Evidence for Letter-Specific Position Coding Mechanisms

Abstract

The perceptual matching (same-different judgment) paradigm was used to investigate precision in position coding for strings of letters, digits, and symbols. Reference and target stimuli were 6 characters long and could be identical or differ either by transposing two characters or substituting two characters. The distance separating the two characters was manipulated such that they could either be contiguous, separated by one intervening character, or separated by two intervening characters. Effects of type of character and distance were measured in terms of the difference between the transposition and substitution conditions (transposition cost). Error rates revealed that transposition costs were greater for letters than for digits, which in turn were greater than for symbols. Furthermore, letter stimuli showed a gradual decrease in transposition cost as the distance between the letters increased, whereas the only significant difference for digit and symbol stimuli arose between contiguous and non-contiguous changes, with no effect of distance on the non-contiguous changes. The results are taken as further evidence for letter-specific position coding mechanisms.

On the time-course of adjacent and non-adjacent transposed-letter priming

“…We would point out, however, that single mechanism accounts have other hurdles to cross, such as evidence that the size of TL effects is greater for letters compared with digit and symbol stimuli (Duñabeitia et al, 2012; Massol et al, 2013). Most important is that the Duñabeitia et al (2012) study also demonstrated qualitative differences in TL effects as revealed in ERP recordings.…”

“…There is a long history of research attesting to the fact that letter strings that differ by transposing two letters are perceptually more similar than letter strings that differ by changing two letters (see Duñabeitia, Dimitropoulou, Grainger, Hernández, & Carreiras, 2012; and Massol, Duñabetia, Carreiras, & Grainger, 2013, for two recent studies). For example, more errors are made to real word anagrams as well as nonword anagrams formed by a single transposition (e.g., trial-trail, jugde-judge) compared with non-anagram controls (Andrews, 1996; Bruner & O’Dowd, 1958; Chambers, 1979; O’Connor & Forster, 1981; Perea, Rosa, & Gómez, 2005).…”

Orthographic and phonological processing in developing readers revealed by ERPs

“…TL effects are also found in the same-different matching task, where participants have to decide whether or not two successive briefly presented stimuli are the same or not. In this task, it is harder to detect a transposition change than a substitution change, and this is particularly difficult with letter stimuli (e.g., PGFMR– PGMFR; Duñabetia, Dimitropoulou, Grainger, Hernandez, & Carreiras, 2012; Massol, Dunabeitia, Carreiras, & Grainger, 2013). Importantly, a longitudinal investigation of beginning readers using the same-different matching task with random consonant strings (Duñabeitia, Lallier, Paz-Alonso, & Carreiras, 2015) revealed robust TL effects only in children who had acquired basic literacy skills, in line with the observation that illiterate adults do not show TL effects in this task (Dunabeitia, Orihuela, & Carreiras, 2014).…”

The Impact of Literacy on Position Uncertainty

“…Furthermore, it is worth noting that these studies used the masked-priming version of the same/different task, which has been shown to be insensitive to critical differences in transposed-character effects for known visual elements. In the classic (unprimed) version of the same/ different task, which was also used in Duñabeitia et al (2014), transposition effects are significantly different for letters, digits, and symbols (see Duñabeitia, Dimitropoulou, Grainger, Hernández, & Carreiras, 2012;Massol, Duñabeitia, Carreiras, & Grainger, 2013).…”