September 2017
Volume 17, Issue 10
Open Access
Vision Sciences Society Annual Meeting Abstract  |   August 2017
Temporal modulation of signal/noise reveals processing units of a scale greater than letters in visual word recognition.
Author Affiliations
  • Simon Fortier-St-Pierre
    CERNEC, Département de psychologie, Université de Montréal
  • Martin Arguin
    CERNEC, Département de psychologie, Université de Montréal
Journal of Vision August 2017, Vol.17, 1035. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Simon Fortier-St-Pierre, Martin Arguin; Temporal modulation of signal/noise reveals processing units of a scale greater than letters in visual word recognition.. Journal of Vision 2017;17(10):1035. doi:

      Download citation file:

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

  • Supplements

Visual word recognition is largely based on the identification of the letters making up the stimulus (e.g. Pelli et al., 2003). Some studies, however, suggest that units of a greater scale than letters (ranging from transletter features to whole words) also contribute to performance. This claim rests on the negative impact of aLtErNaTiNg CaSe across consecutive letters on reading performance (e.g. Mayall et al., 1997; Pelli & Tillman, 2007). One drawback of case alternation is that it may require observers to switch perceptual tuning between the recognition of uppercase and lowercase letters for every consecutive letter. This means that the cost of case alternation may lie in the letter identification process itself, thus making the result inconclusive with respect to the size of the processing units involved in word recognition. Here, we assess whether units of a scale greater than the letter effectively contribute to visual word recognition. We used a random temporal modulation of the signal-to-noise ratio of the stimulus that was either applied simultaneously throughout all the letters in the word (homogeneous condition) or separately and independently for each letter (heterogeneous condition). Temporal sampling functions were made from the integration of sine waves of 5, 10, 15, and 20 Hz, each with a random amplitude and phase. Stimuli were displayed on a 120 Hz monitor, their exposure (i.e. target + noise) lasted 200 ms and the duration of visibility of each letter was equated across conditions. Our results show significantly better word recognition performance in the homogeneous condition (72.5 %) in comparison to the heterogeneous one (59.3 %), a finding that is verified in each individual participant. The advantage for a temporal modulation of signal/noise that applies simultaneously across all the letters in a word demonstrates a significant contribution of processing units of a scale larger than the single letter.

Meeting abstract presented at VSS 2017


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.