Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
Contents lists available at SciVerse ScienceDirect
Neuroscience and Biobehavioral Reviews
journal homepage: www.elsevier.com/locate/neubiorev
Review
Distilling the neural correlates of consciousness
Jaan Arua,b,∗
, Talis Bachmannc
, Wolf Singera,b,d
, Lucia Mellonia,d,∗
a
Max-Planck Institute for Brain Research, Deutschordnerstr. 46, 60528 Frankfurt am Main, Germany
b
Frankfurt Institute for Advanced Studies, Max-von-Laue-Str. 1, 60438 Frankfurt am Main, Germany
c
University of Tartu, Kaarli puiestee 3, 10119 Tallinn, Estonia
d
Ernst Strüngmann Institute in Cooperation with Max Planck Society, Deutschordnerstr. 46, 60528 Frankfurt am Main, Germany
a r t i c l e i n f o
Article history:
Received 21 June 2011
Received in revised form 7 December 2011
Accepted 8 December 2011
Keywords:
Neural correlates of consciousness
Conscious perception
Contrastive analysis
a b s t r a c t
Solving the problem of consciousness remains one of the biggest challenges in modern science. One key
step towards understanding consciousness is to empirically narrow down neural processes associated
with the subjective experience of a particular content. To unravel these neural correlates of consciousness
(NCC) a common scientiﬁc strategy is to compare perceptual conditions in which consciousness of
a particular content is present with those in which it is absent, and to determine differences in measures
of brain activity (the so called “contrastive analysis”). However, this comparison appears not to reveal
exclusively the NCC, as the NCC proper can be confounded with prerequisites for and consequences of
conscious processing of the particular content. This implies that previous results cannot be unequivocally
interpreted as reﬂecting the neural correlates of conscious experience. Here we review evidence
supporting this conjecture and suggest experimental strategies to untangle the NCC from the prerequisites
and consequences of conscious experience in order to further develop the otherwise valid and
valuable contrastive methodology.
© 2011 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
1.1. Contrastive analysis as the key methodological strategy in consciousness research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
2. The 3 NCCs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
2.1. NCC for processes directly corresponding to conscious experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
2.2. Prerequisite processes carrying no actual conscious contents (NCC-pr) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
2.3. Consequences following actual conscious experience (NCC-co). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 739
2.4. The 3 NCCs: conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 739
3. Distilling consciousness: disentangling the NCCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 740
3.1. Disentangling NCC-pr from NCC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 741
3.2. Disentangling NCC-co from NCC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742
4. The 3 NCCs and previous results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742
4.1. The 3 NCCs and the temporal course of conscious perception. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 742
4.2. The 3 NCCs and the activity of different brain areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 743
5. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 744
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745
∗ Corresponding authors at: Max Planck Institute for Brain Research, Department of Neurophysiology, Deutschordenstrasse 46, 60528 Frankfurt am Main, Germany.
Tel.: +49 69 96769 268; fax: +49 69 96769 327.
E-mailaddresses:jaan.aru@gmail.com (J. Aru), lucia.melloni@brain.mpg.de(L. Melloni).
0149-7634/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neubiorev.2011.12.003
738 J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
1. Introduction
1.1. Contrastive analysis as the key methodological strategy in
consciousness research
The problem of consciousness could be approached scientiﬁcally
only after concrete research strategies were suggested and research
problems were deﬁned (Baars, 1989; Crick and Koch, 1990). Since
then, an important research goal has been to ﬁnd the minimal set
of neural processes that are together sufﬁcient for the conscious
experience of a particular content – the neural correlates of consciousness
(NCC). This purpose is complementary to the research
tradition about general states of awareness (e.g., awake, NREM
sleep, vegetative state) where consciousness and lack of consciousness
are contrasted (Koch, 2004; Laureys and Tononi, 2010). In
this paper we deal with the former aspect where NCC are studied
in relation to the content of perception as opposed to treating
consciousness as a state variable.
When investigating the neural correlates that underlie the content
of our consciousness a common and widely accepted strategy
has been to hold the stimulus conditions similar while conscious
perception varies as the dependent measure – on some trials
the subject perceives the target consciously, on other trials not.
This experimental approach, known as the contrastive analysis
(Baars, 1989), is implemented in various experimental paradigms
(Bachmann et al., 2011; Kim and Blake, 2005) and combined with
measures of neural activity has brought about important insights
into the NCC (Dehaene and Changeux, 2011; Koivisto and Revonsuo,
2010; Rees, 2007; Tononi and Koch, 2008). The asserted advantage
of the contrastive analysis is the following: by comparing conscious
and non-conscious perception, the NCC can be unraveled without
confounding it with unconscious processes involved in target perception
that take place in both conditions. However, despite the
apparently straightforward logic of this approach the results are
inconclusive and contradictory. For instance, some studies report
differences between conscious and non-conscious conditions starting
already at stimulus onset or around 100 ms post-stimulus (Aru
and Bachmann, 2009a,b; Melloni et al., 2007; Pins and Ffytche,
2003), others have only found differences in later phases (>300 ms)
(Del Cul et al., 2007; Gaillard et al., 2009). These contradictory
results can be understood when considering a severe methodological
pitfall: the result of the contrast between trials with and
without conscious perception of a target is not only the NCC proper
but could also reﬂect processes that in a particular experiment
paradigm, regularly and lawfully, precede and/or follow conscious
perception without directly corresponding to the subjective experience
(Bachmann, 2009; de Graaf et al., 2012; Melloni and Singer,
2010). These processes need to be distinguished and disentangled
from each other before we can understand the crucial neural
mechanisms underlying conscious experience. Fig. 1 illustrates the
development of enthusiastic proposals as well as critical ideas in the
science of consciousness. The recent criticism on the contrastive
analysis is expected to pave the way for further positive developments
and experimental paradigms that will eventually lead us
closer to unraveling the neural processes directly associated with
conscious experience.
2. The 3 NCCs
In what follows we recapitulate the two new types of NCCs proposed
previously (Bachmann, 2009; de Graaf et al., 2012; Melloni
and Singer, 2010): the prerequisites for and the consequences of
conscious perception, in short NCC-pr and NCC-co, respectively. We
also highlight arguments that support the existence of these processes
confounding the search for the NCC. For the sake of clarity
we start with the existing deﬁnition of the NCC.
2.1. NCC for processes directly corresponding to conscious
experience
NCC is a neural process that directly corresponds to the phenomenal
experience of the target. NCC is the “minimal set of neural
events jointly sufﬁcient for a speciﬁc conscious experience (given
the appropriate enabling conditions)” (Koch, 2004). In other words,
if we would stimulate or generate these neural events, a particular
conscious experience would happen. Therefore, NCC is the process
we need to study in order to understand how conscious experience
of a particular content is related to the neuronal processes of the
brain. Previously, it has been thought that the contrastive analysis,
as commonly used, directly reveals the NCC (Fig. 2a). Recently,
however it has been argued that the result of the contrast between
trials with and without conscious perception can also be processes
that precede or follow conscious perception (Fig. 2b).
