August 2009
Volume 9, Issue 8
Free
Vision Sciences Society Annual Meeting Abstract  |   August 2009
I like what I see: Using eye-movement statistics to detect image preference
Author Affiliations
  • Tim Holmes
    Department of Psychology, Royal Holloway, University of London
  • Johannes Zanker
    Department of Psychology, Royal Holloway, University of London
Journal of Vision August 2009, Vol.9, 385. doi:10.1167/9.8.385
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Tim Holmes, Johannes Zanker; I like what I see: Using eye-movement statistics to detect image preference. Journal of Vision 2009;9(8):385. doi: 10.1167/9.8.385.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

The preferential looking paradigm is most commonly used with participants such as infants who are unable or unwilling to express their preference consciously. The immediacy of eye-movements as a measure of preference offers the potential to explore typically noisy subjective evaluations, such as aesthetic preference with reliable objective measures in normal adults (Schirillo, 2007, Perception, 36:19). We presented a variety of simple (geometric shapes) and complex (everyday objects, buildings and commercial products) images in sets of 2, 4 or 8 items and recorded oculomotor statistics such as dwell time, returns to location and fixation sequence while participants searched for their preferred image. After eye-tracking for 1500, 2500 or 5000ms, we asked participants to indicate their preference using a button press. The amount of time spent looking at an image correlates increasingly well with preference over the three presentation durations. For short presentations the first and last fixation provide a more reliable predictor of image preference. All statistics become increasingly noisy as the amount and complexity of images presented are increased. By combining the information from these measures the signal to noise ratio can be significantly improved to provide a reliable predictor, which could be used for the subjective evaluation of stimuli in visual psychophysics. Its role as a fitness function in visually driven evolutionary algorithms (Holmes & Zanker, 2008, Perception, 37:148) and potential for application to commercial product testing is discussed.

Holmes, T. Zanker, J. (2009). I like what I see: Using eye-movement statistics to detect image preference [Abstract]. Journal of Vision, 9(8):385, 385a, http://journalofvision.org/9/8/385/, doi:10.1167/9.8.385. [CrossRef]
Footnotes
 Supported by EPSRC - Grant Number 05002329.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×