FORUM: Remote testing for psychological and physiological acoustics.

Acoustics research involving human participants typically takes place in specialized laboratory settings. Listening studies, for example, may present controlled sounds using calibrated transducers in sound-attenuating or anechoic chambers. In contrast, remote testing takes place outside of the laboratory in everyday settings (e.g., participants' homes). Remote testing could provide greater access to participants, larger sample sizes, and opportunities to characterize performance in typical listening environments at the cost of reduced control of environmental conditions, less precise calibration, and inconsistency in attentional state and/or response behaviors from relatively smaller sample sizes and unintuitive experimental tasks. The Acoustical Society of America Technical Committee on Psychological and Physiological Acoustics launched the Task Force on Remote Testing (https://tcppasa.org/remotetesting/) in May 2020 with goals of surveying approaches and platforms available to support remote testing and identifying challenges and considerations for prospective investigators. The results of this task force survey were made available online in the form of a set of Wiki pages and summarized in this report. This report outlines the state-of-the-art of remote testing in auditory-related research as of August 2021, which is based on the Wiki and a literature search of papers published in this area since 2020, and provides three case studies to demonstrate feasibility during practice.

Cross-frequency weights in normal and impaired hearing: Stimulus factors, stimulus dimensions, and associations with speech recognition.

Previous studies of level discrimination reported that listeners with high-frequency sensorineural hearing loss (SNHL) place greater weight on high frequencies than normal-hearing (NH) listeners. It is not clear whether these results are influenced by stimulus factors (e.g., group differences in presentation levels, cross-frequency discriminability of level differences used to measure weights) and whether such weights generalize to other tasks. Here, NH and SNHL weights were measured for level, duration, and frequency discrimination of two-tone complexes after measuring discriminability just-noticeable differences for each frequency and stimulus dimension. Stimuli were presented at equal sensation level (SL) or equal sound pressure level (SPL). Results showed that weights could change depending on which frequency contained the more discriminable level difference with uncontrolled cross-frequency discriminability. When cross-frequency discriminability was controlled, weights were consistent for level and duration discrimination, but not for frequency discrimination. Comparing equal SL and equal SPL weights indicated greater weight on the higher-level tone for level and duration discrimination. Weights were unrelated to improvements in recognition of low-pass-filtered speech with increasing cutoff frequency. These results suggest that cross-frequency weights and NH and SNHL weighting differences are influenced by stimulus factors and may not generalize to the use of speech cues in specific frequency regions.

Sub-optimal construction of an auditory profile from temporally distributed spectral information.

When spectral components of a complex sound are presented not simultaneously but distributed over time, human listeners can still, to a degree, perceptually recover the spectral profile of the sound. This capability of integrating spectral information over time was investigated using a cued informational masking paradigm. Listeners detected a 1-kHz pure tone in a simultaneous masker composed of six random-frequency tones drawn on every trial. The spectral profile of the masker was cued using a precursor sound that consisted of a sequence of 50-ms bursts, separated by inter-burst intervals of 100 ms. Each burst in the precursor consisted of pure tones at the masker frequencies with tones appearing at each of the masker frequencies at different presentation probabilities. As the presentation probability increased in different conditions, the detectability of the target improved, indicating reliable precursor cuing regarding the spectral content of the masker. For many listeners, performance did not significantly improve as the number of precursor bursts increased from 2 to 16, indicating inefficient integration of information beyond 2 bursts. Additional analyses suggest that when intensity of the bursts is relatively constant, the contribution of the precursor is dominated by information in the initial burst.

Neural fluctuation cues for simultaneous notched-noise masking and profile-analysis tasks: Insights from model midbrain responses.

Results of simultaneous notched-noise masking are commonly interpreted as reflecting the bandwidth of underlying auditory filters. This interpretation assumes that listeners detect a tone added to notched-noise based on an increase in energy at the output of an auditory filter. Previous work challenged this assumption by showing that randomly and independently varying (roving) the levels of each stimulus interval does not substantially worsen listener thresholds [Lentz, Richards, and Matiasek (1999). J. Acoust. Soc. Am. 106, 2779-2792]. Lentz et al. further challenged this assumption by showing that filter bandwidths based on notched-noise results were different from those based on a profile-analysis task [Green (1983). Am. Psychol. 38, 133-142; (1988). (Oxford University Press, New York)], although these estimates were later reconciled by emphasizing spectral peaks of the profile-analysis stimulus [Lentz (2006). J. Acoust. Soc. Am. 120, 945-956]. Here, a single physiological model is shown to account for performance in fixed- and roving-level notched-noise tasks and the Lentz et al. profile-analysis task. This model depends on peripheral neural fluctuation cues that are transformed into the average rates of model inferior colliculus neurons. Neural fluctuations are influenced by peripheral filters, synaptic adaptation, cochlear amplification, and saturation of inner hair cells, an element not included in previous theories of envelope-based cues for these tasks. Results suggest reevaluation of the interpretation of performance in these paradigms.

