Facilitating recognition of crowded faces with presaccadic attention

Benjamin A Wolfe,David Whitney

doi:10.3389/fnhum.2014.00103

Benjamin A Wolfe, David Whitney

Open Access

https://doi.org/10.3389/fnhum.2014.00103

Copy DOI

Abstract

In daily life, we make several saccades per second to objects we cannot normally recognize in the periphery due to visual crowding. While we are aware of the presence of these objects, we cannot identify them and may, at best, only know that an object is present at a particular location. The process of planning a saccade involves a presaccadic attentional component known to be critical for saccadic accuracy, but whether this or other presaccadic processes facilitate object identification as opposed to object detection—especially with high level natural objects like faces—is less clear. In the following experiments, we show that presaccadic information about a crowded face reduces the deleterious effect of crowding, facilitating discrimination of two emotional faces, even when the target face is never foveated. While accurate identification of crowded objects is possible in the absence of a saccade, accurate identification of a crowded object is considerably facilitated by presaccadic attention. Our results provide converging evidence for a selective increase in available information about high level objects, such as faces, at a presaccadic stage.

Highlights

Observers make several saccades per second to foveate objects in the world, since objects near other objects are often crowded from our awareness
EXPERIMENT 1 Experiment 1 tested whether presaccadic information facilitated identification of a crowded emotional face in the absence of foveating the face; as described in Section Methods, target-flanker spacing was held constant at 3◦ throughout the experiment
The data were analyzed using a ±1.5◦ window centered on the target face for saccade trials, ensuring that the saccades in question were well-localized to the crowded face and that the saccade on a given trial had not landed at the location where an inverted flanker had been present prior to the saccade

Summary

Introduction

Observers make several saccades per second to foveate objects in the world, since objects near other objects are often crowded from our awareness We can see these objects and we have a sense of where they are in visual space, but we cannot identify them without saccading to and foveating them. Saccadic eye movements have been studied for over a century (beginning with Javal, 1878, who coined the term; translated by Huey, 1908; see Kowler, 2011) for an extensive review of the current state of the art It is only relatively recently, that we have started to ask about what information is acquired prior to the start of a saccade. We make saccades to crowded objects in order to identify them

Methods

Results

Conclusion