Loudness Enhancement Research Articles

Loudness enhancement in man. I. Brainstem−evoked response correlates

This paper correlates electrophysiological responses and psychophysical judgments in subjects performing in a loudness enhancement task. Subjects received a tone burst Tc followed after an interval (ΔT) by a signal S1 whose loudness they were required to match by adjusting the intensity of another signal S2 presented 1.5 sec later. They judged the loudness of S1 to be enhanced by 15 dB or more when ΔT was small. Analysis of the electrophysiological brainstem responses evoked by S1 and S2 (as recorded via scalp electrodes) revealed no changes in the electrophysiological response to S1 that could explain its loudness enhancement. A latency measure decreased and a power spectrum measurement increased for both S1 and S2 as their physical intensity was raised, but presenting Tc prior to S1 had either no effect or the opposite effect by either measure despite the large perceived enhancement of S1 loudness. These results show the brainstem response to reflect stimulus intensity changes accurately, and perceived ...

Evoked potentials in cat auditory nerve: Suppression by prior tonal stimulation

Action potentials (AP) recorded from the eighth nerve of a cat to a tone pip, S1, were suppressed when preceded by a monotic or dichotic conditioning tone, Tc. The suppression increased when (1) the interval between Tc and S1 was decreased, (2) either the intensity or the duration of Tc was increased, or (3) the difference in frequency between the two stimuli was reduced. Thus the nerve action potential amplitude correlates strongly, but inversely, with the loudness enhancement effects seen in man (Galambos, Bauer, Picton, Squires & Squires,Journal of the Acoustical Society of America, 1972, 52). Since Coats and Dickey (Journal of the Acoustical Society of America, 1972, 52) found that the human AP can be diminished by a preceding noise without reducing the loudness of a click, the available results suggest that a central mechanism functions to compensate for the reduction in AP amplitude. The dichotic effects could be accounted for by cross-hearing, thereby providing no evidence for a role of the olivocochlear bundle in the effects observed.

Electrophysiological Responses Evoked during Loudness Enhancement in Man

In a loudness enhancement task, subjects received a tone burst (Tc) followed after an interval (ΔT) by a signal (S1) whose loudness they matched by adjusting the amplitude of another burst (S2) presented 1.5 sec later. All signals were 8 kHz, 0.2 msec, with Tc at 100 dB SPL and 25 dB more intense than S1. As ΔT was decreased in steps from 400 to 100 msec, subjects systematically raised S2 amplitude to a maximum of approximately 15 dB above S1 amplitude. Computer-averaged brain activity evoked by each signal was concurrently recorded between vertex-mastoid electrodes, revealing (1) latencies to the peaks of the brainstem response [which measures activity in auditory nerve and brainstem nuclei; see Jewett and Williston, Brain 84, 681–696 (1971)] for S1 remain constant or increase slightly while those for S2 decrease, and (2) the amplitude of the vertex response (100–300 msec latency) to S1 systematically decreases while that to S2 does not. Hence the electrophysiological results reflect the changes in stimulus parameters, not in the perceptual effects.

Loudness enhancement and summation in pairs or short sound bursts

When a pair of short, temporally spaced sound bursts is presented and the listeners are requested to match the loudness of a third, comparison burst to the overall loudness of the pair, a fundamentally different result is obtained than when they are instructed to match the loudness of the comparison burst to that of the second burst in the pair. Loudness-level changes occurring in the first situation are designated as loudness summation; those occurring in the second situation, as loudness enhancement. Some parameters of both phenomena are studied. The results lead to the suggestion of a principle of maximum similarity in stimulus matching and to a reinterpretation of some earlier data.

On loudness enhancement of a tone burst by a preceding tone burst

The loudness level of a second tone burst in a monotic burst pair is investigated as a function of the intensity and frequency of the first burst relative to the corresponding variables of the second burst and as a function of the interburst time interval. The loudness level is measured with the help of a third, comparison burst whose frequency is the same as that of the second burst. The results, in connection with preceding results, show beyond any reasonable doubt that loudness effects in pairs of sound bursts are controlled by two perceptual processes: loudness enhancement and loudness summation. The first refers to the loudness of the second burst, the second, to the overall loudness of the burst pair. The time and frequency functions of the two processes are fundamentally different.

Loudness Enhancement with Monaural, Binaural, and Dichotic Stimulation

In this study of the loudness enhancement phenomenon [e.g., Galambos et al., J. Acoust. Soc. Amer. 51, 141(A) 1972; Sokolich and Zwislocki, J. Acoust. Soc. Amer. 52, 141–142(A) 1972], a tone burst, Tc, was followed 100 msec later by a 70-dB tone burst, S1. Subjects matched S1 in loudness by varying the intensity of a tone burst, S2, presented 1.5 sec. after S1. Tc, S1, and S2 were all 20-msec 5-kHz signals. Increasing Tc in 10-dB steps from 60 to 100 dB SPL yielded a progressively increased subjective loudness of S1. The monaural and binaural paradigms produced greater and more reliable loudness enhancement than the dichotic. With Tc at 100 dB, loudness enhancements averaged over 15 dB for eight naive subjects tested monaurally and eight binaurally. With Tc less intense than S1, a small loudness decrement resulted. With Tc set at 90 dB and S1 at 40 dB, enhancements of over 30 dB, the largest so far observed, were regularly reported in the monaural paradigm.