2.2. Prerequisite processes carrying no actual conscious contents
(NCC-pr)
These processes are different in the conditions target in
consciousness vs. target not in consciousness and participate
in determining in the particular experimental setup if the target
appears in consciousness, but are not part of the NCC (Bachmann,
2009; de Graaf et al., 2012). There are many different processes that
can act as NCC-prs. Hence, one particular NCC-pr need not be sufﬁcient
or even necessary for generating a conscious experience of
the target, but some kind of NCC-pr might be necessary to bring
about the very processes of NCC.
In a paradigm with transient stimulation an example of NCC-pr
(Fig. 3a and b) is the stochastic ﬂuctuations in the excitability of
neurons. For long, it has been recognized that the ability to perceive
a weak signal ﬂuctuates over time – an effect that has been
exploited by the contrastive method to create conscious vs. nonconscious
conditions. However, only recently it was shown that
ﬂuctuations of ongoing brain activity, as indexed by the phase of
pre-stimulus oscillations, systematically determine these behavioral
dynamics (Busch et al., 2009; Mathewson et al., 2009; Monto
et al., 2008). The phase of ongoing alpha (10 Hz) oscillations also
determines the effectiveness with which a single TMS pulse elicited
a phosphene: with a 15% increase in the likelihood of a perceptual
outcome (phosphene) between opposite phases (Dugue et al.,
2011). This effect was observed over an extended time period
(∼400 ms), showing that processes correlating with conscious perception
of the target clearly precede the experience itself. As all
this takes place in the pre-stimulus interval (being it either before
stimulus presentation or TMS pulse application), where no conscious
experience of the actual target could have emerged, it
constitutes a NCC-pr. Other studies relating pre-stimulus activity
to subjective reports show that NCC-pr can reﬂect spontaneous
excitability as also indexed by oscillatory power (Ergenoglu et al.,
2004; Linkenkaer-Hansen et al., 2004; Romei et al., 2008; van
Dijk et al., 2008) and that the NCC-pr can be linked to attention
(Thut et al., 2006; Worden et al., 2000; Wyart and Tallon-Baudry,
2009), decision bias (Wyart and Tallon-Baudry, 2009) and potentially
other processes.
In the experimental paradigms with sustained epochs of perception
(Fig. 3c), NCC-prs can manifest themselves in various ways.
For example, for binocular rivalry it was proposed more than a
century ago and shown recently (Alais et al., 2010) that adaptation
leading to weakening of reciprocal inhibition determines the
alternations between competing stimuli. More precisely, neurons
coding for the dominant stimulus adapt over time, which in turn
weakens the inhibition of the suppressed stimulus, increasing its
neuronal responses and thus bringing that stimulus into consciousness.
Importantly, (reciprocal) inhibition could be seen as NCC-pr,
J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746 739
Fig. 1. A time-line of the major events in the science of consciousness. The time-line illustrates that fresh optimistic theoretical and empirical insights are always followed
by critical evaluation of these ideas and results. Therefore, the recognition of the problem regarding the prerequisites for and consequences of NCC is a logical corollary of
the popularity of the contrastive method in the science of consciousness. The acknowledgement of the problem should be followed by clever experiments trying to separate
the prerequisites for and consequences of conscious perception from the processes that reﬂect NCC directly. (The selection of the events for the time-line was necessarily
subjective and we apologize to all the co-researchers who feel that their work has been neglected.)
as it contributes to which target will be consciously perceived and
is different between the two contrastive conditions (a particular
target in consciousness vs. not in consciousness). However, it is not
part of the neural processes sufﬁcient for generating or maintaining
conscious experience of that target.
2.3. Consequences following actual conscious experience
(NCC-co)
These processes appear as after-effects when NCC is or just has
been present, but are not a part of the minimally sufﬁcient mechanisms
of the conscious experience of the target. These processes
are the consequences of conscious perception (de Graaf et al., 2012;
Melloni and Singer, 2010). The existence of such processes is a logical
consequence of assigning any function to conscious perception –
if conscious perception enables certain processes that unconscious
perception does not, these processes would inevitably appear in
the contrast between trials with and without conscious perception,
even if they are solely the consequences and not the direct correlates
of consciousness. Importantly, most theories of consciousness
do confer a function to consciousness in the chain of information
processing (Seth, 2009), for instance sustained maintenance
of information, access to long-term memory, novel combinations
of operations, and intentional behavior.
For example, it is known that neurons in the medial temporal
lobe (MTL) respond in all-or-none fashion, closely following the
subjective report of the patient (Quiroga et al., 2008). The ﬁdelity
with which MTL neurons follow the subjective visibility of the stimuli
is so high that conscious versus non-conscious trials can be
distinguished solely based on the neurons’ ﬁring rate: either the
neuron responds when the subject reports to have consciously recognized
the picture, or stay completely silent when the image did
not reach consciousness. However, damage to the MTL-system (or
even its complete resection) does not affect moment-to-moment
conscious perception (Crick and Koch, 1990; Postle, 2009), but
only the formation of a memory trace. That is, subjects will continue
to have subjective experience but will have no memory of it.
Therefore, such all-or-none responses in MTL, even though closely
following the subject’s report, do not correspond to the NCC but are
instead part of the NCC-co reﬂecting processes related to memory
consolidation.
In addition, NCC-co processes can also reﬂect differences in
performance. Usually, conditions in which subjects are conscious
of the stimuli correlate with higher performance (for instance
higher detectability or discriminability) than when subjects are
not conscious of the stimuli. Thus, it has been claimed that a part
of the brain activity observed using the contrast between conscious
and non-conscious perception pertains to differences in
performance instead of conscious experience (Lau, 2008). Direct
evidence for this claim comes from a recent neuroimaging study
(Lau and Passingham, 2006) reporting that when performance was
equalized between conditions, so that only subjective experience
varied, only dorsolateral prefrontal cortex distinguished conscious
from unconscious conditions as opposed to the extensive frontoparietal
network typically reported in previous studies (Dehaene
and Naccache, 2001). Recent studies in which performance was
assessed alongside with subjective experience further conﬁrm that
these two processes can indeed dissociate in time, space and,
importantly, in neural locus (Hesselmann et al., 2011; Lamy et al.,
2009; Schwiedrzik et al., 2011). This highlights that performance
and subjective experience are not interchangeable and that to
investigate NCC proper, performance has to be controlled for.
2.4. The 3 NCCs: conclusions
Thus, despite the principal methodological importance of the
contrastive analysis, the method, as commonly used, seems to
lack the required speciﬁcity to unravel the neuronal processes
exclusively related to the subjective experience. In fact, simply contrasting
conscious to non-conscious conditions could also reﬂect
processes preceding or following the NCC. The problem exists
in all experimental paradigms currently used to investigate consciousness
(Fig. 3), regardless of whether differences in conscious
perception result from internal switches in brain states, such as
in binocular rivalry, or from external manipulations of visibility,
as for example achieved by varying the SOA in masking experiments.
