A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Experimental Psychology Learning Memory and Cognition
This content is subject to copyright. Terms and conditions apply.
RESEARCH REPORT
The Role of Attention for Context–Context Binding of Intrinsic and
Extrinsic Features
C. Dennis Boywitt and Thorsten Meiser
University of Mannheim
There is converging evidence that the feeling of conscious recollection is usually accompanied by the
bound retrieval of context features of the encoding episode (e.g., Meiser, Sattler, & Weier, 2008).
Recently, however, important limiting conditions have been identified for the binding between context
features in memory. For example, focusing on the semantics of the stimuli during encoding eliminates
binding between perceptual context features (Meiser & Sattler, 2007). In the present research, we
investigated the interplay of the focus of attention during encoding and stimulus characteristics in
context–context binding. In particular, it has been suggested that context features differ in the degree to
which they can be regarded as intrinsic or extrinsic to the items and that intrinsic features might be given
more attentional processing during encoding than extrinsic features (e.g., Ecker, Zimmer, & Groh-
Bordin, 2007a). In two experiments, we manipulated the “intrinsicality” of context features to investigate
whether context–context binding might be limited to features that are in the focus of processing.
Multinomial modeling analyses revealed that while context–context binding was eliminated for inciden-
tally processed extrinsic context features (Experiment 1), it was preserved for intentionally processed
extrinsic context features (Experiment 2).
Keywords: binding, source memory, remember/know, multinomial modeling
Episodic memories are inherently multidimensional, including a
variety of different perceptual context features such as background
color and shape or semantic features such as gender or academic
status of a speaker. In addition, recent research has shown that
episodic remembering goes together with a more integrated rein-
stantiation of the encoding episode (Meiser & Bröder, 2002;
Meiser, Sattler, & Weier, 2008). This integrated retrieval of
context details has been termed context–context binding, indicat-
ing that context features are bound to the item and to each other,
forming a holistic representation of the feature configuration of the
encoding episode (Meiser & Bröder, 2002).
In laboratory research, context–context binding is operational-
ized as stochastically dependent memory for context features.
1
Memory for context features is stochastically dependent if the
probability of retrieving one context feature is higher when the
other context feature is also retrieved than when the other context
feature is not retrieved. For instance, Meiser and Bröder (2002)
had participants study items that varied in position on the screen
(upper vs. lower) and font size (large vs. small). Subsequently,
participants’ memory was probed for their recognition of the items
and for the context features position and font size. In addition,
participants were queried about whether they vividly recollected
the prior presentation of the items (i.e., remembered [R]) or
whether recognition was accompanied by a feeling of familiarity
(i.e., knew [K]). Over a variety of studies, it has consistently been
demonstrated that for items that are consciously recollected, re-
trieval of context features is stochastically dependent, whereas for
items that are merely familiar, context features are retrieved inde-
pendently (Meiser & Bröder, 2002; Meiser et al., 2008). This result
is not due to higher overall levels of context memory in R re-
sponses than in K responses. Meiser et al. (2008) equated overall
context memory between R and K responses experimentally and
still observed context binding uniquely for R responses. This
particular result also raises questions concerning the interpretation
that R responses differ from K responses only in terms of overall
memory strength. Although R/K judgments have also been ex-
plained by models assuming only one underlying process (see
Wixted & Mickes, 2010, for a recent critique of the dual-process
interpretation of RK data), the qualitative difference in context–
context binding associated with R/K judgments cannot easily be
accommodated by unidimensional models.
In line with the conceptualization of episodic memories as rein-
stantiations of the encoding episode, the feeling of conscious recol-
lection is thus associated with bound context memory of the encoding
episode. Nonetheless, previous research has identified important lim-
1
Please note that the term context (or context–context) binding as used
here refers to the phenomenon of stochastic dependency in memory for two
context features, not to the binding of context features to the item.
This article was published Online First January 16, 2012.
C. Dennis Boywitt and Thorsten Meiser, Department of Psychology,
University of Mannheim, Mannheim, Germany.
We thank Beatrice G. Kuhlmann and Jan Rummel for very helpful
comments on an earlier version of this article.
Correspondence concerning this article should be addressed to C. Dennis
Boywitt, Department of Psychology, University of Mannheim, Mannheim
D-68131, Germany. E-mail: boywitt@uni-mannheim.de
Journal of Experimental Psychology: © 2012 American Psychological Association
Learning, Memory, and Cognition
2012, Vol. 38, No. 4, 1099–1107 0278-7393/12/$12.00 DOI: 10.1037/a0026988
1099
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.