Loudness enhancement following contralateral stimulation.

The apparent loudness of a tone pip can be increased by 15 dB or more if it is preceded by a tone burst to the contralateral ear. The experiment is done by delaying the pip, S1, by a variable time, ΔT, after the offset of a contralateral tone; the listener assesses the loudness of S1 by adjusting the intensity of a second tone pip, S2, that follows S1 by 1500 msec. Some parametric explorations of the phenomenon are reported here.

On Loudness Enhancement of a Tone Burst by a Preceding Tone Burst

According to preceding experiments, loudness effects in pairs of tone bursts are controlled by two processes, loudness enhancement and loudness summation. This paper deals with the effects of intensity, time, and frequency differences between the bursts on the loudness enhancement of the second burst. Listeners were asked to match the loudness of a third (comparison) burst to the loudness of the second burst of a monotic burst pair. All three bursts had a duration of 10 msec, and the second and third bursts were at 1000 Hz. The intensity of the first burst was either +20, 0, or −20 dB re the intensity of the second burst, which was always at a sensation level of 40 dB. The interval between the onsets of the first and second bursts was varied stepwise between 10 and 500 msec, and the frequency of the first burst, between 200 and 4000 Hz. The results indicate a loudness enhancement that increases with the intensity of the first burst and decreases as the time interval and the frequency difference between the bursts in the burst pair increase. When the first burst was at an intensity level of −20 dB, a loudness decrement occurred.

Loudness Enhancement Following Contralateral Stimulation

The apparent loudness of a tone pip can be increased by a prior toneburst presented to the contralateral ear. A listener receives two tone pips S1 and S2 separated by 1.5 sec, and he adjusts the intensity of S2 to equal S1 in loudness. When a toneburst to the contralateral ear precedes S1 by an interval ΔT, the loudness of S1 is enhanced by as much as 14 dB. This enhancement (1) increases as the frequency of the contralateral toneburst approaches S1 from above or below; (2) decreases to zero as ΔT is increased up to 1 sec; (3) increases as the contralateral stimulus duration is increased up to 200 msecs; and (4) is monotonically related to contralateral intensity greater than 70 dB SPL. In a particular case with S1 at 8 kHz, changing the frequency of the contralateral toneburst from 8 to 8.2 kHz cut the loudness enhancement to one-half. The possible physiological basis and psychological significance of this phenomenon will be discussed.

Loudness Enhancement and Summation in Pairs of Short Sound Bursts

When a pair of short temporally spaced sound bursts are presented monaurally, and the listeners are requested to match the loudness of a third comparison burst to the total loudness of the pair, a fundamentally different result is obtained than when they are instructed to match the loudness of the comparison burst to that of the second burst in the pair. The most striking difference has been obtained thus far with the sound frequencies of the first two bursts at 1 and 4 kHz, respectively, and the frequency of the comparison tone at their geometric mean when the listeners judged the loudness of the burst pair and at 4 kHz when they judged the loudness of the second burst. In the first situation, the loudness level of the pair appeared increased by almost 10 dB over the loudness level of a single burst and was independent of the time interval between the bursts in the pair. In the second situation, the loudness was increased by only 3 dB for very short time intervals, and the increment decreased to zero within 80 msec. Since, at durations greater than 80 msec the tone bursts are heard as distinctly separate entities, we conclude that listeners consciously summate the loudness of the bursts when asked to judge the total loudness of the pair.

Effect of unilateral masking on the lateralization of binaural pulses.

Pulsed tones were presented binaurally to subjects. Pure-tone frequencies were 250, 1000, 4000, and 8000 Hz. Loudness was varied in each ear. One ear was continuously stimulated with an octave band of noise. The subject attempted to lateralize the pulsed tones in either the masked or unmasked ear. Results indicate that the pulsed tone in the masked ear had little or no effect on the lateralization until it exceeded masked threshold. Comparison of the loudness levels in the unmasked ear, required for lateralization in that ear with curves for loudness in noise, indicated that there may well have been centrally mediated loudness enhancement in the unmasked ear produced by the masking stimulus.

Loudness effects in pairs of tone bursts

The loudness of dichotic and monotic pairs of short tone bursts was investigated as a function of the interburst time interval. For short intervals, the loudness was increased relative ty the loudness of a single burst. However, the loudness of a burst pair was equal to the loudness of the second burst in the pair and, therefore, no loudness summation but only a loudness enhancement took place. In dichotic bursts, the loudness enhancement decayed monotonically as the time interval increased, and the rate of decay increased with sound intensity. In monotic bursts, the loudness enhancement decayed to a minimum at about 40 msec, independent of sound intensity. It had a tendency to rebound at longer time intervals and go through a relative maximum in the vicinity of 200 msec. The results are interpreted in terms of an interaction of peripheral and central poststimulatory inhibition with temporal summation.