Loudness Judgments Research Articles

Induced loudness reduction

Induced loudness reduction (ILR) is a phenomenon by which a preceding higher-level tone (an inducer tone) reduces the loudness of a lower-level tone (a test tone). Under certain conditions, ILR can result in loudness reductions of 10 to 15 phons for pure tones. The strength of this effect depends on a number of parameters including: (1) the levels of both the inducer and test tones; (2) the frequency separation between the inducer and test tones; (3) the durations of the inducer and test tones; (4) the time separation between the inducer and test tones; (5) individual differences; and, possibly (6) the number of exposures to inducers. Because of the sensitivity of ILR to procedural conditions, it is quite important to consider the potential effects of ILR when considering any experimental design in which level varies. The understanding of ILR has also given insight into a number of unexplained discrepancies between data sets that were collected using different procedures. In fact, some of the variability known to affect loudness judgments may be due to ILR. [Work supported by NIH/NIDCD Grant R01DC02241.]

Miniseminar: Vocal Fold Medialization: Injection Versus Thyroplasty?

Medialization laryngoplasty restores voice in patients with unilateral laryngeal paralysis. Of question was whether patients with vocal fold paralysis resulting from cancer or its treatment had as good a post-operative voice result as patients with vocal fold paralysis of benign etiology. The purpose of the present study was to compare post-operative perceptual, acoustic, aerodynamic, and quality of life data in these two patient groups. Twenty-eight patients with vocal fold paralysis secondary to malignancy or its treatment were age and gender-matched with patients with paralysis resulting from benign origin. Pre- and post-operative perceptual judgments of pitch, loudness and quality were rated independently by two speech–language pathologists. A digital audiotape of the patient's voice was analyzed using Soundscope software. Fundamental frequency, conversational intensity and perturbation were evaluated. Glottal flow rates in propositional speech, phonation times and extent of pitch and loudness ranges were also measured. Three quality of life surveys, the Short Form-36 general health survey, the Voice Handicap Index, and the Voice Outcomes Study were administered. Results of voice testing indicated that perceptual, acoustic and aerodynamic data were significantly improved 3 months after thyroplasty in all patients regardless of whether they had a history of cancer. Quality of life data, however, distinguished the two groups. In particular, the general health measure found a significant difference in physical functioning and overall vitality, although satisfaction with improved voice was equally appreciated in both patient groups. Of clinical significance is that though general health may differ, patients with cancer-related laryngeal paralysis can expect to have as good a voice outcome following thyroplasty as patients with paralysis of benign etiology.

Temporal and spectral interaction in loudness perception

An experiment was conducted to investigate how changes in spectral content influence loudness judgments. Six listeners were asked to discriminate sounds, which were of one second duration and changing in level every 0.1 s. In one condition the first half of the sound was low-pass filtered and the second half high-pass filtered. In a second condition the opposite order was used. In a third condition no filtering was applied and the frequency spectrum was simply white noise. The results were analyzed using a statistical method, which assigns relative weights to the ten temporal segments. In this way individual weighting curves were obtained for each condition. Listeners tended to emphasize the beginning of the sound in their loudness judgments. When the frequency spectrum changed in the middle of the sound, however, the weighting of the onset of the new spectral content was emphasized as well. This outcome is inconsistent with overall temporal integration, and argues for a cognitive mechanism allocating attention to changes in an event sequence.

Comparisons between Neural Response Imaging thresholds, electrically evoked auditory reflex thresholds and most comfortable loudness levels in CII Bionic Ear users with HiResolution™ sound processing strategies

