Reading and Writing: An Interdisciplinary Journal 15: 233–259, 2002.
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.
Effects of memory load on word recognition:
Are there dual-routers in Norway?
ARNE LERVÅG & IVAR BRÅTEN
University of Oslo, Oslo, Norway
Abstract. The release-from-competition (RFC) effect, in which a difﬁcult concurrent memory
task speeds pronunciation of low-frequency irregular words but slows pronunciation of other
word types, has been interpreted as strong support for the dual-route approach to word recogni-
tion. However, attempts to replicate this effect have not produced consistent results. Besides,
attempts at replication have mostly been limited to skilled readers of English. The present
research attempted to replicate the RFC effect with mature normal readers of Norwegian and
thus tested the generalizability of dual-route models to a considerably more shallow ortho-
graphy than English. There was no evidence that the RFC effect reliably occurred among
Norwegian readers in this study, not even among certain selected readers who were seen
as candidates for possessing a dual-route architecture and suffering within that architecture
the kind of competition in naming low-frequency irregular words that RFC is supposed to
eliminate. Thus, it was not possible to extend the applicability of a dual-route approach to
word recognition to the relatively shallow Norwegian orthography, and the question of the
architectural organization of Norwegian readers was essentially left unanswered by our data.
Key words: Concurrent memory task, Dual-route models, Word recognition
Much research effort has been devoted to understanding word recognition.
Models of word recognition can generally be classiﬁed as either dual-route
or single-route models, and many efforts to answer the question of whether
of word frequency and word regularity in naming tasks. It has generally been
found that high-frequency words are named more quickly than low-frequency
words (e.g., Forster & Chambers 1973; Scarborough, Cortese & Scarborough
1977), and that regular words are named more quickly than irregular or excep-
tion words (e.g., Baron & Strawson 1976; Stanovich & Bower 1978). Further,
frequency and regularity are typically found to interact, with regular words
named more quickly than irregular words only when the words are of low
frequency. Thus, regularity seems to have no effect on naming times when
high-frequency words are named (e.g., Paap, Chen & Noel 1987; Seidenberg,
Waters, Barnes & Tannenhaus 1984; Taraban & McClelland 1987; Waters &