Behavioural Brain Research 214 (2010) 172–179
Contents lists available at ScienceDirect
Behavioural Brain Research
journal homepage: www.elsevier.com/locate/bbr
Basic operations in working memory: Contributions from functional
Christoph Bledowskia, Jochen Kaisera, Benjamin Rahma,b,∗
aInstitute of Medical Psychology, Goethe University, Heinrich-Hoffmann-Strasse 10, D-60528 Frankfurt am Main, Germany
bMedical Psychology and Sociology, Johannes Gutenberg University, Duesbergweg 6, D-55128 Mainz, Germany
a r t i c l e i n f o
Received 23 April 2010
Accepted 23 May 2010
Available online 1 June 2010
a b s t r a c t
Working memory (WM) constitutes a fundamental aspect of human cognition. It refers to the ability to
keep information active for further use, while allowing it to be prioritized, modified and protected from
interference. Much research has addressed the storage function of WM, however, its ‘working’ aspect
still remains underspecified. Many operations that work on the contents of WM do not appear specific
to WM. The present review focuses on those operations that we consider “basic” because they operate
in the service of memory itself, by providing its basic functionality of retaining information active, in a
stable yet flexible way. Based on current process models of WM we review five strands of research: (1)
mnemonic selection of one item amongst others, (2) updating the focus of attention with the selected
item, (3) updating the content of visual WM with new item(s), (4) rehearsal of visuospatial informa-
tion and (5) coping with interference. We discuss the neuronal substrates underlying those operations
obtained with functional magnetic resonance imaging and relate them to findings on “executive func-
attentional and mnemonic functions, with separable neural bases. However, interference processing and
the representation of rule switching in WM may demand an extension of the current WM models by
executive control functions.
© 2010 Elsevier B.V. All rights reserved.
1.1.Cognitive models of working memory....................................................................................................... 173
Neuroimaging studies of working memory operations .............................................................................................
2.1. Selection of an item representation.......................................................................................................... 174
2.2.Separation of selection and updating ........................................................................................................ 175
2.3.Updating the memory content ............................................................................................................... 176
2.5.Coping with interference..................................................................................................................... 176
3.1.Basic operations in WM and their relation to executive functions.......................................................................... 177
Center, Johannes Gutenberg University, Duesbergweg 6, 55128 Mainz, Germany.
Tel.: +49 6131 39 20501; fax: +49 6131 39 22750.
E-mail address: firstname.lastname@example.org (B. Rahm).
Higher-order cognition such as reasoning and problem solving
to reach an optimal solution or a correct conclusion. Herein cogni-
tion strongly relies on the ability to retain some information active
for further use, and to do so in a flexible way allowing information
0166-4328/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
“working memory”, and ever since its introduction by Miller et al.
, cognitive and clinical neuroscientists have attempted to elu-
cidate its components, associated processes and their underlying
In the neurosciences, much of the research on working mem-
ory (WM) has concentrated on one of its most pertinent qualities:
to retain information over some period of time, when the physi-
cal stimulus is no longer present. To accomplish this, the contents
of WM are vastly reduced compared to the wealth of information
continuously streamed by our senses. The most commonly noted
some perceptual details may be lost in the mnemonic stimulus
tion and memory, many processes that operate on sensory stimuli
may operate equivalently on WM contents. For example, judg-
ing which of two simultaneously presented visual stimuli has a
higher contrast can similarly be performed on mental representa-
between memory items, and both processes involve largely over-
lapping brain regions [3,4].
This notion is in accordance with the contemporary view that
of brain regions separate from those subserving perception [5–9].
Instead WM is seen as emergent from the interaction of higher
sensory, attentional and mnemonic component processes involv-
ing their underlying brain regions. For example, higher visual areas
subserving the perception of faces and places are also involved in
retaining information on faces/places in WM . Activity in per-
top-down signals, i.e. attention, from prefrontal and parietal cor-
tex, allowing us to maintain stimulus information in WM for some
roanatomical bases, perception and memory may also share many
operations. Thus, many operations performed on the contents of
WM are most likely not uniquely and specifically dedicated to WM.
As a consequence, simply listing potential operations on the con-
tents of WM is not appropriate for identifying the basic operations
that compose WM (and reviewing the evidence for these). Rather
we seek to review those basic operations that make WM emerge –
even if they may be “borrowed” from e.g. attentional and general
We consider those operations as “basic” that in the context of
memory tasks operate in service of memory itself, to provide its
basic functionality of retaining information active, in a stable yet
flexible way. We focus on operations that go beyond the mere pas-
sive maintenance of stimuli, and also leave aside the principles of
encoding items into WM. In short, the focus of the present review
is on those processes that make WM work, not on all possible oper-
ations that can be applied to the contents of WM. This distinction is
We try to define each of the selected operations in the theoreti-
cal context provided by Cowan and Oberauer’s process model of
WM [11,12] which will be described in some detail below. This
will be followed by summaries and discussions of studies on the
following operations in WM that we consider as “basic” for its
not all the representations held in memory are of equal importance
all the time, thus requiring some sort of selection or prioritization
according to their momentary relevance. For example at an airport,
you may have to remember different pieces of information like the
terminal number, flight number, gate number and your seat. Their
momentary relevance depends on which steps you have already
taken. You will not need your seat number before boarding the
plane, but by then the terminal number has long become unimpor-
prioritizing items is the first basic operation we review. (2) Focus
updating: recently, on the foundation of Cowan’s and Oberauer’s
attentional focusing, which naturally co-occur and in consequence
had not been studied separately in the studies reviewed in the
“selection” section, have dissociable neural substrates. (3) Content
updating: returning to the example, arriving at the check-in you
may find that your ticket has been re-scheduled to another flight
due to overbooking. You therefore have to discard the formerly rel-
literature this type of requirement mostly is termed “updating” of
memory contents. (4) Rehearsal: in order not to have to look up the
flight info repeatedly you may silently rehearse it to prevent for-
getting. (5) Coping with interference: when re-scheduled, or just
by being confronted with announcements for other flights, your
WM contents are easily overwritten or confused with one another.