It is conceivable that the distinction between the 3 NCCs
also applies to studies in which different states of consciousness
are contrasted (e.g. deep sleep vs. waking or vegetative state vs.
minimally conscious states), as these conditions do not exclusively
differ in the consciousness state (which is the intended manipulation),
but could also do so in their prerequisite processes and
740 J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
Fig. 2. Neural processes revealed by the contrastive analysis. In the contrastive analysis, a stimulus (here a blue “splash”) is presented so that it sometimes appears in
conscious experience and sometimes not. On each trial the subject gives a response indicating whether the stimulus was consciously perceived or not. Neural processes
are sorted according to the subjective responses to “aware” and “unaware” conditions, which are compared to each other. What neural processes can be the outcome of
this contrast? (a) The traditional view on contrastive analysis assumes that the only difference between these conditions is the subjective experience of the stimulus. Thus,
the contrastive analysis appears suitable to reveal the neural correlates of subjective experience (NCC). (b) The proposed interpretation of contrastive analysis. According
to this view the outcome of contrastive analysis consists besides the NCC of two other processes – NCC-pr and NCC-co. NCC-pr corresponds to unconscious processes
that though related to conscious perception of a stimulus (e.g., attention that enhances weak information to cross the threshold of consciousness) appear before any
subjective experience emerges. NCC-co represents the consequences of consciously perceiving the stimulus. Consequently, if this view is correct, the traditional contrastive
analysis by itself cannot exclusively reveal the NCC because it confounds NCC with these other processes that do not directly correspond to the conscious experience of the
stimulus.
especially in the consequences (in the conscious state people think,
access long-term memory, have intentions etc.). It will be necessary
to determine if and how much the 3 NCC problem applies
to other areas of research where conscious experience is investigated
(e.g. studies of conscious intentions, introspection, imagery,
dreams, inner speech etc.).
This conjecture implies that previous results based on the contrast
between trials with and without conscious perception of a
particular content cannot be unequivocally interpreted as reﬂecting
the NCC, as they could also represent other processes. Given that
the majority of studies investigating consciousness have used such
contrasts (Dehaene and Changeux, 2011; Koivisto and Revonsuo,
2010; Rees, 2007; Rees et al., 2002; Tononi and Koch, 2008) such
reinterpretation might have a major impact on consciousness
research and on current theories on the neural mechanisms of consciousness.
As long as the NCC proper cannot be clearly dissociated
from these other processes, we should be cautious when relating
any experimental ﬁnding about neural processes correlating with
conscious experience to the NCC. Therefore, if we want to proceed
in understanding consciousness, the NCC has to be experimentally
distilled from the prerequisites for and the consequences of
conscious perception. As mentioned above, the problem of 3 NCCs
may also apply to other areas in the study of conscious experience
besides the investigation of conscious perception. Here we focus
on the latter as it constitutes the best studied aspect of conscious
experience. However, it might prove fruitful to apply the concept
of 3 NCCs also to these other aspects of conscious experience as
alternative solutions to disentangle the 3 NCC conundrum might
emerge from that research.
3. Distilling consciousness: disentangling the NCCs
There is probably no single experiment with today’s research
techniques that would yield a clear separation between NCC and
NCC-pr or NCC and NCC-co. As discussed above, the contrast
between trials with and without conscious perception of a particular
content does not dissociate these processes. In what follows,
we propose research strategies to separate the NCC from the
J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746 741
Fig. 3. Possible temporal relationships between NCC, NCC-pr and NCC-co in various experimental situations. (a) If a single transient stimulus is presented, NCC-pr, NCC and
NCC-co are evoked. Areas with dotted outlines indicate that the temporal differences and overlap between the 3 NCCs is currently unknown. This scenario corresponds to
experimental paradigms like threshold stimulation. (b) If two (or more) transient stimuli are presented sequentially, the different NCCs of these different stimuli can be
overlapping in time. This scenario corresponds to experimental paradigms like masking or attentional blink. (c) In experimental situations with longer epochs of perception,
each epoch can be understood as a succession of the transient events (as in a), where the NCC of time point t can overlap with NCC-co from time point t − 1 and NCC-pr
from time point t + 1. This scenario corresponds to experimental paradigms like binocular rivalry or motion induced blindness. Similar ideas on the temporal overlap of the
processes unfolding within pre-conscious and conscious stages of percept formation have been also proposed previously (Brown, 1988).
NCC-prs and NCC-cos (see also de Graaf et al., 2012; Melloni and
Singer, 2010). Success in understanding the neuronal processes
directly underlying conscious experience rests upon distinguishing
and disentangling its confounds.
3.1. Disentangling NCC-pr from NCC
There is no question that processes differentiating conscious
from non-conscious stimuli already before stimuli onset correspond
to NCC-pr. Beyond that, difﬁculties begin. As NCC-pr can
potentially appear after stimulus onset but also concurrently with
NCC, there is no simple time criterion that separates NCC-pr from
NCC. Furthermore, an interesting theoretical but also empirically
challenging problem arises when considering that the NCC-pr do
not only have to appear before the NCC but can also stay active
during the NCC (Fig. 3). If NCC-pr is there only to “ignite” NCC by preceding
it, the empirical study and measurement of NCC is tractable
because later epochs of target experience in long-duration target
stimuli are not confounded with NCC-pr. However, we currently
lack the required knowledge about the NCC-prs to make any claims
about their duration or temporal structure.
The most straightforward way to disentangle NCC-pr from NCC
is to directly manipulate the NCC-pr processes, and compare the
neural signatures that are common to all of them. We assume that
different NCC-pr would elicit distinct neural activities, while neural
process directly involved in consciousness would be invariantly
present in all conditions. As an example, consider a “consciousness
task” in which a stimulus under identical stimulation conditions
(e.g., a masking experiment with invariant masks and SOAs) is
sometimes consciously perceived and sometimes not. As discussed
above, this approach would lead to the problem of 3 NCCs. However,
now, in an additional step, in the same experimental setup with the
same stimuli, we could vary potential NCC-prs such as for instance
stimulus expectation, adaptation, working memory or allocation
of attention independently. Speciﬁcally, in one condition stimuli
could be brought to consciousness through expectation (Melloni
et al., 2011) and in another through attention (Wyart and TallonBaudry,
2008). Which neural processes resulting from the contrast
consciously perceived vs. not consciously perceived are similar
and which ones are different in these two conditions? Neural signatures
that differ between conditions should belong to the NCC-pr;
neural signatures common to both comparisons are likely to be
related to the NCC. Using such an approach, it was recently shown
that short-latency event-related potentials previously related to
conscious perception (Pins and Ffytche, 2003) most likely represent
signatures of the NCC-pr (Melloni et al., 2011) as opposed to
NCC. Contrasting attention with expectations might be particularly
revealing, as these processes are proposed to have opposite effects
on neural activity – attention increases sensory responses whereas
expectation decreases them (Summerﬁeld and Egner, 2009). Thus,
neural processes that increase with attention but decrease with
expectation under similar subjective experience and objective performance
are unlikely to reﬂect the NCC proper.
In addition to cognitive factors, one could affect NCC-pr in a similar
vein with TMS. In a recent experiment it was observed that when
the visual cortex was stimulated 100–120 ms after a near-threshold
visual stimulus with TMS intensity below the phosphene threshold,
the thresholds for explicit perception of the visual stimulus were
decreased (Abrahamyan et al., 2011). In line with the experimental
strategies proposed above, one could combine this TMS-related
742 J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
improvement of perception with an independent cognitive manipulation
(e.g. of attention or expectation) that also has a beneﬁcial
effect on perception and measure the neural correlates of conscious
perception with the same logic as presented previously: only neural
signatures related to perceptual enhancement common to both
manipulations are likely to be related to the NCC.
