Visual scan adaptation during repeated visual search

There is no consensus as to how to characterize eye fixations during visual search. On the one hand, J. M. Wolfe, G. A. Alvarez, and T. S. Horowitz (2000) have described them as a haphazard sequence of fixations. On the other hand is research that shows systematic repetition of visual patterns when freely viewing a scene (T. Foulsham & G. Underwood, 2008; D. Noton & L. W. Stark, 1971a). Two experiments are reported that demonstrate the repetition and adaptation of visual scans during visual search, supporting an adaptive scanning hypothesis. When trials were repeated in a simple search task, visual scan similarity and search efficiency increased. These increments in similarity and efficiency demonstrate the systematic and adaptive nature of visual scans to the characteristics of the visual environment during search.     

The psychophysics of visual search

Most theories of visual search emphasize issues of limited versus unlimited capacity and serial versus parallel processing. In the present article, we suggest a broader framework based on two principles, one empirical and one theoretical. The empirical principle is to focus on conditions at the intersection of visual search and the simple detection and discrimination paradigms of spatial vision. Such simple search conditions avoid artifacts and phenomena specific to more complex stimuli and tasks. The theoretical principle is to focus on the distinction between high and low threshold theory. While high threshold theory is largely discredited for simple detection and discrimination, it persists in the search literature. Furthermore, a low threshold theory such as signal detection theory can account for some of the phenomena attributed to limited capacity or serial processing. In the body of this article, we compare the predictions of high threshold theory and three versions of signal detection theory to the observed effects of manipulating set size, discriminability, number of targets, response bias, external noise, and distractor heterogeneity. For almost all cases, the results are inconsistent with high threshold theory and are consistent with all three versions of signal detection theory. In the Discussion, these simple theories are generalized to a larger domain that includes search asymmetry, multidimensional judgements including conjunction search, response time, search with multiple eye fixations and more general stimulus conditions. We conclude that low threshold theories can account for simple visual search without invoking mechanisms such as limited capacity or serial processing.     

Situating visual search

Visual search attracted great interest because its ease under certain circumstances seemed to provide a way to understand how properties of early visual cortical areas could explain complex perception without resorting to higher order psychological or neurophysiological mechanisms. Furthermore, there was the hope that properties of visual search itself might even reveal new cortical features or dimensions. The shortcomings of this perspective suggest that we abandon fixed canonical elementary particles of vision as well as a corresponding simple to complex cognitive architecture for vision. Instead recent research has suggested a different organization of the visual brain with putative high level processing occurring very rapidly and often unconsciously. Given this outlook, we reconsider visual search under the broad category of recognition tasks, each having different trade-offs for computational resources, between detail and scope. We conclude noting recent trends showing how visual search is relevant to a wider range of issues in cognitive science, in particular to memory, decision making, and reward.     

The use of visual search to assess attention

Background: Under some conditions, the time required for a visual search increases with the number of elements to be searched. It has been suggested that the overall search time reflects the duration that attention is devoted to each element multiplied by the number of elements. On this basis, it has been proposed that visual search time can be used as a measure of attention capability in dyslexic readers. However, there is evidence to suggest that the search time reflects task difficulty rather than attentional factors. Many dyslexic readers suffer from various sensory deficits. These deficits would effectively increase task difficulty for these readers. Here we use computer simulations to investigate the potential effects of sensory deficits on visual search. Method: Visual search was modelled on a computer within the framework of signal detection theory as a matter of detecting a noisy signal from a series of noisy distractors. Sensory deficits were modelled as decreased discriminability. Results: Consistent with previous observations, we find that discriminability, which decreases with the number of distractors, may have a substantial effect on the search time. With regard to the effects of sensory deficits, we find that under low discriminability conditions, small sensory deficits may cause pronounced increases in search time. Conclusion: The finding that small sensory deficits may cause pronounced increases in search time makes it specifically problematic to use visual search to test attention in individuals who suffer from sensory deficits. This applies particularly to dyslexic individuals, many of whom have been shown to suffer from visual deficiencies.     

