October 2020
Volume 20, Issue 11
Open Access
Vision Sciences Society Annual Meeting Abstract  |   October 2020
Differential grouping affordances of random spatial and temporal information for accurate number estimation
Author Affiliations
  • Frank Durgin
    Swarthmore College
  • Elsie Aubry
    Swarthmore College
  • JJ Balisanyuka-Smith
    Swarthmore College
  • Çiçek Yavuz
    Haverford College
Journal of Vision October 2020, Vol.20, 1659. doi:https://doi.org/10.1167/jov.20.11.1659
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Frank Durgin, Elsie Aubry, JJ Balisanyuka-Smith, Çiçek Yavuz; Differential grouping affordances of random spatial and temporal information for accurate number estimation. Journal of Vision 2020;20(11):1659. https://doi.org/10.1167/jov.20.11.1659.

      Download citation file:

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

  • Supplements

Magnitude estimates of spatial non-symbolic number are thought to follow a power function, like other psychophysical dimensions. However, existing data identifies a common transition point at about 20 dots: Below 20, mean estimates are typically accurate, with a slope of 1. Above 20, underestimation produces a power function with an exponent of 0.5 to 0.9. Here we ask whether the 20-dot transition point reflects a capacity limitation of a general number mechanism or the limit of a spatial grouping strategy. In a first study, we presented temporal sequences of clicks and/or visual flashes with random timing (minimum SOA 75 ms). The average event rates (6 - 8 Hz) precluded verbal counting. Forty-eight participants estimated the numbers of either auditory, visual, or audio-visual events (numbering from 2 to 58). For visual flashes, there was no evidence of special accuracy: Even in the subitizing range, mean estimates followed a power function with an exponent of 0.83 (R2 = .999). For auditory clicks, only the mean estimates for the numbers 2 and 3 were accurate, while a power function exponent of 0.84 fit the numbers beyond 3 (R2 = 1.000). For audiovisual events, mean performance was accurate only up to 4 events; beyond 4, data were fit with an exponent of 0.82 (R2 = .999). A follow-up study with 15 new participants, used hybrid spatiotemporal stimuli consisting of auditory clicks synched with accumulating visual items in a random spatial array that disappeared after the final click. Accurate mean performance in this hybrid condition reached about 11, beyond which estimates had an exponent of 0.70 (R2 = .998). When verbal counting is prevented and intervals are random, temporal numerosity followed a power function above 4. Random temporal processes less easily afford strategic grouping and subitizing than spatial displays, and this may limit accuracy.


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.