Differences between psychoacoustic and frequency following response measures of distortion tone level and masking

A high-density EEG and structural MRI source analysis of the frequency following response to missing fundamental stimuli reveals subcortical and cortical activation to low and high frequency stimuli

Pitch is a perceptual rather than physical phenomenon, important for spoken language use, musical communication, and other aspects of everyday life. Auditory stimuli can be designed to probe the relationship between perception and physiological responses to pitch-evoking stimuli. One technique for measuring physiological responses to pitch-evoking stimuli is the frequency following response (FFR). The FFR is an electroencephalographic (EEG) response to periodic auditory stimuli. The FFR contains nonlinearities not present in the stimuli, including correlates of the amplitude envelope of the stimulus; however, these nonlinearities remain undercharacterized. The FFR is a composite response reflecting multiple neural and peripheral generators, and their contributions to the scalp-recorded FFR vary in ill-understood ways depending on the electrode montage, stimulus, and imaging technique. The FFR is typically assumed to be generated in the auditory brainstem; there is also evidence both for and against a cortical contribution to the FFR. Here a methodology is used to examine the FFR correlates of pitch and the generators of the FFR to stimuli with different pitches. Stimuli were designed to tease apart biological correlates of pitch and amplitude envelope. FFRs were recorded with 256-electrode EEG nets, in contrast to a typical FFR setup which only contains a single active electrode. Structural MRI scans were obtained for each participant to co-register with the electrode locations and constrain a source localization algorithm. The results of this localization shed light on the generating mechanisms of the FFR, including providing evidence for both cortical and subcortical auditory sources.

Electrophysiological and Psychophysical Measures of Temporal Pitch Sensitivity in Normal-hearing Listeners

To obtain combined behavioural and electrophysiological measures of pitch perception, we presented harmonic complexes, bandpass filtered to contain only high-numbered harmonics, to normal-hearing listeners. These stimuli resemble bandlimited pulse trains and convey pitch using a purely temporal code. A core set of conditions consisted of six stimuli with baseline pulse rates of 94, 188 and 280 pps, filtered into a HIGH (3365-4755 Hz) or VHIGH (7800-10,800 Hz) region, alternating with a 36% higher pulse rate. Brainstem and cortical processing were measured using the frequency following response (FFR) and auditory change complex (ACC), respectively. Behavioural rate change difference limens (DLs) were measured by requiring participants to discriminate between a stimulus that changed rate twice (up-down or down-up) during its 750-ms presentation from a constant-rate pulse train. FFRs revealed robust brainstem phase locking whose amplitude decreased with increasing rate. Moderate-sized but reliable ACCs were obtained in response to changes in purely temporal pitch and, like the psychophysical DLs, did not depend consistently on the direction of rate change or on the pulse rate for baseline rates between 94 and 280 pps. ACCs were larger and DLs lower for stimuli in the HIGH than in the VHGH region. We argue that the ACC may be a useful surrogate for behavioural measures of rate discrimination, both for normal-hearing listeners and for cochlear-implant users. We also showed that rate DLs increased markedly when the baseline rate was reduced to 48 pps, and compared the behavioural and electrophysiological findings to recent cat data obtained with similar stimuli and methods.

Using Interleaved Stimulation to Measure the Size and Selectivity of the Sustained Phase-Locked Neural Response to Cochlear Implant Stimulation

We measured the sustained neural response to electrical stimulation by a cochlear implant (CI). To do so, we interleaved two stimuli with frequencies F1 and F2 Hz and recorded a neural distortion response (NDR) at F2-F1 Hz. We show that, because any one time point contains only the F1 or F2 stimulus, the instantaneous nonlinearities typical of electrical artefact should not produce distortion at this frequency. However, if the stimulus is smoothed, such as by charge integration at the nerve membrane, subsequent (neural) nonlinearities can produce a component at F2-F1 Hz. We stimulated a single CI electrode with interleaved sinusoids or interleaved amplitude-modulated pulse trains such that F2 = 1.5F1, and found no evidence for an NDR when F2-F1 was between 90 and 120 Hz. However, interleaved amplitude-modulated pulse trains with F2-F1~40 Hz revealed a substantial NDR with a group delay of about 45 ms, consistent with a thalamic and/or cortical response. The NDR could be measured even from recording electrodes adjacent to the implant and at the highest pulse rates (> 4000 pps) used clinically. We then measured the selectivity of this sustained response by presenting F1 and F2 to different electrodes and at different between-electrode distances. This revealed a broad tuning that, we argue, reflects the overlap between the excitation elicited by the two electrodes. Our results also provide a glimpse of the neural nonlinearity in the auditory system, unaffected by the biomechanical cochlear nonlinearities that accompany acoustic stimulation. Several potential clinical applications of our findings are discussed.