Hierarchical organization of melodic sequences is encoded by cortical entrainment.

Natural speech is organized according to a hierarchical structure, with individual speech sounds combining to form abstract linguistic units, and abstract linguistic units combining to form higher-order linguistic units. Since the boundaries between these units are not always indicated by acoustic cues, they must often be computed internally. Signatures of this internal computation were reported by Ding et al. (2016), who presented isochronous sequences of mono-syllabic words that combined to form phrases that combined to form sentences, and showed that cortical responses simultaneously encode boundaries at multiple levels of the linguistic hierarchy. In the present study, we designed melodic sequences that were hierarchically organized according to Western music conventions. Specifically, isochronous sequences of "sung" nonsense syllables were constructed such that syllables combined to form triads outlining individual chords, which combined to form harmonic progressions. EEG recordings were made while participants listened to these sequences with the instruction to detect when violations in the sequence structure occurred. We show that cortical responses simultaneously encode boundaries at multiple levels of a melodic hierarchy, suggesting that the encoding of hierarchical structure is not unique to speech. No effect of musical training on cortical encoding was observed.

Hierarchy of speech-driven spectrotemporal receptive fields in human auditory cortex.

Existing data indicate that cortical speech processing is hierarchically organized. Numerous studies have shown that early auditory areas encode fine acoustic details while later areas encode abstracted speech patterns. However, it remains unclear precisely what speech information is encoded across these hierarchical levels. Estimation of speech-driven spectrotemporal receptive fields (STRFs) provides a means to explore cortical speech processing in terms of acoustic or linguistic information associated with characteristic spectrotemporal patterns. Here, we estimate STRFs from cortical responses to continuous speech in fMRI. Using a novel approach based on filtering randomly-selected spectrotemporal modulations (STMs) from aurally-presented sentences, STRFs were estimated for a group of listeners and categorized using a data-driven clustering algorithm. 'Behavioral STRFs' highlighting STMs crucial for speech recognition were derived from intelligibility judgments. Clustering revealed that STRFs in the supratemporal plane represented a broad range of STMs, while STRFs in the lateral temporal lobe represented circumscribed STM patterns important to intelligibility. Detailed analysis recovered a bilateral organization with posterior-lateral regions preferentially processing STMs associated with phonological information and anterior-lateral regions preferentially processing STMs associated with word- and phrase-level information. Regions in lateral Heschl's gyrus preferentially processed STMs associated with vocalic information (pitch).

Potential cues for the "level discrimination" of a noise band in the presence of flanking bands.

To evaluate the ability of a restricted range of auditory-nerve fibers to encode a large perceptual dynamic range, Viemeister [(1983). Science 221, 1206-1208] examined the detection of a change in the level of a high-frequency band of noise flanked by more intense fixed-level noise maskers. Here, stimuli and procedures similar to Viemeister's were used, but random manipulations of level and notch cutoff frequency were included to evaluate predictions of energy-based models. The results indicate that cues other than the change in level per se are available, and suggest the potential contribution of changes in pitch/timbre for this task.

The effect of prior knowledge and intelligibility on the cortical entrainment response to speech.

It has been suggested that cortical entrainment plays an important role in speech perception by helping to parse the acoustic stimulus into discrete linguistic units. However, the question of whether the entrainment response to speech depends on the intelligibility of the stimulus remains open. Studies addressing this question of intelligibility have, for the most part, significantly distorted the acoustic properties of the stimulus to degrade the intelligibility of the speech stimulus, making it difficult to compare across "intelligible" and "unintelligible" conditions. To avoid these acoustic confounds, we used priming to manipulate the intelligibility of vocoded speech. We used EEG to measure the entrainment response to vocoded target sentences that are preceded by natural speech (nonvocoded) prime sentences that are either valid (match the target) or invalid (do not match the target). For unintelligible speech, valid primes have the effect of restoring intelligibility. We compared the effect of priming on the entrainment response for both 3-channel (unintelligible) and 16-channel (intelligible) speech. We observed a main effect of priming, suggesting that the entrainment response depends on prior knowledge, but not a main effect of vocoding (16 channels vs. 3 channels). Furthermore, we found no difference in the effect of priming on the entrainment response to 3-channel and 16-channel vocoded speech, suggesting that for vocoded speech, entrainment response does not depend on intelligibility.NEW & NOTEWORTHY Neural oscillations have been implicated in the parsing of speech into discrete, hierarchically organized units. Our data suggest that these oscillations track the acoustic envelope rather than more abstract linguistic properties of the speech stimulus. Our data also suggest that prior experience with the stimulus allows these oscillations to better track the stimulus envelope.