Coping with potentially interfering signals is another basic func-
tion that is of vast importance in real-life settings. While selection,
of coping with interference are main sources of maintaining items
stable in WM.
ations and their potential neural substrates. They are related to
“executive functions” and cognitive control.
1.1. Cognitive models of working memory
According to the seminal “multiple-component model” by Bad-
deley and Hitch , WM has been conceptualized as a system
made up of two specialized temporary memory buffers (a phono-
logical and a visuospatial store), and a supervisory system (the
central executive). While the storage systems hold and refresh
memory traces in their dedicated content domain for a few sec-
onds, the central executive is involved in control and regulation
of WM. Even though the authors did not explicitly assign its
components to structural anatomical regions , much of the
neuroscientific research stimulated by this model did interpret
its components literally and sought for dedicated brain regions.
In contrast to Baddeley and Hitch , Cowan  explicitly
considers WM not as a separate system but as a functional
state that allows a direct access to information. Specifically, his
“embedded-processes model” combines hierarchically arranged
faculties comprising long-term memory (LTM), the subset of LTM
that is currently activated and the subset of activated memory that
tations of incoming stimuli and previous cognitive operations. The
portion of activated memory that is not in the focus of attention is
but their number is tightly limited. The focus of attention may hold
up to four of the activated representations. As commonly assumed
for attention, the focus can be driven by exogeneous stimuli or be
voluntarily controlled and directed.
Oberauer  extended Cowan’s model  by adding a fur-
ther component, a more narrow focus of attention that holds only
four-element focus refer to two different functional states of acces-
sibility for cognitive processes. Whereas the four-element focus
single representation that is actually selected as the object of the
next cognitive operation. Thus, Oberauer  argued that the role
but to hold a single object already selected for processing.
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
Fig. 1. Oberauer’s model of working memory (adapted from ). Nodes and lines
represent a network of long-term memory representations, some of which are acti-
vated (black nodes). A subset of these items is held in a region of direct access (gray
field). Within the region of direct access, one item is selected for processing by the
focus of attention (dark gray circle).
2. Neuroimaging studies of working memory operations
2.1. Selection of an item representation
Most cognitive tasks require simultaneously holding a number
in WM become transiently more important than others. Hence,
all items are to be retained active, one of them has to be selected
ity and becoming accessible to further cognitive operations. In this
section we review studies that have examined these processes.
gated which brain regions were involved in the process of selecting
an item in mind. In their studies, selection has been operationally
defined as the mental capacity to choose one out of three spatial
positions held in WM on the basis of an indirect cue. Specifically,
participants maintained three dot positions over a delay phase of
varying length (9.5–18.5s), after which a cue line indicated which
position should be retrieved to guide a response (moving a cursor
to the selected position using a joystick) . In control trials, the
visual and motor components of the task were similar but the par-
ticipants were not required to remember spatial locations or select
a position in memory, but the position was indicated directly. In a
second study  participants were asked to maintain both spatial
the first, second or third position had to be selected for response.
tions but did not demand mnemonic selection. Brain activations
associated with the selection of an item in WM were observed
in prefrontal (dorsolateral, orbitofrontal and ventral) and parietal
(medial parietal and intraparietal) cortex. In contrast, the mainte-
nance of the positions (contrasted with equivalent non-memory
periods in control trials in the first study, with baseline in the
second study) was associated with bilateral activation of superior
prefrontal and intraparietal cortex. Based on these results Rowe et
al. concluded that in particular the activity of the prefrontal cortex
(PFC) is closely tied to the process of selection and does not reflect
memory storage as such.
Johnson et al. [17–19] studied the neural correlates of bringing
a memory to one’s mind with a different approach: they defined
a refreshing operation as an act of thinking back to a single just-
seen stimulus, by which this item is “foregrounded”, or – in model
terms – is selected and moved in the focus of attention. Most of
the studies by Johnson and colleagues used verbal materials but
some of them also included visual stimuli. For example, in a study
by Johnson et al.  participants read a word or viewed a visual
stimulus (line drawing of an object or abstract pattern). After a
short delay (550ms) subjects were cued to think back to the just-
presented item. Similar to Rowe et al. [15,16], greater activity in
the left PFC was associated with the refreshing operation as com-
pared to a control condition in which subjects viewed a new item
or the same item again. In addition, some differences in the distri-
bution of activity across left PFC were observed depending on the
type of material being refreshed. These results were confirmed by
a meta analysis  that identified similar regions in left PFC (dor-
solateral, anterior, and ventrolateral) associated to varying degrees
with refreshing different types of information (visual and auditory
by Johnson et al. [17,18] focused on the role of the PFC; however,
refreshing-related activities were also present in the parietal lobes
(see for example Fig. 2 in ). In a more recent study using a
slightly modified paradigm  participants saw slides including
faces and scenes and, after a short delay, either viewed another
image or saw a cue that instructed them to perform a refresh of
the just-seen picture (face or house). Again, refresh-related activ-
ity in PFC was observed. Interestingly, refreshing faces and scenes
also modulated activity in the fusiform face area and the parahip-
pocampal place area, respectively. These regions are known to
be selectively involved in face versus scene perception. Refresh-
ing thus may lead to a re-initiation of activation in the respective
feature-coding region of extrastriate cortex that may be similar to
attentively encoding a stimulus into WM.
