Role of noise in image processing by the human perceptive system

Two psychophysics experiments are described, pointing out the significant role played by stochastic resonance in recognition of capital stylized noisy letters by the human perceptive apparatus. The first experiment shows that an optimal noise level exists at which the letter is recognized for a minimum threshold contrast. A simple two-parameter model that best fits the experimental data is also discussed. In the second experiment we show that a dramatically increased ability of the visual system in letter recognition occurs in an extremely narrow range of increasing noise. Possible interesting future investigations suggested by these experimental results and based on functional imaging techniques are discussed.

Ambiguity and structural information in the perception of reversible figures

The perspective reversals elicited by a set of drawings based on the Mach truncated pyramid are examined. We obtained each pattern of the set from the previous one by adding to it some graphic cues, which were easily integrated into one of the two competing interpretations, thus reducing step by step the ambiguity of the basic pattern. The phenomenological model, proposed to link the mean times of both alternative interpretations with their complexities, is in close agreement with our experimental data. Furthermore, different aspects of such data are well described by the model equations: The measure of the prevalence of the supported interpretation is well correlated with the difference in complexity between the two alternative interpretations; the two different trends of the mean time of the unfavored interpretation, found in the data obtained from different observers as a function of the various patterns of the set, are well fitted by the model without the need for any specific additional hypothesis.

Perceptual ambiguity and stability of reversible figures

In this work, the results of two experiments on ambiguous patterns are reported, which have been obtained by presenting a series of stimuli designed, in both cases, to reduce gradually the ambiguity of the patterns. Such reduction has been performed by respectively increasing or decreasing the amount of graphic details in the experiments. Data of both experiments show a lengthening of mean reversal time. The increase in the stability of one percept can be regarded as associated with the increasing difficulties encountered by an observer in organizing and restating the alternative "hypothesis" through the perceptual mechanisms. The loss of balance in the durations of percepts is discussed in terms of their different informational contents. Finally, in Exp. 1 an analysis is made to evaluate to what extent an addition of perceptual cues, designed to reinforce a three-dimensional interpretation of a pattern, can influence its figure-ground alternation.

Effect of luminance on perceptual alternation of ambiguous patterns

The perceptual alternation of the Necker cube, in its stationary phase, is studied as a function of the stimulus luminance down to the extreme condition in which the cones are completely inactivated so that the pattern of neural excitation reaching the primary visual cortex is very different from the normal one. No evident effect of luminance was found under passive observation either on the reversal rate or on the features of data distributions. Moreover, the complementary situation in which perception is based only on the cones does not affect the phenomenon either. These results permit the assumption that the perceptual alternation in its stationary phase, is a passive and automatic process, that is affected neither by a satiation of the first levels of the visual system nor by the psychological attitude of the subject consequent on dazzling the retina.

Effects of visual angle on perspective reversal for ambiguous patterns

Reversal rates of an ambiguous figure (the Necker cube) were studied for different pattern sizes covering a range of visual angles theta from approximately 1 to 62 deg. A large number of reversals was obtained for each observer and each pattern in order to examine the statistical distributions of reversal times. A pronounced flattening of the statistical distributions (represented throughout by a gamma distribution) and a growth of the mean duration of each percept, with increasing pattern size was found. A plateau in the range of theta between 5 and 20-30 deg was observed. For larger values of theta two kinds of observers have been identified: for 'fast' observers the inversion rate is little affected by theta, whilst for 'slow' observers, the mean reversal time increases strongly with theta. A tentative model, based on three different contributions to the duration of the alternation process, is proposed: a constant term, independent of theta, and two terms dependent on theta--a retinal term, and a cortical one. The last term is interpreted as due to the spreading of excitation with the characteristic of a filling-in process.