The control of saccadic adaptation: implications for the scanning of natural visual scenes

Accurate scanning of natural scenes depends on: (1) attentional selection of the target; (2) spatial pooling over the attended target to compute the precise landing position; and (3) adaptive modification of saccades to ensure saccadic accuracy. The present experiments studied adaptation. Adaptive modifications were induced by displacing the target during saccades. Adaptation was found to be: (1) similar for a small target point and a large target circle, despite the differences in the spatial pattern of landing position errors for each; (2) unaffected by instructions to look part way to the target, even though such instructions altered landing position error relative to the target; and (3) insensitive to symbolic cues disclosing the direction of the intra-saccadic displacement. Briefly delaying the presentation of the post-saccadic target greatly reduced adaptation. Neither corrective saccades, nor the position errors that trigger corrections, were involved in adaptation because corrective saccades rarely occurred with a large target circle even though the circle produced as much adaptation as the single point. Taken together, the results do not support the traditional notion that post-saccadic retinal position error controls adaptation. We propose that adaptation relies on a comparison of the actual post-saccadic retinal image with the post-saccadic image that would be predicted based on a representation of the planned saccade. Such a comparison: (1) is consistent with our results; (2) may be more effective than retinal position error in controlling adaptation in natural visual scenes containing large targets and backgrounds; and (3) is similar to the motion-based adaptive mechanisms associated with the VOR. Similarity between the adaptive control of saccades and adaptive control of the VOR raises the possibility that the most important role of saccadic adaptation may be the coordination of eye and head movements during shifts of gaze.     

Light adaptation for visual pattern recognition in flies

Flies were found to possess the light adaptation phenomena of vertebrates whereby the eyes achieve optimal sensitivity to an intensity increment on a given background illumination. This was found in the photoreceptors and several higher levels including a behavioral response. It has a time constant varying directly with the sudden change of background intensity that ranges up to about 20 sec for large changes in background. This was distinguished from another adaptation phenomena of about 6 min duration, part of which appears to be due to a change in ommatidial fields produced by shielding pigment migration effects.     

The effects of orientation-specific adaptation on the duration of short-term visual storage

The hypothesis that the perceived duration of a briefly presented stimulus could be modified as a result of adapting subjects to square-wave gratings of a particular orientation was tested in two experiments. Perceived durations to horizontal and vertical gratings were estimated by having subjects judge when the intermittent stimuli were experienced as continuously present. In experiment 1, it was found that the apparent duration of a grating decreased following adaptation to a grating of the same orientation, and increased after adaptation to gratings perpendicular in orientation to the test stimulus. The interocular transfer of the effect was demonstrated in experiment 2. These results suggest that the outputs of cortical orientation-sensitive populations of neural units are involved in duration judgments and the features available in short-term visual storage are based on these cortical populations.     

Light adaptation and visual latency

The temporal resolving properties of the eye are considered in terms of binocular latency differences. Two experiments demonstrate that temporal resolution is a function of visual adaptation. The effects of target-background contrast on visual latencies appear to be consistent with the effects of field adaptation. Thus, it is suggested that the mechanism of visual adaptation is sufficient to account for the relation between intensity and latency.     

An invariant for timing of saccades during visual search

The variable latency of a saccade to the onset of a single target reveals our brain's hypothesis testing about the target's presence. Search in complex scenes involves multiple objects that compete to become fixated. The initiation of a saccade in this case involves two hypotheses: (1) a potential target is present outside the fovea and (2) the currently fixated object is not the target. Previous models suggest that these hypotheses are evaluated independently, each involving a decision signal that races towards threshold. We show here that the skewed latency distributions during search comply with strong competition between these decision signals rather than independence. Moreover, the thresholds for the two competing processes are not independent either but conform to an invariant that suggests that saccades in complex scenes are made when the odds for the target's presence outside the fovea versus within the fovea are about four.     

The human visual evoked cortical potential and dark adaptation

The relationship between the amplitude of the human visual evoked cortical potential (VECP) and dark adaptation was examined in three normal subjects. In addition, the VECP dark adaptation functions were compared with psychophysical functions obtained in the same sessions under identical stimulus conditions. In order to separate the rod and cone systems two chromatic stimuli were used, one from the short wavelength and the other from the long wavelength portion of the visible spectrum. The results indicate that the VECP is a sensitive correlate of both rod and cone adaptation since the VECP dark adaptation changes parallel the psychophysical threshold changes. It was concluded that the VECP is an excellent tool for investigating both cone and rod dark adaptation when the appropriate methods and conditions for study are used.     