Auditory "bubbles": Efficient classification of the spectrotemporal modulations essential for speech intelligibility.

Speech intelligibility depends on the integrity of spectrotemporal patterns in the signal. The current study is concerned with the speech modulation power spectrum (MPS), which is a two-dimensional representation of energy at different combinations of temporal and spectral (i.e., spectrotemporal) modulation rates. A psychophysical procedure was developed to identify the regions of the MPS that contribute to successful reception of auditory sentences. The procedure, based on the two-dimensional image classification technique known as "bubbles" (Gosselin and Schyns (2001). Vision Res. 41, 2261-2271), involves filtering (i.e., degrading) the speech signal by removing parts of the MPS at random, and relating filter patterns to observer performance (keywords identified) over a number of trials. The result is a classification image (CImg) or "perceptual map" that emphasizes regions of the MPS essential for speech intelligibility. This procedure was tested using normal-rate and 2×-time-compressed sentences. The results indicated: (a) CImgs could be reliably estimated in individual listeners in relatively few trials, (b) CImgs tracked changes in spectrotemporal modulation energy induced by time compression, though not completely, indicating that "perceptual maps" deviated from physical stimulus energy, and

Psychometric functions for sentence recognition in sinusoidally amplitude-modulated noises.

Listeners' speech reception is better when speech is masked by a modulated masker compared to an unmodulated masker with the same long-term root-mean-square level. It has been suggested that listeners take advantage of brief periods of quiescence in a modulated masker to extract speech information. Two experiments examined the contribution of such "dip-listening" models. The first experiment estimated psychometric functions for speech intelligibility using sentences masked by sinusoidally modulated and unmodulated speech-shaped noises and the second experiment estimated detection thresholds for a tone pip added at the central dip in the masker. Modulation rates ranging from 1 to 64 Hz were tested. In experiment 1 the slopes of the psychometric functions were shallower for lower modulation rates and the pattern of speech reception thresholds as a function of modulation rate was nonmonotonic with a minimum near 16 Hz. In contrast, the detection thresholds from experiment 2 increased monotonically with modulation rate. The results suggest that the benefits of listening to speech in temporally fluctuating maskers cannot be solely ascribed to the temporal acuity of the auditory system.

A MATLAB toolbox for the efficient estimation of the psychometric function using the updated maximum-likelihood adaptive procedure.

A MATLAB toolbox for the efficient estimation of the threshold, slope, and lapse rate of the psychometric function is described. The toolbox enables the efficient implementation of the updated maximum-likelihood (UML) procedure. The toolbox uses an object-oriented architecture for organizing the experimental variables and computational algorithms, which provides experimenters with flexibility in experimental design and data management. Descriptions of the UML procedure and the UML Toolbox are provided, followed by toolbox use examples. Finally, guidelines and recommendations of parameter configurations are given.

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.

Temporal modulation transfer function for efficient assessment of auditory temporal resolution.

Two common measures of auditory temporal resolution are the temporal modulation transfer function (TMTF) and the gap detection threshold (GDT). The current study addresses the lack of efficient psychophysical procedures for collecting TMTFs and the lack of literature on the comparisons of TMTF and GDT. Two procedures for efficient measurements of the TMTF are proposed: (1) A Bayesian procedure that adaptively chooses the stimulus modulation rate and depth to maximize the information gain from each trial and (2) a procedure that reduces the data collection to two adaptive staircase tracks. Results from experiments I and II showed that, for broadband carriers, these approaches provided similar results compared to TMTFs measured using traditional methods despite taking less than 10 min for data collection. Using these efficient procedures, TMTFs were measured from a large number of naive listeners and were compared to the gap detection thresholds collected from the same ears in experiment III. Results showed that the sensitivity parameter estimated from the TMTF measurements correlated well with the GDTs, whereas the cutoff rate is either uncorrelated or positively correlated with the gap detection threshold. These results suggest caution in interpreting a lower GDT as evidence for less sluggish temporal processing.

Bayesian adaptive estimation of the auditory filter.

A Bayesian adaptive procedure for estimating the auditory-filter shape was proposed and evaluated using young, normal-hearing listeners at moderate stimulus levels. The resulting quick-auditory-filter (qAF) procedure assumed the power spectrum model of masking with the auditory-filter shape being modeled using a spectrally symmetric, two-parameter rounded-exponential (roex) function. During data collection using the qAF procedure, listeners detected the presence of a pure-tone signal presented in the spectral notch of a noise masker. Dependent on the listener's response on each trial, the posterior probability distributions of the model parameters were updated, and the resulting parameter estimates were then used to optimize the choice of stimulus parameters for the subsequent trials. Results showed that the qAF procedure gave similar parameter estimates to the traditional threshold-based procedure in many cases and was able to reasonably predict the masked signal thresholds. Additional measurements suggested that occasional failures of the qAF procedure to reliably converge could be a consequence of incorrect responses early in a qAF track. The addition of a parameter describing lapses of attention reduced the likelihood of such failures.

