Volume 30, Issue 5, May 1958
Index of content:
30(1958); http://dx.doi.org/10.1121/1.1909626View Description Hide Description
The relation between the subjective intensity of a vibration on the skin and the physical amplitude of the vibration is different for the different parts of the skin, for example, the finger tip and the shoulder. The reason for this is that the sensation of vibration “loudness” is a complex interaction of summation and inhibition along the stimulated skinsurface. Both summation and inhibition vary with the amplitude of the vibration, the distance between the stimulated areas, the slope of the spatial distribution of the stimulus along the surface of the skin, its time pattern, and the density of the innervation. I have tried to describe these relations as variations in “funneling action” of the nervous tissues in the hope that this terminology will lead to a better understanding of the various phenomena.
A series of experiments was designed to establish rules for the estimation of the funneling action. For this purpose (1) the vibration loudness produced by a point vibrator was compared with the loudness produced by stimulation of a large area of skin (which may serve to explain why the curves of equal loudness for pure tones in hearing are compressed at the low frequencies); (2) the difference limen for amplitude variations was investigated under different conditions; (3) the lateral spread of the skin sensations was determined; and (4) all these phenomena were compared with analogous sensations in hearing. Through the concept of funneling, rotating skin sensations are brought into very close relation with rotating tones (Drehtön), and Mach's law of contrast in vision with a similar phenomenon of the “funneling” action observed on the skin. Finally, it is demonstrated by experiments that once a funneling action is established on the skin, there is a certain period of time during which it inhibits all other sensation, exactly as in hearing. It is very similar to the refractory period of the nerves. These refractory times are considerably longer when the funneling action takes place at the higher levels of the brain.
30(1958); http://dx.doi.org/10.1121/1.1909628View Description Hide Description
This paper is an investigation of the phenomenon which was observed by Huggins in 1953. Huggins found that a binaural stimulus gives a fairly clear perception of pitch although the separate stimuli to the two ears give no such perception. The basic stimulus consists of white noise introduced into one ear while the same white noise, phase transformed in a narrow band of frequencies, is introduced into the other ear. A practiced subject listening to this stimulus reports a faint pitch quality which is judged to sound about the same as narrow‐band filtered noise. A forced‐choice technique was used in which six subjects were asked to judge the direction of the pitch change when the frequency band over which the phase shift occurs was changed. The control consisted of the same stimulus presented to the two ears. Data are presented indicating the relationship between the percent of correct judgments and the three experimental variables, frequency, band width, and intensity level.
30(1958); http://dx.doi.org/10.1121/1.1909630View Description Hide Description
It is proposed that the limen of loudness ΔS for pure tones is equal or proportional to the root‐mean‐square deviation in the number of pulses produced by the sound in the time T of observation. By making a plausible assumption concerning the deviation and by assuming that loudnessS is proportional to the pulse rate one obtains . Here A is a constant which must be a function of frequency if the equation is to fit experimental data. The fit is encouraging but not completely convincing. If valid, the hypothesis might be used to explain other acoustic phenomena.
30(1958); http://dx.doi.org/10.1121/1.1909632View Description Hide Description
In an attempt to derive a limited number of descriptive adjectives with which to characterize passive sonarsounds, 50 U.S. Navy sonarmen (median sonar experience of one year) rated 20 different passive sonarsounds on 50 seven‐point scales defined by polar‐opposite adjectives (e.g., heavy‐light). The intercorrelations between scales (calculated over both subjects and sounds) were factor analyzed by the Thurstone complete centroid method. Seven interpretable orthogonal factors (accounting for approximately 40% of the variance of judgments) were extracted from the data.
The generality of the methodology for psychological research on auditory perception is considered and suggestions for research along these lines are presented.
30(1958); http://dx.doi.org/10.1121/1.1909634View Description Hide Description
A procedure for transmitting the letters of the alphabet by tone‐coded signals was examined in quiet and against a noise background. The procedure employed successive selections, each from among a small number of alternatives, in order to transmit a target vocabulary of 25 letters. Four stimulus variables: tonal frequency, sound level, location, and duration were examined, one at a time. Successive selections were made among two, three, and five alternatives per variable. The highest reception rate was obtained with a three‐alternative frequency‐coded display. Reception of tone‐coded signals in noise was nearly equivalent to that in the quiet, when the tonal signals were about 3 db above masked threshold.