Color in visual search.

Colored targets pop out of displays under conditions in which the standard red-green, yellow-blue and black-white mechanisms cannot directly mediate detection. Experimental evidence suggests that observers possess chromatic detection mechanisms tuned to intermediate hues such as orange as well as to hues characterizing the standard color-opponent mechanisms and that these mechanisms, as a group, form a fine-grained representation of hue within the central visual field. Spatially-parallel search is mediated by a single such mechanism that is spectrally sensitive to the target chromaticity but insensitive to the distractor chromaticities; different mechanisms are used to detect a single target in a way that depends on distractor chromaticities.     

The effect of target foreknowledge on visual search for categorically separable orientation targets

This research note assesses the role of target foreknowledge in visual search for categorically defined orientation targets, as first described by Wolfe et al. [Wolfe, J. M., Friedman-Hill, S. R., Stewart, M. I., & O'Connell, K. M. (1992). The role of categorisation in visual search for orientation. Journal of Experimental Psychology: Human Perception and Performance, 18, 34-49]. We compared search with known versus unknown (respond to the odd item) targets. An RT advantage for categorical search only emerged with known targets. The evidence points to an important role for top-down processes in search for categorically--defined orientation targets.     

Distance-from-target dynamics during visual search

Tseng, Y. C., & Li, C. S. (2004). Oculomotor correlates of context-guided learning in visual search. Perception & Psychophysics, 66, 1368-1378 noted that visual search with eye movements may be characterized by a search phase in which fixations do not move towards the target, followed by a phase in which fixations move steadily towards the target. They speculated that the phases are related to memory and recognition processes. Human visual search and Monte Carlo simulations are described towards an explanation. Distance-from-target dynamics were demonstrated to be sensitive to geometric constraints and therefore do not provide a solution to the question of memory in visual search. Finally, it is concluded that the specific distance-from-target dynamics noted by Tseng, Y. C., & Li, C. S. (2004). Oculomotor correlates of context-guided learning in visual search. Perception & Psychophysics, 66, 1368-1378 are parsimoniously explained by random walks that were initialized at the centre of their stimulus displays.     

A technique for estimating the contribution of photomechanical responses to visual adaptation

This paper presents a technique for isolating and quantifying the contribution of photomechanical responses to visual adaptation. The technique is developed in the context of the pupillary mechanism that is active in the primary visual cells of the insect eye. Using a feedback-control model to represent the combination of retinular cell and pupil, it is shown that the effect of pupillary movements on retinal illumination can be inferred by analysing an intensity-response function of the pupil. When the technique is applied to the pupils of the butterfly and the fly, the results indicate that, in each case, the pupil decreases retinal illumination by approximately 0.7 log units when the ambient light level is increased from pupillary threshold to a level 2.5 log units higher. The validity of the technique is examined by applying it to the human pupil. The results predict changes of retinal illumination which are in close agreement with those expected on the basis of changes in iris diameter, including the Stiles-Crawford effect. The procedure presented here is simple and can, in principle, be applied to many forms of photomechanical adaptation.     

A Bayesian model for efficient visual search and recognition

Humans employ interacting bottom-up and top-down processes to significantly speed up search and recognition of particular targets. We describe a new model of attention guidance for efficient and scalable first-stage search and recognition with many objects (117,174 images of 1147 objects were tested, and 40 satellite images). Performance for recognition is on par or better than SIFT and HMAX, while being, respectively, 1500 and 279 times faster. The model is also used for top-down guided search, finding a desired object in a 5×5 search array within four attempts, and improving performance for finding houses in satellite images.     

The eye movements of dyslexic children during reading and visual search: impact of the visual attention span

The eye movements of 14 French dyslexic children having a VA span reduction and 14 normal readers were compared in two tasks of visual search and text reading. The dyslexic participants made a higher number of rightward fixations in reading only. They simultaneously processed the same low number of letters in both tasks whereas normal readers processed far more letters in reading. Importantly, the children's VA span abilities related to the number of letters simultaneously processed in reading. The atypical eye movements of some dyslexic readers in reading thus appear to reflect difficulties to increase their VA span according to the task request.