Level dominance for the detection of changes in level distribution in sound streams.

Sound streams were generated by randomly choosing the levels of tone pips from two different distributions, A and B. Of the 18 tone pips, the first nine were drawn from distribution A and the second nine from distribution B, or the opposite. The listeners' task was to indicate order, A-B or B-A. In two conditions the A and B distributions differed in mean (condition 1) or variance (condition 2). In contrast to an ideal observer, listeners' strategies were consistent across the two conditions. Analyses suggest that listeners relied primarily on the more intense tone pips in making their decisions.

Effects of sound segregation cues on multi-sound intensity discrimination.

The effects of sound segregation cues on the sensitivity to intensity increments were explored. Listeners indicated whether the second and fourth sounds (harmonic complexes) within a five-sound sequence were increased in intensity. The target sound had a fundamental frequency of 250 Hz. In different conditions, nontarget sounds had different fundamental frequencies, different spectral shapes, and unique frequency regions relative to the target. For targets more intense than nontargets, nontarget characteristics did not affect thresholds. For targets less intense than the nontargets, thresholds improved when the targets and nontargets had unique frequency regions.

Investigating the auditory enhancement phenomenon using behavioral temporal masking patterns.

A narrowband signal is subjected to less masking from a simultaneously presented notched masker if it is preceded by a precursor that occupies the same spectral region as the masker, a phenomenon referred to as enhancement. The present study investigated (i) the amount of enhancement for the detection of a narrowband noise added to a notched masker, and (ii) masking patterns associated with the detection of tone pips added to the narrowband signal. The resulting psychophysical data were compared to predictions generated using a model similar to the neural adaptation-of-inhibition model proposed by Nelson and Young [(2010b). J. Neurosci. 30, 6577-6587]. The amount of enhancement was measured as a function of the temporal separation between the precursor and masker in Experiment I, and as a function of precursor level in Experiment II. The model captured the temporal dynamics of psychophysical enhancement reasonably well for both the long-duration noise signals and the masking patterns. However, in contrast to psychophysical data which indicated reliable enhancement only when the precursor and masker shared the same levels, the model predicated enhancement at all precursor levels.

Spectral processing of two concurrent harmonic complexes.

In a concurrent profile analysis task, each of the two observation intervals was the sum of two harmonic complexes. In the first interval one of the harmonic complexes had a flat spectrum and the other had a broad spectral peak at 1 kHz. In the second interval, the association between the spectral profiles and the complexes was either consistent with the first interval, or inconsistent so that profile changes (flat versus peaked) could be created in both of the complexes. In two experiments, thresholds and psychometric functions for detecting the profile change were measured in terms of the spectral peak's magnitude as functions of three types of segregation cues: Difference in fundamental frequency, onset asynchrony, and difference in interaural time difference between the two complexes. Decreasing the magnitude of each cue led to higher thresholds, and shallower psychometric functions whose upper asymptotes often failed to reach 100% correct. The patterns of the threshold and psychometric functions varied across cue types and across individual listeners. The results suggest that informational masking is present in the concurrent profile analysis task. Segregation cues appear to contribute to the release from informational masking, but the process depends on listening strategies adopted by individual listeners.

A maximum-likelihood procedure for estimating psychometric functions: thresholds, slopes, and lapses of attention.

Green [J. Acoust. Soc. Am. 87, 2662-2674 (1990)] suggested an efficient, maximum-likelihood-based approach for adaptively estimating thresholds. Such procedures determine the signal strength on each trial by first identifying the most likely psychometric functions among the pre-proposed alternatives based on responses from previous trials, and then finding the signal strength at the "sweet point" on that most likely function. The sweet point is the point on the psychometric function that is associated with the minimum expected variance. Here, that procedure is extended to reduce poor estimates that result from lapses in attention. The sweet points for the threshold, slope, and lapse parameters of a transformed logistic psychometric function are derived. In addition, alternative stimulus placement algorithms are considered. The result is a relatively fast and robust estimation of a three-parameter psychometric function.

Toward an objective measure for a "stream segregation" task.

A procedure to estimate the relative contribution of "A" and "B" tones for a stream-segregation task is described. Listeners detected a delay in the penultimate A tone in an A-B-A-B sequence of tones. For small A-B frequency separations, for most listeners, classification models based on both the A and B tones were superior to models based on just the A tones. For large frequency separations, models based on just the A tones were superior, indicating the A and B tones were segregated. The results also revealed individual differences in the strategies adopted to complete the task.