30(1958); http://dx.doi.org/10.1121/1.1909636View Description Hide Description
The threshold of audibility as a function of frequency is determined by the transmission characteristic of the ear and by the sensitivity of the nervous system. In order to separate the two factors, the sound transmissioncharacteristic calculated on the basis of known experimental data and compared to the threshold of audibility. As a final result a simple curve is obtained which describes the sensitivity of the nervous system as a function of frequency.
30(1958); http://dx.doi.org/10.1121/1.1909638View Description Hide Description
The statistical decision model, which has achieved outstanding success in describing the detection of signals in noise was applied to the reception of filtered speech. A confidence rating was added to the articulation test procedure in order to obtain additional information about the listener's criterion of message acceptance and message rejection of filtered speech. The relation between correct confirmations and false alarms—the Receiver Operating Characteristic—obtained with filtered speech corresponds with that typically obtained with noise interference. It is suggested that the “noise” of the decision model may be extended to a wide range of operations which perturb the signal.
30(1958); http://dx.doi.org/10.1121/1.1909640View Description Hide Description
A psychoacoustic experiment to determine the just‐discriminable changes in the fundamental frequency of synthetic vowels is described. The experimental parameters investigated are several combinations of four vowels (|i, æ, ɑ, u|), three sound pressure levels (60, 70, and 80 db re 0.0002 dyne/cm2) and two fundamental frequencies (80 and 120 cps). The results indicate that the just‐discriminable changes in fundamental frequency are of the order of 0.3 to 0.5 cps, and are, in general, slightly less than the frequency changes discriminable a pure tone of the same frequency and sound pressure level. Application of the the results to speech compression is suggested.
30(1958); http://dx.doi.org/10.1121/1.1909642View Description Hide Description
Interest in acquiring more definite knowledge of the mechanism of sound absorption in electrolyticsolutions led to this experimental investigation of the pressure dependence of this phenomenon. A technique was evolved to measuresound absorption in liquids as a function of pressure at frequencies of 100–600 kc. The following properties of a 0.5‐molar solution of were measured up to 20 000 psi at 26°C: absorption, relaxation frequency and velocity. The absorption at 20 000 psi is less than one quarter that at atmospheric pressure and the relaxation frequency remains constant at 1.2×106 rad/sec over this pressure interval within the estimated experimental error of ±10%. The velocity increases linearly with pressure at the rate 0.1588 m/atmos. From these data the chemical and adiabatic compressibilities are computed.
In an effort to secure quantitative agreement with experimental results some new aspects of the dissociation theory are discussed.
30(1958); http://dx.doi.org/10.1121/1.1909644View Description Hide Description
The radiation efficiency of a linear array of point sources with linear phase variation is solved using different approximations for different ranges of the phase parameter. The results are compared with experimental results on small models and also on full‐scale suction rolls in paper mills with different drilling patterns. The experiments agree quite closely with theory and indicate that a noise reduction of 15 to 30 db can be obtained by the choice of a substitute drilling pattern that can easily be drilled by multiple spindle machines. Design charts for choosing and evaluating drilling patterns are given.
30(1958); http://dx.doi.org/10.1121/1.1909646View Description Hide Description
The results of a study of the effect of shape and orifice edge conditions the nonlinear acousticproperties of orifices are presented. The samples studied consisted of orifices of circular, rectangular, and square shape in thin orifice plates. Varied edge conditions considered are square, round, and symmetric and nonsymmetric beveled edges for circular orifices in thick plates. These samples were studied for sinusoidal volume velocity in the orifice. It is found that for all the symmetric samples the pressure contains odd harmonics while for the nonsymmetric sample a steady component and second harmonic component are superimposed on the odd harmonies. An equation for the pressure‐flow relation is given which is applicable to both symmetric and nonsymmetric orifices.
30(1958); http://dx.doi.org/10.1121/1.1909648View Description Hide Description
Most narrow‐band‐width wave analyzers currently available on the market were designed for producing spectra of periodic signals. When they are applied to producing spectra for low‐frequency random signals such as occur in vibration, a longer averaging time must be provided on the output side of the detector circuit to average out fluctuations and yield data of satisfactory statistical significance. The application of standard statistical techniques in relation to the band width of the selective circuit provides design data for averaging circuits to be added as modifications to existing analyzers. Proper averaging time carries with it a requirement of a slower sweep rate and hence greatly increases the time required for producing a spectrum. Techniques and apparatus for producing a spectrum in a decreased time interval are discussed.