Personal semantics: at the crossroads
of semantic and episodic memory
Louis Renoult1,2
, Patrick S.R. Davidson2
, Daniela J. Palombo1,3
, Morris Moscovitch1,3
,
and Brian Levine1,3,4
1
Rotman Research Institute, Baycrest, Toronto, Ontario M6A 2E1, Canada
2
School of Psychology, University of Ottawa, Ontario K1N 6N5, Canada
3
Department of Psychology, University of Toronto, Ontario M5S 3G3, Canada
4
Department of Medicine, Division of Neurology, University of Toronto, Ontario M5S 1A1, Canada
Declarative memory is usually described as consisting
of two systems: semantic and episodic memory.
Between these two poles, however, may lie a third
entity: personal semantics (PS). PS concerns knowledge
of one’s past. Although typically assumed to be
an aspect of semantic memory, it is essentially absent
from existing models of knowledge. Furthermore, like
episodic memory (EM), PS is idiosyncratically personal
(i.e., not culturally-shared). We show that, depending
on how it is operationalized, the neural correlates of
PS can look more similar to semantic memory, more
similar to EM, or dissimilar to both. We consider three
different perspectives to better integrate PS into existing
models of declarative memory and suggest experimental
strategies for disentangling PS from semantic
and episodic memory.
PS: an intermediate form of declarative memory?
Declarative memory is typically described as consisting of
two systems, EM and semantic memory. EM entails recollecting
unique events within their specific spatio-temporal
context, imbued with a sense of the self mentally traveling
through time to re-experience the original event [1]. By
contrast, semantic memory pertains to culturally-shared
general knowledge (including facts and vocabulary), detached
from its context of acquisition and devoid of any
subjective sense of mental time travel [1–3].
Although drawing a stark contrast between episodic and
semantic memory has proved useful in spurring cognitive
neuroscience research, these systems have largely been
investigated in isolation, leading to an under-appreciation
of their interactions and of potentially intermediate forms
of memory. One such form of memory is personal semantics
(PS). In early case studies, this type of memory was described
in patients who could retrieve few or no EMs, yet
showed knowledge of events from their personal past [4–7],
resembling a ‘skeleton autobiography’ [5] that seemed ‘to
be drawn entirely from a personal pool of generalized
knowledge’ [4].
The paradox of PS is that it is highly personal (like EM),
yet, at the same time, devoid of any subjective sense of
recollection and detached from its context of acquisition
(like semantic memory). Crucially, although PS is typically
assumed to be a form of semantic memory, formal studies
of PS have been rare and PS is not well integrated into
existing models of semantic memory and knowledge (e.g.,
[3,8,9]). In the present review, we describe four ways in
which PS has been operationalized in previous studies and
show that how one defines PS can influence how it compares
to EM and general semantic memory (GS). We then
consider three different perspectives to better integrate PS
into existing models of episodic and semantic memory, and
offer suggestions for future work.
When PS appears similar to semantic memory (and
dissimilar to EM): autobiographical facts
The most common conceptualization of PS is as a set of
autobiographical facts (e.g., ‘Nicholas is the name of my
youngest brother.’) in contrast to general knowledge of the
world (e.g., ‘Nicholas was a 4th
-century saint.’) and EMs of
unique events (‘I could see that my brother Nicholas’ hands
were shaking during his speech at my wedding’). Autobiographical
facts share with EM the notion of personal
content. However, in PS this personal content refers to a
common organizing theme, whereas in EM it refers to a
subjective first-person perspective [10]. In addition, autobiographical
facts, like GS facts but unlike EMs, are
detached from their original context of acquisition [11].
When PS is studied using autobiographical facts, it
appears similar to GS and dissimilar to EM (Table 1). In
priming studies with healthy people, both autobiographical
and general facts can be primed by related knowledge
and can be verified by participants more quickly than can
EMs [12,13]. In neuropsychological studies using the
Autobiographical Memory Interview (AMI; [14], Box 1),
the typical finding is of impaired EM with relatively preserved
PS and GS [7,15–18], especially in patients with
medial temporal lobe (including hippocampal) lesions.
A few studies have described the opposite pattern, that
is, impaired PS and GS despite relatively preserved EM
[19,20], most often following extensive lesions or degeneration
in lateral temporal regions.
Although the aforementioned neuropsychological studies
suggest that PS and GS are usually impaired or preserved
together, the few extant neuroimaging studies that
have compared PS, GS, and EM have reported similar, but
distinct neural correlates for PS and GS. In a series of
Review
Corresponding author: Levine, B. (blevine@research.baycrest.org).
550 1364-6613/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tics.2012.09.003 Trends in Cognitive Sciences, November 2012, Vol. 16, No. 11
studies using sentence verification, Maguire and colleagues
reported that, even though PS and GS-related activations
overlapped in lateral temporal and medial prefrontal
regions [Brodmann area (BA) 10], the PS condition showed
greater activation relative to GS in a left-lateralized network
including the medial prefrontal cortex, retrosplenial
cortex, temporal pole, and temporoparietal junction
[21,22]. A concordant structural MRI study in Alzheimer’s
patients reported a correlation between temporopolar
volumes (especially on the left side) and PS as measured
by the AMI [23]. Autobiographical facts are typically relatively
idiosyncratic (e.g., I know that my brother owns an
old blue station wagon), which is compatible with the
presumed role of the temporal pole in the retrieval of
information associated with unique objects and individuals
(e.g., [24,25]). The greater medial prefrontal activation for
PS than for GS in Maguire’s neuroimaging studies is
consistent with the putative role of this brain region in
self-referential processing [26]. Interestingly, compared to
EM, PS has been associated with lesser activity in these
three brain regions (i.e., medial prefrontal cortex, temporal
pole, and retrosplenial cortex), as well as in the hippocampus
[21,22,27]. Taken together, these studies suggest that
this form of PS has more in common with GS than with EM
and are compatible with the idea of PS being a sub-domain
of GS, with overlapping but partly distinct neural bases
(Box 2). On the other hand, the fact that some brain regions
(i.e., the medial prefrontal cortex, temporal pole and retrosplenial
cortex) are differentially responsive to the three
types of memory, with a decreasing pattern of activation
from EM to PS to GS, is compatible with the idea of PS
being an intermediate system on a continuum of abstraction
from EM to GS (Box 3).
