On peripheral and central processes in vision: Inferences from an information-processing analysis of masking with patterned stimuli.

Abstract

The masking of briefly exposed letter forms by a preceding or succeeding stimulus may originate in either peripheral or central visual mechanisms. The question of how masking varies with origin was examined in a series of experiments which made use of stimuli that masked the target forms only monoptically (or binocularly), or both monoptically and dichoptically. Peripheral forward and backward masking were described by a simple relation between target stimulus energy and the minimal interval between target offset and mask onset permitting evasion of masking: the minimal interval multiplied by the target energy equals a constant. Peripheral forward masking, however, was more sensitive to mask intensity than was peripheral backward masking. On the other hand, central masking, which was primarily backward, was relatively unaffected by stimulus energy and was determined by the interval elapsing between the onsets of the two stimuli. The multiplicative rule and the onset-onset rule characterized, respectively, peripheral and central visual processes. The peripheral processes were viewed as a set of parallel systems or nets signaling crude features of the stimulus and the central processes as a series of decisions conducted, in part, on these features and resulting in stimulus recognition. It was hypothesized that the peripheral and central processes were related in a concurrent and contingent fashion: apparently, the two occur in parallel, with the central decisions contingent on the output of the peripheral systems which signal different features at different rates.Perceptual interference results when two stimuli are delivered to an observer in rapid succession. The term "forward masking"

A model of character recognition and legibility.

“…The preceding filtering step is a linear transformation of the stimulus, but the square-root transformation is nonlinear, causing a greater fractional attenuation of the higher intensity values in the filtered output than of the lower intensities. The decision to include this nonlinear step was made because spatial masking of targets by uniform and structured backgrounds is a common finding in both target detection and target recognition paradigms, both for touch and vision (Breitmeyer & Ganz, 1976;Craig, 1974Craig, , 1982Eriksen, 1966;Legge & Foley, 1980;Turvey, 1973). Although there may be multiple causes of masking for recognition of targets on structured backgrounds, masking of a target on a uniform background in a detection task virtually requires some sort of compressive nonlinearity (see Legge & Foley, 1980).…”

The use of a visual mask may seriously confound your experiment

“…The evidence for integration theories is impressive , 1970;Turvey, 1973). In these accounts, integration masking is recognized as occurring at shorter stimulus onset asynchronies (as long as 150 msec) and interruption masking as occurring at longer asynchronies.…”

“…Only the former is reasonably compatible with the assumption that a mask, following stimulation, effectively limits the time available for stimulus processing. Interruption theories, of which there are several versions (Spencer & Shuntich, 1970;Turvey, 1973), have in common the assumption that a second arriving stimulus takes priority in processing and thus diverts the processing mechanisms from the first stimulus to the mask.…”

Picture perception: Effects of luminance on available information and information-extraction rate.

“…In some masking experiments, target luminance has been varied (Turvey, 1973;Cheatham, 1952;Kinsbourne & Warrington, 1962a, 1962b and in others, mask luminance has been varied (Eriksen, 1966;Eriksen & Lappin, 1964;Loftus & Ginn, 1984;Spencer & Shuntich, 1970;Thompson, 1966). Irrespective of whether it is target or mask energy that is varied, performance has been found to increase with increases in target-mask energy ratio.…”