Effects of inter-character spacing on saccade programming in beginning readers and dyslexics

ABSTRACT

The present study investigated the impact of inter-character spacing on saccade programming in beginning readers and dyslexic children. In two experiments, eye movements were recorded while dyslexic children, reading-age, and chronological-age controls, performed an oculomotor lateralized bisection task on words and strings of hashes presented either with default inter-character spacing or with extra spacing between the characters. The results of Experiment 1 showed that (1) only proficient readers had already developed highly automatized procedures for programming both left- and rightward saccades, depending on the discreteness of the stimuli and (2) children of all groups were disrupted (i.e., had trouble to land close to the beginning of the stimuli) by extra spacing between the characters of the stimuli, and particularly for stimuli presented in the left visual field. Experiment 2 was designed to disentangle the role of inter-character spacing and spatial width. Stimuli were made the same physical length in the default and extra-spacing conditions by having more characters in the default spacing condition. Our results showed that inter-letter spacing still influenced saccade programming when controlling for spatial width, thus confirming the detrimental effect of extra spacing for saccade programming. We conclude that the beneficial effect of increased inter-letter spacing on reading can be better explained in terms of decreased visual crowding than improved saccade targeting.

KEYWORDS:

• Saccade programming
• inter-character spacing
• oculomotor bisection task
• developmental dyslexia
• beginning readers

Acknowledgments

The authors are grateful to Dr. Marie-Odile Livet and all the clinical team of the pediatric service of the Hospital of Aix-en-Provence (France) for their precious help in recruiting dyslexic children. We would like to thank Mr. Claude Auger, Inspector of French National Education- Aix Est (Aix-Marseille Academy, France) for allowing us to run this research in some of his mainstream elementary schools. Furthermore, we are grateful to Dr. Pom Charras for the helpful discussions and comments on early drafts of the general discussion of the present manuscript. Finally, we would like to thank Laetitia Filippi for her help in collecting part of the data presented in this paper and all the children and their parents who agreed to take part in this research.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Developmental dyslexia is diagnosed when no sensory and intellectual deficits can explain reading and/or writing disorders and when adequate instruction and socio-cultural opportunities are available but fail to result in an adequate level of performance (World Health Organisation [WHO], 1992). Even if one of the most widely accepted explanations of developmental dyslexia posits a core deficit at the phonological level of processing, a number of studies highlighted alternative deficits associated to reading disorder. The general idea is that a lower-level deficit can be linked to dyslexia together with the phonological one (e.g., Goswami, Power, Lallier, & Facoetti, 2014). In particular, an important number of evidences showed the presence of visual deficits in developmental dyslexia, affecting both high-level visuo-attentional skills or low level visual-perceptual skills (e.g., Bellocchi, Muneaux, Bastien-Toniazzo, & Ducrot, 2013b; Bellocchi et al., 2017; Bosse, Tainturier, & Valdois, 2007; Jainta & Kapoula, 2011).

² We used Courier New, a non-proportional fixed-width font, meaning that all characters are of equal width. This font ensures that the probability of any given character attracting a fixation is equal. Moreover, in order to facilitate legibility to beginning readers (Grade 1 and 2), we used a bigger font size than previous studies on adult readers (e.g., Perea & Gómez, 2012a; 2012b; Perea et al., 2012; see Ducrot et al., 2013).

³ In order to test this particular extra-spacing condition, we run a preliminary study where we tested 18 beginning readers (Grades 1 and 2) and 18 children skilled readers (Grade 5) in a lexical decision task. Each stimulus was presented with either extra-spacing or with default spacing. We found a significant lexicality effect [F(1,104) = 246.0, p < .001] and longer lexical decision times for beginning readers than for skilled ones [F(1,34) = 16.3; p < .001]. These findings suggest that, this particular extra-spacing condition did not totally disrupt word reading processing, allowing the LDT to be sensitive to the lexical knowledge of the children.

⁴ The by item analysis revealed Spacing, Group and Visual Field effects [Spacing: F2(1,131) = 35.57; p < .001, η² = .197; Group: F2(2,262) = 160.01, p < .001, η² = .527; Visual Field: F2(1,131) = 9.91, p < .05, η² = .055].

⁵ The by item analysis revealed a Spacing effect [F2(1,92) = 86.82; p < .001, η² = .462] showing that initial fixation duration was longer in extra-spacing condition (M = 312 ms, SE = 26 ms) than in default one (M = 650 ms, SE = 26 ms).

Additional information

Funding

This work, carried out within the Labex BLRI [ANR-11-LABX-0036,ANR-16-CONV-0002] and the Institut Convergence ILCB [ANR-16-CONV-0002], has benefited from support from the French government, managed by the French National Agency for Research (ANR) and the Excellence Initiative of Aix-Marseille University (A*MIDEX).