3.2. Disentangling NCC-co from NCC
As with NCC-pr, one cannot straightforwardly rely on a strict
temporal distinction between NCC and NCC-co. Probably correlates
that appear 2 s after the onset of a 30 ms stimulus indeed represent
NCC-co, however, the critical question refers to neural correlates
appearing earlier than 500 ms post-stimulus as the bulk of studies
have related those to NCC (Del Cul et al., 2007; Fisch et al.,
2009; Gaillard et al., 2009; Koivisto et al., 2008; Melloni et al., 2011;
Sergent et al., 2005). Neurons in the MTL have a latency of about
300 ms MTL (Quiroga et al., 2008). If we are right in stating that the
activity of MTL neurons, even when apparently closely following
the subjective report, is not the NCC but constitutes the NCC-co, it
becomes evident that the NCC-co can be present already in a time
window that is often investigated for the NCC.
According to the deﬁnition, the NCC-co corresponds to the aftereffects
of conscious perception. Thus, it is logically possible to
observe NCC even when NCC-co is not elicited. In the normal brain
such a dissociation might not occur very often; however, lesion
studies might be revealing: if an area can be lesioned or removed
without any effect on conscious perception, then activation of this
area can be regarded as NCC-co. This approach allowed, for instance,
ruling out activity in MTL as a direct NCC (Crick and Koch, 1990;
Postle, 2009). Importantly, the tests applied to identify NCC-cos
have to be sensitive enough to capture subtle effects on conscious
perception. For instance, although the view that prefrontal cortex
(PFC) is causally relevant for conscious perception (Dehaene and
Changeux, 2011) could at ﬁrst have been rejected based on lesion
studies (Pollen, 1999), it has gained support from carefully conducted
studies exploiting the masking paradigm in patients with
PFC lesions (Del Cul et al., 2009). Nonetheless, the fact that the psychophysical
threshold for conscious access is elevated in patients
with PFC lesions cannot be taken as direct evidence of PFC having
a central role in conscious perception. Instead, PFC could provide
top-down support for either strengthening or maintaining the (cognitive
results of the) conscious percept (Gazzaley et al., 2007; Miller
et al., 2011). This is a likely possibility as patients could perform the
difﬁcult masking task relatively well and were reporting conscious
perception despite their PFC lesion.
Lesion studies can be complemented with interventional techniques
in healthy subjects, such as TMS (see also de Graaf et al.,
2012) and/or transcranial direct or alternating current stimulation
(tDCS or tACS). This approach offers several advantages over lesion
studies. First, it allows for comparisons within the same individual,
as TMS can be applied in an on–off manner. Second, due to
the transient nature of the intervention, profound reorganization
of brain networks is not to be expected. This allows dissociating
effects related to lesioning a particular area from alterations at the
level of the network resulting from compensatory plasticity. Third,
virtual lesions can be made with a precision not comparable even
when performing probabilistic lesion mapping studies in patients.
Furthermore, virtual lesions can be made not only of a single region,
but also in several regions at the same or at varying times. Combined
with neuroimaging, one could ﬁrst ﬁnd a neural signature X
that correlates with conscious perception in the contrastive analysis
(e.g. the activity in the frontal cortex) and then in a second step
test whether perturbation of this signature X (e.g. by TMS) has an
effect on conscious perception. If such perturbation produces no
effects on conscious perception, the respective process or area is a
prime candidate for NCC-co. However, to ﬁrmly conclude that, evidence
has to be provided that such manipulation has a behavioral
effect. Otherwise, it is unclear whether the effect was not observed
because the perturbation is simply ineffective, or it speciﬁcally
does not affect conscious perception. The feasibility of such selective
manipulation has been recently shown: theta-burst TMS to
bilateral dorsolateral PFC interfered with metacognitive aspects of
visual awareness but not with discrimination performance (Rounis
et al., 2010).
Besides interventional approaches, much progress could be
made by (i) obtaining comprehensive data on the sequence
of cognitive events distinguishing conscious from unconscious
processing, assuming that causes precede effects, and by (ii) considering
theoretical approaches which clearly disclose some cognitive
processes (e.g. working memory) as dependent on conscious perception
(Baars, 1989). This is particularly relevant as it allows for
contrasts in which not only consciousness is manipulated (e.g.,
aware vs. unaware trials) but also its consequences (e.g., conditions
in which encoding in working memory is present or absent).
Following such an approach, it was recently shown that late electrical
signatures of consciousness, in particular the P3 event-related
potential typically associated with conscious access (Dehaene and
Changeux, 2011), does not follow conscious perception when subjects
already have a conscious working memory representation of
the target stimulus (Melloni et al., 2011). This result points to the
tantalizing possibility that late waves of EEG activity such as the P3
might reﬂect NCC-cos, and not the NCC itself.
4. The 3 NCCs and previous results
In light of the ideas proposed here, an important question is
whether previous ﬁndings reﬂect NCC or rather NCC-pr, NCC-co
or even a compound of these different processes. Since little is
known about the NCC-pr and NCC-co and how they interact with
the NCC proper, it is currently not straightforward to determine
which results indeed relate to NCC and which ones do not. A ﬁrst
step is to identify and recognize the neuronal signatures of processes
that might not reﬂect the NCC.
Importantly, many current theories of consciousness are
founded on empirical ﬁndings, assuming that these genuinely
reﬂect the NCC. As shown, this assumption might not hold true.
Thus, it is possible that some theories about the NCC are actually
based on NCC-prs or NCC-cos. Clarifying this issue will hopefully
lead to better agreement among theories of consciousness. Note
that we do not suggest that all previous ﬁndings are necessarily
wrong or doubtable, nor do we claim that all theories are based on
the wrong assumptions about the previous results; we simply point
to the worrisome possibility that follows the theoretical distinction
between the 3 NCCs. In the next paragraphs we link this theoretical
problem to existing results from the contrastive analysis with the
hope that the recognition of the 3 NCC problem will contribute constructively
to the disputes about the timing and neuroanatomical
locus of conscious perception.
4.1. The 3 NCCs and the temporal course of conscious perception
The distinction between the three NCCs could prove valuable
to resolve some controversies regarding the question whether
conscious perception happens early (Aru and Bachmann, 2009b;
Melloni et al., 2007; Pins and Ffytche, 2003) or late (Del Cul
et al., 2007; Gaillard et al., 2009; Sergent et al., 2005), as some of
the studies observing early correlates might reﬂect NCC-pr while
those reporting late correlates might reﬂect NCC-co. In a recent
study, such an approach was purposely followed: the processes
related to NCC-pr and NCC-co were manipulated in an attempt
J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746 743
−100 0 100 200 300 400 500
−1
−0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
Time [ms]
GlobalFieldPower
Bottom-up stimulation
Seen
Unseen
Seen − Unseen
−100 0 100 200 300 400 500
−1
−0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
Time [ms]
GlobalFieldPower
Bottom-up stimulation + Top-down Expectations
Fig. 4. Neural signatures of conscious perception: NCC-pr, NCC and NCC-co. In the study of Melloni et al. (2011), both the pre-requisites and the after-effects of conscious
perception were manipulated. Subjects had to rate the visibility of a stimulus embedded in noise. Visibility could either arise because of the strength of sensory evidence (left
panel) or because of the presence of perceptual expectations (right panel). In these conditions, the prerequisites (NCC-pr) differed, but conscious perception was identical.
