Academic journal article Perception and Psychophysics

Perceptual Learning in Monocular Pattern Masking: Experiments and Explanations by the Twin Summation Gain Control Model of Contrast Processing

Academic journal article Perception and Psychophysics

Perceptual Learning in Monocular Pattern Masking: Experiments and Explanations by the Twin Summation Gain Control Model of Contrast Processing

Article excerpt

We investigated practice effects on contrast thresholds for target patterns. Results showed that practice decreased contrast thresholds when targets were presented on maskers. Thresholds tended to decrease more at the higher end of the masker contrast range. At least partially, learning transferred to stimuli of the untrained phase. We simulated changes in threshold versus contrast functions using a contrast-processing model and then fit the model to pre- and posttraining data. The simulation results and model fit suggest that learning in pattern masking can be accounted for by changes in nonlinear transducer functions for divisive inhibitory signals.

Perceptual learning refers to "any relatively permanent and consistent change in the perception of a stimulus array following practice or experience with this array" (Gibson, 1963, p. 29). Since the 1970s, many studies have examined perceptual learning based on simple stimuli and tasks (for reviews, see Ahissar & Hochstein, 1998; Fahle & Poggio, 2002). These studies investigated stimulus change in relation to various dimensions, including orientation, retinal position, spatial frequency, and the eye to which stimuli were presented. For example, Ball and Sekuler (1987) reported that the ability to discriminate similar directions of motion gradually improved with practice, but also noted that this improvement did not transfer to the direction opposite that of the training condition. Shiu and Pashler ( 1992) reported that the ability to discriminate the orientation of a pair of straight lines improved with practice; however, this improvement did not transfer to any untrained retinal position. Results from several similar studies also showed that improvements were specific to simple features of stimuli, suggesting that an aspect of perceptual learning reflects changes only in the low-level perceptual processing by which fundamental properties of visual stimuli are processed. Luminance contrast is considered one such fundamental property.

Luminance contrast affects performance in various visual tasks, such as the discrimination of motion speed (Stone & Thompson, 1992). It is relevant, therefore, to consider how contrast processing changes with practice. Several studies have shown that contrast thresholds for target patterns decreased with practice when target gratings were presented on masker gratings (Swift & Smith, 1983), when Gabor targets were presented on identical patterns (Yu, Klein, & Levi, 2004), ' when Gabor targets were presented on identical patterns with flanking Gabor maskers (Adini, Sagi, & Tsodyks, 2002), and when Gabor targets were presented on a combination of two Gabor maskers, the orientations of which were different from the target's orientation (Dorais & Sagi, 1997; Zenger & Sagi, 1996). As for learning on contrast detection thresholds (with no masker), two studies reported that contrast detection thresholds decreased with practice (Sowden, Rose, & Davies, 2002; Yu et al., 2004), whereas two other studies found no learning effect (Dorais & Sagi, 1997; Zenger & Sagi, 1996).

Previous studies have used different training durations. Swift and Smith ( 1983) trained observers for about 15 trials, whereas others trained observers over several days (Adini et al., 2002; Dorais & Sagi, 1997; Yu et al., 2004). There appear to be at least two types of learning: within-day and interday learning. Within-day learning occurs during a session lasting several minutes or hours, whereas interday learning gradually improves performance over the course of several days. The present study focused on interday learning only.

Contrast thresholds can be plotted as functions of the contrast of maskers and pedestals. Such functions are called threshold versus contrast (TvC) functions. TvC function has been found to have a dipper shape; that is, thresholds are lower at the lower ends of masker contrast ranges, but elevated at the higher ends of masker contrast ranges (Legge & Foley, 1980; Nachmias & Sansbury, 1974; Stromeyer & Klein, 1974). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.