ll

Review
Attention in flux
Anna C. Nobre1,2,* and Freek van Ede3,*
1Department of Experimental Psychology, University of Oxford, Oxford OX2 6GG, UK
2Oxford Centre for Human Brain Activity, Wellcome Centre for Integrative Neuroimaging, Department of Psychiatry, University of Oxford,
Oxford OX3 7JX, UK
3Institute for Brain and Behavior Amsterdam, Department of Experimental and Applied Psychology, Vrije Universiteit Amsterdam, Amsterdam

1081BT, the Netherlands

*Correspondence: [email protected] (A.C.N.), [email protected] (F.v.E.)
https://doi.org/10.1016/j.neuron.2023.02.032

SUMMARY

Selective attention comprises essential infrastructural functions supporting cognition—anticipating,

prioritizing, selecting, routing, integrating, and preparing signals to guide adaptive behavior. Most
studies have examined its consequences, systems, and mechanisms in a static way, but attention is at
the confluence of multiple sources of flux. The world advances, we operate within it, our minds
change, and all resulting signals progress through multiple pathways within the dynamic networks of
our brains. Our aim in this review is to raise awareness of and interest in three important facets of how
timing impacts our understanding of attention. These include the challenges posed to attention by the
timing of neural processing and psychological functions, the opportunities conferred to attention by
various temporal structures in the environment, and how tracking the time courses of neural and behavioral
modulations with continuous measures yields surprising insights into the workings and principles of
attention.

INTRODUCTION CHALLENGES BY TIME

Picking out useful content from the environment and from Given the inescapable importance of timing for attention, it is
memory to guide adaptive behavior is a core prerequisite for puzzling how narrowly the topic has been addressed. One may
all neural and cognitive systems. The functions responsible wonder why.
for this infrastructural task make up ‘‘selective attention’’— Time is strange. Even assuming a well-behaved Newtonian
responsible for anticipating, prioritizing, selecting, routing, conception of sequentially advancing time, various challenges sur-
integrating, and preparing signals from external and internal face immediately. Transduction in the various senses follows
sources to inform the performance of current and upcoming different time courses, resulting in asynchronous stimulation in
tasks.1,2 the brain. Within the senses, constituent features of items are
The primordial temporal dimension frames all our interactions broken down further and processed and integrated through selec-
with the environment. To do its job effectively, attention must tive specializations in networks containing many forward, lateral,
orchestrate the confluence of dynamics from various sources and re-entrant connections. Processing at each stage takes
(Figure 1). As active agents, we move through environments. time, and information flows between stations at different rates de-
The environment, too, is always changing. Our brains process pending on the length, width, and myelination of the fibers. The
the resulting stimulation from our senses and actions over time result is a perplexing temporal dimension to the binding problem.
in dynamic networks. Finally, as cognitive beings, the disposi- Many basic questions remain, such as how to determine when the
tions of our minds—our goals, expectations, memories, and mo- computational transformations within cell assemblies yield infor-
tivations—also shift regularly and strongly influence how sense mative signals and how to piece together the resulting asynchro-
turns to experience. nous signals originating from a common event within and across
The multiple sources of flux pose extraordinary challenges modalities, while also separating relevant and competing events.
to attention. We turn to these challenges in the first part of Thus, it is baffling enough to consider how the functional archi-
this review. Concomitantly, recurring temporal structures tecture of the brain manages to build a cohesive construct from
ground predictions and bring opportunities for anticipating even one single static multisensory source in the environment.
and prioritizing incoming relevant events. We turn to these The problem is greatly amplified by considering the dynamics
opportunities in the second part of this review. Finally, in the in the external environment, our active agency through it, and
third part, we reflect on how methods that enable tracking the mutability of mental dispositions and contents. To find a
behavior and neural activity continuously have proven essen- way through these challenges, theoreticians and researchers
tial for revealing the temporal choreography of attention have considered ways of reducing the mind-boggling complexity
functions. resulting from the multitude of simultaneous temporal fluxes.

Neuron 111, April 5, 2023 ª 2023 Elsevier Inc. 971

 ll
Review

Figure 1. A confluence of dynamic sources

Selective attention functions must maneuver, orchestrate, and capitalize on temporal dynamics in the world, agent, brain, and mind.

Parcellating continuous time into brief sequential chunks remain far from explicating the multiple concurrent timescales
could simplify the problem by providing sufficient periods for governing active cognition up to several seconds and more.
sampling and integrating signals.3–6 Along these lines, some re- Given its scope and sophisticated methodology, the attention
searchers propose that oscillatory properties of sensory systems research field should embrace the temporal dimension more
play an important role in pacing perceptual analysis, such as the firmly. By manipulating predictions and goal relevance of stimuli
prevalent alpha oscillations in visual circuits.7,8 Rhythmic proper- in static sensory displays, attention research has produced
ties of circuits have also been suggested as fundamental for impressive insights into the myriad of cell-, circuit-, and
coordinating the communication required to integrate and route network-level modulatory mechanisms that can operate from
features of relevant stimuli.9 Another related area of research ex- early sensory to later motor stages of information processing in
plores the behavioral manifestations of the potential rhythmic the service of adaptive behavior.20–22 Newer studies manipu-
pickup of sensory information.10–12 According to the active- lating predictions and goals concerning memoranda in simple
sensing perspective, perception is an active exploration of the working-memory displays have extended the enquiry to modula-
sensory world, paced by the natural rhythms of motor systems tory mechanisms operating on internal representations.23,24
gating the sensory sampling (e.g., eye movements, whisking).13 Many sources of modulatory signals have also been identified,
This notion of strong and mutually reinforcing loops between including automatic consequences of differences in sensory
perception and action has rich roots.14 salience, top-down consequences of current task goals, and in-
Alternatively, or in addition, different temporal windows over fluence of memories of various types and time scales.25
which neurons within and between areas integrate signals The fundamental next step is to situate attention tasks in an
may help coordinate neural processing by nesting faster local ecologically valid temporal context—encompassing dynamics
computations within slower, higher-level ones.15–17 At the and temporal structures, both in the environment and in the
network level, the sweep of re-entrant feedback activity has agent. Making headway on how the brain focuses on useful stim-
also been proposed to facilitate integration and to mark a ulation to guide behavior in this dynamic way requires using
separate stage of processing from the faster preceding bot- powerful continuous measures of brain activity, as well as contin-
tom-up sweep.18,19 uous readouts of behavior, to track the degree and timing of con-
Yet, we are far from a consistent set of observations and mutu- tent processing at various levels of organization. Studies on the
ally informed lines of investigation that can generate unifying temporal orienting of attention, or temporal expectations,26
theoretical proposals. For example, reconciling findings from take the first step in this effort, but there is further to travel.
the observed cyclic variation of performance at the alpha range8
and rhythmic sampling at various estimates in the theta ATTENTION IN TIME
range10–12 is not straightforward. How the timings of neuronal
computation vs. communication relate and influence behavior Cast of timings
is also still unclear. Even considering longer time constants of Before considering how attention research can embrace timing,
high-level associative networks and feedback sweeps, we brief comments on some pertinent facets of timing may help.

972 Neuron 111, April 5, 2023

 ll
Review

An important and useful distinction exists between explicit and ring targets in the same sensory modality,39 identifying a target
implicit timing,27 which is often overlooked or confused. The occurring shortly after another in a rapidly presented stream
distinction between explicit and implicit timing is about the (referred to as the attentional blink),40,41 noticing repeated items
objective of the task and is independent of whether participants occurring in close temporal proximity (repetition blindness),42
are aware of temporal parameters in the task. Explicit timing re- and performing different tasks in rapid succession (psychologi-
fers to situations in which timing is the primary objective of a task cal refractory period).43 The precise explanations for each of
and needs to be explicitly reported. Explicit timing tasks include these well-documented phenomena are still being debated,
reproducing a stimulus duration—for example, using a manual but they share a common thread of proposing bottlenecks in
response—or estimating a stimulus duration, for example, by specific stages of processing due to ongoing processing
judging its temporal midpoint or comparing its duration to a pre- impeding other simultaneous analyses. Studies containing
vious stimulus. In contrast, implicit timing tasks utilize temporal rapidly presented stimuli have also highlighted confusability be-
processing but do so for purposes other than timing itself. No tween temporally adjacent targets and preceding or succeeding
awareness or explicit report of timing is required, but the pro- distractor items (distractor intrusion)44–47 as well as the incorrect
cessing of timing impacts performance, nevertheless. Temporal combination of features between items (illusory conjunctions in
attention-orienting tasks are a typical example of implicit timing the temporal domain).48 The swapping and meshing of tempo-
tasks in which predictive temporal intervals lead to improve- rally proximal stimuli suggest a temporal buffer for active
ments in performance in perceiving or responding to target processing as well as a limitation within which it is difficult to indi-
stimuli. viduate and select target items. These phenomena are often
Explicit and implicit timing tasks both necessitate clocking considered properties of attention. Technically, however, they
mechanisms, but these need not be the same.27 The brain sys- are better understood as inherent limitations of perceptual and
tems and mechanisms supporting explicit temporal judgments cognitive systems and are not part of selective attention in the
may overlap only partially—or not at all—with those that regulate sense of functions that prioritize and select signals to guide
the impact of temporal durations on sensory processing, atten- adaptive behavior.
tion, decision-making, or action. To understand the relationship
between attention and timing, the brain systems and mecha- Varieties of temporal expectations
nisms that support implicit timing are of the essence. As a counterpoint, in addition to challenging the brain, the contin-
How the brain keeps time is of growing research interest. In our uous dynamic influx of stimulation from the environment also pro-
opinion, the field is still wide open, and many possibilities exist. vides abundant predictive information that is useful for guiding
The first and remaining question concerns the functional anat- behavior. The burgeoning field of temporal orienting of attention,
omy of timing. Proposals vary from dedicated internal clocks28 or temporal expectation, shows how various types of predictable
to fully distributed timing properties intrinsic to neural systems.29 temporal structures in the environment support the selective tem-
Evidence for the local coding of timing within various specialized poral anticipation, prioritization, processing, and action-prepara-
circuits is strong, but these do not negate the existence of one or tion of task-relevant stimuli.26,49 Temporal orienting builds on
more general timing systems. Local timing computations could the rich and longstanding foreperiod literature that revealed sys-
benefit from circuits that coordinate timing functions across sen- tematic variations in response times according to the variability,
sory, associative, and action systems. These general circuits length, and sequence of intervals (foreperiods) preceding an
need not be entirely dedicated to timing but could instead imperative stimulus (see Niemi and Na €a€ta
€n50 and Los51).
perform important functions for organizing (scaffolding) neural Temporal-orienting tasks add the element of selective atten-
activity in the service of behavior that involves high temporal res- tion by specifically manipulating predictions or goals concerning
olution, such as sensorimotor integration and motor preparation the temporal intervals of task-relevant stimuli. Benefits in perfor-
(for discussion see Muller and Nobre30). mance occur for target stimuli occurring at intervals predicted or
Complementing the localization question is the quest for instructed by informative temporal cues (e.g., Coull and Nobre,52
timing mechanisms. A range of possible cellular and systems- Denison et al.,53 and Griffin et al.54) or by learned temporal asso-
level mechanisms may operate and differ in their contribution ciations (e.g., Cravo et al.,55 Los et al,56 Zimmermann et al.,57
at different time scales. For example, coding for interval dura- and Olson and Chun58). Benefits from specific learned associa-
tions has been linked to: temporal tuning of receptive field prop- tions extend to incidentally learned sequences of temporal inter-
erties of neurons in sensory cortices31 and the hippocampus32; vals, at least when spatial or action predictions are also available
firing rates of neurons in sensory,33 sensorimotor,34 and motor35 (e.g., O’Reilly et al.59 and Heideman et al.60). In the extreme,
neurons; prediction-error signals within reward circuits36; delay- items appearing rhythmically at fixed intervals have a strong
line circuits in the cerebellum37; oscillations in the basal advantage over those appearing at variable intervals (e.g., Riess
ganglia28; neural sequences38; and short-term synaptic plasticity Jones61 and Jones62). Temporal expectations can also be
reinforcing states within dynamical neuronal circuits.29 derived from the probability of an event occurring at a given
time point within a context, given that it has not occurred yet.
Temporal limitations Performance advantages ebb and flow in tandem with changes
The temporally extended nature of neural processing constrains in the probability function for the target to appear (e.g., Ghose
information processing and results in various refractory phenom- and Maunsell,33 Janssen and Shadlen,34 Cravo et al.,63 and
ena limiting perception and performance over short intervals. Ex- Grabenhorst et al.64). Other, more complex, types of predictive
amples include deficits in identifying two simultaneously occur- or recurring temporal patterns may also facilitate behavior. For