The studies by Rowe et al. and Johnson et al. [15–19] have
However, these studies also included the act of focusing attention
on this item. A series of studies conducted by Nobre and cowork-
ers [20,21] addressed the orienting of attention to perceptual as
opposed to memory items. Specifically, Nobre et al.  examined
whether the same neural system that is involved in shifting atten-
spatial attention between locations stored in WM. They presented
spatial cues before (precues) or after (retrocues) the sample array
embedded in a visual delayed-match-to-sample task. While the
precue enabled participants to shift attention toward the relevant
location in the upcoming sample array, the retrocue required par-
ticipants to shift attention toward the relevant location in the array
held as a mental representation in WM. Both cues predicted the
likely location of the probe stimulus and led to significant ben-
efits in accuracy and reaction time compared to uninformative
cues. Brain areas supporting spatial shifts of attention to exter-
nal or internal events were identified by comparing activations in
response to precues and retrocues with those elicited by neutral
cues. Nobre and colleagues  found common activations bilat-
erally in the posterior parietal cortex (including precuneus [PCN],
superior parietal lobule, intraparietal sulcus) and frontal cortex
(including caudal superior frontal sulcus [cSFS]) that overlapped
frequently been observed in typical spatial attention paradigms
that do not impose strong memory demands [22–24]. In addition,
[15–19]. In a follow-up study Lepsien et al.  additionally var-
ied the number of items in the sample array in order to separate
the contributions of retro-cueing versus simple memory storage.
from attentional tasks were activated during orienting the focus of
attention to an internal representation.
used a parametric manipulation of directing attention to spatial
positions within WM. They confronted participants with a variable
number of cue arrows (0, 3, 6, 9, and 12) during the delay phase
indicating to which of the four neighboring cells within a 4 by 4
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
Fig. 2. (a) Paradigm by Bledowski et al.  for an independent manipulation of selection and focus updating. Subjects saw four dots (black) and retained their locations in
working memory (indicated here by white squares for illustration only). After a variable interval, a cue line ran through the vicinity of one or two of the dots. Participants
were asked to select in their mind the location of the dot that was closest to the cue line and maintain it as a specific target item. The number of cue presentations varied
from two up to five. At the end of a trial subjects judged whether a probe matched the dot position that had been cued by the immediately preceding line and received a
feedback. (b) The sequence of cues determined the demands for updating: during the high-selection/update condition (+S+U), the cue line ran through the vicinity of two of
the encoded dots requiring the selection of the closer dot position. In addition, the cue indicated a different target dot as on the immediately preceding cue presentation, thus
needing an update of the focus of attention. In the high selection/no-update condition (+S−U), the cue was ambiguous too, but indicated the same target as the preceding
cue, hence not requiring a change of the focus of attention. In the low-selection/update condition (−S+U), the cue line ran through the vicinity of only one dot and required
an update of the focus of attention. In the low-selection/no-update condition (−S−U), a cue unambiguously denoted the same target as in the preceding presentation. (c)
Distinct set of activations sensitive to the manipulation of selection (yellow) and focus updating (red) projected onto the surface cortex reconstruction of the SPM canonical
single-subject brain. Areas of overlap are shown in orange. For details see Figure 3 and Table 1 in Bledowski et al. .
grid a target position should be mentally moved. With increasing
number of attentional shifts activity increased monotonically, in
particular in the superior frontal and posterior parietal cortex, i.e.
in the very same set of frontal and parietal regions as observed by
Nobre and colleagues [20,21].
In summary, studies on the selection of an item representation
in memory have yielded activations in diverse brain regions, such
as medial and lateral prefrontal and parietal cortex. On the model
level, these are likely to represent a mixture of activations due
to the selection of an item representation per se and the ensu-
ing attentional focusing of this item. A differential weighting of
these two operations across the referenced studies may explain
some seemingly contradictory results. Nobre et al.  concluded
Rowe et al. [15,16] had termed their process under examination
“attentional selection”, and found that prefrontal cortex, but nei-
ther superior frontal nor parietal cortex played a significant role in
the selection process. However, in the studies by Nobre et al. ,
the analyses aimed at elaborating the commonalities between per-
ceptual and mnemonic shifting, in consequence their results may
be based more on attentional focusing than on memory selection
(that had not been demanded in the perceptual condition). In con-
trast, in the pioneering study by Rowe and Passingham , both
of attention to the target position, but memory selection had only
been necessary in the experimental condition, in which they found
increased dlPFC activation.
Thus, most likely a mixture of two different processes was
studied by Nobre and Rowe [15,21] using similar labels, lead-
ing to apparently discrepant results. These may be reconciled by
assuming that mnemonic selection and attentional updating had
been involved in varying proportions across studies. We sought
to gather evidence on these two processes and identify their sub-
strates by varying them independently in a single study which will
be described in the next section.
