Orthographic processing
Saturday, October 01st, 2011 [17:20 - 19:20]
PS_2.086 - Is the activaction of homographic stems affected by phonological recoding?
Bracco, G. & Laudanna, A.
Department of Communication Sciences. University of Salerno. Italy.
A debated issue in Psycholinguistics is whether or not a pre-lexical conversion of written words into a phonological code is needed to achieve lexical access: we addressed this issue in Italian, a language with a fairly transparent orthography. Three priming experiments using different prime durations (150 ms, 250 ms, 350 ms) were carried out to observe the putative time course of phonological code activation in the recognition of Italian words, by comparing homographic to homographic but non homophonic stems. In each experiment 48 verbal forms were used as targets in three priming conditions: A) prime/target containing homographic and homophonic stems (e.g., sparato/sparito, shooted/disappeared); B) prime/target containing homographic but not homophonic stems (e.g., pregi (/prεdʒ/)/prega (/prεg/), virtues/ he/she prays); C) prime/target sharing the initial orthographic pattern (e.g., tornio/tornavo, lathe/ I returned). In the control condition an unrelated prime was used for each target. Experimental and control lists were all matched for the degree of orthographic overlap between primes and targets. The results show that the effect of homographic stem is still robust until prime duration of 250 ms, whereas inhibitory orthographic priming requires a 350-ms prime presentation, when priming is also modulated by phonological information.
PS_2.087 - Top-down modulation of the crowding effect in reading
Montani, V. 1 , Facoetti, A. 2 & Zorzi, M. 2
1 Department of Developmental Psychology and Socialization-University of Padua
2 Department of General Psychology-University of Padua
The reading rate depends on the number of letters we take in at each fixation (visual span). This span seems limited by crowding: beyond some eccentricity the reader’s critical spacing exceeds the spacing of the text and the letters crowd each other, spoiling recognition. We used the progressive demasking task, a degraded stimulus presentation procedure, to study the crowding effect during reading of isolated words. Stimuli were familiar words or pronounceable non-words, and spacing between letters was manipulated. We used standard letter and decreased letter spacing (1,03 x letter length). Our results show that the identification of decreased spaced strings was slower than normally spaced strings (crowding effect). More importantly, decreasing the lateral distance between letters impaired non-word more than word identification, thereby revealing a top-down modulation of the crowding effect. Lexical, whole-word representations would convey top-down signals that interactively help to extract spatial details in the reduced spacing condition. This feedback mechanism is not available for unfamiliar letter strings. Since sub-lexical information and phonological decoding are crucial for reading development, increased crowding could be an important factor underlying reading difficulties in dyslexic children.
PS_2.088 - Written language processing in Hearing and Deaf readers
Barca, L. , Castrataro, M. , Rinaldi, P. & Caselli, M. C.
Institute for Cognitive Sciences and Technologies, ISTC-CNR, Rome, Italy
A number of studies indicate that covert word recognition is mediated by phonological activation. We investigated visual word recognition in hearing and deaf readers to test for an effect of auditory deprivation in processing written language. Deaf readers have more difficulties in forming phonological representations of words and letters, which might affect the functioning of brain areas typically associated with phonological processing and their interaction with other brain regions. In this behavioral study, a lexical decision task comparing response to Italian 5 letters words and consonant strings was performed by tree groups of adults: Hearing participants; Deaf Signers with Italian Sign Language as primary language; Deaf Non-Signer, which use spoken language and lip reading for communicate. Overall, Deaf readers were faster than Hearing participants in performing the task. Deaf Signers were slower than Deaf Non-Singers in processing consonant strings, whereas no difference emerged for words. Significant Lexicality by Group interaction indicates that the Lexicality effect was restricted to Deaf Signer, which responded faster to words than consonant strings. Results will be discussed within classical models of visual word recognition and taking into account effects of early training intervention on deaf participants’ written language processes.
PS_2.089 - The development of sub-lexical spelling mechanisms in a shallow orthographic system (Spanish)
Carrillo Gallego, M. S. 1 & Alegria Iscoa, J. 2
1 Dpto. Psicologia Evolutiva y de la Educacion. Universidad de Murcia (UMU) - Spain
2 Lab. Cognition Langage et Dévelopment. Université Libre de Bruxelles (ULB) - Belgium
Word spelling involves, besides grapheme to phoneme translation rules, two purely orthographic processes, one based on whole-lexical representations and another one exploiting sub-lexical regularities. Data collected in English and French, two opaque orthographic systems, show that young children do use sub-lexical regularities. The present study examines the acquisition of lexical and sub-lexical abilities in Spanish, a transparent system. To examine these question Spanish-speaking participants were asked to spell high and low frequency words containing sequences like “va, ve, vi...” in witch phoneme-grapheme translation rules are useless. Spelling would therefore reflect both word frequency and sub-lexical regularity effects (i.e. “vi” is more frequent than “bi” at word beginning while “va” is less frequent than “ba”). The results show strong lexical and sub-lexical effects from first grade onwards. In a second experiment the nature of units stored in orthographic memory was examined. The results were compatible with the notion that this information was purely orthographic and it didn’t take into account the syllabic structure of words. Results are discussed in the context of specific features of orthographic learning in transparent systems and on the role of implicit mechanisms involved in learning to read and spell.