Visual search in peripheral vision: learning effects and set-size dependence.

Feature search for a light bar with one orientation (or color) embedded in an array of bars with a very different orientation (or color) is quick, easy and independent of the number of array elements. In contrast, search for a conjunction target has a linear response time dependence on the number of distractors. Training can improve performance of both these tasks. We report that these properties may not be valid for eccentric stimulus presentation. In general, the two hemifields are not equally suited to search, and training is most effective in the weaker hemifield. In addition, the feature-search independence of set-size may not always be valid for stimulus arrays that are presented peripherally. Subjects were tested on orientation and color feature tasks, and on orientation-color conjunction search with 3 array sizes presented at fixation or eccentrically in the right or left hemifield. During a second testing session, improvement was so much greater for the non-preferred hemifield that sometimes the preference was switched. Surprisingly, preferred hemifield performance actually declined for some subjects. Thus, the hemifield preference effect seems related to competition, and perhaps an automatic attention-directing mechanism. We confirmed the central presentation set-size independence for feature search but found a great difference between large and small arrays when presentation was lateral. There are two sources of this array size effect: 1. Target eccentricity, demonstrated by comparing performance for different target locations with the same array size. 2. Target location uncertainty, seen by comparing performance for different size arrays when the target elements appeared at the same locations. Training also affected the array-size dependence, changing search performance from set-size dependent to independent or vice versa at the point of greatest training effect.

Contrast dependence of perceptual grouping in brain-damaged patients with visual extinction.

Extinction is manifested in conditions of bilateral simultaneous stimulation, as a failure to detect the stimulus contra-lateral to the side of a cerebral lesion, while the same stimulus is correctly detected there when presented in isolation. The phenomenon is usually interpreted in terms of impaired mobilization of attention from an attended to an unattended object. We have recently shown, using pairs of Gabor patches as stimuli, that pair detection is maximally improved in conditions where the two stimuli presented simultaneously to the two halves of the visual field are co-oriented and co-axial and their location is not too eccentric. Here we add new information by showing that contrast isotropy of the stimulus pair is important in producing this orientation-similarity gain. The further advantage of co-oriented co-linear stimuli over co-oriented parallel (vertical) stimuli was shown exclusively with iso-contrast stimulus pairs, and was significantly enhanced when the contrast level of the stimulus pair was low. Stimulus properties producing reduced extinction seem to correlate with the selectivity pattern and contrast dependence of (a) spatial lateral facilitation observed in psychophysical studies with normal observers, and (b) long-range interactions observed in the primary visual cortex. Thus, two remote visual stimuli seem to be processed as a single object when the corresponding neuronal activities are linked via long-range lateral interactions. The present demonstration of contrast dependency in such processing, strengthens our previous conjecture that even in the presence of significant, extinction producing, parietal damage, the primary visual cortex preserves the capacity to encode, using long-range lateral interactions, an image description in which visual objects are already segregated from background.

Visual extinction and cortical connectivity in human vision.

Visual extinction is a common, poorly understood, consequence of unilateral cerebral damage, where a patient fails to detect one of two simultaneously presented stimuli (the one more contralateral to the lesion), despite the fact that each stimulus is correctly detected when presented in isolation. The phenomenon implies a failure of shifting attention from an attended object to an unattended one. We show here that pair detection is improved in conditions where the two stimuli presented to the two halves of the visual field are proximal, co-oriented and co-axial. It is further shown that stimulus properties producing reduced extinction correlate with the selectivity pattern of spatial lateral interactions observed in the primary visual cortex. We suggest that neuronal activity in early stages of cortical visual processing encodes, using long-range lateral interactions, an image description in which visual objects are already segmented and marked. Segmentation seems to function properly even in the presence of significant destruction of the parietal cortex leading to extinction.

Coordinate frame for pattern recognition in unilateral spatial neglect.

The present research examines the effect of spatial (object-centered) attentional constraints on pattern recognition. Four normal subjects and two right-hemisphere-damaged patients with left visual neglect participated in the study. Small, letterlike, prelearned patterns served as stimuli. Short exposure time prevented overt scanpaths during stimulus presentation. Attention was attracted to a central (midsagittal) hation point by precuing this location prior to each stimulus presentation. Minute (up to 1.5° of visual angle) rightward and leftward stimulus shifts caused attention to be allocated each time to a different location on the object space, while remaining in a fixed central position in viewercentered coordinates. The task was to decide which of several prelearned patterns was presented in each trial. In the normal subjects, best performance was achieved when the luminance centroid (LC; derived from the analysis of low-spatial frequencies in the object space) of each pattern coincided with the spatial position of the precue. In contrast, the patients with neglect showed optimal recognition performance when precuing attracted attention to locations within the object space, to the left of the LC. The normal performance suggests that the LC may serve as a center of gravity for attention allocation during pattern recognition. This point seems to be the target location where focal attention is normally directed, following a primary global analysis based on the low spatial frequencies. Thus, the LC of a simple pattern may serve as the origin point for an object-centered-coordiate-frame (OCCF), dividing it into right and left. This, in turn, serves to create a prototype description of the pattern, in its own coordinates, in memory, to be addressed during subsequent recognition tasks. The best match of the percept with the stored description may explain the observed advantage of allocating attention to the LC. The performance of the brain- damaged patients can be explained in terms of neglect operating in the OCCE.

Facilitation of pattern recognition by cuing foveation with the luminance centroid as origin of the frame of reference.

The strategy for visual information processing must vary with the specific situation. We assume that in recognition of pre-learnt letter-like patterns under time-pressure conditions, mechanisms of selective attention are involved. We propose that, with simple stimuli, foveation is to the luminance centroids of such patterns, and if normally the latter has to be computed by the brain, cuing such information should improve performance. This assumption was tested on three subjects with five stimuli. In confirmation of the working hypothesis, we found significant improvements in performance (P = 0.01 or better) for each of the stimuli, with cuing to 'relevant' as compared with 'irrelevant' sites, and also when the point of foveation was nearer to the 'nodal' site. It is concluded that in pattern recognition processes, nodal regions are computed, the luminance centroid for example, for simple pattern stimuli.