Mechanisms of attentional priority.

Abstract

When a single abrupt onset occurs in a multielement visual display, it captures attention, possibly by generating an attentional interrupt that designates onsets as being of high priority. In 3 experiments, the mechanisms subserving attentional priority setting were investigated. Subjects searched for a prespecified target letter among multiple distractor letters, half of which had abrupt onsets and half of which did not. The target, when present, was equally often an onset element and a no-onset element. Several models for attentional priority, differing in how many onset elements have priority over no-onset elements, were assessed. The data support a model in which approximately 4 onset stimuli are processed before any no-onset stimuli are processed. Two attentional priority mechanisms are proposed: (a) queuing of a limited number of high-priority elements and (b) temporally modulated decay of attentional priority tags.

Blink and Shrink: The Effect of the Attentional Blink on Spatial Processing.

“…This result is important because (a) it shows that the attentional blink was over by the time the search display appeared, and thus, the blink did not affect the search directly but solely the processing of the cues; and (b) it justifies keeping the RT intercept (I) and slope (b) estimates the same for both the short and long lag conditions when fitting the model (see Method section above). Yantis and Johnson's (1990) model appears to provide a good fit for the data. The spatial capacity estimates derived from this model suggest that the number of spatial cues being processed was halved under attentionally demanding conditions.…”

Spatiotemporal Segregation in Visual Search: Evidence From Parietal Lesions.

“…The present study offers additional validation of Yantis and Johnson's (1990) serial search model. The model was developed to estimate how many abrupt new onsets would be prioritized for selection in their attentional capture paradigm.…”

“…Note that the model offers a rather straightforward account of visual search, containing a strong serial component. Visual attention seems to first select those locations that have been marked by a cue (as here) or a rapid onset (as in Yantis and Johnson, 1990) before it returns to the remainder of the items. I return to this model in the General Discussion, below.…”

“…If participants could successfully perceive and retain the locations indicated by the cues, then this should have benefited subsequent search reaction times (RTs). The observers' capacity to use the spatial cues in search was then estimated from the RTs on the basis of a relatively simple serial search model proposed by Yantis and Johnson (1990;see below). The crucial question then was how this spatial capacity varied depending on whether the cues were presented inside or outside the blink.…”

Object substitution masking: When does mask preview work?

“…More recently, Tata and Giaschi (2004) proposed that OSM occurs when attention is selectively deployed to the masking object. The visual system fundamentally assigns priority tags to objects in a scene (Yantis & Jonides, 1984), and attention is selectively deployed to high-priority objects (new items in the display) before low-priority objects (older items) (Yantis & Johnson, 1990). Under this view, OSM occurs when the novel masking object appears simultaneously with the target.…”