Simultaneous dichotic loudness balance (SDLB): Why loudness "fatigues" with two ears but not with one.

In the laboratory method called simultaneous dichotic loudness balance (SDLB), the contribution-to-loudness that arises from the listener's continually exposed "fatiguing" ear is required to be matched (balanced) by the listener, by adjusting the intensity of a noncontinuous stimulus at the other ("comparison") ear. The latter intensity usually declines, allegedly indicating "fatigue" of the contribution-to-loudness from the "fatiguing" ear. However, no "fatigue" is found when one ear alone (with the other ear in quiet) experiences a continuous well-suprathreshold stimulus. This is a quandary that remains unresolved. The present article offers a resolution, through a novel conceptual model in which any ear experiencing stimuli acts through a well-characterized physiological structure, the olivocochlear bundle, to "turn down the volume" at the opposite ear. The model explains how "fatigue" varies in eight different SDLB conditions, some having several subconditions. Altogether, the model demonstrates that "fatigue" is an artifact of SDLB itself.

Effects of external noise on detection of intensity increments.

The detection of an intensity increment in a longer duration sinusoid or pedestal is often used as a measure of intensity resolution, but the decision processes underlying this measure are poorly understood. Thresholds were obtained for detection of an increment in a 370-ms, 4-kHz pedestal in quiet or in noise to determine the relative contributions of background noise level and pedestal level, the effect of increment duration, and the effect of different noise spectra. Increment detection thresholds expressed in units of DeltaL[10 log(1+DeltaI/I)] decreased as pedestal levels increased. At low pedestal levels, increment detection was limited by the masking effect of the noise and was similar across noise conditions for pedestals of equal sensation level. At high pedestal levels, the noise had no effect and increment detection was determined by the pedestal level in dB SPL (sound pressure level). Increment detection improved with increasing increment duration and was altered less by a noise band above the pedestal/increment frequency than by a broadband noise that produced equal masking at the pedestal/increment frequency. The quadratic-compression model described by Neely and Jesteadt [(2005). Acta Acust. Acust. 91, 980-991] provided a better approximation to the data than a model based on excitation patterns.

The intensity-difference limen for 6.5 kHz: an even more severe departure from Weber's law.

Intensity-difference limens (DLs) were obtained for tones of 6.5 kHz over levels of 30-90 dB SPL. The tones had Gaussian-shaped envelopes whose duration was expressed as equivalent rectangular duration D, that of a rectangle enclosing the same area as the envelope. D ranged from 0.314 to 30 msec. DL behaviors such as the "severe departure from Weber's law," the "midlevel hump," or the "midduration hump"--that is, a rise in the DL over a range of short durations--were identified using trend analysis. The DL versus level followed a "severe departure" that increased as duration dropped to D approximately 1-2 msec. That midlevel hump then fell with further shortening of the tone pip, producing a mid-duration hump. Materials for this article may be accessed through the Psychonomic Society's Norms, Stimuli, and Data archive, at www.psychonomic.org/archive.

Dynamic range relations for auditory primary afferents.

Dynamic range is one of four attributes typically assigned to the plot of firing rate vs. stimulus level of an auditory primary afferent. Dynamic range is generally understood to be the contiguous range of sound-pressure-level over which the neuron can indicate some small level change. Typically, however, dynamic range has been quantified as the width in decibels between the endpoints of the rate-level plot, which is not a measure of sensitivity to level change. A sensitivity measure is provided here by first deriving an equation for the intensity-difference limen (DL) in terms of attributes of the rate-level curve. The result is a generally U-shaped curve of DL vs. level. Any given criterion DL corresponds to a horizontal line cutting the DL curve at two points, with the separation in decibels between those points providing a dynamic range for that DL criterion. Plotting the dynamic ranges vs. the respective DLs yields a dynamic range curve. These were made for 62 afferents from the cat. The dynamic ranges of sloping-saturating rate-level plots do not exceed those for sigmoidal plots until the DL criterion reaches 50 dB, supporting the conclusion of Palmer and Evans [Cochlear fibre rate-intensity functions: no evidence for basilar membrane nonlinearities, Hearing Research 2 (1980) 319-326] that sloping saturation is not a reflection of cochlear nonlinearity.

Intensity-difference limens predicted from the click-evoked peripheral N1: the mid-level hump and its implications for intensity encoding.