Neuron 111, April 5, 2023 973

 ll
Review

example, benefits are observed when intervals separating ments.78–80 To date, however, as far as we know, studies that
events in a stream increase or decrease systematically in a pre- consider dynamic temporal structures in both the agent and
dictable fashion.65 the environment are still largely missing.
The various types of commonly studied predictive temporal
structures—associative cues, sequences, probabilities, and Behavioral consequences of temporal expectations in
rhythms—are not mutually exclusive (see Nobre and van Ede26 perception and action
and Nobre and Rohenkohl49). The extent to which they can How temporal expectations impact performance depends not
interact, or act independently, deserves further investigation. only on the type of predictive structure but also on the stimulus
Functional dissociations have been noted between benefits from properties and response demands of a given task.26 Temporal
rhythms and cues (e.g., Breska and Ivry,66 Breska and Deouell,67 expectations driven by a particular type of predictive structure
and Rohenkohl et al.68). For example, participants with basal can influence multiple stages of processing across sensory,
ganglia damage show a selective impairment in benefiting from sensorimotor, and motor areas. The resulting behavioral conse-
rhythms, whereas participants with cerebellum damage show a quences are multifaceted and heavily depend on factors such as
selective impairment in benefiting from cued temporal associa- the strength of sensory stimuli, degree of sensory competition,
tions66 (Figure 2A). Strong interactions have been highlighted be- conflict among sensorimotor mappings, and emphasis on accu-
tween the effects of predictive cues and probabilities, such that racy or response time. For example, cue-based temporal expec-
cueing benefits are stronger when the temporal probability is tations about relevant targets in tasks that emphasize accuracy
otherwise low or moderate.69 Furthermore, the effects of contin- under demanding sensory conditions often yield improvements
uous temporal probability have been shown to be separable and in perceptual sensitivity and accuracy-related measures. In
independent from the discrete probability of the imperative stim- contrast, temporal expectations in tasks that emphasize
ulus happening at all during a trial.70 Mapping the full set of relation- speeded responses with little sensory challenge mainly impact
ships among all temporal structures is still wanting, and their na- response times (see e.g., van Ede et al.71 and Zokaei et al.81).
tures may also depend on other task variables and demands. A representative example is depicted in Figure 2B, in which the
Interestingly, despite the continuous and dynamic nature of same temporal-cueing procedure was applied in two tasks,
time, most temporal-expectation tasks use simple discrete trial characterized either by predominantly visual demands (tilt iden-
designs. Much of our knowledge so far, therefore, comes from tification) or motor demands (speeded response).71 Temporal
measuring performance in trials with single imperative stimuli expectations in the two tasks did not only drive (latch onto)
that vary in their temporal predictability or relevance. Notable ex- distinct neural modulations but also had distinct consequences
ceptions include continuous performance tasks manipulating on performance: affecting predominantly accuracy in the visual
the temporal sequence,59,60 learned probability,72,73 or rhythm task but reaction time in the motor task.
of targets.74 Interestingly, such results also demonstrate that Temporal expectations have been observed to influence several
learning of temporal associations can transcend simple one- psychological functions and the list continues to grow (Figure 3 for
to-one timing relations between successive stimuli. Controlling a non-exhaustive overview). Examples include: improving
for all applicable experimental variables in extended dynamic visual discriminability,82 enhancing visual contrast sensitivity,83,84
tasks can be challenging. For example, several possible tempo- increasing processing speed,85 modulating non-decision pro-
ral structures can combine in tasks that manipulate the temporal cesses,86 expediting the onset of decision-related processes,87
probabilities of the targets in addition to their recurring temporal facilitating multisensory coupling,88 lowering motor-response
probabilities: the likelihood that a target will occur, sequential ef- thresholds,89,90 offsetting distraction,47,91 diminishing the atten-
fects from temporal priming by the previous interval, the time tional blink,92,93 increasing fixational stability,94,95 focusing encod-
since the last response, and the passage of time within a trial ing into working memory,96,97 improving encoding into long-term
(e.g., Shalev and Nobre74 and Salet et al.75). Facing the resulting memory,98,99 and enabling phasic modulation of arousal.74,100
complexity is a necessary step for discovering the workings of Distilling the principles regulating the conditions for the
attention in its natural dynamic context, but it is not the only one. expression of given effects still lies ahead. There are multiple var-
The missing final step is to incorporate the natural dynamics of iables to consider. The nature of the temporal structure is likely to
the agent. Participants in typical laboratory studies remain facing contribute. For example, the effects of rhythmic expectations
a display, usually with their eyes fixed, limiting their movements appear to be stronger and more automatic than those of cues
to delivering responses. Adding natural movement greatly com- (e.g., Breska and Deouell67 and Rohenkohl et al.68). The sensory
plicates experimental control and is simply not possible with quality of stimulation also matters. In addition to the amount of
most non-invasive methods for measuring brain activity. This sensory difficulty or competition, the sensory modality can also
may soon change. Advances in virtual-reality technology are play a role (see Nobre101). The different senses likely evolved
paving the way to rigorous investigations in immersive set- for detecting different types of information in the environment.
tings.76 Tracking eye, head, hand, and body motion with good For example, the auditory system is intrinsically temporal and
resolution is already possible and can bring new insights to has a much higher temporal resolution than the visual system,
attention (e.g., Draschkow et al.77). The increasing power and which in turn has a better spatial resolution.102,103 The source
portability of human neurophysiology methods (portable electro- and strength of the temporal prediction are also worth consid-
encephalogram [EEG] and optically pumped magnetometers ering. Learned predictive structures may have long or short life-
[OPMs]) should soon enable rigorous neuroscientific investiga- times and differ in their precision and reliability. The purpose of
tions while active participants interact with dynamic environ- the task, too, is likely to be crucial.71 For example, tasks can

974 Neuron 111, April 5, 2023

 ll
Review

Figure 2. Mechanisms and consequences of temporal expectations depend on temporal structures and task demands
(A) Performance benefits of rhythms and single-interval cued associations rely on distinct brain circuits. Participants with basal ganglia damage are selectively
impaired in benefiting from predictable rhythms, while participants with cerebellum damage are selectively impaired in benefiting from single-interval cued
temporal associations. Error bars denote ±1 SEM. Data from Breska and Ivry66.
(B) Performance benefits and neural modulations by cued temporal expectations depend on task demands. When cueing the timing of an anticipated visual task
(tilt identification at the left or right of fixation), temporal expectations shape neural modulations in the visual cortex and predominantly affect accuracy. When the
same cues inform about the timing of an anticipated motor task (speeded button press with the left or right index finger), temporal expectations shape neural
modulations in motor cortex and predominantly affect reaction time. Error bars denote ±1 SEM. Data from van Ede et al71.

emphasize perceptual sensitivity, learning, decision-making, ac- Behavioral consequences of temporal expectations in
curate deliberation, or speed of responding. The precise nature working memory
of the task goal determines the critical, limiting stage(s) of neural The consequences of temporal expectation go beyond the
processing informing performance. Hence, different task goals immediate sensory-motor arc, extending to internal representa-
may tap into the different types of temporal modulatory mecha- tions within working memory (see van Ede and Nobre23). In work-
nisms within different stages of processing. ing-memory tasks, significant performance benefits result from

Neuron 111, April 5, 2023 975

 ll
Review
Figure 3. Dimensions of temporal attention
Temporal expectations can be driven by multiple
temporal structures, act through multiple stages of
processing, be supported by multiple neural
mechanisms, and result in multiple behavioral
consequences. Entries in this schematic are not
intended to be mutually exclusive nor exhaustive.