When PS appears similar to EM (and dissimilar to GS):
repeated events and autobiographically significant
concepts
Repeated events
A typical EM concerns a single personal experience, whereas
semantic memory is based on the extraction of similarities
across many events. PS is sometimes operationalized
as intermediate between these two poles: repeated personal
experiences (e.g., bringing my brother to school every
day; Table 1). For such events, it is assumed that one
typically does not remember a single instance, but instead
has extracted the common characteristics from the series of
similar but separate episodes ([28]; but see Box 4). Repeated
events are generally remembered less vividly and associated
with lower ratings of personal significance and
emotion than are unique episodes [29–31]. On the other
hand, memory for repeated events appears to be similar to
memory for unique events in some ways: both usually
contain visual imagery [32,33] and contextual information
[34], often from a first-person perspective [35]. Accordingly,
it has been proposed that EM may mediate recollection not
only for unique events, but also for repeated or summarized
events [36,37].
The few extant neuropsychological studies of memory
for repeated events have observed parallel perturbations of
Table 1. The four operational definitions of PS and their main characteristics
Autobiographical facts Self-knowledge Repeated events AS conceptsa
Examples My brother’s
name is Nicholas
I am a stubborn person I brought my brother
to school every day
Knowledge that Barack
Obama is President +
Recollection of an
argument with my
brother as to whether
he should be re-elected
Typical neural
correlates
MPFC, retrosplenial
cortex, temporal pole,
posterior temporal cortex
MPFC, retrosplenial cortex,
precuneus, middle and inferior
temporal gyri, inferior
parietal lobe
MPFC, hippocampus,
para-hippocampal gyrus,
temporo-parietal junction,
fusiform gyrus, inferior
temporal cortex
MTL, temporo-parietal
junction, ventrolateral
prefrontal cortex, fusiform
gyrus
Neuropsychological
patterns
Similar to GS Different from GS and EM Similar to EM Similar to EM
a
Abbreviations: AS concepts, autobiographically significant concepts; GS, general semantic memory; EM, episodic memory; MPFC, medial prefrontal cortex; MTL, medial
temporal lobe.
Box 1. Two common methods for assessing PS in
autobiographical memory: the Autobiographical Memory
Interview and the Autobiographical Interview
The Autobiographical Memory Interview (AMI) [14] refers to PS as
‘factual knowledge about a person’s own past’, which corresponds
best to the category of autobiographical facts in our classification
scheme of operational definitions. In the AMI, PS is evaluated using
questions about decontextualized facts, such as names of schools
one attended, names of friends and colleagues, and addresses
where one lived. One of the AMI’s main goals is to assess PS and
EM in a similar manner across the same three time periods:
childhood, early adult life, and recent years. Although the AMI
enables the comparison of PS and EM, the standard version does
not contain a GS condition. For example, in the original study [14],
PS was implicitly considered as part of GS, although GS was not
formally assessed by means other than a famous personalities test.
Subsequent neuropsychological studies using the AMI have shown
that PS and GS (assessed by other means) are often preserved or
impaired together in patients (see main text), suggesting that their
neural bases are similar. However, the fact that GS is not examined
formally in the AMI makes it more difficult to compare PS, EM, and
GS with one another.
A different approach is taken in Levine et al.’s [72] Autobiographical
Interview (AI) in which participants’ descriptions of autobiographical
events are scored for episodic and semantic details
(among other types). No attempt is made to differentiate these
types of memory from one another at the time of the interview; this
is done only at the time of scoring. In the standard version, semantic
details are not separated into PS versus GS categories (but see [38]
for a modified version that evaluates PS as repeated events).
However, the AI method could easily be modified to probe memory
for PS details more specifically.
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
551
memory for repeated and unique episodes after medial
temporal lobe lesions [6,38]. Neuroimaging studies indicate
that repeated events differ from all other forms of PS
in that they are associated (as are unique events) with
hippocampal/medial temporal lobe activity [29,31,39].
Note that similar findings have been observed for unique
episodes when they are retrieved repeatedly [40,41]. However,
some differences between repeated and unique
events have been noted in functional neuroimaging studies:
the medial prefrontal cortex, superior temporal gyrus,
right parahippocampal gyrus, right temporoparietal junction,
and precuneus show greater activation for unique
relative to repeated events [29–31,39]. By contrast, these
same studies found greater activity in the left parahippocampal
gyrus, left temporoparietal junction, fusiform gyri
and right inferior temporal cortex for repeated than unique
events. Taken together, these studies suggest that, although
memory for repeated events can be differentiated
from both GS and EM, this form of PS has much more in
common with EM than with GS. Therefore, even though
the dearth of studies on PS has sometimes resulted in our
treating autobiographical facts and repeated events as
belonging to the same category of semantic autobiographical
memory [42], this may not be appropriate. Memory for
unique and repeated events can be thought of as relying on
different weightings of the same component processes
(Box 5). Although they both include spatial and temporal
features, as well as a subjective sense of re-experiencing
events, unique events typically include more elaborated
sensory imagery (perhaps owing to greater precuneus
activation) and self-reflection (perhaps due to greater medial
prefrontal cortex activation) than do repeated events.
Autobiographically significant concepts
A third way to study PS is through autobiographically
significant concepts, that is, semantic concepts that are
Box 2. PS as a sub-domain of semantic memory
Is PS simply an aspect of GS, with similar organizing principles? This
is a parsimonious assumption made by many investigators. On the
face of it, the case in which this argument seems easiest to make is
when PS is operationalized as autobiographical facts, but even here
there are some differences between PS and GS in terms of their
neural correlates. It may be that PS is a sub-domain of GS with
overlapping, but partly distinct neural correlates. This is compatible
with the picture of a complex, distributed semantic system in the
brain, with different categories of objects (e.g., animals, tools, faces)
represented preferentially by different brain regions [9,73,74]. PS may
be one such category. A PS category in the semantic system would
likely have played a useful role in evolution, a criterion that is
considered essential for category specificity [74]. In functional
neuroimaging studies, medial prefrontal, temporal polar, retrosplenial,
and precuneus regions seem to be critical for the representation
of personal aspects of semantic memory (Figure I). The medial
prefrontal cortex, heavily involved in self-referential processing [26],
is active in the most abstract forms of PS, namely autobiographical
facts and self-knowledge (with activity centered around BA 10 in both
cases). The temporal poles, known to support the representation of
unique entities (e.g., [24,25]), are one of the brain regions most
robustly associated with the representation of autobiographical facts.
Finally, the retrosplenial cortex and precuneus, considered to play an
important role in the interface of GS and EM [3], show greater activity
for self-knowledge (as well as for autobiographical facts, in the case of
retrosplenial cortex) than for GS. Importantly, these areas are coactive
with other brain regions associated with GS, such as posterior and
inferior temporal lobe and inferior parietal lobe, which is compatible
with a conception of PS as a sub-domain of semantic memory.
