Guided Search 6.0: An updated model of visual search

Abstract

This paper describes Guided Search 6.0 (GS6), a revised model of visual search. When we encounter a scene, we can see something everywhere. However, we cannot recognize more than a few items at a time. Attention is used to select items so that their features can be "bound" into recognizable objects. Attention is "guided" so that items can be processed in an intelligent order. In GS6, this guidance comes from five sources of preattentive information: (1) top-down and (2) bottom-up feature guidance, (3) prior history (e.g., priming), (4) reward, and (5) scene syntax and semantics. These sources are combined into a spatial "priority map," a dynamic attentional landscape that evolves over the course of search. Selective attention is guided to the most active location in the priority map approximately 20 times per second. Guidance will not be uniform across the visual field. It will favor items near the point of fixation. Three types of functional visual field (FVFs) describe the nature of these foveal biases. There is a resolution FVF, an FVF governing exploratory eye movements, and an FVF governing covert deployments of attention. To be identified as targets or rejected as distractors, items must be compared to target templates held in memory. The binding and recognition of an attended object is modeled as a diffusion process taking > 150 ms/item. Since selection occurs more frequently than that, it follows that multiple items are undergoing recognition at the same time, though asynchronously, making GS6 a hybrid of serial and parallel processes. In GS6, if a target is not found, search terminates when an accumulating quitting signal reaches a threshold. Setting of that threshold is adaptive, allowing feedback about performance to shape subsequent searches. Simulation shows that the combination of asynchronous diffusion and a quitting signal can produce the basic patterns of response time and error data from a range of search experiments.

Keywords: Attention; Bottom-up; Errors; Guided search; Reaction time; Selective attention; Top-down; Visual search; Visual working memory.

Figure 1:

Basic laboratory search tasks and RT x set size graphs.

Figure 2:

A representation of Guided Search 2.0

Figure 3:

A representation of Guided Search 6.0

Figure 4:

Feature search based on bottom-up salience, top-down relations, and top-down identity.

Figure 5:

Density effects in search: Feature differences are easier to detect when items are closer together.

Figure 6:

Look at the star and report the color of ovals.

Figure 7:

A) Which boxes could hide a sheep? B) Find sheep. The scene is on the grounds of Chatsworth House, a stately home in Derbyshire, England

Figure 8:

The search process in GS6.

Figure 9:

The GS6 search termination process

Figure 10:

Results of a simulation of the aspects of GS6, illustrated in Figure 9: A: RT x set size functions (c.f. Fig 1f), B: Error rates

Figure 11:

RT distributions: A) Data from Wolfe, Palmer, & Horowitz (2010). B) GS6 simulation data. Each distribution represents one set size. Lighter curves are smaller set sizes (the four set sizes are 5, 10, 15, & 20, prevalence is 0.5). Green shows target-present. Purple shows target-absent.

Figure 12:

Simulation of prevalence effects. A) RT as a function of prevalence, B) D’ and criterion, ’c’ as function of prevalence, C) ROC derived from variation in prevalence. Blue number values within the graph show prevalence associated with each datapoint. D) zROC derived from variation in prevalence. See the supplement to Wolfe & VanWert (2010) for comparison data.

Figure 13:

Simulation of different miss error goals. A) Quitting thresholds as a function of time. Each color represents a different Error Goal from 3% (red-top) to 15% (yellow-bottom). Lower functions in 13A are the diffuser starting point values that produce false alarms. B) Error rates as a function of Error Goal. Each function is for a different set size: Top (blue) line = set size 20, teal=15, green=10, brown=5.

Figure 14:

Three different types of functional visual fields (FVF) that need to be considered in visual search. A) Resolution FVF, B) Exploratory FVF, C) Attentional FVF

Figure 15:

Two templates in visual search: A guiding template and a target template.