Furthermore, a conscious working memory representation of the target could already be present (left panel) or absent (right panel). Here, conscious perception was again
equalized, but the consequences (NCC-co) differed, since in the latter case, a new representation needed to be encoded in working memory. The red bars indicate the
periods of signiﬁcant difference between the aware and unaware trials. The only component differentiating consciously perceived from not consciously perceived trials
independently of differences in NCC-pr or NCC-co was a mid latency component, likely reﬂecting neuronal processes directly involved in the NCC. In contrast, early processes
differentiating seen from unseen stimuli were not present when consciousness resulted from an interaction between sensory evidence and top-down expectations. Also,
the late ERP component was only present when a working memory representation had to be established. The fact that some neural signatures of consciousness disappeared
when either the prerequisites or consequences of consciousness were manipulated questions the general involvement of these processes in conscious perception.
to more directly address neuronal processes related to the NCC
itself (Melloni et al., 2011). To investigate NCC-prs, Melloni et al.
(2011) examined how sensory evidence and top-down expectations,
respectively, inﬂuence the threshold of awareness, and
whether the two factors modulate brain activity differently. This
allowed contrasting brain states with and without expectations and
with perceived and non-perceived stimuli for identical stimulation
conditions. To investigate NCC-cos, conditions based on whether
the target was or was not present in working memory were compared.
As in previous studies (Mathewson et al., 2009; Pins and
Ffytche, 2003) it appeared that in the absence of expectations an
early event-related potential (ERP), already present around 100 ms
in occipital sensors, distinguished conscious from unconscious trials
(Fig. 4). Perception under those circumstances is known to
depend on bottom-up stimulation (the amount of sensory evidence)
and stochastic ﬂuctuations in the prestimulus period which
jointly determine whether a target will or will not be consciously
perceived (Busch et al., 2009; Mathewson et al., 2009). However, in
the presence of expectations, that is, when perception depends not
only on bottom-up information but also on top-down expectations,
this early component was not different between the consciously
perceived and not perceived targets. As both conditions resulted in
similar rates of visibility, this ﬁnding suggests that early processes
differentiating seen from unseen stimuli – previously interpreted
as a NCC – were actually not directly related to subjective experience,
but reﬂect differences in NCC-pr (Fig. 4). Along the same
lines, a late ERP (P300) distinguishing consciously perceived from
not consciously perceived trials was only observed in trials where
the target had to be encoded in working memory but not when
subjects already had a conscious working memory representation
of the target. This suggests that late effects like the P300 can index
processes related to the NCC-co instead of the NCC proper under
certain conditions (Fig. 4). At threshold, the component most consistently
related to conscious perception was the P200, arising at
about 200 ms over occipitoparietal sensors. This study also revealed
that the electrophysiological signatures of conscious perception are
not bound to processes with a strict latency, but depend on how
consciousness comes about: earlier electrophysiological signatures
were observed in the presence of top-down expectations (P200)
than in their absence (P300). This further complicates the search of
the NCC, as they might change in time depending on which NCC-pr
determines perception.
Converging evidence for the proposed temporal relationships
between NCC-pr, NCC and NCC-co comes from studies with intermittent
binocular rivalry. In this paradigm, epochs of binocular
rivalry stimulation alternate with blank intervals, which allows to
study neural processes locked to the stimulus onset and to delineate
the sequence of neural events related to bistable perception.
Interestingly, Pitts and Britz (2011) when reviewing recent studies
employing such methodology come to similar three time windows
related to conscious perception as we did based on a different
experimental paradigm described above (Melloni et al., 2011): (1)
early processes around 130–160 ms after stimulus onset that vary
with conscious perception but probably constitute a preconscious
state (in the current terms the NCC-pr), as although this component
varies with the subjective percept (Pitts et al., 2010), the stable perceptual
representation of the target has not emerged yet (Pitts and
Britz, 2011), (2) the “reversal negativity” around 200–300 ms that
is a “primary candidate for a neural correlate of awareness” (Pitts
and Britz, 2011), as there the stable representation has been established
and (3) the late positive complex around 400–600 ms that
might reﬂect further processing of the perceptual information or
processes related to working-memory maintenance (NCC-co in the
current terms).
4.2. The 3 NCCs and the activity of different brain areas
Regarding the neural locus of consciousness it is also debated
whether conscious perception is associated with activity in sensory
areas (Hesselmann et al., 2011; Lamme, 2006; Zeki, 2001)
or whether higher non-sensory areas are directly involved in
conscious perception (Dehaene et al., 2006; Lau and Rosenthal,
2011). In this context, several previous studies are directly relevant
for understanding whether particular brain areas such as
the prefrontal cortex serve a crucial role in conscious perception,
as proposed in some theories (Dehaene and Naccache, 2001;
Lau, 2008), or whether they reﬂect other cognitive processes,
such as top-down control, report, or performance on a task. The
744 J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
available evidence relating PFC activity with consciousness has
been obtained by contrasting conscious vs. non-conscious trials
(Dehaene and Changeux, 2011). The key message of our proposal
is that by solely relying on the contrast between trials with and
without conscious perception of a particular content it is undetermined
whether such activation indeed constitutes an NCC or rather
a consequence of conscious perception (NCC-co). In line with this
conjecture, when subjects do not need to report the stimuli, differential
activity between conscious and non-conscious stimuli is
not observed in the prefrontal cortex but only in occipital visual
areas downstream of V1/V2 (Tse et al., 2005). In addition, activity
in prefrontal cortex is suppressed when subjects are engaged
in a demanding perceptual categorization task, although they are
presumably fully conscious of their rapidly changing visual or auditory
world during that time (Goldberg et al., 2006). Finally, although
TMS over frontal cortex affected voluntary control of bistable perception,
TMS had no inﬂuence on passive bistable perception (de
Graaf et al., 2011), which conforms nicely with the approach proposed
in Section 3.2: TMS had a measurable effect on one process
(voluntary control) but not on conscious perception. These results
directly support the thesis that prefrontal activity might not be a
part of the NCC but could rather represent processes that follow
conscious perception as consequences.
In another relevant study, Eriksson and colleagues (Eriksson
et al., 2008) found that prefrontal activity and coupling between
prefrontal and occipital areas decreased after training in an identiﬁcation
task. This result might imply that at least part of
the prefrontal activity observed in contrastive analysis could be
attributable to the fact that in such paradigms, conscious recognition
is often made difﬁcult (e.g. by masking or by lowering the
contrast of the stimuli) and PFC might be recruited to aid the recognition
process under poor sensory evidence (Eriksson et al., 2008).
From this perspective PFC activity could constitute a NCC-pr, as
prefrontal activity would determine whether the target reaches
consciousness or not. This idea agrees with theories and empirical
evidence suggesting that short-latency prefrontal activity facilitates
perception (Bar, 2003; Bar et al., 2006). Alternatively, it could
be argued that PFC activity represents NCC-co (instead or in addition
to the NCC-pr) that change as a function of perceptual training.
In particular, extensive experience with the stimulus could lead to
established memory representations such that there is no need to
create working memory representations on each trial, leading to
decreased prefrontal activity with training.
