Reading and Writing: An Interdisciplinary Journal 15: 15–42, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
Asynchrony of visual-orthographic and auditory-phonological
word recognition processes: An underlying factor in dyslexia
Laboratory for Neurocognitive Research, Faculty of Education, University of Haifa, Israel
Abstract. This study investigated whether asynchrony of speed of processing (SOP) between
visual-orthographic and auditory-phonological modalities can account for word recogni-
tion deﬁcits among dyslexic readers. SOP among elementary school dyslexic readers was
compared to that of chronologically age-matched normal readers. SOP was assessed using
nonlinguistic and linguistic auditory and visual low-level tasks and higher-level orthographic
and phonological tasks. Behavioral and electrophysiological (ERP) measures of SOP were
obtained. Data indicated that dyslexic readers were signiﬁcantly slower than control readers
in most of the experimental tasks. Moreover, dyslexics revealed a systematic SOP gap between
the auditory-phonological and the visual-orthographic modalities. This gap was found in both
P200 and P300 latencies, and explained most of the variance in word recognition. A theory
is proposed suggesting that asynchrony between the processing rates of the visual and the
auditory modalities may be an underlying cause of dyslexia.
Keywords: Asynchrony, Auditory, Dyslexia, Brain activity, Electrophysiology, ERP, Modal-
ities, Orthography, Phonology, Reading, Speed of processing, Timing, Visual
Word recognition is a fundamental reading skill, and provides the foundation
upon which other aspects of reading ability develop. Deﬁcits in word recog-
nition skills are typical of most dyslexic readers (Adams 1990; Perfetti 1992;
Stanovich 1991). The explication of this phenomenon has been the subject
of ongoing investigation. Two decades of research point to a phonological-
core deﬁcit as a primary explanatory factor (for a review, see Liberman &
Shankweiler 1991). Recent work suggests that orthographic processing deﬁ-
cits may also be implicated (Barker, Torgesen & Wagner 1992; Cunningham
& Stanovich 1990; Stanovich & West 1989; Zecker 1991).
While empirical work suggests that both phonological and orthographic
processes contribute to effective decoding, the way in which they interact
during reading remains a subject of theoretical speculation. According to
the current connectionist approach to reading, phonological and ortho-
graphic processes operate in unison to activate semantic representations of