a long interval. Errors and response times

for reproducing the probed orientation
diminished according to the predicted
probe time. Furthermore, item prioritiza-
tion shifted from favoring the likely-early
item to favoring the likely-late item
purely based on internal states tracking
the passage of time. Thus, attention can
be allocated reversibly within memory
contents without notable trade-offs be-
tween the information available between
items. Similar behavioral benefits were
observed for another task employing a
similar arrangement and logic. In this
case, left and right items had complemen-
tary probabilities of being probed after a
short or long interval.108 Items were light
and dark orientation gratings. Probes
indicated the side of the item for repro-
ducing its orientation. Response preci-
informative cues presented during the delay period that retroac- sion and response times improved at the predicted probe time,
tively predict which stimulus is likely to be goal-relevant (retro- favoring first the likely-early and then the likely-late item. Inter-
cues) (for reviews see van Ede and Nobre23 and Souza and estingly, the internal focus of attention could also be tracked
Oberauer104). When retrocues also indicate the order in which by the pupillary response. The size of the pupil adjusted accord-
an orientation item from the memory array will be probed, bene- ing to the brightness of the attended stimulus within the internal
fits are temporally specific.105 In the absence of order informa- memory array, even though stimulus brightness was irrelevant
tion, prioritizing one item leads to biases in the report of the for task performance and no change in luminance occurred dur-
orientation of other (uncued) items. When participants are ing the working-memory delay. The findings reinforce the flexible
probed on an uncued item first, they report a distorted orienta- reversibility of attention in working memory according to internal
tion away from the retrocued item. However, if participants states of temporal expectation. They also suggest embodied
know that the cued item will be probed last, no biasing of orien- mechanisms for attention in working memory that may have
tations occurs during the first probing interval. The findings sug- evolved for anticipating interactions with events in the environ-
gest the temporally specific prioritization of items. More direct ment. In this laboratory task, no other change in luminance
demonstrations of temporally specific prioritization of contents was involved during the delay or probe phase but, in everyday
in working memory come from studies manipulating the tempo- situations, we often retain items in working memory to interact
ral probability of items being probed at certain intervals, in the with them shortly (see van Ede and Nobre23 and Draschkow
absence of retrocues. In a change-detection task using two in- et al.109).
tervals for comparing the sample and probe arrays, both accu- Working-memory tasks incorporating interfering events have
racy and response times improved at the short interval when further broadened the set of modulatory mechanisms by tempo-
this duration was more probable within the block,106 akin to ral expectations. Interference during a working-memory delay
what is typically observed in perceptual tasks.26 typically impairs performance, whether it is just a distracting
Tasks manipulating the evolving temporal probability for prob- stimulus to be ignored or a secondary task that needs to be per-
ing different contents of a memory array have revealed striking formed in the interim. Temporal expectation about the onset of
flexibility in the dynamic prioritization of internal contents. For the interfering distracting or interrupting event significantly re-
example, using two-item arrays, we varied the probability of duces the degree of this interference and suggests shielding of
each item being probed early vs. late. Items were colored ori- the memory contents.91 After encoding two orientation stimuli,
ented bars, and the probe color prompted the report of the item’s an interfering tilted stimulus occurred at fixed or variable tempo-
orientation.107 One color consistently had a high probability ral intervals, which could be ignored (distraction) or required a
(80%) of being probed after a short interval and the other after simple leftward/rightward response (interruption). At the end of

976 Neuron 111, April 5, 2023

 ll
Review

the trial, a color prompted the reproduction of the corresponding strengthened analysis of the target orientation, with little effect
orientation. Temporal expectation about the interfering stimulus on distractor processing. As previously mentioned, temporal
counteracted its disruptive effects on reproducing the orienta- expectation can also shield working-memory representations
tion from memory. The protective effects occurred for both types from interference,91 though the neural counterparts of the effect
of interference and, furthermore, also improved response times have not yet been revealed.
in the interruption task itself. The effects, therefore, show that Predictive rhythmic structures with fixed durations can entrain
temporal expectations do more than dampen the processing natural endogenous slow brain rhythms in the delta and theta
of interfering sensory stimuli because the improved performance bands112,113 (cf. Breska and Deouell114 and Haegens and Zion
of the interruption task confirms the ability to engage flexibly with Golumbic115). Excitability for sensory stimuli varies with the
interfering events when necessary. Instead, the findings show phase of the entrained neural rhythm.84 Rhythmic stimulation in
that temporal expectations can additionally shield working- one modality can also entrain excitability in other modal-
memory contents from sensory stimuli.91 In a related follow-up ities.88,116 For example, a rhythmic-frequency-modulated sound
study, the timing of the interruption task was held constant, while resulted in related changes in the discrimination of a brief visual
temporal expectation about the working-memory report was stimulus, depending on its timing relative to the phase of the
manipulated. Temporal expectation about the working-memory auditory rhythm. Neural entrainment occurred in both auditory
report diminished both errors and response times for the orien- and visual cortices.88
tation reproduction. In addition, performance on the interruption In addition to bringing new mechanisms to the table, temporal
task also improved, suggesting the ability of temporal expecta- expectations can potentiate the effects of other attention-related
tion to organize the setting of priorities across multiple tasks.110 modulatory mechanisms. The first demonstration came in a task
that manipulated the predictable location and/or interval for a vi-
Neural mechanisms of temporal expectations sual target to reappear after moving across the display in
When hypothesizing about the mechanisms through which discrete steps and disappearing behind an occluding band.111
temporal structures influence neural processing, two strong The amplitude of the early visual P1 potential was increased by
temptations are best avoided. The first is to look for THE mech- spatial expectation—when the circle moved along a linear
anism. Gone are the days of pinpointing the singular bottleneck compared with a meandering trajectory. This increase in P1
of capacity limitation in the brain or arbitrating between simple was greatly amplified by temporal expectation—when the target
dichotomies regarding the locus of attention.1 Instead, as we moved in fixed rhythmic steps compared with variable-duration
have learned from the rich literature on visual spatial attention, steps. The strong synergy between temporal and spatial expec-
changes in many cellular, ensemble, and communication param- tations was not reflected in the behavioral performance mea-
eters can contribute. The same applies to effects of selective sures, probably because perceptual demands were not a limiting
temporal expectations (Figure 3 for a non-exhaustive overview). factor for performance. A follow-up behavioral study designed to
The second trap is to confine the search to the known modu- stress perceptual demands did instead reveal the hypothesized
latory mechanisms for spatial, object-based, and feature-based synergy between temporal and spatial expectations at the
attention. It is instinctive to draw from the repertoire of familiar behavioral outcome level.117
mechanisms, such as increased firing rates related to antici- The interaction between temporal and other expectations also
pated stimuli, filtering of irrelevant stimuli competing for process- plays out during attention-related anticipation. In the occlusion
ing, different types of gain control (to prioritize coding of relevant task, the alpha-band marker of visual excitability follows the
vs. irrelevant stimulation), changes in the intrinsic and correlated cadence of the preceding predictable rhythm during the occlu-
noise of neurons, and synchronization of neuronal activity (see sion period in anticipation of the target re-appearance.118 In tem-
Maunsell,20 Moore and Zirnsak,21 and Buschman and Kast- poral-cueing tasks, cues combining temporal and spatial expec-
ner22). Yet, incorporating timing into the equation should also tation pace oscillatory markers of visual71,119 (see Figure 2B,
prompt us to consider additional ways in which the brain may visual task) and somatosensory120 excitability. When temporally
anticipate, prioritize, and gate events occurring at temporally predictive cues indicate the response hand required to report a
predicted or relevant moments. simple central stimulus in a speeded reaction-time task, they
Human studies instead pace the timing of beta-band activity related to motor
Research in humans using neurophysiology has revealed that preparation71 (see Figure 2B, motor task). Together, the findings
spatial vs. temporal expectation can yield similar patterns of reinforce the notion that predictive temporal information can
behavioral improvement through divergent neural effects.54,111 work through other attention mechanisms related to different
In addition to changing the strength of neural processes, as types of information. They also highlight the plurality of mecha-
often observed during spatial attention, the timing of neural pro- nisms and systems through which temporal expectations can
cesses can also be affected. Temporal expectations can result in improve behavior (Figure 3).
faster progression through successive stages of processing.101 Studies in animal models
Changes in the duration of processing stages also apply to Studies of temporal expectation in animal models are still relatively
protecting relevant information from distraction. Temporal prep- few, but they vindicate the plurality of possible sites and mecha-
aration for a target orientation improved performance when the nisms of modulation. Investigations have probed brain areas
target was shortly followed by another interfering orientation.47 and cellular mechanisms during visual, auditory, and reward pro-
Multivariate decoding of target and distractor orientations re- cessing in non-human primates and rodents. Most studies have
vealed that temporal preparation led to prolonged as well as used temporal probability manipulations,33,34,121,122 but others

Neuron 111, April 5, 2023 977

 ll
Review

have relied on specific learned temporal associations between TIME IN ATTENTION

stimuli,36,123,124 temporally structured sequences,125 and
rhythms.126,127 Neural changes following temporal expectations So far, we have reflected on how the temporal dimension both
have been observed in the motor cortex35; sensorimotor regions, challenges and guides attention. In this final section, we consider
such as in posterior parietal areas34; and subcortical and cortical how studying the temporal progression in neural and behavioral
regions linked to reward.36,125 Modulatory effects are also preva- measures during attention control and modulation can illuminate
lent in sensory areas, such as the auditory,121,128 somatosen- the workings of these core scaffolding functions (cf. van Was-
sory,129 and visual33,122–124,130,131 cortices. A study inactivating senho et al.133). By using continuous measures to examine the
A1 in rodents suggests that sensory areas may be causally timings of unfolding neural activity and behavioral outputs,
involved in generating or expressing temporal expectations.121 some surprising insights are beginning to emerge.
In addition, the hippocampus was shown to be necessary to
mediate temporal-expectation effects in the visual cortex during Time course and stages of attention
sequence learning,130 and dopamine modulation in prefrontal Since the earliest investigations, scientists have naturally
and brainstem areas was causally implicated in temporal expec- wondered about the time course of attention. One prominent
tation during simple reaction-time tasks.125 The emerging picture question has been: how long does it take to shift attention? Early
is thus one of the widely distributed temporal-expectation effects behavioral studies systematically manipulated the time between
in which heteromodal and limbic regions may further play a prom- predictive attention cues and sensory targets, discovering that
inent role in coordinating effects in specialized sensory and motor hundreds of milliseconds elapse until full behavioral benefits
regions.132 from voluntary goal-directed shifts of attention (e.g., Posner
Relevant cellular changes noted so far include changes in the et al.134 and Mu €ller and Rabbitt135). In contrast, automatic
strength and timing of neuronal firing and local field poten- capture of attention follows a distinct temporal profile with a
tials,33–36,121–125,128–131 improvements in signal-to-noise ra- bi-phasic pattern of early facilitation followed by a slower inhibi-
tios,129 changes in higher-frequency oscillatory signals,35,123,124 tion-of-return.136 Interestingly, the answer to the time course of
and strengthening of the spike-field coherence.123 Within rhyth- attention may depend on which behavioral outcome measure
mic contexts, relevant stimulus streams can entrain intrinsic low- you use for interrogation. For example, eye-movement measures
frequency oscillations.126 Phase alignment within sensory areas can reveal periods in which attention is neither driven by goals
can act crossmodally to impact neural excitability in other mo- nor by saliency,137 and attentional benefits in accuracy and
dalities126 and can help to differentiate target vs. distracting reaction times can develop at different rates and match different
stimuli in different streams across the tonotopic map in the neural modulatory mechanisms.138,139
auditory cortex, according to phase alignment.127 Another sticky question has been: how early does attention act?
A long-lasting debate was based heavily on how sensory and se-
Practical and theoretical implications mantic manipulations changed performance in dichotic listening
It is worth noting that temporal structure exists in every attention tasks.140 In such tasks, participants hear two competing auditory
experiment. Inevitably, choices are made about the durations of streams and attend to one to repeat the words as they come
stimuli, the intervals separating them, the responding window, through. Interestingly, here too, the answer could change depend-
and so on. Over trials, these result in temporal probability func- ing on the behavioral outcome measures, for example, varying be-
tions about stimulus onsets and offsets and response deadlines. tween explicit reports and automatic responses concerning the
In turn, temporal predictions may influence the strength, latency, contents in the target vs. distracting stream (see Nobre1).
and duration of processing at many different stages, including Neurophysiology methods, with their high temporal resolution,
sensory, motor, and mnemonic. Understanding how temporal can track the workings of selective attention continuously
orienting influences neural processing is therefore indispensable through modulation of neural activity.141 They made it possible
for clear interpretations concerning many other variables of inter- to map the stages of attention that result in behavioral advan-
est that may be under investigation. tages. The first event-related-potential (ERP) investigations in
Acknowledging the temporal limitations in neural systems humans tested when attention acted, providing definitive evi-
as well as the role of temporal structuring of attention in dence for early sensory modulations.142,143 More subtly, ERP
turn poses major challenges and constraints for theories con- studies also indicated that multiple stages could be affected in
cerning the routing, coordination, or integration of brain sig- functionally dissociable ways. For example, successive early
nals at the network level. Such theories need to account for P1 and N1 visual potentials were sensitive to different experi-
endogenous rhythms possibly pacing the sampling of sensory mental manipulations, such as having to inhibit distractors of
stimuli, salient events possibly resetting neural activity, discriminate target features.144 In animal models, studies have
external rhythmic stimulation pushing around endogenous identified attention-related modulation across the visual hierar-
rhythms, temporal predictions changing the timing of process- chy and charted their respective timings145,146 (see also Micha-
ing stages, temporal expectations interacting with other lareas147). Complementary studies examined the sites and tim-
biasing signals, and more. Then, theories must be placed in ings of attentional control signals (e.g., Buschman and Miller148
their actual dynamic setting, in which incoming stimulation and Siegel149). Such investigations have helped delineate the
from the changing environment is streaming in through non- ‘‘stages’’ and ‘‘networks’’ of attention. They have reinforced
synchronized modality channels while the agent engages in how many sites can undergo attention modulation and revealed
active behavior. that relative timing in areas controlling attention depends on the

