October 2020
Volume 20, Issue 11
Open Access
Vision Sciences Society Annual Meeting Abstract  |   October 2020
Habitual scanning bias in complex problem solving
Author Affiliations & Notes
  • Frances G. McCormick
    University of Toronto
  • Dirk B. Walther
    University of Toronto
    Samsung Artificial Intelligence Center Toronto
  • Karen F. Bernhardt-Walther
    York University
  • Footnotes
    Acknowledgements  We thank the University of Toronto and the University of Toronto Excellence Award for funding this work. We also thank Chengcheng Huang, Agnes Priscilla Layarda, Rida Aamer, Chris Dobronyi, Jiongtian Guo, and Professor Wil Cunningham for their continued support.
Journal of Vision October 2020, Vol.20, 926. doi:https://doi.org/10.1167/jov.20.11.926
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Frances G. McCormick, Dirk B. Walther, Karen F. Bernhardt-Walther; Habitual scanning bias in complex problem solving. Journal of Vision 2020;20(11):926. doi: https://doi.org/10.1167/jov.20.11.926.

      Download citation file:

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

  • Supplements

Habitual scanning direction biases performance on perceptual tasks such as characterizing the shading of an object (Smith et al. 2015) or the symmetry of a scene (Afsari 2016).  But how does this spatial bias affect complex problem solving? More generally, how do perceptual patterns affect cognitive outcomes? We employ Ravens-like puzzles as a new experimental paradigm to reflect problem solving with directional bias. Based on work of Matzen et al. (2010), we consistently vary complexity and puzzle attributes, including direction of change.  In a preliminary eye-tracking study, we find that scanning patterns for our Ravens-like puzzles are predominantly horizontal, but also vary by directionality of puzzle attributes. Do participants solve puzzles more correctly when puzzle directionality corresponds to habitual scanning direction?  If so, this suggests an extension of models for predicting Ravens-like puzzle performance to include directional factors (Hayes et al. 2011). If scanning direction does not affect correctness, this suggests a disjoint between habitual behaviour and top-down attention in spatial problem solving. We conduct a behavioural experiment to isolate the effects of habitual scanning patterns on puzzle-solving performance.  We analyze correctness for two groups of Ravens-like puzzles: puzzles with counter-habitual bias, or attribute directionalities which elicit more vertical scanning, and puzzles without counter-habitual bias.  To ensure that variation in other puzzle attributes is identical between these two groups, we generate a set of transposed puzzles where each puzzle and its transpose have opposite directional biases. We then compare performance on all puzzles with and without counter-habitual bias.  We find that puzzles with counter-habitual bias correspond with significantly lower correctness. This finding indicates that people perform better on complex visual problems where bottom-up directional bias matches habitual trends of perception. More broadly, it suggests continuity between perceptual patterns and cognitive outcomes in a complex task setting.


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.