Information about personal relevance could thus be appended to
existing semantic information [12]. Alternatively, for other types of
PS, associated EMs could be added to semantic knowledge. This is
most clearly seen in autobiographically significant concepts, which
seem to be represented both in semantic and episodic memory
[44,45], and thus constitute an amalgamation between GS and EM
rather than a sub-system of GS.[(Box_2)TD$FIG]
Lateral view
Medial view
Ventral view
Anterior
Posterior
TRENDS in Cognitive Sciences
Figure I. Personal semantics (PS) as a sub-domain of semantic memory.
Schematic view of the brain with lateral, medial, and ventral orientations.
Areas in red show the approximate locations of the regions typically implicated
in general semantic (GS) processing of concrete objects ([2,9]; see also the
semantic convergence zones in [3]). In yellow are the approximate locations of
regions that show greater activity for the more abstract forms of PS
(autobiographical facts and self-knowledge) than for GS. In orange are the
approximate locations where GS and PS overlap.
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
552
associated with vivid EMs from our lives. Conceptually,
this form of PS is at the intersection of GS and EM
(Table 1). In contrast to other forms of PS, it does not relate
to knowledge of personal events or the self, but rather to
semantic information, which, by its tight association with
specific episodes, has high degrees of personal significance
and emotional salience [43].
In initial work on this phenomenon with healthy people,
Westmacott and Moscovitch [44] showed that famous
names that brought to mind personally-significant EMs
Box 3. The continuum perspective
A number of continua have been proposed to describe how
declarative memory, autobiographical memory, and knowledge are
organized. In each of these domains, the continuum ranges from the
specific to the general, in terms of the kind and/or quality of
information represented: in declarative memory models, the range
is from episodic to semantic memory [75–79]; in autobiographical
memory models it is from specific episodes to summarized/repeated
events to abstract knowledge (‘life time periods’; [34]; see also [80]); in
knowledge/semantic memory models [81–84], it is from more specific
and context-bound semantic representations to more abstract and
context-free ones. Perhaps PS should be thought of as an intermediate
step between the episodic and the semantic memory
systems (Figure I). In all of the aforementioned models, the most
abstract information is also the least personally relevant. According to
this perspective, the personal aspects of semantic memory may
generally be less abstracted than GS (e.g., my collie vs the general
category of collies) and have privileged interactions with specific
episodes. Indeed, Conway ([85]) found that only less abstracted
semantic categories could prime the retrieval of EMs. These results
are complemented by recent neuroimaging studies that show that
certain inherently autobiographical semantic categories (‘names of
friends’) are associated with hippocampal/medial temporal activations
[86,87], brain regions usually associated with episodic retrieval.
If one considers PS as an intermediate system in a continuum of
abstraction from GS to EM, one might expect to see quantitative rather
than qualitative variations in brain activity across these memory types.
Indeed, several neuroimaging studies have found a graded decreasing
pattern of activity from EM to autobiographical facts to GS in medial
prefrontal, temporal polar and retrosplenial cortex ([21,22,27], see also
[42]). The precuneus also shows a graded reduction of activity from EM
to self-knowledge to GS [63]. However, the interpretation of these
observations is complicated by the fact that very few studies have
included all three types of memory (i.e., PS, GS and EM). A further
problem with the continuum perspective is that there may still be a
qualitative gap between the less abstracted forms of personal knowledge
(i.e., repeated events) and EM, notably in term of time frame
(shorter for episodes), related awareness (noetic vs autonoetic awareness,
Box 4), and visual imagery (generic for PS vs detailed for EM).
[(Box_3)TD$FIG]
RecollecƟon of
watching last
week’ s dog show
Superordinate level
categories of knowledge:
Animal
Basic level categories of knowledge:
Dog
Subordinate level
categories of knowledge:
Collie
General event knowledge/scripts:
A typical dog show
Specific event knowledge:
The dog show I went to last week
Self-knowledge:
Dogs are my favorite animals
Autobiographical facts:
I have a collie named Rex
Repeated events:
I bring Rex to the dog show every year
Autobiographically significant concepts:
Knowledge of
dog shows
Moreexperience-near
GreaterprobabilityofeliciƟngrecollecƟon
TRENDS in Cognitive Sciences
Figure I. The continuum perspective. This figure represents a hypothetical continuum of content in general semantics (GS, in blue) and personal semantics (PS, in
orange). In both GS and PS, representations become less and less abstracted and more and more context-dependent, from the higher to the lower levels of
representation (see [81]). According to this perspective, the personal aspects of semantic memory may be less abstracted than GS (but still related in content, as
indicated by the dotted lines). Moreover, in GS itself, less abstracted representations are also more experience-near and thus more personal, while still being part of
semantic knowledge (i.e., knowledge of events such as ‘The dog show I went to last week.’ [34]). Autobiographically significant concepts are semantic concepts that are
associated with vivid episodic memories (EMs). In this example, the concept of ‘dog shows’ becomes autobiographically significant by virtue of its association with
episodic recollection of last week’s dog show (in purple). Episodic recollection can be associated with concepts at a less abstracted level of representation, although this
becomes less likely as abstraction increases.
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
553
were associated with superior performance on tests of
semantic and episodic memory, compared to less personally-significant
(but equally familiar) famous names. Westmacott
et al. [45] (see also [43]) found that this performance
advantage was absent in Alzheimer’s and amnesic
patients, two groups with severe EM impairment due to
medial temporal lobe damage. By contrast, the advantage
was preserved in semantic dementia patients, who have
marked deficits in the ability to remember semantic concepts,
stemming from lateral temporal lobe deterioration
(see also [46] for similar results using public events instead
of famous names). These results suggest that, although
famous names are thought to be represented in semantic
memory, the performance advantage seen for autobiographically
significant names may derive from the additional
influence of EM.
A number of neuroimaging studies have investigated
the impact of self-relevance on neural activity, comparing,
for example, personally known versus unknown (or
famous) faces, names, objects, or scenes (reviewed in
[42]). However, in these studies, personally relevant stimuli
typically differ from control stimuli not only by their
association with EMs, but also by a greater degree of
familiarity and associated knowledge. This confound was
taken into account in an fMRI study by Denkova, Botzung
and Manning [47], in which names and faces of famous
Box 4. Participants in psychology experiments do not necessarily access the representations we assume they do
It is always difficult to be certain that one is making the correct
inferences about underlying cognitive processes based on behavior.