All these results challenge the notion that prefrontal cortex is
directly involved in visual awareness but suggest instead that it
could reﬂect executive functions, top-down facilitation and/or topdown
control under conditions of increased effort. As a number
of theories of consciousness (Crick and Koch, 2003; Dehaene and
Changeux, 2011; Dehaene and Naccache, 2001; Lau, 2008) assert
that activation of prefrontal cortex is part of the NCC, it is clear that
the question whether PFC activity is indeed a NCC or rather a NCCpr
or NCC-co is central for further research on the neural correlates
of consciousness.
Similar caveats hold true for the involvement of parietal cortex
in consciousness. In fact, when using the contrastive method
in visual versus auditory tasks, it was found that activity in parietal
cortex is correlated with visual but not with auditory conscious
perception (Eriksson et al., 2007), questioning its general, modalityindependent
role in conscious perception. Even further, it is known
from the neurological literature that bilateral lesions of the parietal
cortex do not abolish conscious perception: although patients
exhibiting such lesions are only able to perceive one object at a time
(simultagnosia), this one object is nevertheless consciously experienced
(Friedman-Hill et al., 1995; Robertson, 2003). These results
suggest that the activity of the parietal cortex might not be a part of
the NCC. As substantial empirical evidence links the activity of the
parietal cortex to perceptual alternations in paradigms with multistable
perception (Kleinschmidt et al., 1998; Lumer et al., 1998),
it is likely that the processes in the parietal cortex constitute an
essential prerequisite for NCC. This claim is substantiated by recent
evidence that TMS to parietal cortex affects the dominance durations
during binocular rivalry (Carmel et al., 2010; Zaretskaya et al.,
2010) and that the structure of the parietal cortex is correlated with
the intraindividual differences in the perceptual alternation rate of
a bistable stimulus (Kanai et al., 2010).
Finally, there is a long-standing controversy about the role of
primary visual cortex (V1) in conscious perception. Even if V1
lesions lead to loss of awareness in the corresponding parts of the
visual ﬁeld, it constitutes no strong argument for the NCC being
in V1, as V1 lesions also disrupt information ﬂow to higher order
visual areas. Empirical and theoretical arguments in the mid-1990s
suggested that V1 might not be necessary for visual conscious
experience (Barbur et al., 1993; Crick and Koch, 1995) but further
research and theoretical positions generally favored the idea
that cortical feedback to V1 is indeed necessary for visual conscious
perception and therefore part of the NCC (Lamme, 2001;
Pascual-Leone and Walsh, 2001; Tong, 2003). We believe that the
division between the three types of NCC can be fruitful in solving
such controversies, as it offers the possibility that although
V1 activity correlates with conscious perception (Tong, 2003), it
might constitute a NCC-pr rather than be a part of the NCC. Indeed,
a recent experiment revisited the issue whether patients with V1
lesion lack visual consciousness (Ffytche and Zeki, 2011) and contrary
to what is commonly believed the authors showed that these
patients do have visual experiences. Furthermore, clinical studies
have shown that conscious vision can recover after a V1 lesion
(Silvanto and Rees, 2011). These results imply that neither activity
in V1 nor the cortical feedback to V1 is necessary for visual conscious
perception. In the current terminology, intact V1 would be
an important prerequisite for visual conscious experience, but not
part of the NCC. However, it is possible that V1 is not necessary for
all visual experiences but those involving the ﬁne-tuned spatial contrast
distribution in the experienced image due to the functional
capabilities of V1. Thus, whether some brain area is necessarily
involved in NCC may depend not only on its neuroanatomical locus,
but also on the speciﬁc details of the experience (Haynes, 2009).
Taken together, the tripartite distinction of NCCs calls for a reappraisal
of previous results and theories, but also highlights the need
of further studies that put emphasis on investigating and distinguishing
the pre-conscious determinants of conscious perception
(NCC-pr) and its consequences (NCC-co) from the NCC proper.
5. Conclusions
Despite its original simplicity and appeal, the traditional
method, contrasting trials with and without conscious perception
of a particular target, by itself does not appear to have the necessary
speciﬁcity to reveal the NCC. Instead, processes that precede
(NCC-pr) and follow (NCC-co) conscious perception are confounded
with the NCC proper. Here we relied on the theoretical distinction
between those processes (Bachmann, 2009; de Graaf et al., 2012;
Melloni and Singer, 2010) and suggested tentative experimental
paradigms aimed at disentangling these processes from each other.
An important way to circumvent the speciﬁcity problem altogether
is to move away from mere correlates and focus the search
on mechanisms of conscious experience, as correlates which do
not have explanatory power can be excluded a priori (Melloni
and Singer, 2010). However, the key to success in unraveling the
NCC will ultimately lie in combining such mechanistic models
that generate testable predictions (Bachmann, 2007; Dehaene and
Changeux, 2011; Lamme, 2006; Melloni and Singer, 2010) with
J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746 745
suitable experimental paradigms. Only then we will be capable of
putting theoretical proposals to the test. We thus hope that future
work, adopting the distinction between NCC, NCC-pr and NCC-co
and following the empirical strategies proposed here, will allow
us to re-assess previous studies and theories, focus on the NCC,
and distill the crucial neural processes that underlie our conscious
experience.
Acknowledgments
This work was supported by the Max Planck Society, the
Ernst Strüngmann Foundation, and the targeted ﬁnancing #2717
from the Estonian Scientiﬁc Competency Council. The authors are
indebted to Caspar M. Schwiedrzik for helpful discussions and comments
on a previous version of this manuscript. We also thank
Felipe Aedo-Jury for suggesting the title of the paper.
References
Abrahamyan, A., Clifford, C.W., Arabzadeh, E., Harris, J.A., 2011. Improving visual
sensitivity with subthreshold transcranial magnetic stimulation. J. Neurosci. 31,
3290–3294.
Alais, D., Cass, J., O’Shea, R.P., Blake, R., 2010. Visual sensitivity underlying changes
in visual consciousness. Curr. Biol. 20, 1362–1367.
Aru, J., Bachmann, T., 2009a. Boosting up gamma-band oscillations leaves targetstimulus
in masking out of awareness: explaining an apparent paradox.
Neurosci. Lett. 450, 351–355.
Aru, J., Bachmann, T., 2009b. Occipital EEG correlates of conscious awareness when
subjective target shine-through and effective visual masking are compared:
bifocal early increase in gamma power and speed-up of P1. Brain Res. 1271,
60–73.
Baars, B.J., 1989. A Cognitive Theory of Consciousness. Cambridge University Press,
New York.
Bachmann, T., 2007. Binding binding: departure points for a different version of the
perceptual retouch theory. Adv. Cogn. Psychol. 3, 41–55.
Bachmann, T., 2009. Finding ERP-signatures of target awareness: puzzle persists
because of experimental co-variation of the objective and subjective variables.
Conscious. Cogn. 18, 804–808 (discussion pp. 809–810).
Bachmann, T., Breitmeyer, B.G., Ö˘gmen, H., 2011. The Experimental Phenomena of
Consciousnes: a Brief Dictionary Revised Edition. Oxford University Press, New
York.
Bar, M., 2003. A cortical mechanism for triggering top-down facilitation in visual
object recognition. J. Cogn. Neurosci. 15, 600–609.
