Letter Visibility and the Optimal Viewing Position Effect of Isolated Connected and Un-Connected Letters in Arabic

Abstract

Th e present study provides a further exploration of the role of Arabic lett er visibility as a possible cause of the Optimal Viewing Position (OVP) eff ect. We used isolated connected and un-connected Arabic lett ers of diff erent shapes (basic, initial, medial, fi nal) placed at the center of fi xation (Experiment 1) and at various possible positions in isolated presentation (Experiment 2). In order to investigate whether performance in the visual identifi cation task is modulated by lett er type, we presented each of the isolated connected and un-connected lett er targets in each of the eleven stimulus positions across the array to produce a mean RT (ms) for each of the lett er types. Using the initial fi xation paradigm enabled us to compare reaction times with correctly identifi ed lett er targets appearing in the diff erent possible positions. Th e fi ndings of the present experiments demonstrated that visual lett er recognition is infl uenced by: (i) the isolated lett ers' type (connected, un-connected), as connected lett ers are easier to recognize than un-connected lett ers; (ii) isolated lett ers' shape (basic, initial, medial, fi nal), as medial and fi nal are harder to recognize than basic and initial lett er shapes; (iii) visual fi eld, as reading rates were longer for lett er stimuli that were presented in LVF compared to RVF; and (iv) eccentricity, as lett er reading rates were correlated with their eccentric placement.

Insights from the study of Arabic reading

“…This was attributed to the complexity of Arabic letter visual representations, including letter connectedness. However, a series of experiments have also indicated that, for skilled Arabic readers, the identification of words containing semi‐connected letters is more prone to error than words containing connected letters (Ganayim, 2015; Khateb et al, 2013; Taha, Ibrahim, & Khateb, 2013). These latter findings suggest that, although semi‐connected letters might be visually simpler, ligatures facilitate word identification accuracy.…”

Letter connectedness and Arabic visual word recognition

“…This study further indicated that skilled Arabic readers were slower at letter identification when target letters were presented in different allographic forms compared with when they were presented only in their typical isolated forms. Ganayim (2015) also found that letter recognition in Arabic is influenced by both letter type and allographic form. In this study, skilled Arabic readers had longer identification times for unconnecting letters than for connecting letters, likely due to the fact that unconnecting letters occur less frequently in this script.…”

“…These results were interpreted to indicate that, contrary to the idea that visual complexity causes processing difficulty during word recognition, there appeared to be processing costs for unconnected words—arguably the least visually complex of the word types that were tested. This somewhat unexpected result was attributed to the fact that the unconnecting letters that made up these words occur less frequently than connecting letters in Arabic (again, see Ganayim, 2015) and that most Arabic words contain multi-letter graphemic chunks. The same word types were also examined in a lexical decision task with both skilled and dyslexic Arabic readers (Khateb et al, 2013).…”

“…In cases where pronunciation is important, international phonetic alphabet transcriptions are provided.) When these unconnecting letters occur in an Arabic word, they have the effect of separating its letters into two or more strings, sometimes referred to as sub-units (Ganayim, 2015) or graphemic chunks (Yakup et al, 2014, 2015). For example, although the word منخفض (mnxfD, low ) consists of a single, five-letter graphemic chunk, the word غضبان (gDbAn, angry ) is made up of a four-letter chunk (غضبا) and a one-letter chunk (ن), while the word موافق (mwAfq, agree ) consists of three chunks—a two-letter chunk (مو), a one-letter chunk (ا), and another two-letter chunk (فق).…”

“…The first of these questions is important because it has been argued that letter connectedness might be a source of reading difficulty in Arabic, in part because it contributes to the complexity of visual word forms (see, for example, Abu-Rabia, 2001; Azzam, 1989; Eviatar & Ibrahim, 2004; Ibrahim et al, 2007). However, although letter identification studies seem to indicate a cost for processing the different letter shapes that arise because of connectedness (see, for example, Asaad & Eviatar, 2013; Eviatar et al, 2004; Ganayim, 2015; Ibrahim et al, 2002), the influence of letter connectedness on word-level reading is less clear (see, for example, Khateb et al, 2014; Taha et al, 2013). Therefore, the first aim of this study was to address this issue in more detail by examining the processing of a tightly controlled set of words and nonwords that instantiate different levels of letter connectedness.…”