Low vision affects the reading process quantitatively but not qualitatively.

In this article, the authors compare the reading behavior of students with low vision to that of two groups of students with normal vision (reading-match and age-match students). In Experiment 1, students identified the first letter in words and nonwords and the researchers measured latency and accuracy. No group differences were found for latency, but the reading-match group differed from the other two groups on accuracy. Ali three groups identified the first letter of words faster than they did for nonwords. In Experiment 2, students named words with typical and atypical letter-sound relationships; the researchers measure latency and accuracy. Both the low-vision and the reading-match groups were slower than the age-match group; only the reading-match group made more errors on typical stimuli than on atypical stimuli. The absence of significant interactions between group and any experimental-word variable justified the conclusion that low vision affects the reading process quantitatively but not qualitatively.

**********

In The Netherlands, children are considered visually impaired, and therefore eligible for institutional support, if their functional vision is less than 3/10 and/or their visual field is less than 30 degrees. If the visual impairment is such that the child is unable to read print and must resort to Braille, he or she is considered to be blind. If, despite a reduced acuity or limited visual field, the visual impairment is such that the child is still able to read print, he or she is considered to have low vision. Children who have low vision but sufficient residual vision to read print are the focus of this study.

A substantial amount of research has indicated that the reading development of children with low vision lags behind that of children with normal vision (e.g., Daugherty, 1977; Fellenius, 1999; Gompel, van Bon, Schreuder, & Adriaansen, 2002). Despite a similar level of intelligence and equal education, children with low vision generally do not succeed in attaining the same reading speed as their peers with normal vision (but see Fridal, Jansen, & Klindt, 1981). Several explanations have been put forward for the relatively slow reading speed of children with low vision. Generally, these explanations pertain to the potential problems that readers with low vision encounter when extracting visual information from a page or a computer screen.

One important explanation for the slow reading speed of children with low vision concerns the ease with which text can be fixated. During reading, the eyes do not slide evenly over the paper, but longer breaks (fixations, varying from 200-250 ms) are alternated with relatively short eye movements (saccades, varying from 20-35 ms). The main function of a saccade is to bring a new piece of text into the central visual field. Information from the text can only be extracted during a fixation (Rayner & Pollatsek, 1989). For example, central visual-field loss due to macular degeneration (damage to a small area near the center of the retina) or central scotomas (regions on the retina that have no or reduced acuity as a result of damage; the retinal field is nonfunctional) will hinder fixation of texts. Individuals with damage to the fovea (the region on the retina with maximal visual acuity) tend to develop eccentric viewing behavior; that is, an eccentric part of the retina, rather than the fovea, is used for fixation. This is called the preferred retinal locus (see Raasch & Rubin, 1993; Watson, 2000). In people with normal vision, text presented to an eccentric part of the retina is read more slowly than text presented to the fovea, even when the letter size is magnified to compensate for reduced resolution at the eccentric location. According to Raasch and Rubin (1993), this finding suggests that eccentric locations are simply not as fast as the fovea is in recognizing words. At the same time, these authors implied that for people with low vision who have central scotomas, reading rates will be depressed simply by virtue of the fact that they need to read with an eccentric location. In sighted people with artificially created scotomas, the reading rates were reduced when the scotoma was displaced to either the left or the right of the fovea but not when above or below it (Cummins & Rubin, 1991). Reading rates were most depressed when the scotoma was to the right of the fixation. In this condition, it seems that the scotoma leads the eye movements and obscures the word upon which the eye is about to be fixated. There is no reason to assume that true scotomas in people with low vision would have a different effect. However, because central visual-field deficits are rare in children, these deficits alone cannot explain slow reading in children with low vision.

A second reason why readers with low vision might read more slowly is related to the process of extracting information from the peripheral visual field. Although exact identification of letters in the peripheral area is not possible for readers with normal vision, information retrieved from the peripheral visual field can guide the reading process. For instance, short words on the right side of the fixated word can sometimes be identified, which in some cases makes fixation on that word unnecessary. The size of the peripheral visual field is restricted when readers suffer from, for example, tunnel vision (peripheral vision limited to 10 degrees), which forces the reader to make more fixations. In the case of a short distance between the eyes and the paper--for instance, as a result of text enlargement or the use of a reading aid--the reader is also forced to make more fixations because fewer letters or words can be seen at close distance than at a farther distance, given a fixed size of the visual field. Thus, a short-reading distance might cause lower reading speed as well (Watson, 2000).

Results obtained with the RSVP technique (rapid serial visual presentation) shed some light on the role of saccadic eye movements in people with low vision. With the RSVP technique, in which words are presented sequentially, one at the time, at a uniform rate, and in the same location on a monitor, saccadic eye movements become unnecessary. According to Arditi (1999), people with normal vision appeared to benefit substantially from RSVP, as demonstrated by three to four times faster reading rates, but people with low vision did not. People with central visual-field loss read only 1.5 times faster with RSVP than with conventionally presented text (Rubin & Turano, 1994). This suggests that people with low vision have difficulty with saccadic eye movements. The preceding explanations, except for eccentric fixating, are mainly concerned with the reading of sentences and text and are less applicable to the reading of single words.