On the Existence of Semantic Working Memory:
Evidence for Direct Semantic Maintenance
Geeta Shivde
West Chester University
Michael C. Anderson
Medical Research Council Cognition and Brain Sciences Unit,
Cambridge, England
Despite widespread acknowledgment of the importance of online semantic maintenance, there has been
astonishingly little work that clearly establishes this construct. We review the extant work relevant to
short-term retention of meaning and show that, although consistent with semantic working memory, most
data can be accommodated in other ways. Using a new concurrent probe paradigm, we then report
experiments that implicate a semantic maintenance capacity that is independent of phonological or visual
maintenance that may build on a mechanism of direct semantic maintenance. Experiments 1 through 5
established that while subjects maintain the meaning of a word, a novel delay-period marker of semantic
retention, the semantic relatedness effect, is observed on a concurrent lexical decision task. The semantic
relatedness effect refers to slowed response times when subjects make a lexical decision to a probe that
is associatively related to the idea they are maintaining, compared to when the probe is unrelated. The
semantic relatedness effect occurred for semantic but not for phonological or visual word-form main-
tenance, dissipated quickly after maintenance ends, and survived concurrent articulatory suppression. The
effect disappeared when subjects performed our immediate memory task with a long-term memory
strategy rather than with active maintenance. Experiment 6 demonstrated a parallel phonological
relatedness effect that occurs for phonological but not semantic maintenance, establishing a full double
dissociation between the effects of semantic and phonological maintenance. These findings support a
distinct semantic maintenance capacity and provide a behavioral marker through which semantic working
memory can be studied.
Keywords: semantics, working memory, inhibition, attention
Much of human thought relies on the ability to temporarily store
and retain information in an active, highly accessible state. Al-
though considerable research in cognitive psychology and cogni-
tive neuroscience has addressed this ability in the context of
storing phonological and visual information, much of our mental
life concerns the processing of meaning. The ability to actively
maintain semantic representations underlies our success and effi-
ciency in nearly all complex cognitive activity, whether we are
solving a problem, devising a plan, deciding between options,
learning a new fact, comprehending an utterance, or preparing the
next thought for translation into written words. In this article, we
refer to the ability to maintain semantic representations in a stable,
highly accessible state as semantic working memory. Our primary
aims in this article are to build an empirical case for semantic
working memory and to argue that this capacity is supported by a
system that is functionally distinct from other known working
memory subsystems.
Despite a dramatic expansion of research on working memory
over the last two decades, surprisingly little attention has been
devoted to how people temporarily store semantic content. On
computational grounds, a system that maintains semantics in an
active state is seen as a necessary component to theories of higher
level cognition and is included in many computational models, and
this type of maintenance follows from the frameworks of several
authors (e.g., Cowan, 1995; Martin & Saffran, 1997; Ruchkin,
Grafman, Cameron, & Berndt, 2003). Theoretically targeted stud-
ies examining semantic maintenance are limited, despite its puta-
tive importance. Some attention was devoted to the issue in early
research on short-term memory, mainly in reaction to the claim
that short-term memory was fundamentally acoustic (Raser, 1972;
Shulman, 1970, 1972). But research on this topic never took firm
root in the classic literature, and later decades have seen little work
on the subject. The notable exception comes from studies with
neuropsychological patients showing that performance on phono-
logical and semantic maintenance tasks is anatomically dissociable
(Martin & Romani, 1994; Martin, Shelton, & Yaffee, 1994). These
findings provide the most focused evidence to date for a distinct
system supporting semantic maintenance (see also Haarmann &
Usher, 2001, for another approach to semantic maintenance).
This article was published Online First August 15, 2011.
Geeta Shivde, West Chester University; Michael C. Anderson, Memory
Group, Medical Research Council Cognition and Brain Sciences Unit,
Cambridge, England.
Portions of this work were completed in fulfillment of the requirements for
the dissertation of Geeta Shivde or were supported by a National Institute of
Mental Health grant to Michael C. Anderson. We are grateful to Douglas
Hintzman for his helpful advice, Timothy Carman and Sarah Parrish for help
running the experiments, and Nathan Foster for help in preparation of the
manuscript. We thank Douglas Hintzman, Ben Levy, Ulrich Mayr, and Mike
Posner for helpful comments on initial drafts of this paper.
Correspondence concerning this article should be addressed to Geeta
Shivde, Department of Psychology, West Chester University, West Ches-
ter, PA 19383. E-mail: gshivde@wcupa.edu
Journal of Experimental Psychology: © 2011 American Psychological Association
Learning, Memory, and Cognition
2011, Vol. 37, No. 6, 1342–1370
0278-7393/11/$12.00 DOI: 10.1037/a0024832
1342
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.