Conclusions. The data collected in this study indicated that first Neural Response Imaging (NRI) thresholds had a better correlation with HiResolution™ most comfortable loudness (M) levels than tNRI thresholds. Electrically evoked auditory reflex thresholds (EARTs) had a higher correlation with HiResolution M levels than tNRI thresholds and a lower correlation than first NRI thresholds. NRI is a very useful method for programming the cochlear implants of young children who cannot demonstrate a reliable judgment of loudness. Objective. To investigate how HiResolution sound processing, designed to deliver high-rate stimuli, relates to EARTs and electrically evoked compound action potential measurements produced by low-rate stimuli. Material and methods. Nine profoundly hearing-impaired children and adults aged 6–29 years participated in the study. NRI responses were elicited using pulse trains consisting of biphasic pulses at a pulse width per phase of 32 µs delivered at a frequency of 30 Hz using SoundWave programming software. Stimuli were delivered to the odd electrodes (1, 3, 5, 7, 9, 11, 13 and 15) along the array. tNRI (NRI threshold) and first NRI thresholds were recorded for each stimulating electrode. “Speech bursts” stimuli used in EARTs recording were delivered to four electrodes at a time and stapedial reflexes were recorded from the impedance bridge. The M levels used were those used by each patient in their everyday HiResolution programs. Results. For 8 patients (53 stimulating electrodes) the correlation between tNRI threshold and M level was r = 0.675 (p = 0.000) and that between first NRI thresholds and M level was r=0.741 (p = 0.000). On average the M-level value was 20 CU (Current Unit) lower than the first NRI threshold value and 12 CU higher than the tNRI threshold value. The M-level patterns across the electrode array overall were similar to the tNRI or first NRI threshold patterns. For 7 patients (112 stimulating electrodes) the correlation between EART and M levels was r=0.710 (p = 0.000). On average the EART value was 14 CU higher than the M-level value.

Investigations into electrically evoked stapedius reflex measures and subjective loudness percepts in the MED-EL COMBI 40+ cochlear implant

For several years there has been interest in using objective measures to set channel-specific upper programming limits when programming the speech processor of cochlear implant users. The present study aims to add to previous reports by examiningcorrelations between electrically evoked stapedius reflex threshold (ESRT) and a range of psychophysical loudness estimates in a group of 22 adult users of the MED-EL COMBI 40+ system. Thirteen of the 15 subjects (87%) had recordable stapedius reflexes.Psychophysical measures of threshold, maximum comfort level (MCL) and maximum acceptable loudness (MAL) were recorded. Results showed that mean ESRT was closest to the MCL using 500 ms burst ('MCL500'), with MCL50 (MCL using 50 ms burst) and MAL500 some 2 dB and MAL50 3 dB higher. Correlations between ESRT and the behavioural loudness judgements were highest for MCL500 (R = 0.69, p < 0.001) and slightly less for MAL500. These results confirm the ease of measuring ESRT in a clinicalsetting and that a high level of confidence can be placed on the use of these measures for setting processor maps in the absence of behavioural data.

Brighter noise: sensory enhancement of perceived loudness by concurrent visual stimulation.

Two experiments investigated the effect of concurrently presented light on the perceived loudness of a low-level burst of white noise. The results suggest two points. First, white noise presented with light tends to be rated as louder than noise presented alone. Second, the enhancement in loudness judgments is resistant to two experimental manipulations: varying the probability that light accompanies sound and shifting from a rating method to a forced choice comparison. Both manipulations were previously shown to eliminate a complementary noise-induced enhancement in ratings of brightness. Whereas noise-induced enhancement of brightness seems to reflect a late-stage decisional process, such as a response bias, the present results suggest that light-induced enhancement of loudness may reflect an early-stage sensory interaction.

Standards for acoustical instruments

Prior to 1930, sounds were classified from subjective judgments of loudness. General-purpose instruments that could make objective measurements of sounds were not yet developed, nor were there any national or international standards with specifications for the performance characteristics of such instruments. In recognition of the need for instruments that could provide objective measurements of a sound, as well as the need for a national consensus on the characteristics of a sound to be measured, the newly founded Acoustical Society of America established Sectional Committees in 1930 to develop such standards under the procedures of the American Standards Association (now the American National Standards Institute). A draft of a standard for sound level meters prepared by the subcommittee for Sound Levels and Sound Level Meters was considered at a meeting of the Sectional Committee in May 1934. After incorporating revisions in response to comments and suggestions, the draft standard for sound level meters was approved for publication by the American Standards Association on 17 February 1936. This paper describes the development of sound level meters and other instruments; along with related national and international standards, in the years since that auspicious beginning.

