Letter and symbol misrecognition in highly legible typefaces for general, children, dyslexic, visually impaired and ageing readers

Böhm, Thomas

doi:10.1075/idj.21.1.05boh

Cited by 4 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the research question could be put this way: Would the detectors of a given letter (e.g., “m”) be activated early in processing when a word not containing this specific letter—but containing a multi-letter homoglyph (e.g., docurnent )—is briefly presented? This question is not only important at a theoretical level (i.e., it would help refine the perceptual feature-letter front-end of models of printed word recognition); at an applied level, the potential confusability across letters is a matter of serious concern when accessing Internet websites (Davis & Suignard, 2012; see also Bohm, 2015). As acknowledged on the Microsoft website when discussing security in Internet domain names, “rnicrosoft.com looks much like microsoft.com,” 3 and this may lead to spoofing attacks (see Gabrilovich & Gontmakher, 2002; Krammer, 2006).…”

mentioning

confidence: 99%

Can I order a burger at rnacdonalds.com? Visual similarity effects of multi-letter combinations at the early stages of word recognition.

Marcet

Perea

2018

Journal of Experimental Psychology: Learning, Memory, and Cogni

View full text Add to dashboard Cite

Previous research has shown that early in the word recognition process, there is some degree of uncertainty concerning letter identity and letter position. Here, we examined whether this uncertainty also extends to the mapping of letter features onto letters, as predicted by the Bayesian Reader (Norris & Kinoshita, 2012). Indeed, anecdotal evidence suggests that nonwords containing multi-letter homoglyphs (e.g., rn→m), such as docurnent, can be confusable with their base word. We conducted 2 masked priming lexical decision experiments in which the words/nonwords contained a middle letter that was visually similar to a multi-letter homoglyph (e.g., docurnent [rn-m], presiclent [cl-d]). Three types of primes were employed: identity, multi-letter homoglyph, and orthographic control. We used 2 commonly used fonts: Tahoma in Experiment 1 and Calibri in Experiment 2. Results in both experiments showed faster word identification times in the homoglyph condition than in the control condition (e.g., docurnento-DOCUMENTO faster than docusnento-DOCUMENTO). Furthermore, the homoglyph condition produced nearly the same latencies as the identity condition. These findings have important implications not only at a theoretical level (models of printed word recognition) but also at an applied level (Internet administrators/users). (PsycINFO Database Record

show abstract

mentioning

confidence: 99%