Dichotic phase effects on frequency following responses reveal phase variant and invariant harmonic distortion products

The dichotic frequency following responses (FFR) have been used in studies to infer about dichotic auditory processing. In the present study, we hypothesize that the proximity of the binaural neural generators of the FFR would result in interference of the volume-conducted electrical fields. This might lead to contamination of the scalp-recorded dichotic FFRs due to which it might be difficult to infer about true dichotic processing in the putative neural generators. We investigated this by recording FFRs to binaurally presented 200 Hz pure tone with graded dichotic phase offsets (0°, 90°, 180° and 270°) in normal hearing young adults. Spectral analysis of the FFRs was performed for the estimation of the magnitude and phase at the component frequencies. FFR spectra were compared using non-parametric paired randomizations within the subjects. We found that the brainstem responses to a 200 Hz pure tone consisted of prominent peaks at 200 Hz, and at frequencies corresponding to the harmonics of 200 Hz. The FFR spectral magnitude at 200 Hz diminished with a phase offset of 180°. Phase offsets of 90° and 270° showed reduced spectral magnitudes at 200 Hz than those in the 0° condition. Our findings, in line with the hypothesis, show that the dichotic FFRs do not reflect true dichotic processing and that they are contaminated during volume conduction. Additionally, we found harmonic distortion products (HDP) in the FFRs. We found that the response at 200 Hz and the 3rd HDP systematically varied with a change in phase of the stimulus, while the even HDPs (2nd and 4th) were phase-invariant. Based on our findings, and modeling FFRs using auditory models, we propose a rectification process as the contributors for the generation of HDPs. We also discuss the implications of this HDP generating mechanism in understanding the pitch represented in FFRs.

Specificity of the Human Frequency Following Response for Carrier and Modulation Frequency Assessed Using Adaptation

The frequency following response (FFR) is a scalp-recorded measure of phase-locked brainstem activity to stimulus-related periodicities. Three experiments investigated the specificity of the FFR for carrier and modulation frequency using adaptation. FFR waveforms evoked by alternating-polarity stimuli were averaged for each polarity and added, to enhance envelope, or subtracted, to enhance temporal fine structure information. The first experiment investigated peristimulus adaptation of the FFR for pure and complex tones as a function of stimulus frequency and fundamental frequency (F0). It showed more adaptation of the FFR in response to sounds with higher frequencies or F0s than to sounds with lower frequency or F0s. The second experiment investigated tuning to modulation rate in the FFR. The FFR to a complex tone with a modulation rate of 213 Hz was not reduced more by an adaptor that had the same modulation rate than by an adaptor with a different modulation rate (90 or 504 Hz), thus providing no evidence that the FFR originates mainly from neurons that respond selectively to the modulation rate of the stimulus. The third experiment investigated tuning to audio frequency in the FFR using pure tones. An adaptor that had the same frequency as the target (213 or 504 Hz) did not generally reduce the FFR to the target more than an adaptor that differed in frequency (by 1.24 octaves). Thus, there was no evidence that the FFR originated mainly from neurons tuned to the frequency of the target. Instead, the results are consistent with the suggestion that the FFR for low-frequency pure tones at medium to high levels mainly originates from neurons tuned to higher frequencies. Implications for the use and interpretation of the FFR are discussed.

Losing the music: aging affects the perception and subcortical neural representation of musical harmony

When two musical notes with simple frequency ratios are played simultaneously, the resulting musical chord is pleasing and evokes a sense of resolution or "consonance". Complex frequency ratios, on the other hand, evoke feelings of tension or "dissonance". Consonance and dissonance form the basis of harmony, a central component of Western music. In earlier work, we provided evidence that consonance perception is based on neural temporal coding in the brainstem (Bones et al., 2014). Here, we show that for listeners with clinically normal hearing, aging is associated with a decline in both the perceptual distinction and the distinctiveness of the neural representations of different categories of two-note chords. Compared with younger listeners, older listeners rated consonant chords as less pleasant and dissonant chords as more pleasant. Older listeners also had less distinct neural representations of consonant and dissonant chords as measured using a Neural Consonance Index derived from the electrophysiological "frequency-following response." The results withstood a control for the effect of age on general affect, suggesting that different mechanisms are responsible for the perceived pleasantness of musical chords and affective voices and that, for listeners with clinically normal hearing, age-related differences in consonance perception are likely to be related to differences in neural temporal coding.