978 Neuron 111, April 5, 2023

 ll
Review
Figure 4. Temporal profiles in neural and
behavioral measurements can reveal
insights into the principles of attention, such
as parallel processing in internal selective
attention
Intuited vs. observed temporal profiles in the four
studies described in the text. Data in (A) from van
Ede et al.150 Data in (B) from Boettcher et al.151
Data in (C) from van Ede et al.152 Data in (D) from
van Ede et al.153 Data in (A) and (B) consist of
continuous EEG recordings, while data in (C) and
(D) consist of continuous behavioral gaze-position
measurements.

and task demands (see Maunsell,20

Moore and Zirnsak,21 and Buschman
and Kastner22).

Informative temporal profiles of

attention modulations
Tracing the timing of attentional modula-
tions can yield knowledge beyond just
the staging of attention effects. As
methods advance for revealing the con-
tents of information processing and
measuring behavioral outcomes with
high temporal resolution, surprising find-
ings lead to novel insights and fresh theo-
rizing. In what follows, we highlight four
recent examples from our own work on in-
ternal selective attention. This is intended
to bring to life how the temporal profiles of
the modulatory effects of attention can
shed new light on our understanding of
core attention functions—such as the
parallel workings of attentional functions
inside working memory (Figure 4).
Visual and motor selection unfold
concurrently in working memory
The timing of sensory and motor neural
markers in the study described in van
source driving attention orienting. For example, changes in pos- Ede et al.150 (Figure 4A) revealed that selecting motor action
terior parietal activity lead when sensory salience drives atten- plans occurs concurrently with, and not after, selecting the visual
tion to a visual display, but changes in frontal activity lead attributes of memoranda in working memory. Participants main-
when working memory drives attention.148 Invasive recordings tained two peripheral visual oriented bars in working memory
studies can further shed light on the extent to which attention- that were each linked to a different manual action (Figure 4A,
related mechanisms are shared across different situations, for left). We were interested in the cascade of the selection of rele-
example, when directing attention externally to the sensory vant visual and motor memory attributes at the end of the mem-
world or internally to the contents of working memory (e.g., Pan- ory delay, when one of the two memoranda became relevant for
ichello and Buschman24). behavior.
Understanding the various stages of attentional modulations The intuitive model is that participants would maintain the
in the brain is not merely a matter of cataloging. It is critical sensorial visual representations of both items and that, when
for understanding how distinct behavioral outputs of attention probed, they would first select the visual information and only
come about, such as the processing stages that contribute to then derive the motor plan for the appropriate action
changes in detection thresholds, distractor filtering, or response (Figure 4A, middle). Instead, we observed concurrent selection
speeds. Understanding the relationship between brain net- of both the visual attribute and the appropriate action plan
works, modulatory mechanisms, and behavioral outcomes (Figure 4A, right). The lesson was that participants encode and
can help draw out the principles for attentional effects maintain more than just the visual attributes during delay. They
across different situations, depending on stimulus parameters also prepare potential action plans for each memorandum, and

Neuron 111, April 5, 2023 979

 ll
Review

these are ready to go when either memorandum becomes ciated with directional biases in microsaccades. We previously
selected for behavior. The implication is a much richer concep- demonstrated that these gaze biases provide a sensitive and
tualization of the nature of visual working-memory representa- continuous behavioral read-out of selective attention within the
tions, in which sensory items are poised for action, highlighting spatial layout of in working memory156 (see also Liu et al.157).
the future-facing and pragmatic, behavior-enabling role of work- Participants held two peripheral, colored, oriented bars in work-
ing memory.23 ing memory until a cue indicated which item was required for
Upon reflection, our sensory-then-motor serial intuition likely behavior. We independently manipulated voluntary and involun-
stemmed from projecting what happens during perception to tary sources of internal attention by manipulating whether the
working memory. When processing incoming stimulation, action cues were informative or not (voluntary factor) and whether the
plans necessarily depend upon, and therefore follow, the sen- cues matched the color of either the relevant or the irrelevant
sory attributes. In contrast, once visual information has been en- item in working memory (involuntary factor) (Figure 4C, left).
coded in working memory, plans for potential actions can be Based on classic spatial orienting studies in perception,134,135
assembled and held available for joint visual-motor selection, we imagined that involuntary shifts would occur quickly and that
whenever memory contents become relevant for behavior. voluntary shifts would lag behind (Figure 4C, middle). Instead,
Anticipation in working memory is bi-phasic the onset of both voluntary and involuntary shifts of attention
The timing of anticipatory neural modulations in the study occurred early and overlapped (Figure 4C, right). Furthermore,
described in Boettcher et al.151 (Figure 4B) revealed how plans their effects were additive, suggesting that they acted indepen-
for future actions are formed surprisingly early and made dently and in tandem.
clear how signals related to anticipation can follow a dynamic, From this, we learned that multiple attention operations can
bi-phasic profile. When anticipating delivering a response, the run concurrently. Previously, we observed how attention can
intuition is that motor preparation builds gradually toward the influence different types of content at the same time in finding
anticipated moment (Figure 4B, middle), as is also typical of concurrent sensory and motor selection.150 The current study
other signatures of anticipation such as the contingent negative additionally shows how multiple sources of attention control
variation (CNV; e.g., Cravo et al.,55 Jin et al.,106 and Miniussi can modulate the same memory content in tandem. Together,
et al.154) or the ramping of neuronal firing rates.34 the findings point to a highly parallel nature of attention opera-
In the study, participants selected and encoded visual con- tions. These results also present yet another example of how
tents into working memory that would become relevant several classic findings from the external-attention literature do not al-
seconds later (Figure 4B, left). As in the study described above, ways translate straightforwardly to internally directed attention.
visual contents were linked to specific manual actions. We were Attention looks ahead in working memory during
interested in when action plans would be formed during the sequential behavior
working-memory delay. The pattern of action anticipation that Our final example (Figure 4D), described in van Ede et al.,153 re-
we observed did not build gradually. Rather, it was bi-phasic vealed that the focus of attention in working memory moves onto
(Figure 4B, right). Early action planning overlapped with the next relevant content before responding to the currently rele-
attentional selection of visual items during memory encoding vant content is complete. Building on our earlier demonstration
(putatively linked to the formation of the action plan in working of flexible and reversible focusing of attention in working
memory). This was followed by a later phase of getting ready memory,107 this study again used gaze biases to track how
for action execution. Interestingly, unlike the late phase, the early participants juggle the contents of working memory when these
phase did not scale with temporal expectation but nevertheless are required in sequence. Participants viewed two oriented
correlated with the speed and quality of performance in the colored bars and later, when probed, had to reproduce them
working-memory task. in sequence, with each report taking approximately 1 s
These findings highlight again how the temporal profile of (Figure 4D, left).
anticipatory neural signals themselves can prompt new insights Given the serial nature of the task, we expected participants to
into the processes that constitute anticipation. Using other neu- switch attention to the next memory content only after they had
ral readouts in neuronal recordings, others too have advocated finished reporting the preceding content (i.e., after the reproduc-
moving beyond the gradual ramping of activity during neural tion ‘‘report 1’’ in Figure 4D, middle). Surprisingly, what we
preparation and considering, for example, evolving attractor observed instead was that attention switched to the next mem-
dynamics.155 ory content before participants completed the initial report
Voluntary and involuntary attention start concurrently in (Figure 4D, right). The gaze biases showed that, by the time par-
working memory ticipants finished their first report (making a button click with the
Importantly, temporal profiles that reveal relevant properties of mouse), their attention had already switched all the way to the
attention are not limited to neural measurements. Time-resolved next memory content.
behavioral markers of attention can also be highly revealing, as The findings make clear how internal attention is fundamen-
the final two studies make clear. tally ‘‘forward looking,’’ anticipating the next event while still
The study described in van Ede et al.152 (Figure 4C) tracked the guiding behavior by the previous. This is reminiscent of starting
timing of voluntary and involuntary shifts of internal selective to look for the next instruction while still working on the previous
attention and revealed that these kick off simultaneously and step when assembling a piece of furniture and is consistent with
compete with one another. To track both types of attention ori- a wider literature on looking-ahead fixations.158 More important
enting, this time we measured small shifts in gaze location asso- than the take-home that attention switches earlier than imagined