2.2. Separation of selection and updating
As in the studies described in the previous section, selection
and updating mostly work hand in hand: the relevant item is
retrieved and the focus of attention is directed to this retrieved
item. We recently investigated whether despite their high rate
of co-occurrence these hypothetically different operations have
separable neural substrates . In our spatial WM paradigm par-
ticipants held four locations in WM while they were confronted
with a series of cues, the position and sequence of which were
designed to impose variable demands for selection among alter-
native candidate items, independent of the requirement to update
ipants were asked to focus in mind the memorized position closest
to the line in order to be able to perform rapidly a subsequent and
unpredictable match-to-sample test. Each of the series of line cues
determined a low or high need for selection depending on whether
requiring the retrieval of a single versus two items. With respect
to updating, the very same position could be cued by two succes-
sive cues. Here the focus would not be updated but remain on the
same item. Alternatively a different position could be indicated,
thus demanding an update of the attentional focus.
We found that selection and updating are subserved by distinct
brain regions supporting the hypothesis that they may represent
distinct WM operations (Fig. 2c). In particular, the set of brain areas
exclusively sensitive to the manipulation of selection comprised
left rostal superior frontal sulcus (rSFS) and posterior cingulate
cortex/precuneus (PCC/PCN). In contrast, whenever an update of
the focus of attention was needed we observed activation of the
bilateral cSFS and posterior parietal cortex (PPC), regardless of the
between effects of selection and updating.
Thus, reconciling the previously heterogenous results we could
show that, in keeping with Nobre, Leung and colleagues [21,26],
cSFS and PPC are implicated in focusing attention on a memory
representation. The selection operation that retrieves the most
relevant memory item, on the other hand, was subserved by a dis-
tinct set of fronto-parietal areas, rSFS and PCC/PCN. Even though
selection-related activity in PCC/PCN was reported by Rowe and
Passingham , our results were in contrast to Rowe’s and John-
son’s [15–19] interpretation who had assigned dlPFC a key role
in the selection of a memory representation. In our study, dlPFC
was activated above baseline in all four selection/updating condi-
tions. This may suggest that the dlPFC activation for selection is
present even with relatively low selection demands and does not
depend on the number of re-activated or attended items between
which one has to select (see paradigm descriptions for Rowe’s and
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
Johnson’s studies above). Thus it may possibly not reflect selec-
tion as such, but may be elicited by being prepared for selection
already. In contrast, activity of the rSFS and PCC/PCN varied with
demands on selection between items in memory. These areas are
known for their sensitivity to retrieval processes in episodic mem-
ory, in particular when recall of a specific item is required .
This corresponds well to both Oberauer’s model  and behav-
ioral data indicating that the access to representations stored in
LTM is mediated by the same cue-based mechanism that retrieves
representations that are outside of the focus of attention [29,30].
Moreover, there is also a high degree of overlap between our
selection-sensitive regions and the activation foci from studies on
perceptual decision making [31,32]. We acknowledge that because
our task also involved a decision of which of the remembered posi-
tions is closest to the cue line, we cannot definitely decide whether
representations is most relevant or the selection as such.
2.3. Updating the memory content
tion in the preceding paragraph, several studies have investigated
the “updating” of the WM contents. While the former defini-
tion refers to a change of the attentional focus, making another
item accessible to operation while leaving the contents of WM
untouched, the latter refers to updating in the sense of a change
of the WM contents. In most studies this involved the replacement
of a stored, but now irrelevant item with a new input.
Earlier studies have examined updating of WM contents with
the n-back task, in which participants are presented a continu-
ous series of items (typically letters) and are asked to indicate
whether the given item matches the one from n steps earlier
in the sequence. Performance on the n-back task was associated
with activity changes broadly distributed over the brain including
mainly the prefrontal regions (dlPFC, inferior frontal and anterior
cingulate cortex) as well as posterior parietal cortex (for exam-
ple see review by Collette et al. ). However, the results of
these studies cannot be interpreted unambiguously because each
item presentation not only requires recognition, evaluation and
encoding activity but also the replacement (content updating) and
jects also had to inhibit a positive response when an item matched
a memory representation but had a different sequential position
(e.g. one instead of two steps ago).
In order to overcome these limitations Roth and colleagues 
have developed a new task in which participants viewed a contin-
uous stream of either faces or houses and indicated whether the
current stimulus matched a sample stimulus defined as the first
object in the task block. In addition, every 4–10s, participants were
cued with one of two well-memorized faces or houses (learned
before the experiment) either to continue to maintain the current
target or to replace it with the next picture as the new sought-
after target (update of the WM contents). Contrasting the update
cue events with the maintenance cue events revealed a transient
increase of activity in attention-related regions (cSFS and PPC).
Additional activations were observed in several areas including
dlPFC, inferior frontal junction, anterior cingulate cortex, inferior
occipitotemporal and parahippocampal gyrus.
In consistence with findings described in the previous para-
graphs, the results presented by Roth et al.  fit to the view that
cSFS and PPC subserve a basic operation that directs the focus of
attention either towards a memory representation or towards new
frontal cortex may mediate visual processing and encoding of new
information into WM as required when updating the contents of
The selection and updating processes described in the previous
sections provide flexible use of WM by allowing for prioritization
of memory items and updating the stored information. Hereby
these operations promote the adaptation to varying task contexts
and changes. They may, however, additionally subserve represen-
tational stability. For spatial information the hypothesis has been
put forward that it is retained by attention-based rehearsal, that
is by serial shifts of attention to the to-be-remembered positions
[3,6,35,36]. Thus, rehearsal of spatial memoranda could be con-
ceived of as a series of attentional updating events. In accordance
with this notion, fMRI recordings have revealed activation in cSFS
and PPC during visuospatial rehearsal [37–39].
