1 Readers with Visual Impairment

Plass and Yager (1995) tested 61 patients with visual impairment due to a variety of causes. Almost all preferred to read text shown on a computer monitor rather than printed on paper. Their reading rates were significantly faster when the individual words were presented on a screen using the RSVP procedure than when they were presented as a single paragraph. Although they had presented the text using both the serif typeface Times Roman and the sans serif typeface Arial, they found no significant differences in reading speed between the two typefaces (see Yager et al., 1998). Four of their patients were adults with a history of congenital nystagmus; as mentioned in Sect. 8.2, this causes visual impairment by disrupting the normal pattern of eye movements. A follow-up study by Aquilante et al. (1997) confirmed that these four patients tended to read faster with RSVP than when reading the same material as a single paragraph of text. This suggested that techniques which eliminated the need for eye movements might be useful for improving the reading ability of people with nystagmus or other visual disorders. However, Aquilante et al. did not compare their patients’ performance using different typefaces.

2 Readers with Dyslexia

Research on people with dyslexia reading from print was described in Sect. 8.7. It was noted that the British Dyslexia Association (2018) had for many years recommended that materials for people with dyslexia should use sans serif typefaces, and this advice seems intended to apply both to material shown on screens and to material printed on paper.

Rello and Baeza-Yates (2013) recruited 48 participants between the ages of 11 and 50 who had been clinically certified as dyslexic and asked them to read 12 extracts of 60 words from a Spanish novel. These were presented in 12 different typefaces, including three serif typefaces (Computer Modern Unicode, Garamond, and Times) and four sans serif typefaces (Arial, Helvetica, Myriad, and Verdana). Assignment of the different typefaces to different extracts was counterbalanced across different participants. The extracts were presented on a liquid crystal display (LCD) screen in 14-point size, and the readers’ eye movements were monitored using a corneal reflection system, both to record the duration of their fixations and to record their total reading time. After each extract, the participants answered a multiple-choice question to check their comprehension of the text, and at the end of the experiment they rated their preference for each typeface on a 5-point scale.

There was significant variation among the 12 typefaces in the participants’ reading time, but there was no overall difference between the serif typefaces and the sans serif typefaces in this regard. There was significant variation among the 12 typefaces in the participants’ fixation duration, and this was slightly longer for the serif typefaces than for the sans serif typefaces. There was significant variation among the 12 typefaces in the participants’ preference ratings, but there was no overall difference between the serif typefaces and the sans serif typefaces in this regard. Rello and Baeza-Yates argued that shorter fixations reflected a reduced processing load and hence greater legibility, and they concluded that sans serif typefaces enhanced reading performance, even though this was not reflected in an enhanced reading time.

However, Rello and Baeza-Yates had not controlled the spacing of their different typefaces. It was noted in Sect. 8.6 that readers who are dyslexic benefit from increased spacing when reading from paper, and that this explains differences in their performance across different typefaces. It was also noted in Sect. 10.1 that the addition of serifs leads to an increase in inter-letter spacing, at least when letters are presented on screen. Perea et al. (2012) found that a small increase in inter-letter spacing could lead to enhanced performance when reading from computer screens, especially in children with dyslexia. Consequently, any suggestion in Rello and Baeza-Yates’ results in favour of sans serif typefaces is likely to be due to the confounding of typeface with inter-letter spacing. Otherwise, the results of their study indicate no difference in legibility between serif and sans serif typefaces in dyslexic readers, and once again this contradicts the advice of the British Dyslexia Association (2018).

3 Readers with Age-Related Macular Degeneration

Several studies have investigated people with age-related macular degeneration (AMD). As was mentioned in Sect. 8.2, this causes impaired vision in the centre of the visual field. Section 11.1 described a study in which Arditi and Cho (2005) measured size thresholds when random five-letter strings were presented on a cathode-ray tube (CRT) screen as black letters against a white background. They used software to construct lowercase typefaces of uniform thickness with slab serifs extending 0% (sans serif), 5%, or 10% of the cap height (the height of capital or uppercase letters). They also used an inter-letter spacing of 0%, 10%, or 40% of the cap height, yielding a 3 × 3 design. In addition to four participants with normal vision, Arditi and Cho also tested two people with AMD. They normalised the data to each participant’s best performance by dividing each data point by the participant’s minimum threshold. Both the participants with normal vision and the participants with AMD showed the same pattern of performance. There was a large effect of spacing such that closely spaced letters yielded higher thresholds (i.e., poorer performance). There was also a significant but small effect of serif size, such that serifs of 5% or 10% led to lower thresholds (i.e., better performance) than a sans serif typeface, which Arditi and Cho ascribed to the concomitant increase in spacing required to accommodate them. There was no evidence that the presence or absence of serifs made any difference to the performance of the participants with AMD.

Subsequent researchers tried to devise typefaces that might be helpful for people with AMD. Bernard et al. (2016) developed Eido, a monospaced sans serif typeface that emphasised the distinctive shapes of different letters. They presented normally sighted volunteers with letters, words, and sentences in either Eido or the monospaced slab serif typeface Courier using a CRT screen. They carried out six different experiments with varying print sizes. In each case, the participants used their dominant or preferred eyes, but an eye-tracking system superimposed a mask over the centre of their visual field to simulate AMD. The participants made fewer errors when stimuli were presented in Eido than when presented in Courier, but there was no difference in their reading speed between the two typefaces.

Xiong et al. (2018) used both Eido and Maxular Rx, a proportionally spaced slab serif typeface developed by Steven Skaggs that employed extra spacing between successive letters and lines. For comparison, they also used Courier, Helvetica, and Times Roman. (Helvetica is a proportionally spaced sans serif typeface, whereas Times Roman is a proportionally spaced serif typeface.) Individual sentences were presented on an LCD screen controlled by a Macintosh computer. The participants consisted of 19 individuals with AMD, 14 age-matched individuals with normal vision, and 26 young adults with normal vision. The researchers measured their fastest speed to read the sentences aloud, the smallest print size to achieve that speed, and the smallest print size that could just be read. The reading speed varied significantly across the five typefaces for the individuals with AMD, but not for the control participants. All three groups showed significant variations in the measures of print size. However, there were no systematic differences between the three serif typefaces (Courier, Maxular Rx, and Times Roman) and the two sans serif typefaces (Eido and Helvetica).

4 Conclusions

As mentioned in Chap. 8, any differences in the legibility of serif and sans serif typefaces might become more apparent in readers whose visual systems are challenged as the result of disablement. One study evaluated a heterogeneous sample of patients with visual impairment and found no difference in reading speed between a serif typeface and a sans serif typeface, regardless of whether text was presented on screen as a single paragraph or using the RSVP procedure. Another study evaluated a large sample of readers with dyslexia. This too found no difference in their reading speed between serif and sans serif typefaces. Any differences in the participants’ preferences or in their eye movements could be attributed to the researchers’ failure to control the inter-letter spacing of the different typefaces. Differences in inter-letter spacing also explain differences in reading speed in a study which simulated AMD in readers with normal vision. A study which compared reading in people with and without AMD found no systematic differences between serif typefaces and sans serif typefaces.