980 Neuron 111, April 5, 2023

 ll
Review

is the lesson that attention functions can overlap in time with Studies using continuous neurophysiological and behavioral
other concomitant processing. methods are beginning to chart the dynamics in neural process-
The intuition that processing stages unfold serially is hard to ing and their effects on performance. Clever task designs have
shake, i.e., first we do one thing, then the next. Yet, this seems started to separate the time courses of various modulatory func-
not to be the case in the brain. Just like the extensive amount tions and their impact on different stimulus content. However,
of concurrent processing during processing of single events, important coding and modulatory effects may also exist under
processing of successive events may also overlap significantly. the surface of palpable brain activity, such as changes in
This psychologically non-intuitive degree of parallelism is likely synaptic weights that link neural ensembles and regulate
to have ecological advantage.159,160 Nevertheless, it also raises excitability.164 Developing new tools to examine more aspects
difficult questions about how the brain achieves the necessary of neural and behavioral states will surely deepen the enquiry.
temporal coordination of all the relevant signals within and be- Although the field is moving in a good direction, one source of
tween events to create cohesive information to guide behavior. dynamics is often sidestepped—the active agent. In the typical
This requires not only resolving the temporal dimension of the laboratory task, participants remain still and have limited means
binding problem but also incorporating the issues brought by to interact with the (experimental) world around them. In life, we
overlapping analysis of sequential events within dynamic con- attend to things while moving and engaging in behavior (Figure 1).
texts. In addition to parallel processing, neural activity within We walk and run and dance and jump and ride; we turn and look
and between separate processing stages may also draw upon and point and reach and grab. Controlled laboratory experi-
serial mechanisms of information sampling to facilitate coordina- ments have exposed the tight link between attention and action
tion and integration, such as rhythmic sampling of externally10–12 systems, such as eye and reach movements,165,166 but few have
or internally161–163 derived contents. actually investigated attention in the context of full body move-
The examples chosen for illustration used innovative experi- ments and natural forms of behavior (see Tatler and Land167).
mental designs and incorporated continuous brain and behav- Setting the agent free is an important next step. Virtual reality
ioral measures to track the neural processing of multiple items provides a propitious means to combine agency with experimental
and follow the focus of attention. Together, they also revealed control76 and increasingly allows for simultaneous tracking of neu-
how the temporal coordination of internal attention functions dif- ral processing80 and various behavioral outcomes, such as subtle
fers from that in external attention. Selection of internal sensory patterns of gaze behavior that reveal shifts of attention (e.g.,
and motor attributes runs in parallel150 and different sources of Draschkow et al.77). Placing moving agents in extended dynamic
internal attention take off together.152 These studies also incre- environments with embedded temporal structures, while
mentally push our experimentation toward more ecological situ- measuring brain activity and behavior as mental dispositions shift,
ations, bringing action back into visual working memory,150,151 will get us much closer to tapping into the fundamental workings of
considering multiple sources of attention together,152 and selective attention.
imposing sequential task demands.153 Studies can go even As ever, multiple methods will need to be combined for a full
further, using extended and dynamic stimulus displays with understanding, as each reveals only clues about the full answer.
various forms of embedded temporal structures.72,73 As the field We also stand to gain from a much better partnership between
embraces more dynamic and naturalistic situations, using time- human and animal experimentation. Studies on humans can
resolved neural and behavioral measurements will become introduce conceptual innovation regarding the psychological
increasingly important and will continue to generate unexpected factors shaping attention and point toward some of the neural
findings that defy intuition and prompt new ways of thinking. systems involved, but understanding functional mechanisms at
the circuit and cellular levels requires granularity that is only
Concluding remarks available in animal models.141
The work discussed in our (non-exhaustive) review shows how The moment is ripe to step into the fluxes of attention. The time
the attention field is edging toward the temporal dimension by dimension, in its glorious strangeness, will surely offer new in-
embracing various sources of relevant dynamics. sights about the essential scaffolding functions of adaptive
So far, studies manipulating temporal expectations have cognition and behavior.
mainly addressed dynamics in the environment. Most tasks so
far use simple and static displays, and future studies can go ACKNOWLEDGMENTS
further by using natural environments and extended task
contexts. This research was supported by a Wellcome Trust Senior Investigator Award
The dynamics in mental dispositions are starting to be (104571/Z/14/Z) and a James S. McDonnell Foundation Understanding Hu-
man Cognition Collaborative Award (220020448) to A.C.N., an ERC Starting
explored in internal-attention studies using retrocues, sequential Grant from the European Research Council (MEMTICIPATION, 850636) and
probes, and extended displays. It has been possible, therefore, a Vidi Grant from the Dutch Research Council (NWO grant 14721) to F.v.E.,
to show flexible and reversible shifts in the prioritization of and by the NIHR Oxford Health Biomedical Research Centre. The Wellcome
Centre for Integrative Neuroimaging is supported by core funding from the
different items within working memory based on internal Wellcome Trust (203139/Z/16/Z). We further wish to thank Assaf Breska for
states107 and to reveal the proactive nature of these shifts.153 providing the raw data to create Figure 2A and Sage Boettcher, Irene
Future studies can enrich this line of enquiry by developing Echeverria-Altuna, and Daniela Gresch for thoughtful comments on the draft.
experimental frameworks to investigate temporal dynamics This research was funded in part by the Wellcome Trust (grant numbers
104571/Z/14/Z, 203139/Z/16/Z). For the purpose of open access, the author
when shifting between task sets and by including multi-tasking has applied a CC BY public copyright license to any Author Accepted Manu-
situations. script version arising from this submission.

Neuron 111, April 5, 2023 981

 ll
Review
AUTHOR CONTRIBUTIONS 18. Lamme, V.A.F., and Roelfsema, P.R. (2000). The distinct modes of vision
offered by feedforward and recurrent processing. Trends Neurosci. 23,
571–579. https://doi.org/10.1016/S0166-2236(00)01657-X.
Conceptualization, A.C.N. and F.v.E.; writing – original draft, A.C.N. and F.v.E.;
writing – review & editing, A.C.N. and F.v.E.; visualization, A.C.N. and F.v.E. 19. Mashour, G.A., Roelfsema, P., Changeux, J.P., and Dehaene, S. (2020).
Conscious processing and the global neuronal workspace hypothesis.
DECLARATION OF INTERESTS Neuron 105, 776–798. https://doi.org/10.1016/j.neuron.2020.01.026.

20. Maunsell, J.H.R. (2015). Neuronal mechanisms of visual attention. Annu.

A.C.N. is on the advisory board of Neuron. Rev. Vis. Sci. 1, 373–391. https://doi.org/10.1146/annurev-vision-
082114-035431.
REFERENCES 21. Moore, T., and Zirnsak, M. (2017). Neural mechanisms of selective visual
attention. Annu. Rev. Psychol. 68, 47–72. https://doi.org/10.1146/an-
1. Nobre, A. (2018). Chapter 6: attention. In Stevens’ Handbook of Experi- nurev-psych-122414-033400.
mental Psychology and Cognitive Neuroscience, J. Wixted and J. Seren-
ces, eds. (John Wiley & Sons). 22. Buschman, T.J., and Kastner, S. (2015). From behavior to neural dy-
namics: an integrated theory of attention. Neuron 88, 127–144. https://
2. Nobre, A.C., and Kastner, S. (2014). The Oxford Handbook of Attention doi.org/10.1016/j.neuron.2015.09.017.
(Oxford University Press).
23. van Ede, F., and Nobre, A.C. (2023). Turning attention inside out: how
3. Craik, K.J.W. (1948). Theory of the human operator in control systems; working memory serves behavior. Annu. Rev. Psychol. 74, 137–165.
Man as an element in a control system. Br. J. Psychol. Gen. Sect. 38, https://doi.org/10.1146/annurev-psych-021422-041757.
142–148. https://doi.org/10.1111/j.2044-8295.1948.tb01149.x.
24. Panichello, M.F., and Buschman, T.J. (2021). Shared mechanisms un-
4. Harter, M.R. (1967). Excitability cycles and cortical scanning: a review of derlie the control of working memory and attention. Nature 592,
two hypotheses of central intermittency in perception. Psychol. Bull. 68, 601–605. https://doi.org/10.1038/s41586-021-03390-w.
47–58. https://doi.org/10.1037/h0024725.
25. Nobre, A.C., and Stokes, M.G. (2019). Premembering experience: a hier-
5. VanRullen, R., Carlson, T., and Cavanagh, P. (2007). The blinking spot- archy of time-scales for proactive attention. Neuron 104, 132–146.
light of attention. Proc. Natl. Acad. Sci. USA 104, 19204–19209. https://doi.org/10.1016/j.neuron.2019.08.030.
https://doi.org/10.1073/pnas.0707316104.
26. Nobre, A.C., and van Ede, F. (2018). Anticipated moments: temporal
6. Holcombe, A.O. (2009). Seeing slow and seeing fast: two limits on structure in attention. Nat. Rev. Neurosci. 19, 34–48. https://doi.org/10.
perception. Trends Cogn. Sci. 13, 216–221. https://doi.org/10.1016/j. 1038/nrn.2017.141.
tics.2009.02.005.
27. Coull, J.T., and Nobre, A.C. (2008). Dissociating explicit timing from tem-
7. Busch, N.A., Dubois, J., and VanRullen, R. (2009). The phase of ongoing poral expectation with fMRI. Curr. Opin. Neurobiol. 18, 137–144. https://
EEG oscillations predicts visual perception. J. Neurosci. 29, 7869–7876. doi.org/10.1016/j.conb.2008.07.011.
https://doi.org/10.1523/JNEUROSCI.0113-09.2009.
28. Buhusi, C.V., and Meck, W.H. (2005). What makes us tick? Functional
8. VanRullen, R. (2016). Perceptual cycles. Trends Cogn. Sci. 20, 723–735. and neural mechanisms of interval timing. Nat. Rev. Neurosci. 6,
https://doi.org/10.1016/j.tics.2016.07.006. 755–765. https://doi.org/10.1038/nrn1764.

9. Fries, P. (2015). Rhythms for cognition: communication through coher- 29. Karmarkar, U.R., and Buonomano, D.V. (2007). Timing in the absence of
ence. Neuron 88, 220–235. https://doi.org/10.1016/j.neuron.2015. clocks: encoding time in neural network states. Neuron 53, 427–438.
09.034. https://doi.org/10.1016/j.neuron.2007.01.006.

10. Fiebelkorn, I.C., and Kastner, S. (2019). A rhythmic theory of attention. 30. Muller, T., and Nobre, A.C. (2014). Perceiving the passage of time: neural
Trends Cogn. Sci. 23, 87–101. https://doi.org/10.1016/j.tics.2018. possibilities. Ann. NY Acad. Sci. 1326, 60–71. https://doi.org/10.1111/
11.009. nyas.12545.

11. Landau, A.N., and Fries, P. (2012). Attention samples stimuli rhythmically. 31. Theunissen, F.E., Sen, K., and Doupe, A.J. (2000). Spectral-temporal
Curr. Biol. 22, 1000–1004. https://doi.org/10.1016/j.cub.2012.03.054. receptive fields of nonlinear auditory neurons obtained using natural
sounds. J. Neurosci. 20, 2315–2331. https://doi.org/10.1523/JNEURO-
12. Tomassini, A., Spinelli, D., Jacono, M., Sandini, G., and Morrone, M.C. SCI.20-06-02315.2000.
(2015). Rhythmic oscillations of visual contrast sensitivity synchronized
with action. J. Neurosci. 35, 7019–7029. https://doi.org/10.1523/ 32. Eichenbaum, H. (2014). Time cells in the hippocampus: a new dimension
JNEUROSCI.4568-14.2015. for mapping memories. Nat. Rev. Neurosci. 15, 732–744. https://doi.org/
10.1038/nrn3827.
13. Schroeder, C.E., Wilson, D.A., Radman, T., Scharfman, H., and Lakatos,
P. (2010). Dynamics of active sensing and perceptual selection. Curr. 33. Ghose, G.M., and Maunsell, J.H.R. (2002). Attentional modulation in vi-
Opin. Neurobiol. 20, 172–176. https://doi.org/10.1016/j.conb.2010. sual cortex depends on task timing. Nature 419, 616–620. https://doi.
02.010. org/10.1038/nature01057.