Role of short-term memory in loudness comparisons

In an earlier study of the auditory discrimination of time-varying noise bursts [Ellermeier and Schrödl, J. Acoust. Soc. Am. 108, 2596 (2000)], listeners were found to place greater weight on the beginning and end of the sounds than on the middle portion. To investigate whether this outcome is due to primacy and recency effects in short-term memory which tend to be sensitive to manipulations of the inter-stimulus interval (ISI), the ISI separating the two noise bursts in a 2IFC task was varied systematically. Six participants performed loudness comparisons on 1-s samples of white noise randomly changing in level every 100 ms. In different blocks of trials, the two noise bursts to be compared were either separated by a 500-ms or a 2-s ISI. COSS analysis [Berg, J. Acoust. Soc. Am. 86, 1743–1746 (1989)] of the overall loudness judgments revealed elevated weights for the beginning and end of the noises, as in the earlier study. These weighting patterns were largely unaffected by the manipulation of ISI, suggesting that the temporal weights found characterize loudness integration in general, and are not just due to idiosyncrasies of the timing used in the 2IFC procedure.

Speech Loudness and Quality 12 Months after Intensive Voice Treatment (LSVT®) for Parkinson’s Disease: A Comparison with an Alternative Speech Treatment

Thirty-five individuals with idiopathic Parkinson’s disease were enrolled in speech treatment. Twenty-two were enrolled in a high-effort phonatory-respiratory treatment program (Lee Silverman Voice Treatment, LSVT^®) and 13 were enrolled in a high-effort respiratory treatment program (RET). Perceptual judgments of speech loudness and quality were made independently by 6 listeners on recordings of the ‘Rainbow Passage’. These recordings had been obtained just before treatment (pre) and at 12 months’ follow-up (FU12). The speech samples in the LSVT group, but not in the RET group, were significantly more likely to be perceived ‘louder’ and ‘better quality’ at FU12 than at pre (p < 0.0001). These findings, along with others, suggest that the long-term effects of the LSVT are perceptible, clinically significant and treatment-specific.

A model of a nucleus 24 cochlear implant fitting protocol based on the electrically evoked whole nerve action potential.

The goal of this study was to model a cochlear implant fitting protocol based on the electrically evoked whole nerve action potential (EAP) measured using the Neural Response Telemetry capabilities of the Nucleus C124M cochlear implant. The model and data were based on the significant correlations found between the EAP threshold and growth function slope and psychophysical threshold and, dynamic range, respectively, in 12 subjects (Franck, Reference Note 3; Franck & Norton, 2001). Using a retrospective split-half study design, these correlations found between psychophysical mapping levels and EAP data from six of the subjects were used in a model to predict psychophysical mapping levels from EAP data of the other six subjects. Predicted psychophysical mapping levels from the model of the EAP-based fitting protocol closely approximated measured cochlear implant fitting psychophysics. The close approximation of the measured data to the model data indicates the feasibility of the clinical use of an EAP-based cochlear implant fitting protocol. The realization of this model would only require two loudness judgments from the patient, whereas traditional fitting requires 44, and would be fit in a live-voice mode, accounting for across-electrode loudness summation.

Voice outcome following thyroplasty in patients with cancer-related vocal fold paralysis

Medialization laryngoplasty restores voice in patients with unilateral laryngeal paralysis. Of question was whether patients with vocal fold paralysis resulting from cancer or its treatment had as good a post-operative voice result as patients with vocal fold paralysis of benign etiology. The purpose of the present study was to compare post-operative perceptual, acoustic, aerodynamic, and quality of life data in these two patient groups. Twenty-eight patients with vocal fold paralysis secondary to malignancy or its treatment were age and gender-matched with patients with paralysis resulting from benign origin. Pre- and post-operative perceptual judgments of pitch, loudness and quality were rated independently by two speech–language pathologists. A digital audiotape of the patient's voice was analyzed using Soundscope software. Fundamental frequency, conversational intensity and perturbation were evaluated. Glottal flow rates in propositional speech, phonation times and extent of pitch and loudness ranges were also measured. Three quality of life surveys, the Short Form-36 general health survey, the Voice Handicap Index, and the Voice Outcomes Study were administered. Results of voice testing indicated that perceptual, acoustic and aerodynamic data were significantly improved 3 months after thyroplasty in all patients regardless of whether they had a history of cancer. Quality of life data, however, distinguished the two groups. In particular, the general health measure found a significant difference in physical functioning and overall vitality, although satisfaction with improved voice was equally appreciated in both patient groups. Of clinical significance is that though general health may differ, patients with cancer-related laryngeal paralysis can expect to have as good a voice outcome following thyroplasty as patients with paralysis of benign etiology.