For instance, single episodes can later be retrieved as autobiographical
facts: I may know what occurred during my brother’s speech at
my wedding, without having a perceptually rich, first-person reexperiencing
of it. In other words, this single event could be retrieved
with noetic awareness, that is, the type of awareness that is
associated with knowing about the world and retrieving semantic
knowledge, rather than with autonoetic awareness, that is, the type
of awareness associated with subjectively re-experiencing events
from EM [1]. As Tulving cautioned, there is always the ‘possibility of
[our] answering questions directed to episodic memory on the basis
of our general knowledge of the world’ ([91]; see also[32]). According
to this view, even if the probability of autonoetically re-experiencing
events in their original episodic context is greater for unique
instances, both single and repeated events can be associated with
either autonoetic or noetic remembering. To overcome this difficulty,
the remember/know procedure [92], in which participants
report on the type of awareness that they have experienced during
retrieval, can be used, as can other methods (e.g., [93]). Alternatively,
it would be interesting to train participants to retrieve memories
voluntarily in either the noetic or the autonoetic mode. A prior
selection of vivid memories of personal experiences would be
necessary. The recall phase would then entail focusing either on the
meaning and knowledge associated with a personal experience
(third-person remembering) or on attempting to re-experience it in
its original context (first-person recollection). In addition to allowing
for a better control of retrieval mode, it would also allow for
comparisons between very similar memories (e.g, unique events
retrieved in the noetic vs autonoetic mode) matched in content and
age between the two conditions (for a similar approach with the cueword
technique see [94]). This enterprise, however, may face other
challenges. Among them is the possibility that EM processes and
information may be activated initially relatively automatically
[44,86,87], with the participant’s awareness of this retrieved information
emerging only at a later stage of processing.
Box 5. The component processes view
The effective encoding and retrieval of declarative memories depends
on multiple mnemonic and other processes, including attention, selfreflection,
emotion, spatial and temporal processing, sensory-perceptual
imagery, and executive functions [78,88,89]. One way to
conceptualize how PS, GS, and EM fit together is to assume that
these three types of memory simply rely on different weightings of
the aforementioned component processes [90].
According to this view, EM might involve a greater degree of selfreflection,
self-projection in time and space, emotion, and sensoryperceptual
imagery than PS and GS (Table I). In turn, PS might differ
from GS in terms of the former’s greater involvement of selfreflection,
and possibly a greater degree of emotional involvement.
The precise differences between PS, EM, and GS will depend on how
PS is operationalized. For example, some forms of PS, such as
memory for repeated events, appear to contain spatial and temporal
features, and may share with EM a chronological re-experiencing of
the event from a first-person perspective. However, they might still
differ from EM in terms of sensory imagery and self-reflection.
Neuroimaging studies have indeed found greater activity for unique
than repeated events in the precuneus and medial prefrontal cortex
[30,31,39], involved respectively in visuo-spatial processing and selfreflection.
These forms of PS (e.g., bringing my brother to school
every day) would be interesting to compare to elaborated scripts (e.g.,
going to school), which are considered part of GS [81]. It is likely that
they would differ mainly in terms of a greater reliance on personal
reflection for PS than for GS and possibly also in terms of more
elaborated sensory imagery (instead of generic images). Finally, selfknowledge
can be considered to involve a more abstract form of selfreflection
than EM, with less sensory imagery and less rich subjective
projection in time and space. Self-knowledge is the only form of PS
which shows similar activation in medial prefrontal cortex to EM. In
contrast, EM is associated with greater activity than self-knowledge in
brain areas related to visuospatial processing, including the precuneus
and superior and inferior parietal lobules.
Table I. The component processes view. The component processes view conceptualizes the different types of PS, GS, and EM as
representing different weighting of various cognitive processes. The ‘+’ sign represents the weighting of the relevant cognitive
processes (more pluses correspond to greater weight).
GS Autobiographical facts Self-knowledge Repeated events AS concepts EM
Self-reflection _ + + ++ ++ +++
Sensory-perceptual imagery Poor Poor Poor Generic Detailed Detailed
Spatial/temporal features + + + ++ ++ +++
Emotional valence _ + ++ + ++ +++
Possibility of eliciting recollection Poor Poor
(may vary with recency)
Poor Good Good Systematic
Typical perspective 3rd
Person 3rd
Person 3rd
Person 1st
Person 1st
/3rd
Person 1st
Person
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
554
people high or low in autobiographical significance – but
equal in familiarity – were compared. The results of this
study converged with the neuropsychological data to suggest
that autobiographically significant concepts do indeed
depend on the medial temporal lobe (see also [48]),
although they also identified other brain regions, such
as the left ventrolateral prefrontal cortex, the temporoparietal
junction, the fusiform gyrus, and the lingual gyrus.
These results were interpreted as indicating that
autobiographically relevant concepts may involve more
effortful retrieval (left prefrontal activation), the activation
of richer autobiographical information (temporo-parietal
junction), and the retrieval of sensory perceptual details
(lingual gyrus) [47]. A recent event-related potential (ERP)
study (Renoult et al., unpublished data) provides further
evidence of the episodic nature of autobiographically significant
concepts: compared to famous names for which
participants had only general knowledge, autobiographically
significant names were associated with increased
amplitude of the late positive component (LPC), an ERP
index of episodic recollection [49]. In contrast, the N400
component, an ERP index of semantic processing [50], did
not differentiate the two kinds of names, but instead was
related to the amount of semantic knowledge participants
had of each famous person. These results suggest that
autobiographically significant concepts differ from other
concepts by being represented in EM, in addition to semantic
memory. Crucially, as in the original paradigm [44,45],
autobiographical significance was incidental to task performance
as it was determined by a separate sample of participants
or by the subjects themselves but at the end of the
experiment. This suggests that the activation of relevant
episodic information by these concepts is automatic and,
more generally, that this information is an intrinsic part of
autobiographically significant concepts.
Although not as well-studied as the other forms of PS,
autobiographically significant concepts constitute a promising
way to study interactions between GS and EM. The
data available so far suggest that this form of PS, like
memory for repeated events, has neural correlates that are
similar to those of EM.