The intensity-difference limen (DL) for an acoustic click rises at moderate click levels, a feature called the 'mid-level hump'. It has long been hypothesized that, because a click does not evoke sustained firing in any primary afferent, the DL must therefore originate from the initial burst of synchronized spikes in the eighth nerve. That burst causes the N1 component of the peripheral compound action potential (CAP). It should therefore be possible to predict click DLs from N1 potentials. Here, a Signal Detection model, using a series expansion, was used to derive equations in N1 for the level-dependence of the DL. The first-order equation predicts a dependence on the standard deviation of N1, and an inverse dependence on the rate-of-growth of the mean N1. The second-order equation is more complicated. Both approximations were applied to N1s from the cat. Both produced a mid-level hump; at its peak, the DLs from the second-order approximation were the smaller ones, and were of the same order of magnitude as the empirical DLs. Overall, the computations show that the rate-of-growth of the mean N1, not the standard deviation of N1, determines the hump in the empirical DL.

Threshold vs duration for Gaussian-shaped tone-pips of one to four periods duration.

Detection thresholds were obtained for Gaussian-shaped tone-pips of 1, 2, 3, or 4 periods duration for 20 frequencies spanning 50-3,000 Hz, in quiet, and in high-pass noise, for a single exceptionally patient and experienced listener. Thresholds were fitted by straight lines in decibels SPL versus the logarithm of duration. Slopes fell into 3 distinct regimes of frequency-dependence under both listening conditions. Models of Garner and of Formby, et al. do not account for this relation.

A measure of internal noise based on sample discrimination.

Internal noise is often inferred from the difference between observed performance and optimum performance in detection and discrimination tasks. It can be measured directly in some cases by observing the extent to which a change in external variability impacts performance. In the studies reported here, external variability was added to an intensity discrimination task by adding a Gaussian random variable with zero mean to the overall level presented in each interval of a two-interval forced-choice task. The standard deviation of the random variable was set to half the mean difference between the levels in the two intervals, resulting in d'(ideal) = 2. As the mean difference and the corresponding standard deviation of the random variable decreased in size, performance was increasingly limited by internal noise, permitting a reliable estimate of internal noise to be obtained. This can be viewed as a sample discrimination task, with one component per sample. In the first study, performance was measured using 2-kHz tones presented at an average level of 70 dB SPL, with mean differences between distributions ranging from 0.1 to 2.2 dB in steps of 0.3 dB. The distributions were either Gaussian in level or in power. Conditions with no external variability were used to obtain a psychometric function. In the second study, performance was measured using 2-kHz tones presented at average levels of 50 and 90 dB SPL, with mean differences ranging from 0.4 to 2.2 dB in steps of 0.6 dB. In both studies, the measure of internal noise was highly reliable and in good agreement with the intensity difference limen (DL) estimated from the psychometric function. Analyses suggest that this measure could be used to estimate the mean difference between the decision distributions as well as the amount of internal noise in cases where the mean difference between the distributions is unknown.

Effects of peripheral nonlinearity on psychometric functions for forward-masked tones.

Psychometric functions (PFs) for forward-masked tones were obtained for conditions in which signal level was varied to estimate threshold at several masker levels (variable-signal condition), and in which masker level was varied to estimate threshold at several signal levels (variable-masker condition). The changes in PF slope across combinations of masker frequency, masker level, and signal delay were explored in three experiments. In experiment 1, a 2-kHz, 10-ms tone was masked by a 50, 70 or 90 dB SPL, 20-ms on-frequency forward masker, with signal delays of 2, 20, or 40 ms, in a variable-signal condition. PF slopes decreased in conditions where signal threshold was high. In experiments 2 and 3, the signal was a 4-kHz, 10-ms tone, and the masker was either a 4- or 2.4-kHz, 200-ms tone. In experiment 2, on-frequency maskers were presented at 30 to 90 dB SPL in 10-dB steps and off-frequency maskers were presented at 60 to 90 dB SPL in 10-dB steps, with signal delays of 0, 10, or 30 ms, in a variable-signal condition. PF slopes decreased as signal level increased, and this trend was similar for on- and off-frequency maskers. In experiment 3, variable-masker conditions with on- and off-frequency maskers and 0-ms signal delay were presented. In general, the results were consistent with the hypothesis that peripheral nonlinearity is reflected in the PF slopes. The data also indicate that masker level plays a role independent of signal level, an effect that could be accounted for by assuming greater internal noise at higher stimulus levels.