Subcortical representation of musical dyads: individual differences and neural generators

When two notes are played simultaneously they form a musical dyad. The sensation of pleasantness, or "consonance", of a dyad is likely driven by the harmonic relation of the frequency components of the combined spectrum of the two notes. Previous work has demonstrated a relation between individual preference for consonant over dissonant dyads, and the strength of neural temporal coding of the harmonicity of consonant relative to dissonant dyads as measured using the electrophysiological "frequency-following response" (FFR). However, this work also demonstrated that both these variables correlate strongly with musical experience. The current study was designed to determine whether the relation between consonance preference and neural temporal coding is maintained when controlling for musical experience. The results demonstrate that strength of neural coding of harmonicity is predictive of individual preference for consonance even for non-musicians. An additional purpose of the current study was to assess the cochlear generation site of the FFR to low-frequency dyads. By comparing the reduction in FFR strength when high-pass masking noise was added to the output of a model of the auditory periphery, the results provide evidence for the FFR to low-frequency dyads resulting in part from basal cochlear generators.

Phase locked neural activity in the human brainstem predicts preference for musical consonance

When musical notes are combined to make a chord, the closeness of fit of the combined spectrum to a single harmonic series (the 'harmonicity' of the chord) predicts the perceived consonance (how pleasant and stable the chord sounds; McDermott, Lehr, & Oxenham, 2010). The distinction between consonance and dissonance is central to Western musical form. Harmonicity is represented in the temporal firing patterns of populations of brainstem neurons. The current study investigates the role of brainstem temporal coding of harmonicity in the perception of consonance. Individual preference for consonant over dissonant chords was measured using a rating scale for pairs of simultaneous notes. In order to investigate the effects of cochlear interactions, notes were presented in two ways: both notes to both ears or each note to different ears. The electrophysiological frequency following response (FFR), reflecting sustained neural activity in the brainstem synchronised to the stimulus, was also measured. When both notes were presented to both ears the perceptual distinction between consonant and dissonant chords was stronger than when the notes were presented to different ears. In the condition in which both notes were presented to the both ears additional low-frequency components, corresponding to difference tones resulting from nonlinear cochlear processing, were observable in the FFR effectively enhancing the neural harmonicity of consonant chords but not dissonant chords. Suppressing the cochlear envelope component of the FFR also suppressed the additional frequency components. This suggests that, in the case of consonant chords, difference tones generated by interactions between notes in the cochlea enhance the perception of consonance. Furthermore, individuals with a greater distinction between consonant and dissonant chords in the FFR to individual harmonics had a stronger preference for consonant over dissonant chords. Overall, the results provide compelling evidence for the role of neural temporal coding in the perception of consonance, and suggest that the representation of harmonicity in phase locked neural firing drives the perception of consonance.

Cochlear neuropathy and the coding of supra-threshold sound

Many listeners with hearing thresholds within the clinically normal range nonetheless complain of difficulty hearing in everyday settings and understanding speech in noise. Converging evidence from human and animal studies points to one potential source of such difficulties: differences in the fidelity with which supra-threshold sound is encoded in the early portions of the auditory pathway. Measures of auditory subcortical steady-state responses (SSSRs) in humans and animals support the idea that the temporal precision of the early auditory representation can be poor even when hearing thresholds are normal. In humans with normal hearing thresholds (NHTs), paradigms that require listeners to make use of the detailed spectro-temporal structure of supra-threshold sound, such as selective attention and discrimination of frequency modulation (FM), reveal individual differences that correlate with subcortical temporal coding precision. Animal studies show that noise exposure and aging can cause a loss of a large percentage of auditory nerve fibers (ANFs) without any significant change in measured audiograms. Here, we argue that cochlear neuropathy may reduce encoding precision of supra-threshold sound, and that this manifests both behaviorally and in SSSRs in humans. Furthermore, recent studies suggest that noise-induced neuropathy may be selective for higher-threshold, lower-spontaneous-rate nerve fibers. Based on our hypothesis, we suggest some approaches that may yield particularly sensitive, objective measures of supra-threshold coding deficits that arise due to neuropathy. Finally, we comment on the potential clinical significance of these ideas and identify areas for future investigation.

Pitch coding and pitch processing in the human brain

Neuroimaging studies have provided important information regarding how and where pitch is coded and processed in the human brain. Recordings of the frequency-following response (FFR), an electrophysiological measure of neural temporal coding in the brainstem, have shown that the precision of temporal pitch information is dependent on linguistic and musical experience, and can even be modified by short-term training. However, the FFR does not seem to represent the output of a pitch extraction process, and this raises questions regarding how the peripheral neural signal is processed to produce a unified sensation. Since stimuli with a wide variety of spectral and binaural characteristics can produce the same pitch, it has been suggested that there is a place in the ascending auditory pathway at which the representations converge. There is evidence from many different human neuroimaging studies that certain areas of auditory cortex are specifically sensitive to pitch, although the location is still a matter of debate. Taken together, the results suggest that the initial temporal pitch code in the auditory periphery is converted to a code based on neural firing rate in the brainstem. In the upper brainstem or auditory cortex, the information from the individual harmonics of complex tones is combined to form a general representation of pitch. This article is part of a Special Issue entitled Human Auditory Neuroimaging.