Loudness scaling in children

Loudness measurements in children have been carried out using different psychoacoustical methods. Besides absolute magnitude estimation (AME) and cross modality matching (CMM) category and non-category loudness scalings have been performed. However, there is a lack of systematic evaluation in loudness scalings with children. A clinically feasible categorical method is presented which was evaluated in 10 normally hearing children of 7 to 8 years. Furthermore, measurements with hearing disabled children were performed in the framework of hearing aid evaluation. Normal hearing children scaled slightly louder than a group of adult subjects with normal hearing. Regarding the shape of the loudness functions measured they were very similar. Particularly, they showed a steeper course at higher levels than for low input signals. Intraindividual scatter of the loudness judgements was larger for children than for adults. However, high correlation appeared between the data across test and retest. Hence, the outcome of the scalings appears to be sufficient reliable. Clinical measurements confirmed that the method is appropriate for children from approximately 5-10 years. Loudness judgements became more consistent and more subtly differentiated with increasing age and decreasing hearing loss. Especially with view of hearing aid evaluation the method can give valuable indications because of the possibility to directly measure individual loudness over a broad range of levels and frequencies.

Effects of Familiarity Level and Repetition on Recognition Accuracy

Theories of recognition memory based on signal detection theory posit that a recognition decision is similar to a psychophysical judgment. Like a judgment of stimulus brightness or loudness, a recognition judgment is based on the value of a unidimensional signal computed for the test item. This signal has been called the strength or familiarity value. One prediction of these models is that the ability to discriminate between a studied and nonstudied test item depends on the ability to detect the difference in their familiarities. This ability in turn is influenced by the items' familiarity levels; discrimination should become more difficult as the familiarity of both items increases. This prediction was supported in 3 experiments using a forced-choice procedure. Also, accuracy was higher when the list contained repeated items rather than a comparable number of distinct items.

Effect of Interstimulus Interval on Subjective Categorical Loudness Judgments

AbstractThe Independent Hearing Aid Fitting Forum uses the visual input/output locator algorithm and the Contour Test of Loudness Perception to achieve the goal of restoring normal loudness perception with amplification. This method presupposes that subjective categorical loudness judgments are valid and reliable when using the procedure outlined by the test developers. There is no indication in the instructions of the Contour Test of a specific duration for interstimulus interval (ISI). The effect of ISI on loudness perception is important to establish because of potential time-error effects. Time-error refers to the extent to which the judged magnitude of a second stimulus varies with the time interval by which it follows the first stimulus. Past research has indicated that judgments of a second stimulus are shifted in the direction of the intensity of the preceding stimulus and that this effect intensifies with shorter ISIs. The current experiment was designed to examine whether a change in ISI produced a change in the loudness rating of the subsequent stimulus for subjective categorical loudness judgment testing. A trend toward lower, median dB values for ratings 2 to 5 at 500 Hz in a group of subjects with normal hearing was noted when 1–sec intervals were used in comparison with longer ISIs. No trends for the effect of ISI were noted at 3000 Hz. The findings provide ISI recommendations for loudness judgment test administration. Abbreviations: DAT = digital audiotape, IHAFF = Independent Hearing Aid Fitting Forum, ISI = interstimulus interval, KEMAR = Knowles Electronics' Mannikin for Auditory Research, VIOLA = visual input/output locator algorithm, WDRC = wide dynamic range compression

Noise labeling for consumer products

Currently, ISO 4871-1996 (Acoustics–Declaration and verification of noise emission values of machinery and equipment) uses A-weighted or spectral sound power levels as metrics of choice for noise labeling. Also included is a measure of the likely variations in levels that result from machine to machine variability, measurement to measurement variability, and site to site variability. This labeling scheme is adequate to label machinery for use in commercial or industrial settings. Here, plant engineers or audiologists, etc., can readily interpret and use data. But a two-number label using the sound power level in decibels is not optimum for conveying information to the consumer. This paper suggests an alternate form for the data. Sound power is still the physical quantity measured, and, as suggested by others, a one-number system is used. The statistical variation is included by reporting the value that will not be exceeded in 19 out of 20 measurements. But the metric reported is one that is roughly proportional to judgments of loudness, and, as such, should provide greater meaning and clarity to the user. Hopefully, with comprehensible noise labeling, the consumer will be able to include noise in his/her choice of products.

