Reading at Low Contrast: A Model for Reading With Macular Disease?

Abstract

Purpose: : Patients with macular disease (MD) read more slowly than their normally–sighted peers, even after correcting for the difference in visual acuity by enlarging print size. Those with MD also suffer from reduced contrast sensitivity. Several hypotheses have been suggested to explain this slower reading in low vision, including a reduction in the visual span (that is the number of characters which can be identified in one fixation); prolonged fixation time; and reduced saccade length.^1,2 The current study measures visual span, fixation duration, saccade length and reading speed at low contrast for subjects with normal vision and reading ability. These data were compared to those already collected for patients with MD² in order to determine the extent to which reading at low contrast can model reading with MD.

Methods: : Eight subjects with no history of eye disease and corrected vision of 0.0 logMAR or better were recruited. An infrared eyetracker was used to measure forward saccade length, fixation duration and reading speed whilst subjects read MN–Read style sentences presented at 90% and 5% Michelson contrast. A staircase procedure was used to measure visual span for trigrams (strings of three letters) presented for 125, 250 and 500 msec at the same contrast levels whilst fixation was monitored. All text was of x–height 0.5^o.

Results: : Contrast attenuation reduced reading speed (one–sided matched pairs t–test, p<0.005). Visual span was significantly reduced at low contrast. At low contrast, increased presentation time resulted in larger visual spans whereas no relationship was found between visual span and presentation time at high contrast. Forward saccade length was not reduced, and fixation duration increased at low contrast. These results are discrepant with those from our macular disease patients who displayed reduced forward saccade length.²

Conclusions: : At low contrast, subjects with good vision used longer fixations. We presume this is in order to maximise visual span and maintain a normal saccade length. This is corroborated by data linking visual span and exposure duration. Conversely, patients with macular disease do not increase fixation duration yet have a reduced forward saccade length, consistent with the reduced visual span hypothesis. Contrast attenuation does not provide a good model for reading with macular disease. 1. Legge et al, Vision Res 1997;37:1999. 2. Crossland & Rubin, Vision Res 2006;in press.

Keywords: reading • eye movements