2.5. Coping with interference
In real-life situations the stability of item representations is
threatened by tempting but irrelevant information from various
sources: (1) external stimuli may act as distracters and have a
detrimental effect on the stable maintenance of items in memory
by items that had been relevant previously (“proactive interfer-
ence”). (3) Items may be confused with one another (“inter-item
interference”). We are not aware of any functional imaging study
concerning interference amongst memory items, so the following
paragraphs will only deal with distraction and proactive interfer-
In the model, distraction may operate on multiple levels: First, a
WM item may be overwritten by a distracter. Second, an irrelevant
item may capture attention in a bottom-up manner and thereby
direct attention away from items that should be held active in
WM, leaving them prone to decay. Third, an attention-demanding
task (in dual task paradigms) interrupts WM processing, both by
diverting attention and by disrupting potential efforts of rehearsal.
WM processing in the presence of interfering external stim-
uli has consistently been found to elicit activation of the dlPFC
[40–44]. In an influential study Sakai et al.  used a delayed-
match-to-sample task which incorporated an additional distracter
task within the delay period. In particular, participants memorized
a sequence of five red spatial positions. During the delay period the
distracter task, a spatial delayed-match-to-sample task, required
them to encode, remember and recognize the positions of five blue
dots. After completing distracter probe recognition, participants
were tested on their memory for the serial order of the red posi-
tions from the main task. Sakai et al.  showed that dlPFC was
selectively involved in protecting WM representations against dis-
traction. More specifically, they found sustained activity in dlPFC
in the distracter condition (as compared to control condition with-
out distracting task) during the memory delay on correct trials
but not on error trials. They argued that in contrast to “simple
maintenance” the dlPFC is involved in “active maintenance”, i.e.
in protecting memory against interference from distracting and
irrelevant stimuli. In that study, interference may have displayed
its effects either via capturing attention, hereby leaving WM items
prone to decay, or by the interruption of attentional processing in
coping with the dual task situation. A recent study by Clapp et al.
 suggests dissociable bases of interruption of delay-spanning
WM processing by multi-tasking as opposed to mere distraction by
irrelevant external stimuli. In their study, dlPFC was more strongly
activated by multi-tasking than by distraction, and showed a dis-
tinct connectivity pattern with visual association cortex between
The second type of interference, “proactive interference” (PI),
is mostly operationalized by presenting a probe stimulus in a
delayed-match-to-sample task, that is not part of the current set
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
of items held in WM, but was an item of the previous trial (“recent
probe”), leading to increased reaction times and error rates when
compared to “non-recent probe” trials. In Cowan’s model  the
previous contents of WM and perception are thought to remain
activated portions of memory, that gradually fall back into their
inactivated state, as they are not actively attended.
Converging results from a number of neuroimaging studies
implicate mid-ventrolateral PFC (vlPFC) and – less consistently
– anterior prefrontal cortex and the anterior cingulate cortex in
the processing of PI [45–48]. From increased activation on “recent
probe” trials alone we cannot tell whether this reflects the activa-
tion of the memory trace, conflict resolution, or a response conflict
due to the need to respond negatively to a familiar item. Nelson et
al.  investigated whether this activation is specific for PI in the
context of WM by imposing conflict in a WM task and a semantic
verb generation task. Activation of vlPFC was found in both, sug-
from both WM and semantic memory. Strong evidence for a causal
role of vlPFC in PI resolution has been provided by a recent study
by Feredoes et al. . The authors used TMS to interfere with con-
flict processing, and found stimulation of the vlPFC but not control
sites to disrupt the accuracy for PI trials. Last, Jimura et al.  used
a modified paradigm to dissociate PI from response requirements.
They found that PI led to stronger activation in vlPFC independent
of response requirements.
ing, PI seems to impose a basic, possibly semantic mechanism of
interference resolution subserved by vlPFC. However, it should be
noted that in contrast to the other studies reviewed here, all stud-
ies mentioned above have used verbal instead of visual or spatial
including verbal, spatial and visual-nonspatial types into a single
study, and did not find significant activation of vlPFC. However,
the different materials showed vast differences in their capabil-
ity to raise behavioral effects of PI, and this variability may have
eliminated potential effects in vlPFC for some of the non-verbal
materials. Similarly, Badre and Wagner  did not find reliable
vlPFC activation for abstract visual pattern stimuli. However, in
their study, matches were not correctly recognized in on average
more than 30% of the trials, which strongly limits the conclusions
that may be drawn from this result: it is hard to imagine that a
the following trial. A study by Rahm et al.  has found conflict-
related activation at a similar location using color stimuli. In that
study, stronger vlPFC activation was observed both when a target
color was presented as a probe that was flanked by a non-target
distracter stimulus or when a non-target probe was flanked by a
nism of disambiguation. However, due to the apparent differences
in the tasks used and their exact requirements, it remains specula-
tive to assume that processing of PI in the visual or spatial domain
will entail activation of vlPFC.
Much evidence has been accumulated on the storage aspect of
WM, however, its ‘working’ aspect is still underspecified. Many
but are not functions of WM as such. Here we sought to isolate
those operations that are “basic” for WM functioning, i.e. that keep
momentarily relevant information in a stable, yet flexibly accessi-
ble state. These operations are not necessarily self-contained but
may be “borrowed” from other cognitive functions, in particular
attention and memory. Based on Cowan  and Oberauer’s 
updating the focus of attention with the selected item, (3) updat-
ing the content of visual WM with new item(s), (4) rehearsal of
visuospatial information and (5) coping with interference. The first
two operations underlie the flexible prioritization of items already
held in WM. Updating of WM contents additionally demands the
encoding of a new item. Rehearsal and diverse mechanisms of pro-
especially in the presence of other, (now) irrelevant information.
