Gaze Contingent Experiments:

A tool to help us learn about the role of attention in visual processing

<See Geisler, W.S., Perry, J.S., Najemnik, J. (2006) for a descrioption of gaze-contingent methodology,
properties of central vs. peripheral vision, and (Bayesian) application to visual search.
>

 

Illustrations (below) show search for an item containing a red-horizontal line .  What makes this (conjunction) task challenging are the features shared by both target and distractors, (i.e., the red color and the horizontal line): the greater the number of shared features, the slower the search and the greater the likelihood of false detections. A similar conjunction task is described in Wolfe, J (2005) in a summary of Bichot's (2005) animal studies. These and many other behavioral & physiological studies establish how conjunction tasks require the use of attention, and produce slower and more sequential (capacity limited) forms of scanning. Further, these studies illustrate how our goals and attention modulate performance, and that an associated neuronal response (in area V4 in the case of Bichot et al's '05 work) precedes eye-movements and fixations. (Many other studies have shown activity in FEF, SC & other thalamo-cortical areas preceding eye-movements depending upon type of search task).

The important implication here is that the mechanism guiding eye-movements in these more complex search tasks, because they are dominated by (an endogeneous form of) attention, may be revealed through patterns that emerge in the sequence of eye-movements. To further examine the role of attention in visual search we will perform various manipulations of its known properties and evaluate how these influence the dynamic of search over different time periods. A critical set of manipulations will involve gaze-contingent (GC) blurring and GC scaling which will help us tease apart a number of the (confounded) characteristics of attention, and help us understand whether attentional improvements in performance are due to improved stimulus clarity, increased scaling or to its central locale in the visual field.



References:
Bichot, N. P.; (2005) Parallel and Serial Neural Mechanisms for Visual
Search in Macaque Area V4. Science, 2005, vol. 308, issue 5721, p 529,

Geisler, W.S., Perry, J.S., Najemnik, J. (2006) Visual search: The role of peripheral information measured using gaze-contingent displays. Journal of Vision, 6, 858-873. [PDF]

Wolfe, J (2005) Neuroscience: Watching Single Cells Pay attention. Science, 308:503

Conditions

Blurred region

& video links

Video trials

Description of what you'll see in the videos

Additional  Manipulations

(not illustrated here )


Some of the clips start with recording of my eye-movements during calibration. Although this may seem a bit creepy, it illustrates the sensitivity of the tracker's IR camera.
   
Gaussian Blur (pixels)
 
Gaussian Filter (pixels)-->

0

8
12

see video description --->

 

 

 

  • item size

    large vs small

  • window size

    large vs small

  • # items:

         ~25, 50, 75

  • # targets:

         1, 3 & 5

  • item location/ context:

        fixed vs. random

 


Search Task

Find:

Peripheral G8 Blur

 

 

periphery

 

 

Conditions:

T1) No Blur

T2) 8 pixel gaussian blur in the periphery

T3) 12 pixel gaussian blur in the periphery

 

 

foveal8GBlur

 

 

foveal

(i.e., central)

T1) No Blur

T2) 8 pixel gaussian blur in central vision

T3) 12 pixel gaussian blur in central vision

 
Additional Gaze contingent projects: Search for letters
     
 

 

|D.J. Aks | Eye-tracking research | System noise | Time series & fractal analysis | Visual search | Attention | Satellite Imagery | Tumor detection | Web eval & info search | Illusions