Spatial-frequency dependence of eetter recognition in central and peripheral vision

Purpose. In normal fbveal vision, letter recognition is mediated by a single spatialfrequency channel, which peaks at about 2 to 3 c/letter. The purpose of this study was to determine whether the spatial channel governing tetter recognition shifts toward lower object spatial frequencies in normal peripheral vision, and/or when the letter size approaches the acuity-limit. Methods. We measured contrast thresholds for recognizing single, Times-Roman lower-case letters that were spatially band-pass filtered. Each of the 26 letters was digitally filtered with a set of 9 cosine log filters, with peak object spatial frequencies ranging from 0.63 to 10 c/letter, in half-octave steps. Bandwidth of the filters was 1 octave. Stimuli were presented for 150 ms on a monochrome monitor using the VideoToolbox software. Three observers with normal vision were each tested monocularly at the fovea, and at 5 and 10 deg in the inferior visual field. Letter sizes were 0.2, 0.4 and 0.6 log units larger than the "acuity" for recognizing high-contrast, unfiltered letters. Results. At all retinal eccentricities and letter sizes tested, contrast sensitivity for letter recognition peaked between 1.8 to 2.5 c/letter. Plots of contrast sensitivity vs. filter frequency had similar shapes and peak object spatial frequencies across eccentricities, for letter sizes matched in multiples of corresponding letter-acuity. When expressed in terms of retinal spatial frequencies (c/deg), our results show the expected progressive shift in the peak sensitivity toward lower spatial frequencies in peripheral vision. There is a slight but systematic shift in the peak sensitivity toward lower object spatial frequency when the letter size approaches the acuity-limit, and the contrast sensitivity functions become broader in shape. Conclusion. Our results show that the dependence of letter recognition on object spatial frequency is invariant in central and peripheral vision when letter size is expressed as a multiple of letter-acuity. The results are consistent with primary reliance on a single spatial channel for letter recognition, wiih channel selection based on the one with peak frequency near 2 to 2.5 c/letter.