Reading and Writing: An Interdisciplinary Journal 16: 443–453, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
New ﬁndings for concreteness and imagery effects in written
, ERNEST T. GOETZ
, ANDREW G. STRICKER
THOMAS K. BURDENSKI, JR.
Texas A&M University, College Station, Texas, USA;
Vanderbilt University, Nashville,
University of Texas at Arlington, Arlington, Texas, USA
Abstract. This experiment investigated the effects of word concreteness and either imagery,
verbal, or control strategy instructions on the composition of written deﬁnitions. Results
revealed signiﬁcant effects of word concreteness on several quantity and quality variables, but
no signiﬁcant effect of strategy instructions or interaction between concreteness and strategy
instructions. Results of self-ratings of strategies actually used in composing revealed that a
mental imagery strategy was used with concrete words and a verbal strategy was used with
abstract words regardless of strategy instructions. Findings replicated the results of Tirre,
Manelis and Leicht [(1979) Journal of Reading Behavior 11, 99–106] in the production of
written composition on word relationships, and partially replicated the results of Sadoski,
Kealy, Goetz and Paivio [(1997) Journal of Educational Psychology 89, 518–526] in the timed
written production of word deﬁnitions. Results are interpreted from Dual Coding Theory and
levels of processing perspectives.
Key words: Concreteness, Dual Coding Theory, Imagery, Strategy instructions, Written
Language concreteness and mental imagery effects are pervasive in perform-
ing verbal tasks. Concrete language invites the mental capacity of forming
images (e.g., glittering diamond), whereas abstract language has relatively
less capacity to do so (e.g., conceptual thought). These effects have been
theoretically explained by Dual Coding Theory, which maintains that cogni-
tion involves the activity of two separate but interconnected mental codes, the
verbal code and the nonverbal code (Paivio, 1971, 1986; Sadoski & Paivio,
2001). In reading, concrete language has been found to be more imageable,
comprehensible, interesting, and memorable than abstract language even
when contextual variables are controlled (e.g., Paivio, Walsh & Bons, 1994;
Sadoski, Goetz & Avila, 1995; Sadoski, Goetz & Fritz, 1993a, b; Sadoski,
Goetz & Rodriguez, 2000). Research on concreteness and imagery effects in
written composition has received far less attention, however.