Rehearsal and content updating may largely build upon selection
and attentional updating, but comprise additional mechanisms,
and were therefore treated in separate paragraphs.
Despite some variability, distinct sets of brain regions are likely
to support different basic operations in WM. In particular, the acti-
vation of rSFS and PCC/PCN may specifically subserve mnemonic
selection of an item in WM . The role of dlPFC for selection
has been emphasized in several studies, however, it was insensi-
tive to the manipulation of selection demands . In contrast,
updating the focus of attention activates bilateral cSFS and PPC
[21,25–27]. Updating the WM content additionally engages extras-
triate, ventrolateral and medial frontal cortex which may support
encoding of the new input, along with attentional and mnemonic
processes . Rehearsal seems to primarily involve the mecha-
nisms of updating the attentional focus. Finally, studies of coping
with interference have most consistently reported activation of
dlPFC in the presence of distraction [40–44] and of vlPFC during
resolving PI from previous trials .
Taken together, the presented evidence supports the view that
WM emerges from interactions between brain regions support-
ing higher sensory, attentional and mnemonic systems [3,5–9].
Yet, the results accentuate the independent contributions of the
mnemonic and the attentional component functions, and control
over interference does not seem to be fully explained in terms
of a conception taking into account memory and attention only.
Concerning attention, our results  are in line with evidence
presented by Nobre and colleagues  showing that activations
of rSFS and PPC were activated by shifts of attention to both visual
of WM functioning, we have noted that the same brain regions that
retrieving information from episodic memory [27,28]. While here
we reviewed studies on visual WM only, there is some neuroimag-
ing evidence on verbal WM that is explicitly based on Cowan and
Oberauer’s model [11,12], and these studies have yielded compa-
rable results. For example, Nee and Jonides  instructed their
participants to remember short word lists in a delayed-match-to-
sample task, while varying the list position matching the probe.
retrieval from LTM than when probing items in the focus of atten-
tion. Again, these results highlight the separation of mnemonic and
attentional contributions to WM.
3.1. Basic operations in WM and their relation to executive
Here we have introduced the term “basic operations” to charac-
terize those functions that help WM provide its basic functionality.
This term is not currently used in the literature and the main pur-
pose of introducing it was twofold: (1) to put the focus on the
operational – “working” – aspect of WM while at the same time
keeping it at a basic level, not implying more complex cognitive
functions or the multiplicity of operations that may be performed
on the contents of WM. Rather, the term “basic operations” refers
to those functions that effectively form the workspace on which
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
higher cognitive functions may act. (2) We sought to distinguish
them from so-called executive functions, in order to keep a rela-
tively narrow focus on WM and its functions.
The term “executive functions” has been applied to a wide
variety of phenomena, making it difficult to formulate a com-
monly agreed definition. Conceptions range from early lists of
classes of situations in which executive control is required ,
multivariate analyses of behavioral data  to a hierarchical orga-
of these conceptions may be that executive or control functions
can be voluntarily initiated to accomplish controlled sensory, cog-
nitive and motor processing by biasing information flow through
top-down signals at various levels.
In the WM literature, the most popular account of executive
directly on the early conception of an attentional control system
proposed by Shallice and Norman . According to Baddeley 
the central executive has not been localized to a circumscribed
brain area in the model, but empirically, the functions ascribed
to it have mostly been assigned to PFC. Using a latent variable
approach, Miyake et al.  have shown that the concept of exec-
utive functions may not be monolithic but actually form a set of
related, but clearly distinct executive functions comprising “inhi-
bition of prepotent responses”, “shifting between tasks and mental
bition” has later been differentiated into control over PI, resistance
against distracters and prepotent response inhibition, with obvi-
are potentially many mechanisms of how control over interfering
information of diverse sources may be implemented in the brain,
only few of which have been examined so far. Also it remains an
ference control or resolution mechanisms. Miyake’s  executive
function of “updating and monitoring of WM representations”, in
comparison to the content updating studies reviewed here, may
rely to a much greater degree on the continuous evaluation and
implementation of changes in WM, as examined in early neu-
roimaging studies using the n-back task. As described earlier, in
these tasks, resolution of interference, counting, discarding and
encoding of information cannot be dissolved. Therefore content
updating may only be partly related to Miyake’s  “updating
and monitoring” functions. Finally, the relation of WM to “shift-
ing between tasks and mental sets” is of utmost importance for
defining both WM and executive control, as the implementation
of a mental task set crucially involves the durable representation
of rules. The establishment of task sets has been linked to anterior
PFC [58,59], while dlPFC, posterior inferior frontal sulcus, inferior
frontal junction and medial frontal cortex have been repeatedly
reported to underlie switching between task sets and contextu-
ally adequate action [60,61]. Recent models have conceptualized
prefrontal functions as a posterior-to-anterior hierarchy of control,
reflecting increasingly abstract levels of control [55,56]. At each
of these levels rules need to be represented. The question arises
whether rule representations and the selection/updating of rules
draw upon the same limited resources as the selection/updating
of stimulus representations reviewed here. In a pioneering study,
Montojo and Courtney  have examined this question by inde-
pendently varying whether a number stimulus or a mathematical
operation had to be updated in WM. Their results show that rule
ble areas. Number updating yielded stronger activation changes in
changes in the inferior frontal junction. Thus, the representation of
task set and WM contents may rely on differentiable or even partly
separate neural substrates. Yet, we note that this should not be
taken to presume that task set does not play a role in WM tasks: as
in the classical studies by Sternberg  on verbal memory, most
tasks on visual WM use very small sets of stimuli. Given that on
each trial, all stimuli had been displayed rather recently, memory
performance may not solely rely on sensory information. Rather
categorically tagging items as “matches” could invoke activation in
regions representing task set or stimulus-response mappings.