14. Dewey, J. (1896). The reflex arc concept in psychology. Psychol. Rev. 3, 34. Janssen, P., and Shadlen, M.N. (2005). A representation of the hazard
357–370. https://doi.org/10.1037/h0070405. rate of elapsed time in macaque area LIP. Nat. Neurosci. 8, 234–241.
https://doi.org/10.1038/nn1386.
15. Cavanagh, S.E., Hunt, L.T., and Kennerley, S.W. (2020). A diversity of
intrinsic timescales underlie neural computations. Front. Neural Circuits €n, S., Diesmann, M., and Aertsen, A. (1997). Spike syn-
35. Riehle, A., Gru
14, 615626. https://doi.org/10.3389/fncir.2020.615626. chronization and rate modulation differentially involved in motor cortical
function. Science 278, 1950–1953. https://doi.org/10.1126/science.278.
16. Chen, J., Hasson, U., and Honey, C.J. (2015). Processing timescales as 5345.1950.
an organizing principle for primate cortex. Neuron 88, 244–246. https://
doi.org/10.1016/j.neuron.2015.10.010. 36. Hollerman, J.R., and Schultz, W. (1998). Dopamine neurons report an er-
ror in the temporal prediction of reward during learning. Nat. Neurosci. 1,
17. Honey, C.J., Thesen, T., Donner, T.H., Silbert, L.J., Carlson, C.E., Devin- 304–309. https://doi.org/10.1038/1124.
sky, O., Doyle, W.K., Rubin, N., Heeger, D.J., and Hasson, U. (2012).
Slow cortical dynamics and the accumulation of information over long 37. Buonomano, D., and Mauk, M.D. (1994). Neural network model of the
timescales. Neuron 76, 423–434. https://doi.org/10.1016/j.neuron. cerebellum: temporal discrimination and the timing of motor responses.
2012.08.011. Neural Comput. 6, 38–55. https://doi.org/10.1162/neco.1994.6.1.38.

982 Neuron 111, April 5, 2023

 ll
Review
38. Buzsáki, G., and Tingley, D. (2018). Space and time: the hippocampus as 57. Zimmermann, J.F., Moscovitch, M., and Alain, C. (2017). Long-term
a sequence generator. Trends Cogn. Sci. 22, 853–869. https://doi.org/ memory biases auditory spatial attention. J. Exp. Psychol. Learn. Mem.
10.1016/j.tics.2018.07.006. Cogn. 43, 1602–1615. https://doi.org/10.1037/xlm0000398.

39. Duncan, J., Martens, S., and Ward, R. (1997). Restricted attentional ca- 58. Olson, I.R., and Chun, M.M. (2001). Temporal contextual cuing of visual
pacity within but not between sensory modalities. Nature 387, attention. J. Exp. Psychol. Learn. Mem. Cogn. 27, 1299–1313. https://
808–810. https://doi.org/10.1038/42947. doi.org/10.1037//0278-7393.27.5.1299.

40. Broadbent, D.E., and Broadbent, M.H.P. (1987). From detection to identi- 59. O’Reilly, J.X., McCarthy, K.J., Capizzi, M., and Nobre, A.C. (2008). Acqui-
fication: response to multiple targets in rapid serial visual presentation. sition of the temporal and ordinal structure of movement sequences in
Percept. Psychophys. 42, 105–113. https://doi.org/10.3758/BF03210498. incidental learning. J. Neurophysiol. 99, 2731–2735. https://doi.org/10.
1152/jn.01141.2007.
41. Raymond, J.E., Shapiro, K.L., and Arnell, K.M. (1992). Temporary sup-
pression of visual processing in an RSVP task: an attentional blink? 60. Heideman, S.G., van Ede, F., and Nobre, A.C. (2017). Temporal align-
J. Exp. Psychol. Hum. Percept. Perform. 18, 849–860. https://doi.org/ ment of anticipatory motor cortical beta lateralisation in hidden visual-
10.1037//0096-1523.18.3.849. motor sequences. Eur. J. Neurosci. 48, 2684–2695. https://doi.org/10.
1111/ejn.13700.
42. Kanwisher, N.G. (1987). Repetition blindness: type recognition without
token individuation. Cognition 27, 117–143. https://doi.org/10.1016/ 61. Riess Jones, M. (2019). Time Will Tell (Oxford University Press). https://
0010-0277(87)90016-3. doi.org/10.1093/oso/9780190618216.001.0001.

43. Pashler, H. (1994). Dual-task interference in simple tasks: data and the- 62. Jones, M.R. (1976). Time, our lost dimension: toward a new theory of
ory. Psychol. Bull. 116, 220–244. https://doi.org/10.1037/0033-2909. perception, attention, and memory. Psychol. Rev. 83, 323–355. https://
116.2.220. doi.org/10.1037/0033-295X.83.5.323.

44. Goodbourn, P.T., and Holcombe, A.O. (2015). ‘‘Pseudoextinction’’: 63. Cravo, A.M., Rohenkohl, G., Wyart, V., and Nobre, A.C. (2011). Endoge-
asymmetries in simultaneous attentional selection. J. Exp. Psychol. nous modulation of low frequency oscillations by temporal expectations.
Hum. Percept. Perform. 41, 364–384. https://doi.org/10.1037/a0038734. J. Neurophysiol. 106, 2964–2972. https://doi.org/10.1152/jn.00157.2011.

45. Ludowici, C.J.H., and Holcombe, A.O. (2021). The dynamics of buffered 64. Grabenhorst, M., Michalareas, G., Maloney, L.T., and Poeppel, D. (2019).
and triggered selection from rapid serial visual presentation (RSVP) The anticipation of events in time. Nat. Commun. 10, 5802. https://doi.
streams. J. Exp. Psychol. Hum. Percept. Perform. 47, 200–222. https:// org/10.1038/s41467-019-13849-0.
doi.org/10.1037/xhp0000885.
65. Morillon, B., Schroeder, C.E., Wyart, V., and Arnal, L.H. (2016). Temporal
prediction in lieu of periodic stimulation. J. Neurosci. 36, 2342–2347.
46. Zivony, A., and Eimer, M. (2023). The temporal dynamics of selective
https://doi.org/10.1523/JNEUROSCI.0836-15.2016.
attention are reflected by distractor intrusions. Sci. Rep. 13, 408.
https://doi.org/10.1038/s41598-022-26902-8.
66. Breska, A., and Ivry, R.B. (2018). Double dissociation of single-interval
and rhythmic temporal prediction in cerebellar degeneration and Parkin-
47. van Ede, F., Chekroud, S.R., Stokes, M.G., and Nobre, A.C. (2018). De-
son’s disease. Proc. Natl. Acad. Sci. USA 115, 12283–12288. https://doi.
coding the influence of anticipatory states on visual perception in the
org/10.1073/pnas.1810596115.
presence of temporal distractors. Nat. Commun. 9, 1449. https://doi.
org/10.1038/s41467-018-03960-z.
67. Breska, A., and Deouell, L.Y. (2014). Automatic bias of temporal expec-
tations following temporally regular input independently of high-level
48. Botella, J., Suero, M., and Barriopedro, M.I. (2001). A model of the forma-
temporal expectation. J. Cogn. Neurosci. 26, 1555–1571. https://doi.
tion of illusory conjunctions in the time domain. J. Exp. Psychol. Hum.
org/10.1162/jocn_a_00564.
Percept. Perform. 27, 1452–1467. https://doi.org/10.1037//0096-1523.
27.6.1452. 68. Rohenkohl, G., Coull, J.T., and Nobre, A.C. (2011). Behavioural dissocia-
tion between exogenous and endogenous temporal orienting of attention.
49. Nobre, A.C., and Rohenkohl, G. (2014). Time for the fourth dimension in PLoS One 6, e14620. https://doi.org/10.1371/journal.pone.0014620.
attention. The Oxford Handbook of Attention (Oxford University Press).
69. Correa, A., and Nobre, A.C. (2008). Neural modulation by regularity and
50. Niemi, P., and Na€a€ ta
€nen, R. (1981). Foreperiod and simple reaction time. passage of time. J. Neurophysiol. 100, 1649–1655. https://doi.org/10.
Psychol. Bull. 89, 133–162. https://doi.org/10.1037/0033-2909.89.1.133. 1152/jn.90656.2008.
51. Los, S.A. (2012). Foreperiod and sequential effects: theory and data. 70. Grabenhorst, M., Maloney, L.T., Poeppel, D., and Michalareas, G. (2021).
Attention and Time (Oxford Academic). https://doi.org/10.1093/acpro- Two sources of uncertainty independently modulate temporal expec-
f:oso/9780199563456.003.0021. tancy. Proc. Natl. Acad. Sci. USA 118. e2019342118. https://doi.org/
10.1073/pnas.2019342118.
52. Coull, J.T., and Nobre, A.C. (1998). Where and when to pay attention: the
neural systems for directing attention to spatial locations and to time in- 71. van Ede, F., Rohenkohl, G., Gould, I., and Nobre, A.C. (2020). Purpose-
tervals as revealed by both PET and fMRI. J. Neurosci. 18, 7426–7435. dependent consequences of temporal expectations serving perception
https://doi.org/10.1523/JNEUROSCI.18-18-07426.1998. and action. J. Neurosci. 40, 7877–7886. https://doi.org/10.1523/JNEUR-
OSCI.1134-20.2020.
53. Denison, R.N., Heeger, D.J., and Carrasco, M. (2017). Attention flexibly
trades off across points in time. Psychon. Bull. Rev. 24, 1142–1151. 72. Salet, J.M., Kruijne, W., and van Rijn, H. (2021). Implicit learning of tem-
https://doi.org/10.3758/s13423-016-1216-1. poral behavior in complex dynamic environments. Psychon. Bull. Rev.
28, 1270–1280. https://doi.org/10.3758/s13423-020-01873-x.
54. Griffin, I.C., Miniussi, C., and Nobre, A.C. (2002). Multiple mechanisms of
selective attention: differential modulation of stimulus processing by 73. Boettcher, S.E.P., Shalev, N., Wolfe, J.M., and Nobre, A.C. (2022). Right
attention to space or time. Neuropsychologia 40, 2325–2340. https:// place, right time: spatiotemporal predictions guide attention in dynamic
doi.org/10.1016/S0028-3932(02)00087-8. visual search. J. Exp. Psychol. Gen. 151, 348–362. https://doi.org/10.
1037/xge0000901.
55. Cravo, A.M., Rohenkohl, G., Santos, K.M., and Nobre, A.C. (2017). Tem-
poral anticipation based on memory. J. Cogn. Neurosci. 29, 2081–2089. 74. Shalev, N., and Nobre, A.C. (2022). Eyes wide open: regulation of arousal
https://doi.org/10.1162/jocn_a_01172. by temporal expectations. Cognition 224, 105062. https://doi.org/10.
1016/j.cognition.2022.105062.
56. Los, S.A., Kruijne, W., and Meeter, M. (2017). Hazard versus history: tem-
poral preparation is driven by past experience. J. Exp. Psychol. Hum. 75. Salet, J.M., Schlichting, N., Kruijne, W., and van Rijn, H. (2022).
Percept. Perform. 43, 78–88. https://doi.org/10.1037/xhp0000279. Addendum: implicit learning of temporal behavior in complex dynamic

