To characterize contrast detection, noise should be extended, not localized

Short-term monocular pattern deprivation reduces the internal additive noise of the visual system

Previous studies have shown that short-term monocular pattern deprivation can shift perceptual dominance in favor of the deprived eye. However, little is known about the effect of monocular pattern deprivation on contrast sensitivity (CS) and its corresponding mechanisms. Here, contrast sensitivity function (CSF) in the nondominant eye of normal subjects was evaluated before and after 150 min of monocular pattern deprivation. To obtain a CSF with high precision and efficiency before deprivation effect washout, a quick CSF (qCSF) method was used to assess CS over a wide range of spatial frequencies and at two external noise levels. We found that (1) monocular pattern deprivation effectively improved the CS of the deprived eye with larger effect on high spatial frequencies, (2) CS improvement only occurred when external noise was absent and its amount was spatial frequency dependent, and (3) a perceptual template model (PTM) revealed that decreased internal additive noise accounted for the mechanism of the monocular pattern derivation effect. These findings help us better understand the features of short-term monocular pattern deprivation and shed light on the treatment of amblyopia.

Variance-dependent neural activity in an unvoluntary averaging task

Ensemble statistics of a visual scene can be estimated to provide a gist of the scene without detailed analysis of all individual items. The simplest and most widely studied ensemble statistic is mean estimation, which requires averaging an ensemble of elements. Averaging is useful to estimate the mean of an ensemble and discard the variance. The source of variance can be external (i.e., variance across the physical elements) or internal (i.e., imprecisions in the estimates of the elements by the visual system). The equivalent noise paradigm is often used to measure the impact of the internal variance (i.e., the equivalent input noise). This paradigm relies on the assumption that the averaging process is equally effective independently of the main source of variance, internal or external, so any difference between the processing when the main source of variance is internal or external must be assumed not to affect the averaging efficiency. The current fMRI study compared the neural activity when the main variance is caused by the stimulus (i.e., high variance) and when it is caused by imprecisions in the estimates of the elements by the visual system (i.e., low variance). The results showed that the right superior frontal and left middle frontal gyri can be significantly more activated when the variance in the orientation of the Gabors was high than when it was low. Consequently, the use of the equivalent noise paradigm requires the assumption that such additional neural activity in high variance does not affect the averaging efficiency.

Aging affects gain and internal noise in the visual system

Visual functions decline with age, but how aging degrades visual functions remains controversial. In the current study, the mechanisms underlying age-related visual declines were examined psychophysically. We developed an efficient method to quickly explore contrast sensitivity as a function of nine spatial frequencies at three levels of external noise in both young and old subjects. Fifty-two young and twenty-six old subjects have been screened for ophthalmological and mental diseases and participated in the experiment. Contrast sensitivity varied significantly with spatial frequency, age, and level of external noise. By adopting a nonlinear observer model, we decomposed contrast sensitivity into inefficiencies in internal additive noise, internal multiplicative noise, perceptual template gain, and/or system non-linearity. Model analysis revealed that aging impacts both internal additive noise and perceptual template gain, and such age-related degradation is tuned to spatial frequency, which is also a good predictor to discriminate old from young. The quick characterization of contrast sensitivity functions at different noise levels and the accompanying analysis developed in the current study may have profound application in other clinical populations.

Internal noise sources limiting contrast sensitivity

Contrast sensitivity varies substantially as a function of spatial frequency and luminance intensity. The variation as a function of luminance intensity is well known and characterized by three laws that can be attributed to the impact of three internal noise sources: early spontaneous neural activity limiting contrast sensitivity at low luminance intensities (i.e. early noise responsible for the linear law), probabilistic photon absorption at intermediate luminance intensities (i.e. photon noise responsible for de Vries-Rose law) and late spontaneous neural activity at high luminance intensities (i.e. late noise responsible for Weber’s law). The aim of this study was to characterize how the impact of these three internal noise sources vary with spatial frequency and determine which one is limiting contrast sensitivity as a function of luminance intensity and spatial frequency. To estimate the impact of the different internal noise sources, the current study used an external noise paradigm to factorize contrast sensitivity into equivalent input noise and calculation efficiency over a wide range of luminance intensities and spatial frequencies. The impact of early and late noise was found to drop linearly with spatial frequency, whereas the impact of photon noise rose with spatial frequency due to ocular factors.

Maximizing noise energy for noise-masking studies

Noise-masking experiments are widely used to investigate visual functions. To be useful, noise generally needs to be strong enough to noticeably impair performance, but under some conditions, noise does not impair performance even when its contrast approaches the maximal displayable limit of 100 %. To extend the usefulness of noise-masking paradigms over a wider range of conditions, the present study developed a noise with great masking strength. There are two typical ways of increasing masking strength without exceeding the limited contrast range: use binary noise instead of Gaussian noise or filter out frequencies that are not relevant to the task (i.e., which can be removed without affecting performance). The present study combined these two approaches to further increase masking strength. We show that binarizing the noise after the filtering process substantially increases the energy at frequencies within the pass-band of the filter given equated total contrast ranges. A validation experiment showed that similar performances were obtained using binarized-filtered noise and filtered noise (given equated noise energy at the frequencies within the pass-band) suggesting that the binarization operation, which substantially reduced the contrast range, had no significant impact on performance. We conclude that binarized-filtered noise (and more generally, truncated-filtered noise) can substantially increase the energy of the noise at frequencies within the pass-band. Thus, given a limited contrast range, binarized-filtered noise can display higher energy levels than Gaussian noise and thereby widen the range of conditions over which noise-masking paradigms can be useful.