Taken together, the functions subserved by WM draw upon
involve content updating, but performance may mainly depend on
Working memory is currently conceived as emerging from the
interplay of multiple cognitive functions. In consequence, it can-
not be defined autonomously, but only by identification of those
basic functions that together are conceived as “working mem-
ory”. Mnemonic selection and attentional focusing form the core of
Cowan and Oberauer’s [11,12] conceptions, and are subserved by
dissociable neural substrates. They may also be the basic prereq-
uisites for updating the content of WM and rehearsal. Hence, the
interaction of attention and memory that is currently emphasized
nisms and the unknown role of task set/rule representation in WM
by executive functions/cognitive control.
 Miller GA, Galanter E, Pribram KH. Plans and the structure of behavior. New
York: Henry Holt and Company; 1960, 226 pp.
 Cowan N. The magical number 4 in short-term memory: a reconsideration of
mental storage capacity. Behav Brain Sci 2001;24:87–114.
 Awh E, Jonides J. Overlapping mechanisms of attention and spatial working
memory. Trends Cogn Sci 2001;5:119–26.
 Lepsien J, Nobre AC. Cognitive control of attention in the human brain:
insights from orienting attention to mental representations. Brain Res
ing memory. Trends Cogn Sci 2003;7:415–23.
 Jonides J, Lewis RL, Nee DE, Lustig CA, Berman MG, Moore KS. The mind and
brain of short-term memory. Annu Rev Psychol 2008;59:193–224.
 Pasternak T, Greenlee MW. Working memory in primate sensory systems. Nat
Rev Neurosci 2005;6:97–107.
 Postle BR. Working memory as an emergent property of the mind and brain.
 Slotnick SD. Visual memory and visual perception recruit common neural sub-
strates. Behav Cogn Neurosci Rev 2004;3:207–21.
 Ranganath C, Cohen MX, Dam C, D’Esposito M. Inferior temporal, prefrontal,
and hippocampal contributions to visual working memory maintenance and
associative memory retrieval. J Neurosci 2004;24:3917–25.
 Cowan N. Evolving conceptions of memory storage, selective attention, and
their mutual constraints within the information processing system. Psychol
 Oberauer K. Access to information in working memory: exploring the focus of
attention. J Exp Psychol Learn Mem Cogn 2002;28:411–21.
of learning and motivation: advances in research and theory, vol. 8. New York:
Academic Press; 1974. p. 47–89.
 Baddeley A. Working memory. Curr Biol 2010;20:136–40.
 Rowe JB, Passingham RE. Working memory for location and time: activity in
prefrontal area 46 relates to selection rather than maintenance in memory.
 Rowe JB, Toni I, Josephs O, Frackowiak RS, Passingham RE. The prefrontal
cortex: response selection or maintenance within working memory? Science
 Johnson MK, Raye CL, Mitchell KJ, Greene EJ, Anderson AW. FMRI evidence
for an organization of prefrontal cortex by both type of process and type of
information. Cereb Cortex 2003;13:265–73.
C. Bledowski et al. / Behavioural Brain Research 214 (2010) 172–179
 Johnson MK, Raye CL, Mitchell KJ, Greene EJ, Cunningham WA, Sanislow CA.
Using fMRI to investigate a component process of reflection: prefrontal corre-
lates of refreshing a just-activated representation. Cogn Affect Behav Neurosci
modulate activity in extrastriate visual areas: top-down effects of refreshing
just-seen visual stimuli. Neuroimage 2007;37:290–9.
 Lepsien J, Nobre AC. Attentional modulation of object representations in work-
ing memory. Cereb Cortex 2007;17:2072–83.
 Nobre AC, Coull JT, Maquet P, Frith CD, Vandenberghe R, Mesulam MM. Orient-
ing attention to locations in perceptual versus mental representations. J Cogn
 Corbetta M, Shulman GL. Control of goal-directed and stimulus-driven atten-
tion in the brain. Nat Rev Neurosci 2002;3:201–15.
Annu Rev Neurosci 2000;23:315–41.
 Yantis S, Serences JT. Cortical mechanisms of space-based and object-based
attentional control. Curr Opin Neurobiol 2003;13:187–93.
 Lepsien J, Griffin IC, Devlin JT, Nobre AC. Directing spatial attention in mental
representations: interactions between attentional orienting and working-
memory load. Neuroimage 2005;26:733–43.
 Leung HC, Oh H, Ferri J, Yi Y. Load response functions in the human spa-
tial working memory circuit during location memory updating. Neuroimage
 Bledowski C, Rahm B, Rowe JB. What “works” in working memory? Sepa-
rate systems for selection and updating of critical information. J Neurosci
 Wagner AD, Shannon BJ, Kahn I, Buckner RL. Parietal lobe contributions to
episodic memory retrieval. Trends Cogn Sci 2005;9:445–53.