When PS appears dissimilar to both GS and
EM: self-knowledge
A fourth tradition of work on PS focuses on the self, selfimage,
and personal identity. For example, Neisser [51]
proposed that the conceptual self is the most basic form of
self-knowledge, allowing us to reflect on our personal
characteristics and attributes (e.g., ‘I am a fast reader,
but poor at remembering names’). Similarly, Brewer [32]
refers to the self-schema as a ‘cognitive structure that
contains generic knowledge about the self’. Self-knowledge
has most often been operationalized as the summary of
one’s personality traits (e.g., ‘I am a stubborn person’
[52,53]). Because they focus on self-image, these aspects
of PS may be strongly influenced by emotional and social
factors both during encoding and retrieval [54], and could
play a crucial role in maintaining a concept of self that
would fit with the rememberer’s goals and emotions [55]. In
addition, self-knowledge may be differentiated from other
forms of autobiographical knowledge by slower updating
mechanisms [56]. These psychological characteristics, together
with the neuropsychological data (see below), argue
against the idea of self-knowledge being a subset of the
category of autobiographical facts (Table 1).
Behavioral studies by Klein and collaborators have
dissociated self-knowledge from EM and GS. In these
studies, self-knowledge was operationalized as knowledge
of one’s personality traits. Using a priming design, Klein
and Loftus [57] showed that judging the self-relevance of a
trait (PS condition) did not influence the time required for
participants to recollect an event in which they displayed
that personality trait (EM condition). Similarly, judging
whether a trait was self-relevant did not influence the time
it took to define that trait (GS condition). These behavioral
results, therefore, suggest a relative independence of trait
self-knowledge from EM or GS related to these traits ([52]).
In neuropsychological work, Klein and colleagues
showed that self-knowledge can be preserved when EM
and GS are impaired ([58–60], see also [36]) and preserved
even when knowledge of autobiographical facts is impaired
[60,61]. This form of PS was considered to be intact when
reasonable correlations were found between patients’ selfreports
about their personality traits and reports from
family members. Klein and colleagues interpreted their
results as indicating that self-knowledge may be a specialized
subsystem within semantic memory.
Neuroimaging studies have found that self-knowledge,
as operationalized by the self-evaluation of one’s personality
traits or by self-evaluative judgments (‘I enjoy going to
New Year’s parties.’; [62–64]), is associated with neural
correlates that are somewhat distinct from those of EM and
GS. The medial prefrontal cortex (BA 10, extending into
BA 9), posterior cingulate cortex (including the retrosplenial
cortex), and precuneus have consistently shown greater
activity for self-knowledge than for GS (operationalized
as judging the valence of the traits [65–69] or as verifying
statements of factual knowledge [62–64,70]). The same
regions of the medial prefrontal cortex and posterior cingulate
cortex were also found to be activated for EM
(operationalized as verifying sentences about unique
events [63] or as recognizing previously studied pictures
[71]). In contrast, these same studies reported that EM was
associated with greater activity than PS in brain regions
related to visuo-spatial processing, such as the precuneus
and the superior and inferior parietal lobules. Finally, selfknowledge
was also found to activate brain regions often
associated with GS [3], such as the left middle and inferior
temporal gyrus and left inferior parietal lobe [64–66,71],
and the left temporal pole [62]. Together, the neuropsychological
and neuroimaging evidence suggests that selfknowledge
is the form of PS that is the most easily dissociable
from GS and EM, although further evidence is
required. For example, the case for PS being truly independent
of EM and GS would be stronger if the opposite
neuropsychological dissociation from Klein and colleagues’
[52] had ever been reported (that is, a patient with impaired
self-knowledge, but intact EM and GS). In addition,
the evidence of overlapping neuroimaging patterns for selfknowledge
and GS is compatible with Klein et al.’s hypothesis
that the self-knowledge aspect of PS is supported by
semantic memory [52] (Box 2).
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
555
Concluding remarks
Even though the empirical assessment of PS first began 30
years ago [4], this field is still in its infancy. Up to the
present, the study of PS has largely been restricted to the
autobiographical memory literature, as PS is typically not
included in models of semantic memory or knowledge. The
dearth of research on PS may have been due to the assumption
made by many researchers that PS is highly
similar to, or even synonymous with, GS. Here, we have
shown that this is an oversimplification. Whereas
some forms of PS (e.g., autobiographical facts) appear to
have neural correlates that are very similar to GS, others
(e.g., memory for repeated events and autobiographically
significant concepts) have neural correlates that are much
more similar to those of EM. Finally, self-knowledge is a
form of PS that seems to be at least partly dissociable from
GS and EM.
Three broad conceptualizations of PS can be articulated:
the GS, continuum, and component processes views, outlined
in Boxes 2, 3 and 5. Each of these might best describe
one or two of the four different operational definitions of PS
we have outlined, but not others. For instance, knowledge
of autobiographical facts, and perhaps self-knowledge,
might best be viewed as subdomains/categories of semantic
memory (Box 2). The more episodic types of PS, such as
memory for repeated events, could perhaps most easily be
integrated into models of declarative memory by proposing
either a continuum of abstraction in knowledge representations
(Box 3) or a component processes approach (Box 5).
Although a continuum model would be supported by quantitative
variations in brain activity seen in a number of
regions when comparing EM, PS and GS [21,22,27,63], a
component process view may better account for the qualitative
gap that appears to exist between the less abstracted
form of PS (such as memory for repeated events) and EM.
At present, the data are insufficient to adjudicate among
these different models of PS. Future work should distinguish
among the four types of PS we have outlined here,
because they seem to have at least somewhat different
cognitive and neural bases from one another. Careful
comparisons of these four forms of PS with one another
and with EM and GS should help to shed light not only on
the nature of PS but also on declarative memory and
knowledge more broadly (Box 6).
Acknowledgments
Preparation of this manuscript was supported by the Heart and Stroke
Foundation of Canada Centre for Stroke Recovery, fellowship 23710 from
the ‘Fonds de la Recherche en Sante´ du Que´bec’ (FRSQ) to L.R., grants
A8347 (M.M.) and 341649 (P.S.R.D.) from the Natural Sciences and
Engineering Research Council of Canada (NSERC), and grant MOP-
62963 (B.L.) from the Canadian Institutes of Health Research (CIHR). We
thank Patrick Seymour for his assistance with preparing the figures.
