Academic journal article Attention, Perception and Psychophysics

Salient Stimuli Capture Attention and Action

Academic journal article Attention, Perception and Psychophysics

Salient Stimuli Capture Attention and Action

Article excerpt

Published online: 6 August 2013

© Psychonomic Society, Inc. 2013

Abstract Reaction times in a visual search task increase when an irrelevant but salient stimulus is presented. Recently, the hypothesis that the increase in reaction times was due to attentional capture by the salient distractor has been disputed. We devised a task in which a search display was shown after observers had initiated a reaching movement toward a touch screen. In a display of vertical bars, observers had to touch the oblique target while ignoring a salient color singleton. Because the hand was moving when the display appeared, reach trajectories revealed the current selection for action. We observed that salient but irrelevant stimuli changed the reach trajectory at the same time as the target was selected, about 270 ms after movement onset. The change in direction was corrected after another 160 ms. In a second experiment, we compared manual selection of color and orientation targets and observed that selection occurred earlier for color than for orientation targets. Salient stimuli support faster selection than do less salient stimuli. Under the assumption that attentional selection for action and perception are based on a common mechanism, our results suggest that attention is indeed captured by salient stimuli.

Keywords Reaching . Attentional capture . Saliency . Action

Introduction

Selection of visual information is achieved by attentional prioritization of visual stimuli and may be controlled in different ways. Attentional selection is said to be top-down when it reflects expectations and goals of the observer, and it is said to be bottom-up if it reflects the saliency of the stimuli (reviewed by Theeuwes, 2010). We will briefly present the contingent capture and the additional singleton paradigms, which are believed to provide evidence for top-down and bottom-up control, respectively. However, it should be noted that the two paradigms confound bottom-up and top-down factors to some degree and also involve mechanisms, such as intertrial priming, that defy the theoretical dichotomy (Awh, Belopolsky, & Theeuwes, 2012).

The contingent attentional capture paradigm consists of a target display that is preceded by a cue display (e.g., Folk, Remington, & Johnston, 1992). In separate blocks of trials, observers search the target display for a colored item among white distractors (i.e., a color target), or they look for a single white item (i.e., an onset target). Color targets and onset targets are shown in separate blocks, resulting in an attentional set for color or onset, respectively. Over blocks of trials, the cue display contains either a color cue or an onset cue. It was observed that reaction times (RTs) in a speeded discrimination task are shorter when the cue is shown at the same location as the target, but only when the cue characteristics match the target characteristics (i.e., onset cue/onset target or color cue/color target). In sum, only cues that match the current attentional set capture attention, which is strong evidence for top-down control. However, subsequent studies showed that the repetition of target features contributes to the pattern of results (e.g., Ansorge & Horstmann, 2007), which does not support the notion of top-down control, because trial history is not easily classified as bottom-up or top-down (Awh et al., 2012).

In the additional singleton paradigm developed by Theeuwes (1991), observers search for a shape singleton in a circular array of items and perform a discrimination task. In one variant of the paradigm, the target and distractors change roles randomly from trial to trial (Theeuwes, 1991). That is, a diamond target among circle distractors may be followed by a circle target among diamond distractors, and vice versa. On half of the trials, a salient color singleton is shown at a nontarget location, which increases RTs. Presumably, attention is attracted to the salient color distractor before moving to the less salient shape singleton. …

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.