Neuron 111, April 5, 2023 983

 ll
Review
environments. Psychon. Bull. Rev. 29, 2325–2329. https://doi.org/10. 93. Shen, D., and Alain, C. (2012). Implicit temporal expectation attenuates
3758/s13423-022-02194-x. auditory attentional blink. PLoS One 7, e36031. https://doi.org/10.
1371/journal.pone.0036031.
76. Draschkow, D. (2022). Remote virtual reality as a tool for increasing
external validity. Nat. Rev. Psychol. 1, 433–434. https://doi.org/10. 94. Tal-Perry, N., and Yuval-Greenberg, S. (2021). Prestimulus inhibition of
1038/s44159-022-00082-8. eye movements reflects temporal expectation rather than time estima-
tion. Atten. Percept. Psychophys. 83, 2473–2485. https://doi.org/10.
77. Draschkow, D., Nobre, A.C., and van Ede, F. (2022). Multiple spatial 3758/s13414-021-02319-9.
frames for immersive working memory. Nat. Hum. Behav. 6, 536–544.
https://doi.org/10.1038/s41562-021-01245-y. 95. Denison, R.N., Yuval-Greenberg, S., and Carrasco, M. (2019). Directing
voluntary temporal attention increases fixational stability. J. Neurosci.
78. Brookes, M.J., Leggett, J., Rea, M., Hill, R.M., Holmes, N., Boto, E., and 39, 353–363. https://doi.org/10.1523/JNEUROSCI.1926-18.2018.
Bowtell, R. (2022). Magnetoencephalography with optically pumped
magnetometers (OPM-MEG): the next generation of functional neuroi- 96. Wilsch, A., Henry, M.J., Herrmann, B., Maess, B., and Obleser, J. (2015).
maging. Trends Neurosci. 45, 621–634. https://doi.org/10.1016/j.tins. Alpha oscillatory dynamics index temporal expectation benefits in work-
2022.05.008. ing memory. Cereb. Cortex 25, 1938–1946. https://doi.org/10.1093/cer-
cor/bhu004.
79. Souza, R.H.C.E., and Naves, E.L.M. (2021). Attention detection in virtual
environments using EEG signals: a scoping review. Front. Physiol. 12, 97. Sørensen, T.A., Vangkilde, S., and Bundesen, C. (2015). Components of
727840. https://doi.org/10.3389/fphys.2021.727840. attention modulated by temporal expectation. J. Exp. Psychol. Learn.
Mem. Cogn. 41, 178–192. https://doi.org/10.1037/a0037268.
80. Roberts, G., Holmes, N., Alexander, N., Boto, E., Leggett, J., Hill, R.M.,
Shah, V., Rea, M., Vaughan, R., Maguire, E.A., et al. (2019). Towards 98. Thavabalasingam, S., O’Neil, E.B., Zeng, Z., and Lee, A.C.H. (2015).
OPM-MEG in a virtual reality environment. Neuroimage 199, 408–417. Recognition memory is improved by a structured temporal framework
https://doi.org/10.1016/j.neuroimage.2019.06.010. during encoding. Front. Psychol. 6, 2062. https://doi.org/10.3389/
fpsyg.2015.02062.
81. Zokaei, N., Gillebert, C.R., Chauvin, J.J., Gresch, D., Board, A.G., Rolin-
ski, M., Hu, M.T., and Nobre, A.C. (2021). Temporal orienting in Parkin- 99. Jones, A., and Ward, E.V. (2019). Rhythmic temporal structure at encod-
son’s disease. Eur. J. Neurosci. 53, 2713–2725. https://doi.org/10. ing enhances recognition memory. J. Cogn. Neurosci. 31, 1549–1562.
1111/ejn.15114. https://doi.org/10.1162/jocn_a_01431.

82. Fernández, A., Denison, R.N., and Carrasco, M. (2019). Temporal atten- 100. Akdoǧan, B., Balcı, F., and van Rijn, H. (2016). Temporal expectation in-
tion improves perception similarly at foveal and parafoveal locations. dexed by pupillary response. Timing Time Percept. 4, 354–370. https://
J. Vis. 19, 12. https://doi.org/10.1167/19.1.12. doi.org/10.1163/22134468-00002075.

83. Rohenkohl, G., Cravo, A.M., Wyart, V., and Nobre, A.C. (2012). Temporal 101. Nobre, A.C. (2010). How can temporal expectations bias perception and ac-
expectation improves the quality of sensory information. J. Neurosci. 32, tion? In Attention and Time, A. Nobre and J. Coull, eds. (Oxford University
8424–8428. https://doi.org/10.1523/JNEUROSCI.0804-12.2012. Press), pp. 371–392. https://doi.org/10.1093/acprof:oso/9780199563456.
003.0027.
84. Cravo, A.M., Rohenkohl, G., Wyart, V., and Nobre, A.C. (2013). Temporal
expectation enhances contrast sensitivity by phase entrainment of low- 102. King, A.J., and Nelken, I. (2009). Unraveling the principles of auditory
frequency oscillations in visual cortex. J. Neurosci. 33, 4002–4010. cortical processing: can we learn from the visual system? Nat. Neurosci.
https://doi.org/10.1523/JNEUROSCI.4675-12.2013. 12, 698–701. https://doi.org/10.1038/nn.2308.

85. Vangkilde, S., Coull, J.T., and Bundesen, C. (2012). Great expectations: 103. Viemeister, N.F., and Plack, C.J. (1993). Time analysis. In Springer Hand-
temporal expectation modulates perceptual processing speed. J. Exp. book of Auditory Research, W.A. Yost, A.N. Popper, and R.R. Fay, eds.
Psychol. Hum. Percept. Perform. 38, 1183–1191. https://doi.org/10. (Springer), pp. 116–154. https://doi.org/10.1007/978-1-4612-2728-1_4.
1037/a0026343.
104. Souza, A.S., and Oberauer, K. (2016). In search of the focus of attention in
86. Jepma, M., Wagenmakers, E.J., and Nieuwenhuis, S. (2012). Temporal working memory: 13 years of the retro-cue effect. Atten. Percept. Psy-
expectation and information processing: A model-based analysis. Cogni- chophys. 78, 1839–1860. https://doi.org/10.3758/s13414-016-1108-5.
tion 122, 426–441. https://doi.org/10.1016/j.cognition.2011.11.014.
105. Myers, N.E., Chekroud, S.R., Stokes, M.G., and Nobre, A.C. (2018). Ben-
87. van den Brink, R.L., Murphy, P.R., Desender, K., de Ru, N., and Nieuwen- efits of flexible prioritization in working memory can arise without costs.
huis, S. (2021). Temporal expectation hastens decision onset but does J. Exp. Psychol. Hum. Percept. Perform. 44, 398–411. https://doi.org/10.
not affect evidence quality. J. Neurosci. 41, 130–143. https://doi.org/ 1037/xhp0000449.
10.1523/JNEUROSCI.1103-20.2020.
106. Jin, W., Nobre, A.C., and van Ede, F. (2020). Temporal expectations pre-
88. Bauer, A.R., van Ede, F., Quinn, A.J., and Nobre, A.C. (2021). Rhythmic pare visual working memory for behavior. J. Cogn. Neurosci. 32, 2320–
modulation of visual perception by continuous rhythmic auditory stimula- 2332. https://doi.org/10.1162/jocn_a_01626.
tion. J. Neurosci. 41, 7065–7075. https://doi.org/10.1523/JNEUROSCI.
2980-20.2021. 107. van Ede, F., Niklaus, M., and Nobre, A.C. (2017). Temporal expectations
guide dynamic prioritization in visual working memory through attenu-
89. Menceloglu, M., Suzuki, S., and Song, J.H. (2021). Revealing the effects ated a oscillations. J. Neurosci. 37, 437–445. https://doi.org/10.1523/
of temporal orienting of attention on response conflict using continuous JNEUROSCI.2272-16.2016.
movements. Atten. Percept. Psychophys. 83, 1463–1478. https://doi.
org/10.3758/s13414-020-02235-4. 108. Zokaei, N., Board, A.G., Manohar, S.G., and Nobre, A.C. (2019). Modula-
tion of the pupillary response by the content of visual working memory.
90. Correa, A., Cappucci, P., Nobre, A.C., and Lupiáñez, J. (2010). The two Proc. Natl. Acad. Sci. USA 116, 22802–22810. https://doi.org/10.1073/
sides of temporal orienting: facilitating perceptual selection, disrupting pnas.1909959116.
response selection. Exp. Psychol. 57, 142–148. https://doi.org/10.
1027/1618-3169/a000018. 109. Draschkow, D., Kallmayer, M., and Nobre, A.C. (2021). When natural
behavior engages working memory. Curr. Biol. 31. 869.e5–874.e5.
91. Gresch, D., Boettcher, S.E.P., van Ede, F., and Nobre, A.C. (2021). https://doi.org/10.1016/j.cub.2020.11.013.
Shielding working-memory representations from temporally predictable
external interference. Cognition 217, 104915. https://doi.org/10.1016/j. 110. Gresch, D., Boettcher, S.E.P., Nobre, A.C., and van Ede, F. (2022). Conse-
cognition.2021.104915. quences of predictable temporal structure in multi-task situations. Cogni-
tion 225, 105156. https://doi.org/10.1016/j.cognition.2022.105156.
92. Naccache, L., Blandin, E., and Dehaene, S. (2002). Unconscious masked
priming depends on temporal attention. Psychol. Sci. 13, 416–424. 111. Doherty, J.R., Rao, A., Mesulam, M.M., and Nobre, A.C. (2005). Synergis-
https://doi.org/10.1111/1467-9280.00474. tic effect of combined temporal and spatial expectations on visual