Afferent response parameters derived from postmasker probe-detection thresholds: 'the decay of sensation' revisited.

The classical model of forward masking postulates that the detection threshold for a tone probe that follows a stimulus of similar frequency content is elevated relative to the quiet threshold because the probe must evoke a just-detectable increment in a decaying postmasker sensation. That postmasker decay is charted by probe-detection thresholds if the sensation increment is small and constant. This model was examined for a 2-kHz Gaussian-shaped probe and a 2-kHz forward masker, based on the model's assumption that a just-detectable increment in sensation results from a just-detectable increment in level. Psychometric functions for detection were obtained at 2.5-30 ms postmasker. Their means and standard deviations generally decreased with delay. It was assumed that standard deviation is related to the putative just-detectable level increment by a simple monotonic transformation. Thus, if the standard deviation of the psychometric function for probe detection is neither small nor constant, then the corresponding just-detectable increment in level is neither small nor constant, and the just-detectable increment in sensation is neither small nor constant. The classical model also fails to allow for the variability of internal events. The concept of detection threshold as a sensation increment was preserved in a Signal Detection model, that does allow for internal variability. In this model the postmasker residual is the input to a probe detector. The new model produces an equation for the just-detectable level increment as a function of probe delay. Comparison data were generated by again assuming some relation between the standard deviation of the psychometric function for detection, and the just-detectable increment in level. The fit of equation to data yields robust values for the probe detector's maximum firing rate, dynamic range, and spike-counting time. All that is required to account for the decay of sensation, for a pure tone, is a single neuron operating at some higher center.

The mid-level hump at 2 kHz.

Shortening the duration of a Gaussian-shaped 2-kHz tone-pip causes the intensity-difference limen (DL) to depart from the "near-miss to Weber's law" and swell into a mid-level hump [Nizami et al., J. Acoust. Soc. Am. 110, 2505-2515 (2001)]. For some subjects the size of this hump approaches or exceeds the size reported for longer tones under forward masking, suggesting that forward masking might make little difference to the DL for very brief probes. To test this hypothesis, DLs were determined over 30 to 90 dB SPL for a brief Gaussian-shaped 2-kHz tone-pip. DLs were obtained first without forward masking, then with the pip placed 10 or 100 ms after a 200-ms 2-kHz tone of 50 dB SPL, or 100 ms after a 200-ms 2-kHz tone of 70 dB SPL. DLs inflated significantly under all forward-masking conditions. DLs also enlarged under an 80 dB SPL forward masker at pip delays of 4, 10, 40, and 100 ms. The peaks of the humps obtained under forward masking clustered around a sensation level (SL) that was significantly lower than the average SL for the peaks of the humps obtained without forward masking. Overall, the results do not support the neuronal-recovery-rate model of Zeng et al. [Hear. Res. 55, 223-230 (1991)], but are not incompatible with the Carlyon and Beveridge hypothesis [J. Acoust. Soc. Am. 93, 2886-2895 (1993)] that nonsimultaneous maskers corrupt the memory trace evoked by the probe.

Estimating auditory neuronal dynamic range using a fitted function.

To obtain the dynamic range of an auditory afferent, the neuron's firing rate is plotted versus stimulus level, and the dynamic range is taken as the difference between the threshold for evoked firing, and the level at which firing rate saturates. Those dynamic range endpoints are typically defined in terms of the neuron's spontaneous firing rate and its maximum firing rate, according to a plurality of schemes, each of which depends on user-chosen sets of numerical criteria. The dynamic ranges predicted by some of these schemes are compared for the first time, and the resulting estimates can differ by a factor of 2. A step can be taken towards standardizing the measurement of neuronal dynamic range, if dynamic range is incorporated into a rate-level function as a parameter. To build this function, it is first assumed that the neuron's rate-level response reaches half its maximum at a level half-way between the threshold and the level at saturation, i.e. at threshold plus half the dynamic range. Then the firing rates at threshold and at threshold plus dynamic range are defined according to the most popular of the endpoint schemes. The resulting equation produces credible estimates of neuronal properties when fitted, and correctly predicts the behavior of the slope of the empirical rate-level plot [McGee, 1983. M.S. thesis, Creighton University; Ohlemiller et al., 1991. J. Acoust. Soc. Am. 90, 274-287]. Thus, despite not being deterministic, the new equation has remarkable predictive power. When two of the rate-level functions are added and weighted, the resulting equation fits sloping-saturating data better than any functions presently employed.