 McElree B. Accessing recent events. In: Ross BH, editor. The psychology of
learning and motivation, vol. 46. San Diego: Academic Press; 2006. p. 145–99.
 Heekeren HR, Marrett S, Bandettini PA, Ungerleider LG. A general mechanism
for perceptual decision-making in the human brain. Nature 2004;431:859–62.
 Heekeren HR, Marrett S, Ruff DA, Bandettini PA, Ungerleider LG. Involvement
of human left dorsolateral prefrontal cortex in perceptual decision making is
independent of response modality. Proc Natl Acad Sci USA 2006;103:10023–8.
 Collette F, Hogge M, Salmon E, Van der Linden M. Exploration of the neural
substrates of executive functioning by functional neuroimaging. Neuroscience
 Roth JK, Serences JT, Courtney SM. Neural system for controlling the contents
of object working memory in humans. Cereb Cortex 2006;16:1595–603.
 Awh E, Vogel EK, Oh SH. Interactions between attention and working memory.
 Postle BR, Ferrarelli F, Hamidi M, Feredoes E, Massimini M, Peterson M, et
al. Repetitive transcranial magnetic stimulation dissociates working memory
etal, cortex. J Cogn Neurosci 2006;18:1712–22.
 Corbetta M, Kincade JM, Shulman GL. Neural systems for visual orient-
ing and their relationships to spatial working memory. J Cogn Neurosci
 Postle BR, Awh E, Jonides J, Smith EE, D’Esposito M. The where and how of
attention-based rehearsal in spatial working memory. Brain Res Cogn Brain
 Awh E, Jonides J, Smith EE, Buxton RB, Frank LR, Love T, et al. Rehearsal in
spatial working memory: evidence from neuroimaging. Psychol Sci 1999;10:
by external interference. Cereb Cortex 2010;20:859–72.
 Dolcos F, Miller B, Kragel P, Jha A, McCarthy G. Regional brain differences in the
effect of distraction during the delay interval of a working memory task. Brain
 Jha AP, Fabian SA, Aguirre GK. The role of prefrontal cortex in resolving distrac-
tor interference. Cogn Affect Behav Neurosci 2004;4:517–27.
 Sakai K, Rowe JB, Passingham RE. Active maintenance in prefrontal area 46
creates distractor-resistant memory. Nat Neurosci 2002;5:479–84.
 Toepper M, Gebhardt H, Beblo T, Thomas C, Driessen M, Bischoff M, et al.
encoding. Neuroscience 2010;165:1244–53.
ory. Neuroscience 2006;139:181–93.
 Nelson JK, Reuter-Lorenz PA, Persson J, Sylvester CY, Jonides J. Mapping inter-
ference resolution across task domains: a shared control process in left inferior
frontal gyrus. Brain Res 2009;1256:92–100.
 Feredoes E, Tononi G, Postle BR. Direct evidence for a prefrontal contribution
to the control of proactive interference in verbal working memory. Proc Natl
Acad Sci USA 2006;103:19530–4.
 Jimura K, Yamashita K, Chikazoe J, Hirose S, Miyashita Y, Konishi S. A critical
component that activates the left inferior prefrontal cortex during interference
resolution. Eur J Neurosci 2009;29:1915–20.
 Postle BR, Brush LN, Nick AM. Prefrontal cortex and the mediation of proactive
interference in working memory. Cogn Affect Behav Neurosci 2004;4:600–8.
 Badre D, Wagner AW. Frontal lobe mechanisms that resolve proactive interfer-
ence. Cereb Cortex 2005;15:2003–12.
subprocesses of decision-based action in the human frontal lobes. Neuroimage
 Nee DE, Jonides J. Neural correlates of access to short-term memory. Proc Natl
Acad Sci USA 2008;105:14228–33.
 Norman DA, Shallice T. Attention to action: willed and automatic control of
behaviour. In: Davidson RJ, Schwarts GE, Shapiro R, editors. Consciousness
and selfregulation. Advances in research and theory, vol. 4. New York: Plenum
Press; 1986. p. 1–18.
 Miyake A, Friedman NP, Rettinger DA, Shah P, Hegarty M. How are visuospatial
working memory, executive functioning, and spatial abilities related? A latent-
variable analysis. J Exp Psychol Gen 2001;130:621–40.
 Badre D, Hoffman J, Cooney JW, D’Esposito M. Hierarchical cognitive con-
trol deficits following damage to the human frontal lobe. Nat Neurosci
 Koechlin E, Summerfield C. An information theoretical approach to prefrontal
executive function. Trends Cogn Sci 2007;11:229–35.
 Friedman NP, Miyake A. The relations among inhibition and interference con-
trol functions: a latent-variable analysis. J Exp Psychol Gen 2004;133:101–35.
 Rowe JB, Sakai K, Lund TE, Ramsoy T, Christensen MS, Baare WF, et al. Is
the prefrontal cortex necessary for establishing cognitive sets? J Neurosci
 Sakai K, Passingham RE. Prefrontal interactions reflect future task operations.
Nat Neurosci 2003;6:75–81.
 Derrfuss J, Brass M, Neumann J, von Cramon DY. Involvement of the inferior
frontal junction in cognitive control: meta-analyses of switching and Stroop
studies. Hum Brain Mapp 2005;25:22–34.
 Monsell S. Task switching. Trends Cogn Sci 2003;7:134–40.
 Montojo CA, Courtney SM. Differential neural activation for updating rule ver-
sus stimulus information in working memory. Neuron 2008;59:173–82.