both the overall sensitivity of listeners and the general pattern of results survived the presentation of maskers on intermediate channels, and did not vary markedly with changes in the polarity of the pulses applied to one channel. Although the results varied substantially across listeners, it is concluded that they demonstrate a genuine sensitivity to the relative timing of stimulation applied to discrete populations of auditory nerve fibers.

Detecting coherent and incoherent frequency modulation.

Three experiments investigated whether or not the auditory system contains a neural mechanism that is sensitive to differences in the pattern of frequency modulation imposed on widely separated carriers. Experiment 1 measured the discrimination between an unmodulated two-tone complex and one in which either coherent or incoherent frequency modulation was imposed on the two carrier frequencies. These two frequencies were either 1100+1925 Hz or 1100+2000 Hz, and the stimuli were presented against a pink-noise background. The method was based on that used in experiments by Furukawa and Moore (1996), which were previously interpreted as providing evidence in favour of a mechanism sensitive to FM coherence. Discrimination was sometimes better for coherent than for incoherent FM, as reported by Furukawa and Moore, but only for four out of the eight listeners tested. The remaining experiments excluded those subjects who had shown no effect of FM coherence in experiment 1. Experiment 2 showed that detection of a static shift on the carrier frequencies of the two components was better when the carriers were shifted in the same, compared to the opposite, direction. This difference occurred regardless of whether the carriers were modulated coherently, incoherently, or were unmodulated. The experiment also showed that performance was better when the 1100-Hz carrier was shifted down and the 1925-Hz carrier was shifted up, compared to when the 1100-Hz carrier shifted up and the 1925-Hz carrier shifted down. Experiment 3 showed that this difference also applied to dynamic changes: detection of quasi-linear frequency sweeps (0.5 cycles of sinusoidal FM) was better when the higher component glided up and the lower component glided down than vice versa. In the former condition, performance was as good as with same-direction sweeps. It is concluded that the effects observed in experiment 1 and by Furukawa and Moore result from the processing of a global percept arising from the perceptual fusion of the two carriers, and do not represent an across-frequency mechanism sensitive to FM coherence. In addition, it is argued that experiments 2 and 3 demonstrate the existence of perceptual asymmetries in hearing.

Neglect between but not within auditory objects.

Unilateral neglect is frequently characterized by the presence of extinction, which is a lack of awareness to contralesional visual stimuli in the presence of those further towards the ipsilesional side. It has been established that this visual extinction can be reduced if the stimuli are grouped together into a single object. However, attention between and within auditory objects has never before been studied. We demonstrate for the first time that unilateral neglect--hitherto thought primarily to be a disorder of visuospatial processing-- involves a specific deficit in allocating attention between auditory objects separated only in time and not in space. Importantly, this deficit is restricted to comparisons between sounds: the patients' ability to make within-sound comparisons is similar to that of controls. These differences cannot be explained in terms of different time spans over which comparisons must be made. The results suggest unilateral neglect is linked to--if not actually determined by--a reduction in attentional capacity in both the visual and auditory domains, and across the dimensions of both space and time. The findings have potential clinical applications.

Auditory temporal processing impairment: neither necessary nor sufficient for causing language impairment in children.

Fourteen twin pairs, aged 8 to 10 years, were tested 3 times over 12 months; they included 11 children with language impairment (LI), 11 control children matched on nonverbal ability and age, and 6 co-twins who did not meet criteria for LI or control status. Thresholds were estimated for detecting a brief backward-masked tone (BM), detection of frequency modulation (FM), and pitch discrimination using temporal cues (deltaf0). Both BM and FM thresholds improved with training, and by the 2nd test session, FM thresholds were in the adult range. There were marked individual differences on BM and deltaf0 and, for both tasks, performance correlated with Tallal's Auditory Repetition Task administered 2 years previously. However, no auditory measure gave significant differences between LI and control groups; performance was influenced more by nonverbal than language ability. Some children did have a stable pattern of poor performance on certain auditory tasks, but their good FM detection raised questions about whether processing of auditory temporal information is abnormal. We found no evidence that auditory deficits are a necessary or sufficient cause of language impairments.

Center frequency modulation detection for harmonic complexes resembling vowel formants and its interference by off-frequency maskers.

Vowels are characterized by peaks in their spectral envelopes: the formants. To gain insight into the perception of speech as well as into the basic abilities of the ear, sensitivity to modulations in the positions of these formants is investigated. Frequency modulation detection thresholds (FMTs) were measured for the center frequency of formantlike harmonic complexes in the absence and in the presence of simultaneous off-frequency formants (maskers). Both the signals and the maskers were harmonic complexes which were band-pass filtered with a triangular spectral envelope, on a log-log scale, into either a LOW (near 500 Hz), a MID (near 1500 Hz), or a HIGH region (near 3000 Hz). They had a duration of 250 ms, and either an 80- or a 240-Hz fundamental. The modulation rate was 5 Hz for the signals and 10 Hz for the maskers. A pink noise background was presented continuously. In a first experiment no maskers were used. The measured FMTs were roughly two times larger than previously reported just-noticeable differences for formant frequency. In a second experiment, no significant differences were found between the FMTs in the absence of maskers and those in the presence of stationary (i.e., nonfrequency modulated) maskers. However, under many conditions the FMTs were increased by the presence of simultaneous modulated maskers. These results indicate that frequency modulation detection interference (FMDI) can exist for formantlike complex tones. The FMDI data could be divided into two groups. For stimuli characterized by a steep (200-dB/oct) slope, it was found that the size of the FMDI depended on which cues were used for detecting the signal and masker modulations. For stimuli with shallow (50-dB/oct) slopes, the FMDI was reduced when the signal and the masker had widely differing fundamentals, implying that the fundamental information is extracted before the interference occurs.