References
1 Tulving, E. (2002) Episodic memory: from mind to brain. Annu. Rev.
Psychol. 53, 1–25
2 Martin, A. (2001) Functional neuroimaging of semantic memory. In
Handbook of Functional NeuroImaging of Cognition (Cabeza, R.
and Kingstone, A., eds), pp. 153–186, MIT Press
3 Binder, J.R. and Desai, R.H. (2011) The neurobiology of semantic
memory. Trends Cogn. Sci. 15, 527–536
4 Cermak, L.S. and O’Connor, M. (1983) The anterograde and
retrograde retrieval ability of a patient with amnesia due to
encephalitis. Neuropsychologia 21, 213–234
5 Warrington, E.K. and McCarthy, R.A. (1988) The fractionation of
retrograde amnesia. Brain Cogn. 7, 184–200
6 Tulving, E. et al. (1988) Priming of semantic autobiographical
knowledge: a case study of retrograde amnesia. Brain Cogn. 8, 3–20
7 Levine, B. et al. (1998) Episodic memory and the self in a case of
isolated retrograde amnesia. Brain 121, 1951–1973
8 Murphy, G.L. (2002) The Big Book of Concepts, MIT Press
9 Martin, A. (2007) The representation of object concepts in the brain.
Annu. Rev. Psychol. 58, 25–45
10 Tulving, E. (1989) Remembering and knowing the past. Am. Sci. 77,
361–367
11 Larsen, S.F. (1992) Personal context in autobiographical and
narrative memories. In Theoretical Perspectives on
Autobiographical Memory (Conway, M.A. et al., eds), pp. 53–71,
Kluwer Academic Publishers
12 Conway, M.A. (1987) Verifying autobiographical facts. Cognition 26,
39–58
Box 6. Current problems and questions for future research
 Is there a qualitative gap between different forms of PS? To study
PS, researchers have focussed separately on two factors: personal
implication (i.e., comparison between personal and non-personal
facts) and uniqueness (i.e., comparison of unique vs repeated
events). Although, to our knowledge, autobiographical facts and
repeated events have never been compared directly, there
appears to be a qualitative gap between them in terms of their
psychological and neural correlates, perhaps due to the inherent
difficulty of contrasting facts and events. To situate PS better, it
may be useful to study personal implication and uniqueness in
the same experiment to allow for a comparison between the
following categories: personal-unique, personal-repeated, nonpersonal-unique
and non-personal-repeated events. Non-personal
unique events could, for instance, be operationalized as the
retrieval of movies or news from the same time period as the
personal events. Non-personal repeated events could consist of
the retrieval of scripts or of general knowledge associated with
extended events.
 How do the more abstract forms of PS differ from GS? One
interpretation of how autobiographical facts and self-knowledge
differ from GS stems from the notion of expertise [57,95]. All
individuals are experts on themselves, in that these types of
knowledge are overlearned and accessed frequently. This could
explain the resilience of self-knowledge to neural damage [52].
Indeed, neuropsychological data show that pre-morbid expertise
in certain knowledge categories can lead to a selective preservation
of these categories in patients with semantic impairments
[96,97]. Accordingly, differences in brain activation between PS
and GS could be influenced by variations in category expertise. It
would thus be interesting, when comparing self-knowledge and
GS, to include GS categories in which participants are experts.
 Does PS only concern knowledge of facts and events from our
personal past? Recent work suggests that people also possess
general knowledge about facts and events that they anticipate to
happen in their personal future ([98,99], e.g., ‘We plan to record an
album with my friend Phillippe and I imagine that I will spend a lot
of time with him, composing new songs and rehearsing them in
our room.’). It will thus be of interest to determine whether these
future and past-oriented forms of PS have similar neural
correlates.
 Is my table the same as your table? Semantic memory is generally
considered to be culturally shared [1,2]. Even though clear
individual differences in how people categorize the world have
been described (e.g., [100–102]), the general topic of individual
differences in semantic memory has largely been neglected. In
neuroimaging studies, anatomical variations in the organization
of knowledge have been reported and may explain inconsistent
findings in studies of category specificity [103,104]. Thus, bridging
the gap between the personal and non-personal forms of
semantics may also require investigating inter-individual variation
in processing general concepts.
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
556
13 Conway, M.A. and Bekerian, D.A. (1987) Organization in
autobiographical memory. Mem. Cogn. 15, 119–132
14 Kopelman, M.D. et al. (1989) The autobiographical memory interview:
a new assessment of autobiographical and personal semantic memory
in amnesic patients. J. Clin. Exp. Neuropsychol. 11, 724–744
15 Oxbury, S. et al. (1997) Severe amnesia: an usual late complication
after temporal lobectomy. Neuropsychologia 35, 975–988
16 Viskontas, I.V. et al. (2000) Remote episodic memory deficits in
patients with unilateral temporal lobe epilepsy and excisions. J.
Neurosci. 20, 5853–5857
17 Hirano, M. et al. (2002) I cannot remember, but I know my past events:
remembering and knowing in a patient with amnesic syndrome. J.
Clin. Exp. Neuropsychol. 24, 548–555
18 McCarthy, R.A. et al. (2005) Remembering and forgetting of semantic
knowledge in amnesia: a 16-year follow-up investigation of RFR.
Neuropsychologia 43, 356–372
19 Hodges, J.R. et al. (1992) Semantic dementia. Progressive fluent
aphasia with temporal lobe atrophy. Brain 115 (Pt 6), 1783–1806
20 Eslinger, P.J. (1998) Autobiographical memory after temporal lobe
lesions. Neurocase 4, 481–495
21 Maguire, E.A. and Frith, C.D. (2003) Aging affects the engagement of
the hippocampus during autobiographical memory retrieval. Brain
126, 1511–1523
22 Maguire, E.A. and Mummery, C.J. (1999) Differential modulation of a
common memory retrieval network revealed by positron emission
tomography. Hippocampus 9, 54–61
23 Gilboa, A. et al. (2005) Retrieval of autobiographical memory in
Alzheimer’s disease: relation to volumes of medial temporal lobe
and other structures. Hippocampus 15, 535–550
24 Tranel, D. (2009) The left temporal pole is important for retrieving
words for unique concrete entities. Aphasiology 23, 867
25 Ross, L.A. and Olson, I.R. (2011) What’s unique about unique entities?
An fMRI investigation of the semantics of famous faces and
landmarks. Cereb. Cortex 22, 2005–2015
26 D’Argembeau, A. and Salmon, E. (2012) The neural basis of semantic
and episodic forms of self-knowledge: insights from functional
neuroimaging. Adv. Exp. Med. Biol. 739, 276–290
27 Maguire, E.A. et al. (2001) The effects of bilateral hippocampal
damage on fMRI regional activations and interactions during
memory retrieval. Brain 124, 1156–1170
28 Neisser, U. (1981) John Dean’s memory: a case study. Cognition 9, 1–22
29 Addis, D.R. et al. (2004) Recollective qualities modulate hippocampal
activation during autobiographical memory retrieval. Hippocampus
14, 752–762
30 Levine, B. et al. (2004) The functional neuroanatomy of episodic and
semantic autobiographical remembering: a prospective functional
MRI study. J. Cogn. Neurosci. 16, 1633–1646
31 Holland, A.C. et al. (2011) The neural correlates of specific versus
general autobiographical memory construction and elaboration.
