China
Africa
Explore chapters and articles related to this topic
Optical and visual metrics
Published in Pablo Artal, Handbook of Visual Optics, 2017
The CSF is measured using sinusoidal gratings of variable contrast and spatial frequency (Wandell 1995). This can be done using a computer/display system or with printed charts containing stimuli of different frequencies and contrasts. In order to estimate a contrast threshold, the observer is tested over many trials, at various contrasts. Each trial is at some contrast and is scored right or wrong. The proportion of correct responses at each contrast is recorded. The observer’s probability of correct response as a function of contrast is the psychometric function. The inverse of contrast threshold thus determined is the sensitivity. The NCSF can be measured subjectively by an interference fringe technique, which theoretically allows a sinusoidal pattern of very high contrast to be projected directly on the retina. In this way the results are not affected by aberration or diffraction from the eye’s optics.
Psychoacoustic methods
Published in Stanley A. Gelfand, Hearing, 2017
The method of limits is also limited in terms of step size and inefficiently placed trials. Too large a step size reduces accuracy because the actual threshold may lie anywhere between two discrete stimulus levels. For example, a 10-dB step is far less precise than a 2-dB increment; and the larger the increment between the steps, the more approximate the result. Too large a step size may place the highest inaudible presentation at a level with a 0% probability of response, and the lowest audible presentation at a level with a 100% probability of response. The 50% point (threshold) may be anywhere between them! To make this point clear, consider the psychometric functions in Figure 7.2. A psychometric function shows the probability (percentage) of responses for different stimulus levels. Figure 7.2a shows the psychometric function for a particular sound. It is inaudible (0% responses) at 13 dB and is always heard (100% responses) at 21 dB. It is customary to define the threshold as the level at which the sound is heard 50% of the time (0.5 probability). The threshold in Figure 7.2a is thus 17 dB. Suppose we try to find this threshold using a 10-dB step size, with increments corresponding to 14 dB, 24 dB, and so on. Notice that this step size essentially includes the whole psychometric function, so that we do not know where the responses change from 0 to 100%, nor do we know whether they do so in a rapid jump (a step function), or along a function where gradual changes in the proportion of “yes” responses correspond to gradual changes in stimulus level. The result is low precision in estimating the location of the 50% point. However, a large step size is convenient in that it involves fewer presentations (and thus shorter test time), since responses go from “yes” to “no” in very few trials, each of which is either well above or well below threshold.
The development of the Language-Independent Speech in Noise and Reverberation test (LISiNaR) and evaluation in listeners with English as a second language
Published in International Journal of Audiology, 2023
Sharon Cameron, Christian Boyle, Harvey Dillon
The second goal of the study was to investigate the relationship between track variability and test performance. For the first three test conditions – all of which involved spatial cues and produced SRTs at substantially negative SNRs – greater SDs of SNR was associated with the poorest SRTs, consistent with previous observations on child participants (Moore et al. 2010, Dillon and Cameron 2021). The correlation was, however, significant only for the Anechoic condition (r = 0.27, p = 0.014). This significant result in only one of the four conditions, therefore, offers only limited support that a larger than average track variability is associated with poorer performance averaged across all the trials, at least in young adults with no hearing or listening problems. Fluctuating attention will cause increased track variability if the differing levels of attention cause the effective ability of the participant to vary during the measurement. The result should be a shallower psychometric function than would otherwise occur with constant attention and ability. Shallower psychometric functions then lead to larger adaptive track variability (Wetherill 1963). Of course, there may well be causes of decreased psychometric slope and increased track variability other than fluctuating attention.
Hearing aid delay in open-fit devices – coloration-pitch discrimination in normal-hearing and hearing-impaired
Published in International Journal of Audiology, 2023
The mean values of the percent correct scores across the NH and HI participants are shown in Figure 4 as a function of HA delay. The error bars indicate the standard error across participants. In the NH group, the mean correct scores for all IGs are strictly monotonic rising within the range of tested delays. The thresholds, here defined as the 75% point on the psychometric function, are between 0.25 ms and 0.35 ms. For the HI group the psychometric functions are displaced towards higher delays. The 75% correct point is at approximately 0.6 ms for the N2 IG and 1 ms for the N3 IG. Hence, the distance between the psychometric functions for the N2 and N3 condition is larger in the HI group compared to the NH group. That indicates that discrimination abilities are more affected by increased gain in the HI group.
Lack of correlation between medial olivocochlear reflex strength and sentence recognition in noise
Published in International Journal of Audiology, 2023
Ian B. Mertes, Abigail L. Stutz
For practice, the first 10 sentences from list 1 were presented at +3 dB SNR. These results were not included in the analysis. Each participant was then presented six lists, with two lists at each SNR. The order of SNRs was counterbalanced across participants using a balanced Latin square design. The six lists were randomly selected without replacement for each participant. For each list, the total number of words correctly repeated back was recorded onto a score sheet (all words in the sentence were considered key words to be scored). For each SNR, the participant’s score in percentage correct was averaged across the two lists. Slope and threshold of the psychometric function were computed by first computing a logistic function p is percent correct, x is SNR in dB, and a and b are coefficients obtained from a generalised linear regression model in MATLAB. Slope was the SNR at which p = 50% and was expressed in dB. Threshold was the gradient of the function at p = 50% and was expressed in %-per-dB (Mertes, Johnson, and Dinger 2019).