Bar, M., Kassam, K.S., Ghuman, A.S., Boshyan, J., Schmid, A.M., Dale, A.M., Hamalainen,
M.S., Marinkovic, K., Schacter, D.L., Rosen, B.R., Halgren, E., 2006. Top-down
facilitation of visual recognition. Proc. Natl. Acad. Sci. U.S.A. 103, 449–454.
Barbur, J.L., Watson, J.D., Frackowiak, R.S., Zeki, S., 1993. Conscious visual perception
without V1. Brain 116 (Pt 6), 1293–1302.
Brown, J.W., 1988. The Life of the Mind. Erlbaum, Hillsdale.
Busch, N.A., Dubois, J., VanRullen, R., 2009. The phase of ongoing EEG oscillations
predicts visual perception. J. Neurosci. 29, 7869–7876.
Carmel, D., Walsh, V., Lavie, N., Rees, G., 2010. Right parietal TMS shortens dominance
durations in binocular rivalry. Curr. Biol. 20, R799–R800.
Crick, F., Koch, C., 1990. Toward a neurobiological theory of consciousness. Semin.
Neurosci. 2, 263–275.
Crick, F., Koch, C., 1995. Are we aware of neural activity in primary visual cortex?
Nature 375, 121–123.
Crick, F., Koch, C., 2003. A framework for consciousness. Nat. Neurosci. 6, 119–126.
de Graaf, T.A., de Jong, M.C., Goebel, R., van Ee, R., Sack, A.T., 2011. On the functional
relevance of frontal cortex for passive and voluntarily controlled bistable vision.
Cereb. Cortex 21, 2322–2331.
de Graaf, T.A., Hsieh, P.J., Sack, A.T., 2012. The ‘correlates’ in neural correlates of
consciousness. Neurosci. Biobehav. Rev. 36, 191–197.
Dehaene, S., Changeux, J.P., 2011. Experimental and theoretical approaches to conscious
processing. Neuron 70, 200–227.
Dehaene, S., Changeux, J.P., Naccache, L., Sackur, J., Sergent, C., 2006. Conscious, preconscious,
and subliminal processing: a testable taxonomy. Trends Cogn. Sci. 10,
204–211.
Dehaene, S., Naccache, L., 2001. Towards a cognitive neuroscience of consciousness:
basic evidence and a workspace framework. Cognition 79, 1–37.
Del Cul, A., Baillet, S., Dehaene, S., 2007. Brain dynamics underlying the nonlinear
threshold for access to consciousness. PLoS Biol. 5, e260.
Del Cul, A., Dehaene, S., Reyes, P., Bravo, E., Slachevsky, A., 2009. Causal role
of prefrontal cortex in the threshold for access to consciousness. Brain 132,
2531–2540.
Dugue, L., Marque, P., VanRullen, R., 2011. The phase of ongoing oscillations mediates
the causal relation between brain excitation and visual perception. J. Neurosci.
31, 11889–11893.
Ergenoglu, T., Demiralp, T., Bayraktaroglu, Z., Ergen, M., Beydagi, H., Uresin, Y., 2004.
Alpha rhythm of the EEG modulates visual detection performance in humans.
Brain Res. Cogn. Brain Res. 20, 376–383.
Eriksson, J., Larsson, A., Ahlstrom, K.R., Nyberg, L., 2007. Similar frontal and distinct
posterior cortical regions mediate visual and auditory perceptual awareness.
Cereb. Cortex 17, 760–765.
Eriksson, J., Larsson, A., Nyberg, L., 2008. Item-speciﬁc training reduces prefrontal
cortical involvement in perceptual awareness. J. Cogn. Neurosci. 20, 1777–1787.
Ffytche, D.H., Zeki, S., 2011. The primary visual cortex, and feedback to it, are not
necessary for conscious vision. Brain 134, 247–257.
Fisch, L., Privman, E., Ramot, M., Harel, M., Nir, Y., Kipervasser, S., Andelman, F.,
Neufeld, M.Y., Kramer, U., Fried, I., Malach, R., 2009. Neural ignition: enhanced
activation linked to perceptual awareness in human ventral stream visual cortex.
Neuron 64, 562–574.
Friedman-Hill, S.R., Robertson, L.C., Treisman, A., 1995. Parietal contributions to
visual feature binding: evidence from a patient with bilateral lesions. Science
269, 853–855.
Gaillard, R., Dehaene, S., Adam, C., Clemenceau, S., Hasboun, D., Baulac, M., Cohen,
L., Naccache, L., 2009. Converging intracranial markers of conscious access. PLoS
Biol. 7, e61.
Gazzaley, A., Rissman, J., Cooney, J., Rutman, A., Seibert, T., Clapp, W., D’Esposito,
M., 2007. Functional interactions between prefrontal and visual association cortex
contribute to top-down modulation of visual processing. Cereb. Cortex 17
(Suppl. 1), i125–i135.
Goldberg, I.I., Harel, M., Malach, R., 2006. When the brain loses its self: prefrontal
inactivation during sensorimotor processing. Neuron 50, 329–339.
Haynes, J.D., 2009. Decoding visual consciousness from human brain signals. Trends
Cogn. Sci. 13, 194–202.
Hesselmann, G., Hebart, M., Malach, R., 2011. Differential BOLD activity associated
with subjective and objective reports during blindsight in normal observers. J.
Neurosci. 31, 12936–12944.
Kanai, R., Bahrami, B., Rees, G., 2010. Human parietal cortex structure predicts individual
differences in perceptual rivalry. Curr. Biol. 20, 1626–1630.
Kim, C.Y., Blake, R., 2005. Psychophysical magic: rendering the visible ‘invisible’.
Trends Cogn. Sci. 9, 381–388.
Kleinschmidt, A., Buchel, C., Zeki, S., Frackowiak, R.S., 1998. Human brain activity
during spontaneously reversing perception of ambiguous ﬁgures. Proc. Biol. Sci.
265, 2427–2433.
Koch, C., 2004. The quest for consciousness. Roberts & Co, Englewood.
Koivisto, M., Lahteenmaki, M., Sorensen, T.A., Vangkilde, S., Overgaard, M., Revonsuo,
A., 2008. The earliest electrophysiological correlate of visual awareness? Brain
Cogn. 66, 91–103.
Koivisto, M., Revonsuo, A., 2010. Event-related brain potential correlates of visual
awareness. Neurosci. Biobehav. Rev. 34, 922–934.
Lamme, V.A., 2001. Blindsight: the role of feedforward and feedback corticocortical
connections. Acta Psychol. (Amst.) 107, 209–228.
Lamme, V.A., 2006. Towards a true neural stance on consciousness. Trends Cogn. Sci.
10, 494–501.
Lamy, D., Salti, M., Bar-Haim, Y., 2009. Neural correlates of subjective awareness and
unconscious processing: an ERP study. J. Cogn. Neurosci. 21, 1435–1446.
Lau, H., Rosenthal, D., 2011. Empirical support for higher-order theories of conscious
awareness. Trends Cogn. Sci. 15, 365–373.
Lau, H.C., 2008. A higher order Bayesian decision theory of consciousness. Prog. Brain
Res. 168, 35–48.
Lau, H.C., Passingham, R.E., 2006. Relative blindsight in normal observers and
the neural correlate of visual consciousness. Proc. Natl. Acad. Sci. U.S.A. 103,
18763–18768.