Differences in child and adult loudness judgments of rock music

The present study was concerned with the perceptual processing of complex auditory stimuli in ten children (M_ age=8.1) as compared to ten young adults (M_ age=19.3) and ten older adult subjects (M_ age=54.2). The auditory stimulus used was 10 s of rock music [Led Zeppelin, CD Recording No. 19127−2 (Atlantic Recording Group, New York 1969)]. All three groups provided numerical responses to nine intensity levels of the rock music stimulus (10, 20, 30, 40, 50, 60, 70, 80, and 90 dB above threshold). Results showed that the children demonstrated a wider range of numerical responses than both adult groups. The range of mean numerical responses for the children was 0.54 to 54.24. For the young adults the range was 0.76 to 11.37, and for the older subjects it was 1.6 to 23.31. Results suggest that the children were not bound by the same set of rules as the adults in regard to magnitude estimation scaling of the loudness of the rock music stimulus. Their internal scaling mechanisms appeared to be more flexible and broader based than those of the adults who participated in this study.

Motion perception and loudness judgments

Besides its acoustic properties, many other nonacoustic properties have been shown to influence the judgments of loudness, noisiness, and noise effects of sounding events. In the environment, one finds moving as well as nonmoving sound sources. The sound emitted may be either constant or dynamic. In an experiment, the effect of the perceived motion of a sound source or sound event on loudness judgments was investigated. Sounds were presented which were ambiguous and could be perceived both as moving and as nonmoving events. In different experimental conditions, the same sound stimuli were presented and introduced as moving or nonmoving events. The subjects judged the loudness of the events. At the end of the experiment, it was reported whether they had perceived the sound events as moving, nonmoving, or both. The results show an effect of perceived motion on judged loudness: Loudness judgments are higher if the sound events are perceived as moving than if they are perceived as nonmoving.

Estimates of the instantaneous versus overall loudness of noise emissions

Usually when comparing the relations between instantaneous and overall loudness judgments of noise emmissions, the overall loudness is larger than the average of the instantaneous loudness judgments. Obviously, the prominent parts of a noise emmission play a crucial part in overall loudness judgment. Two models were proposed to describe the overall loudness on the basis of the instantaneous loudness judgments [Fastl, DAGA 97]. The merits of the different models in predicting the overall loudness of noise emissions with a different number of events were studied. For emissions with 2 up to 11 events within 15 min, a simple percentile of the instantaneous loudness judgments cannot predict overall loudness. However, when the events are prolonged by an exponential decay with 30-s time constant, the judgment of overall loudness can be predicted with an average accuracy of 6%.

Loudness evaluation of urban soundscapes by a cross-modal matching method

In a previous study, loudness of a time-varying pure tone was evaluated using a cross-modal matching method. In this procedure, the loudness was matched to an equivalent muscular force. Subjects had to judge the instantaneous and global loudness using a proprioceptive input device with force feedback. Using this technique, subjects could respond easily, rapidly, continuously, and with feedback concerning their evaluation. The present study examines the relation between instantaneous and global loudness of 16 urban soundscapes lasting about 20 s using the same method. First, the procedure is individually calibrated with 1-s soundscapes presented at different sound-pressure levels. The data obtained for each subject fit well with a linear relation on a log–log scale between the sound-pressure level in dB and the associated force in log Newtons. Then, the global judgments are performed under two experimental conditions: one with instantaneous loudness evaluation during stimulus presentation and one without. Good correlation was found between these two judgments. The global judgment and the average over the instantaneous matching contour are nearly identical. In addition, a fluction factor (FI) calculated on the loudness judgment contour reflects the degree of movement produced by the subject as well as the presence of discriminable sources.

Effect of time distribution of energy on loudness evaluation

Most work on loudness have been done on stationary sounds. However, environmental sounds are usually temporally variable, and memory effects could be important in loudness evaluation of long-lasting dynamic sounds. The loudness of one-minute sounds containing a local dominating peak of energy was measured. The sounds differed only by the temporal position of the peak. The influence of peak location on the global loudness judgment was evaluated. The global loudness was first evaluated using magnitude estimation. Contrary to expectations, mean estimations were the same for all sounds regardless of peak distribution. In a second experiment, subjects had to judge the loudness, using cross-modal matching. They matched the size of a circle on a computer screen with the loudness of the sound. At the end of the signal, a global estimation was also made using the same cross-modal matching paradigm. This global estimation was made either just after the sound or after a pause of one minute. During the pause, the subject had to perform a distracting task. Relations between global and continuous judgments are discussed. The study can help to design a model of the loudness of temporally variable sounds.