Neuropsychologia 49, 3164–3177
32 Brewer, W.F. (1986) What is autobiographical memory? In
Autobiographical Memory (Rubin, D.C., ed.), pp. 25–49, Cambridge
University Press
33 Rubin, D.C. et al. (2003) Belief and recollection of autobiographical
memories. Mem. Cogn. 31, 887–901
34 Conway, M.A. and Pleydell-Pearce, C.W. (2000) The construction of
autobiographical memories in the self-memory system. Psychol. Rev.
107, 261–288
35 Robinson, J.A. and Swanson, K.L. (1993) Field and observer modes of
remembering. Memory 1, 169–184
36 Tulving, E. (1993) Self-knowledge of an amnesic individual is
represented abstractly. In The Mental Representation of Trait and
Autobiographical Knowledge About the Self (Srull, T.K. and Wyer,
R.S., eds), pp. 147–156, Erlbaum
37 Brewer, W.F. (1996) What is recollective memory? In Remembering
our Past: Studies in Autobiographical Memory (Rubin, D.C., ed.), pp.
19–66, Cambridge University Press
38 St-Laurent, M. et al. (2009) Determinants of autobiographical
memory in patients with unilateral temporal lobe epilepsy or
excisions. Neuropsychologia 47, 2211–2221
39 Addis, D.R. et al. (2004) Characterizing spatial and temporal features
of autobiographical memory retrieval networks: a partial least
squares approach. Neuroimage 23, 1460–1471
40 Nadel, L. et al. (2007) Autobiographical memory retrieval and
hippocampal activation as a function of repetition and the passage
of time. Neural Plast. 2007, 90472
41 Svoboda, E. and Levine, B. (2009) The effects of rehearsal on the
functional neuroanatomy of episodic autobiographical and semantic
remembering: a functional magnetic resonance imaging study.
J. Neurosci. 29, 3073–3082
42 Martinelli, P. et al. (2012) Neural substrates of the self-memory
system: new insights from a meta-analysis. Hum. Brain Mapp.
http://dx.doi.org/10.1002/hbm.22008
43 Westmacott, R. et al. (2001) Different patterns of autobiographical
memory loss in semantic dementia and medial temporal lobe
amnesia: a challenge to consolidation theory. Neurocase 7, 37–55
44 Westmacott, R. and Moscovitch, M. (2003) The contribution of
autobiographical significance to semantic memory. Mem. Cogn. 31,
761–774
45 Westmacott, R. et al. (2004) The contribution of autobiographical
significance to semantic memory: evidence from Alzheimer’s
disease, semantic dementia, and amnesia. Neuropsychologia 42,
25–48
46 Petrican, R. et al. (2010) Recollection and familiarity for public events
in neurologically intact older adults and two brain-damaged patients.
Neuropsychologia 48, 945–960
47 Denkova, E. et al. (2006) Neural correlates of remembering/knowing
famous people: an event-related fMRI study. Neuropsychologia 44,
2783–2791
48 Viskontas, I.V. et al. (2009) Human medial temporal lobe neurons
respond preferentially to personally relevant images. Proc. Natl.
Acad. Sci. U.S.A. 106, 21329–21334
49 Rugg, M.D. and Curran, T. (2007) Event-related potentials and
recognition memory. Trends Cogn. Sci. 11, 251–257
50 Kutas, M. et al. (2006) Psycholinguistics electrified II (1994-2005), In
Handbook of Psycholinguistics (2nd edn) (Matthew, J.T. and Morton,
A.G., eds), pp. 659–724, Academic Press
51 Neisser, U. (1988) Five kinds of self-knowledge. Philos. Psychol. 1,
35–59
52 Klein, S.B. and Lax, M.L. (2010) The unanticipated resilience of trait
self-knowledge in the face of neural damage. Memory 18, 918–948
53 Craik, F.I.M. et al. (1999) In search of the self: a positron emission
tomography study. Psychol. Sci. 10, 26–34
54 Northoff, G. et al. (2006) Self-referential processing in our brain – A
meta-analysis of imaging studies on the self. Neuroimage 31, 440–457
55 Baddeley, A. (1992) What is autobiographical memory? In
Theoretical Perspectives on Autobiographical Memory (Conway,
M.A. et al., eds), pp. 13–29, Kluwer Academic Publishers
56 Wagenaar, W.A. (1992) Remembering my worst sins: How
autobiographical memory serves the updating of the conceptual self.
In Theoretical Perspectives on Autobiographical Memory (Conway,M.A.
et al., eds), pp. 263–274, Kluwer Academic Publishers
57 Klein, S.B. and Loftus, J. (1993) The mental representation of trait
and autobiographical knowledge about the self. In Advances in Social
Cognition (Vol. 5) Srull, T.K. and Wyer, R.S.,eds In pp. 1–49,
Lawrence Erlbaum Associates
58 Klein, S.B. et al. (1999) Independence of episodic and semantic selfknowledge:
the case from autism. Soc. Cogn. 17, 413–436
59 Klein, S.B. et al. (2002) Is there something special about the self? A
neuropsychological case study. J. Res. Pers. 36, 490–506
60 Klein, S.B. et al. (2003) Preserved knowledge of self in a case of
Alzheimer’s dementia. Soc. Cogn. 21, 157–165
61 Klein, S.B. et al. (2002) Memory and temporal experience: the effects
of episodic memory loss on an amnesic patient’s ability to remember
the past and imagine the future. Soc. Cogn. 20, 353–379
62 Schroeter, M.L. et al. (2010) Traumatic brain injury affects the
frontomedian cortex – an event-related fMRI study on evaluative
judgments. Neuropsychologia 48, 185–193
63 Zysset, S. et al. (2002) The anterior frontomedian cortex and
evaluative judgment: an fMRI study. Neuroimage 15, 983–991
64 Zysset, S. et al. (2003) Functional specialization within the anterior
medial prefrontal cortex: a functional magnetic resonance imaging
study with human subjects. Neurosci. Lett. 335, 183–186
65 D’Argembeau, A. et al. (2010) The neural basis of personal goal
processing when envisioning future events. J. Cogn. Neurosci. 22,
1701–1713
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
557
66 D’Argembeau, A. et al. (2010) Modulation of medial prefrontal and
inferior parietal cortices when thinking about past, present, and
future selves. Soc. Neurosci. 5, 187–200
67 Schmitz, T.W. et al. (2004) Metacognitive evaluation, self-relevance,
and the right prefrontal cortex. Neuroimage 22, 941–947
68 Schmitz, T.W. et al. (2006) Neural correlates of self-evaluative
accuracy after traumatic brain injury. Neuropsychologia 44, 762–773
69 Whitfield-Gabrieli, S. et al. (2011) Associations and dissociations
between default and self-reference networks in the human brain.