Laureys, S., Tononi, G., 2010. The Neurology of Consciousness: Cognitive Neuroscience
and Neuropathology. Elservier, Amsterdam.
Linkenkaer-Hansen, K., Nikulin, V.V., Palva, S., Ilmoniemi, R.J., Palva, J.M., 2004.
Prestimulus oscillations enhance psychophysical performance in humans. J.
Neurosci. 24, 10186–10190.
Lumer, E.D., Friston, K.J., Rees, G., 1998. Neural correlates of perceptual rivalry in the
human brain. Science 280, 1930–1934.
Mathewson, K.E., Gratton, G., Fabiani, M., Beck, D.M., Ro, T., 2009. To see or not to see:
prestimulus alpha phase predicts visual awareness. J. Neurosci. 29, 2725–2732.
Melloni, L., Molina, C., Pena, M., Torres, D., Singer, W., Rodriguez, E., 2007. Synchronization
of neural activity across cortical areas correlates with conscious
perception. J. Neurosci. 27, 2858–2865.
Melloni, L., Schwiedrzik, C.M., Muller, N., Rodriguez, E., Singer, W., 2011. Expectations
change the signatures and timing of electrophysiological correlates of
perceptual awareness. J. Neurosci. 31, 1386–1396.
Melloni, L., Singer, W., 2010. Distinct characteristics of conscious experience are
met by large scale neuronal synchronization. In: Perry, E., Collerton, D., LeBeau,
F.E.N., Ashton, H. (Eds.), New Horizons in the Neuroscience of Consciousness.
John Benjamins, Amsterdam, pp. 17–28.
Miller, B.T., Vytlacil, J., Fegen, D., Pradhan, S., D’Esposito, M., 2011. The prefrontal
cortex modulates category selectivity in human extrastriate cortex. J. Cogn.
Neurosci. 23, 1–10.
Monto, S., Palva, S., Voipio, J., Palva, J.M., 2008. Very slow EEG ﬂuctuations predict
the dynamics of stimulus detection and oscillation amplitudes in humans. J.
Neurosci. 28, 8268–8272.
Pascual-Leone, A., Walsh, V., 2001. Fast backprojections from the motion to the
primary visual area necessary for visual awareness. Science 292, 510–512.
Pins, D., Ffytche, D., 2003. The neural correlates of conscious vision. Cereb. Cortex
13, 461–474.
746 J. Aru et al. / Neuroscience and Biobehavioral Reviews 36 (2012) 737–746
Pitts, M.A., Britz, J., 2011. Insights from intermittent binocular rivalry and EEG. Front.
Hum. Neurosci. 5, 107.
Pitts, M.A., Martinez, A., Hillyard, S.A., 2010. When and where is binocular rivalry
resolved in the visual cortex? J Vis 10.
Pollen, D.A., 1999. On the neural correlates of visual perception. Cereb. Cortex 9,
4–19.
Postle, B.R., 2009. The hippocampus, memory, and consciousness. In: Laureys, S.,
Tononi, G. (Eds.), The Neurology of Consciousness: Cognitive Neuroscience and
Neuropathology. Academic Press, London, pp. 326–338.
Quiroga, R.Q., Mukamel, R., Isham, E.A., Malach, R., Fried, I., 2008. Human singleneuron
responses at the threshold of conscious recognition. Proc. Natl. Acad.
Sci. U.S.A. 105, 3599–3604.
Rees, G., 2007. Neural correlates of the contents of visual awareness in humans.
Philos. Trans. R. Soc. Lond. B: Biol. Sci. 362, 877–886.
Rees, G., Kreiman, G., Koch, C., 2002. Neural correlates of consciousness in humans.
Nat. Rev. Neurosci. 3, 261–270.
Robertson, L.C., 2003. Binding, spatial attention and perceptual awareness. Nat. Rev.
Neurosci. 4, 93–102.
Romei, V., Brodbeck, V., Michel, C., Amedi, A., Pascual-Leone, A., Thut, G., 2008. Spontaneous
ﬂuctuations in posterior alpha-band EEG activity reﬂect variability in
excitability of human visual areas. Cereb. Cortex 18, 2010–2018.
Rounis, E., Maniscalco, B., Rothwell, J.C., Passingham, R.E., Lau, H., 2010. Theta-burst
transcranial magnetic stimulation to the prefrontal cortex impairs metacognitive
visual awareness. Cogn. Neurosci. 1, 165–175.
Schwiedrzik, C.M., Singer, W., Melloni, L., 2011. Subjective and objective learning
effects dissociate in space and in time. Proc. Natl. Acad. Sci. U.S.A. 108,
4506–4511.
Sergent, C., Baillet, S., Dehaene, S., 2005. Timing of the brain events underlying access
to consciousness during the attentional blink. Nat. Neurosci. 8, 1391–1400.
Seth, A.K., 2009. Functions of consciousness. In: Banks, W.P. (Ed.), Encyclopedia of
Consciousness. Elservier, Amsterdam, pp. 279–293.
Silvanto, J., Rees, G., 2011. What does neural plasticity tell us about role of primary
visual cortex (V1) in visual awareness? Front. Psychol. 2, 6.
Summerﬁeld, C., Egner, T., 2009. Expectation (and attention) in visual cognition.
Trends Cogn. Sci. 13, 403–409.
Thut, G., Nietzel, A., Brandt, S.A., Pascual-Leone, A., 2006. Alpha-band electroencephalographic
activity over occipital cortex indexes visuospatial attention bias
and predicts visual target detection. J. Neurosci. 26, 9494–9502.
Tong, F., 2003. Primary visual cortex and visual awareness. Nat. Rev. Neurosci. 4,
219–229.
Tononi, G., Koch, C., 2008. The neural correlates of consciousness: an update. Ann.
N. Y. Acad. Sci. 1124, 239–261.
Tse, P.U., Martinez-Conde, S., Schlegel, A.A., Macknik, S.L., 2005. Visibility, visual
awareness, and visual masking of simple unattended targets are conﬁned to
areas in the occipital cortex beyond human V1/V2. Proc. Natl. Acad. Sci. U.S.A.
102, 17178–17183.
van Dijk, H., Schoffelen, J.M., Oostenveld, R., Jensen, O., 2008. Prestimulus oscillatory
activity in the alpha band predicts visual discrimination ability. J. Neurosci. 28,
1816–1823.
Worden, M.S., Foxe, J.J., Wang, N., Simpson, G.V., 2000. Anticipatory biasing of
visuospatial attention indexed by retinotopically speciﬁc alpha-band electroencephalography
increases over occipital cortex. J. Neurosci. 20, RC63.
Wyart, V., Tallon-Baudry, C., 2008. Neural dissociation between visual awareness
and spatial attention. J. Neurosci. 28, 2667–2679.
Wyart, V., Tallon-Baudry, C., 2009. How ongoing ﬂuctuations in human visual cortex
predict perceptual awareness: baseline shift versus decision bias. J. Neurosci. 29,
8715–8725.
Zaretskaya, N., Thielscher, A., Logothetis, N.K., Bartels, A., 2010. Disrupting parietal
function prolongs dominance durations in binocular rivalry. Curr. Biol. 20,
2106–2111.
Zeki, S., 2001. Localization and globalization in conscious vision. Annu. Rev. Neurosci.
24, 57–86.