Does the huamn mnid raed wrods as a wlohe?

Abstract

A recent email message about a purported experiment run at Cambridge University provides a useful illustration of some fundamental mechanisms involved in reading. The message demonstrates that a text composed of words whose inner letters have been re-arranged can be raed wtih qutie anazimg esae! Although some of the readability of this email message is probably due to top-down factors made possible by the fact that almost 50% of the words are not mixed up, we suggest that a significant part of this 'jumbled word effect' is due to the special way in which the human brain encodes the positions of letters in printed words. Recent research using the masked-priming technique has helped to elucidate the mechanisms involved in letter-position coding. Masked primes are briefly presented, pattern-masked letter strings, whose effects on target processing are thought to reflect fast, automatic processing [1,2]. We will briefly describe two phenomena, relative-position priming and transposition priming, that have been observed with this paradigm and that are particularly relevant for understanding letter-position coding.Masked-priming studies [1,2] have shown that targetword recognition is facilitated when primes are composed of a subset of the target word's letters (even when the proportion of shared letters is quite low, and absolute, length-dependent, letter position is violated), as long as the shared letters are in the same order in prime and target stimuli. That is, priming occurs only when relative positions are respected. For example, a six-letter word such as 'garden' is identified more rapidly when preceded by the masked prime 'grdn' compared to the unrelated condition 'pmts', and partly changing the order of letters (gdrn, nrdg) destroys the priming effect. This result would appear to be at odds with the apparent ease with which we can read the jumbled-word email message.However, when primes share all their letters with target words, priming is robust with small changes in letter order, a phenomenon referred to as transposition priming. Thus, primes formed by transposing two adjacent letters in a target word (e.g. gadren -garden) facilitate word recognition compared with appropriate control primes [3,4].Most importantly, these two phenomena have propelled a new approach to letter-position coding that we believe provides a coherent account of the 'jumbled word effect'. We note in passing that current computational models of visual word recognition cannot capture these two key phenomena. For example, the well-known Interactive Activation [5] and Dual-Route Cascaded [6] models use a position-specific slot encoding, which is inconsistent with these experimental results. In response to the obvious shortcoming of these approaches, Whitney [7,8] proposed an encoding based on ordered letter pairs (e.g. the input 'take' is represented by activation of units representing TA, TK, TE, AK, AE, and KE). These units do not contain precise information about letter position, or about which letter is next to which. The same units ...

Tuning of the human left fusiform gyrus to sublexical orthographic structure

“…In the interactive activation model, for example, the word superiority effect could be explained by probabilistic letter-sequence nodes (rather than word nodes) providing feedback to letter nodes. Recent priming experiments provide independent support for such letter-sequence representations (Grainger and Whitney, 2004;Perea and Lupker, 2003;Whitney, 2001). Although it may be possible to account for the perceptual effects observed in the present study by a mechanism involving partial activation of a cohort of lexical codes, we suggest that the stimuli were sufficiently unlike words that any such activation would have been very minimal.…”

Morphemes in their place: Evidence for position-specific identification of suffixes

“…However, the results of the present experiment allow us to reject this explanation. If the participants had classified stimuli nessdeaf, and between quarrel and relquar) was also matched according to theoretical match values derived from spatial coding (e.g., Davis & Bowers, 2006) and open-bigram coding (e.g., Grainger & Whitney, 2004) models of letter position coding. This matching was intended to allow us to detect the effects of morphological similarity above and beyond those of pure orthographic similarity.…”

Sublexical frequency measures for orthographic and phonological units in German

“…The question which of both measures reflects the processing of a stimulus best has not yet been answered to our knowledge. Position specificity is a matter of debate in the current literature that has more than these two solutions (see Dehaene, Cohen, Sigman, & Vinckier, 2005;Goswami & Ziegler, 2006;Grainger & Whitney, 2004). For example, relative positions within a word might be another suitable concept (Peressotti & Grainger, 1999).…”