Neuroimage 55, 225–232
70 Johnson, S.C. et al. (2002) Neural correlates of self-reflection. Brain
125, 1808–1814
71 Sajonz, B. et al. (2010) Delineating self-referential processing from
episodic memory retrieval: common and dissociable networks.
Neuroimage 50, 1606–1617
72 Levine, B. et al. (2002) Aging and autobiographical memory:
dissociating episodic from semantic retrieval. Psychol. Aging 17,
677–689
73 Caramazza, A. and Mahon, B.Z. (2003) The organization of conceptual
knowledge: the evidence from category-specific semantic deficits.
Trends Cogn. Sci. 7, 354–361
74 Mahon, B.Z. and Caramazza, A. (2011) What drives the organization
of object knowledge in the brain? Trends Cogn. Sci. 15, 97–103
75 Kintsch, W. (1980) Semantic memory: a tutorial. In Attention and
Performance VIII (Nickerson, R., ed.), pp. 595–620, Lawrence
Erlbaum Associates
76 Jacoby, L.L. and Craik, F.I.M. (1979) Effects of elaboration of
processing at encoding and retrieval: trace distinctiveness and
recovery of initial context. In Levels of Processing in Human
Memory (Cermak, L.S. and Craik, F.I.M., eds), pp. 1–21, Lawrence
Erlbaum Associates
77 Kopelman, M.D. and Kapur, N. (2001) The loss of episodic memories
in retrograde amnesia: single-case and group studies. Philos. Trans.
R. Soc. Lond. B: Biol. Sci. 356, 1409–1421
78 Cabeza, R. and St Jacques, P. (2007) Functional neuroimaging of
autobiographical memory. Trends Cogn. Sci. 11, 219–227
79 Greenberg, D.L. and Verfaellie, M. (2010) Interdependence of episodic
and semantic memory: evidence from neuropsychology. J. Int.
Neuropsychol. Soc. 16, 748–753
80 Barsalou, L.W. (1988) The content and organization of
autobiographical memories. In Remembering Reconsidered:
Ecological and Traditional Approaches to the Study of Memory
(Neisser, U. and Winograd, E., eds), pp. 193–243, Cambridge
University Press
81 Funnell, E. (2001) Evidence for scripts in semantic dementia:
Implications for theories of semantic memory. Cogn. Neuropsychol.
18, 323–341
82 Rosch, E. and Mervis, C.B. (1975) Family resemblances – studies in
internal structure of categories. Cogn. Psychol. 7, 573–605
83 Medin, D.L. and Schaffer, M.M. (1978) Context theory of classification
learning. Psychol. Rev. 85, 207–238
84 Nelson, K. (1983) The derivation of concepts and categories from event
representations. In New Trends in Conceptual Representation:
Challenges to Piaget’s Theory? (Scholnick, E.K., ed.), pp. 129–149,
Erlbaum
85 Conway, M.A. (1990) Associations between autobiographical memories
and concepts. J. Exp. Psychol. Learn. Mem. Cogn. 16, 799–812
86 Ryan, L. et al. (2008) Hippocampal activation during episodic and
semantic memory retrieval: comparing category production and
category cued recall. Neuropsychologia 46, 2109–2121
87 Sheldon, S. and Moscovitch, M. (2012) The nature and time-course of
medial temporal lobe contributions to semantic retrieval: an fMRI
study on verbal fluency. Hippocampus 22, 1451–1466
88 Svoboda, E. et al. (2006) The functional neuroanatomy of
autobiographical memory: a meta-analysis. Neuropsychologia 44,
2189–2208
89 Rubin, D.C. (2006) The basic-systems model of episodic memory.
Perspect. Psychol. Sci. 1, 277–311
90 Moscovitch, M. (1992) Memory and working-with-memory - a
component process model based on modules and central systems.
J. Cogn. Neurosci. 4, 257–267
91 Tulving, E. (1984) Pre´cis of Elements of Episodic Memory. Behav.
Brain Sci. 7, 223–238
92 Tulving, E. (1985) Memory and Consciousness. Can. Psychol. 26, 1–12
93 Gilboa, A. et al. (2006) Hippocampal contributions to recollection in
retrograde and anterograde amnesia. Hippocampus 16, 966–980
94 Dalgleish, T. et al. (2007) Reduced specificity of autobiographical
memory and depression: the role of executive control. J. Exp.
Psychol. Gen. 136, 23–42
95 Maki, R.H. (1993) Knowledge of the Self: Is it special? In Advances in
Social Cognition (Vol. 5) Srull, T.K. and Wyer, R.S.,eds In pp. 101–
110, Lawrence Erlbaum Associates
96 Jefferies, E. et al. (2011) Premorbid expertise produces categoryspecific
impairment in a domain-general semantic disorder.
Neuropsychologia 49, 3213–3223
97 Funnell, E. and Davies, P.D. (1996) JBR: a reassessment of concept
familiarity and a category-specific disorder for living things.
Neurocase 2, 461–474
98 D’Argembeau, A. and Mathy, A. (2011) Tracking the construction of
episodic future thoughts. J. Exp. Psychol. Gen. 140, 258–271
99 D’Argembeau, A. and Demblon, J. (2012) On the representational
systems underlying prospection: evidence from the event-cueing
paradigm. Cognition 125, 160–167
100 McCloskey, M.E. and Glucksberg, S. (1978) Natural categories: well
defined or fuzzy sets? Mem. Cogn. 6, 462–472
101 Hampton, J.A. (1979) Polymorphous concepts in semantic memory.
J. Verb. Learn. Verb. Behav. 18, 441–461
102 Johnson, K.E. and Mervis, C.B. (1997) Effects of varying levels of
expertise on the basic level of categorization. J. Exp. Psychol. Gen.
126, 248–277
103 Spitzer, M. et al. (1998) Functional magnetic resonance imaging of
category-specific cortical activation: evidence for semantic maps.
Cogn. Brain Res. 6, 309–319
104 Chao, L.L. et al. (1999) Attribute-based neural substrates in temporal
cortex for perceiving and knowing about objects. Nat. Neurosci. 2,
913–919
Review Trends in Cognitive Sciences November 2012, Vol. 16, No. 11
558