984 Neuron 111, April 5, 2023

 ll
Review
attention. J. Neurosci. 25, 8259–8266. https://doi.org/10.1523/JNEURO- 129. Lee, C.C.Y., Clifford, C.W.G., and Arabzadeh, E. (2019). Temporal cueing
SCI.1821-05.2005. enhances neuronal and behavioral discrimination performance in rat
whisker system. J. Neurophysiol. 121, 1048–1058. https://doi.org/10.
112. Rimmele, J.M., Morillon, B., Poeppel, D., and Arnal, L.H. (2018). Proac- 1152/jn.00604.2018.
tive sensing of periodic and aperiodic auditory patterns. Trends Cogn.
Sci. 22, 870–882. https://doi.org/10.1016/j.tics.2018.08.003. 130. Finnie, P.S.B., Komorowski, R.W., and Bear, M.F. (2021). The spatiotem-
poral organization of experience dictates hippocampal involvement in
113. Stefanics, G., Hangya, B., Hernádi, I., Winkler, I., Lakatos, P., and Ulbert, primary visual cortical plasticity. Curr. Biol. 31, 3996–4008.e6. https://
I. (2010). Phase entrainment of human delta oscillations can mediate the doi.org/10.1016/j.cub.2021.06.079.
effects of expectation on reaction speed. J. Neurosci. 30, 13578–13585.
https://doi.org/10.1523/JNEUROSCI.0703-10.2010. 131. Gavornik, J.P., and Bear, M.F. (2014). Learned spatiotemporal sequence
recognition and prediction in primary visual cortex. Nat. Neurosci. 17,
114. Breska, A., and Deouell, L.Y. (2017). Neural mechanisms of rhythm- 732–737. https://doi.org/10.1038/nn.3683.
based temporal prediction: delta phase-locking reflects temporal pre-
dictability but not rhythmic entrainment. PLoS Biol. 15, e2001665. 132. Orsolic, I., Rio, M., Mrsic-Flogel, T.D., and Znamenskiy, P. (2021). Meso-
https://doi.org/10.1371/journal.pbio.2001665. scale cortical dynamics reflect the interaction of sensory evidence and
temporal expectation during perceptual decision-making. Neuron 109,
115. Haegens, S., and Zion Golumbic, E. (2018). Rhythmic facilitation of sen- 1861–1875.e10. https://doi.org/10.1016/j.neuron.2021.03.031.
sory processing: a critical review. Neurosci. Biobehav. Rev. 86, 150–165.
https://doi.org/10.1016/j.neubiorev.2017.12.002. 133. van Wassenhove, V., Herbst, S.K., and Kononowicz, T.W. (2019). Timing
the brain to time the mind: critical contributions of time-resolved neuro-
116. Biau, E., Wang, D., Park, H., Jensen, O., and Hanslmayr, S. (2021). Auditory imaging for temporal cognition. Magnetoencephalography (Springer In-
detection is modulated by theta phase of silent lip movements. Curr. Res. ternational Publishing), pp. 855–905. https://doi.org/10.1007/978-3-
Neurobiol. 2, 100014. https://doi.org/10.1016/j.crneur.2021.100014. 030-00087-5_67.

117. Rohenkohl, G., Gould, I.C., Pessoa, J., and Nobre, A.C. (2014). 134. Posner, M.I., Cohen, Y., and Rafal, R.D. (1982). Neural systems control of
Combining spatial and temporal expectations to improve visual percep- spatial orienting. Philos. Trans. R. Soc. Lond. B Biol. Sci. 298, 187–198.
tion. J. Vis. 14, 8. https://doi.org/10.1167/14.4.8. https://doi.org/10.1098/rstb.1982.0081.

118. Rohenkohl, G., and Nobre, A.C. (2011). Alpha oscillations related to 135. Mu €ller, H.J., and Rabbitt, P.M.A. (1989). Reflexive and voluntary orienting
anticipatory attention follow temporal expectations. J. Neurosci. 31, of visual attention: time course of activation and resistance to interrup-
14076–14084. https://doi.org/10.1523/JNEUROSCI.3387-11.2011. tion. J. Exp. Psychol. Hum. Percept. Perform. 15, 315–330. https://doi.
org/10.1037//0096-1523.15.2.315.
119. Heideman, S.G., Rohenkohl, G., Chauvin, J.J., Palmer, C.E., van Ede, F.,
and Nobre, A.C. (2018). Anticipatory neural dynamics of spatial-temporal 136. Posner, M.I., Rafal, R.D., Choate, L.S., and Vaughan, J. (1985). Inhibition
orienting of attention in younger and older adults. Neuroimage 178, of return: neural basis and function. Cogn. Neuropsychol. 2, 211–228.
46–56. https://doi.org/10.1016/j.neuroimage.2018.05.002. https://doi.org/10.1080/02643298508252866.

120. van Ede, F., de Lange, F., Jensen, O., and Maris, E. (2011). Orienting 137. van Heusden, E., van Zoest, W., Donk, M., and Olivers, C.N.L. (2022). An
attention to an upcoming tactile event involves a spatially and temporally attentional limbo: saccades become momentarily non-selective in be-
specific modulation of sensorimotor alpha- and beta-band oscillations. tween saliency-driven and relevance-driven selection. Psychon. Bull.
J. Neurosci. 31, 2016–2024. https://doi.org/10.1523/JNEUROSCI.5630- Rev. 29, 1327–1337. https://doi.org/10.3758/s13423-022-02091-3.
10.2011.
138. van Ede, F., de Lange, F.P., and Maris, E. (2012). Attentional cues affect ac-
121. Jaramillo, S., and Zador, A.M. (2011). The auditory cortex mediates the curacy and reaction time via different cognitive and neural processes.
perceptual effects of acoustic temporal expectation. Nat. Neurosci. 14, J. Neurosci. 32, 10408–10412. https://doi.org/10.1523/JNEUROSCI.1337-
246–251. https://doi.org/10.1038/nn.2688. 12.2012.

122. Sharma, J., Sugihara, H., Katz, Y., Schummers, J., Tenenbaum, J., and 139. Lupiáñez, J. (2010). Inhibition of return. In Attention and Time (Oxford Uni-
Sur, M. (2015). Spatial attention and temporal expectation under timed versity Press), pp. 17–34. https://doi.org/10.1093/acprof:oso/978019
uncertainty predictably modulate neuronal responses in Monkey V1. 9563456.003.0002.
Cereb. Cortex 25, 2894–2906. https://doi.org/10.1093/cercor/bhu086.
140. Cherry, E.C. (1953). Some experiments on the recognition of speech,
123. Anderson, B., and Sheinberg, D.L. (2008). Effects of temporal context with one and with two ears. J. Acoust. Soc. Am. 25, 975–979. https://
and temporal expectancy on neural activity in inferior temporal cortex. doi.org/10.1121/1.1907229.
Neuropsychologia 46, 947–957. https://doi.org/10.1016/j.neuropsycho-
logia.2007.11.025. 141. Nobre, A.C., and van Ede, F. (2020). Under the mind’s hood: what we
have learned by watching the brain at work. J. Neurosci. 40, 89–100.
124. Lima, B., Singer, W., and Neuenschwander, S. (2011). Gamma responses https://doi.org/10.1523/JNEUROSCI.0742-19.2019.
correlate with temporal expectation in monkey primary visual cortex.
J. Neurosci. 31, 15919–15931. https://doi.org/10.1523/JNEUROSCI. 142. Hillyard, S.A., Hink, R.F., Schwent, V.L., and Picton, T.W. (1973). Electri-
0957-11.2011. cal signs of selective attention in the human brain. Science 182, 177–180.
https://doi.org/10.1126/science.182.4108.177.
125. Parker, K.L., Alberico, S.L., Miller, A.D., and Narayanan, N.S. (2013). Pre-
frontal D1 dopamine signaling is necessary for temporal expectation dur- 143. van Voorhis, S., and Hillyard, S.A. (1977). Visual evoked potentials and
ing reaction time performance. Neuroscience 255, 246–254. https://doi. selective attention to points in space. Percept. Psychophys. 22, 54–62.
org/10.1016/j.neuroscience.2013.09.057. https://doi.org/10.3758/BF03206080.

126. Lakatos, P., Karmos, G., Mehta, A.D., Ulbert, I., and Schroeder, C.E. 144. Luck, S.J., Heinze, H.J., Mangun, G.R., and Hillyard, S.A. (1990). Visual
(2008). Entrainment of neuronal oscillations as a mechanism of atten- event-related potentials index focused attention within bilateral stimulus
tional selection. Science 320, 110–113. https://doi.org/10.1126/sci- arrays. II. Functional dissociation of P1 and N1 components. Electroen-
ence.1154735. cephalogr. Clin. Neurophysiol. 75, 528–542. https://doi.org/10.1016/
0013-4694(90)90139-B.
127. Lakatos, P., Musacchia, G., O’Connel, M.N., Falchier, A.Y., Javitt, D.C.,
and Schroeder, C.E. (2013). The spectrotemporal filter mechanism of 145. Buffalo, E.A., Fries, P., Landman, R., Liang, H., and Desimone, R. (2010).
auditory selective attention. Neuron 77, 750–761. https://doi.org/10. A backward progression of attentional effects in the ventral stream.
1016/j.neuron.2012.11.034. Proc. Natl. Acad. Sci. USA 107, 361–365. https://doi.org/10.1073/pnas.
0907658106.
128. Audette, N.J., Zhou, W., and Schneider, D.M. (2021). Temporally precise
movement-based predictions in the mouse auditory cortex. Preprint at 146. Mehta, A.D., Ulbert, I., and Schroeder, C.E. (2000). Intermodal selective
bioRxiv. https://doi.org/10.1101/2021.12.13.472457. attention in monkeys I: Distribution and timing of effects across visual

Neuron 111, April 5, 2023 985

 ll
Review
areas. Cereb. Cortex 10, 343–358. https://doi.org/10.1093/cercor/10. 157. Liu, B., Nobre, A.C., and van Ede, F. (2021). Functional but not obligatory
4.343. link between microsaccades and neural modulation by covert spatial
attention. Nat. Commun. 13, 3503. https://doi.org/10.1101/2021.11.10.
147. Michalareas, G., Vezoli, J., van Pelt, S., Schoffelen, J.M., Kennedy, H., 468033.
and Fries, P. (2016). Alpha-beta and gamma rhythms subserve feedback
and feedforward influences among human visual cortical areas. Neuron 158. Pelz, J.B., and Canosa, R. (2001). Oculomotor behavior and perceptual
89, 384–397. https://doi.org/10.1016/j.neuron.2015.12.018. strategies in complex tasks. Vision Res. 41, 3587–3596. https://doi.org/
10.1016/S0042-6989(01)00245-0.
148. Buschman, T.J., and Miller, E.K. (2007). Top-down versus bottom-up
control of attention in the prefrontal and posterior parietal cortices. Sci- 159. Cisek, P. (2019). Resynthesizing behavior through phylogenetic refine-
ence 315, 1860–1862. https://doi.org/10.1126/science.1138071. ment. Atten. Percept. Psychophys. 81, 2265–2287. https://doi.org/10.
3758/s13414-019-01760-1.
149. Siegel, M., Buschman, T.J., and Miller, E.K. (2015). Cortical information
flow during flexible sensorimotor decisions. Science 348, 1352–1355. 160. Cisek, P., and Kalaska, J.F. (2010). Neural mechanisms for interacting
https://doi.org/10.1126/science.aab0551. with a world full of action choices. Annu. Rev. Neurosci. 33, 269–298.
https://doi.org/10.1146/annurev.neuro.051508.135409.
150. van Ede, F., Chekroud, S.R., Stokes, M.G., and Nobre, A.C. (2019). Con-
current visual and motor selection during visual working memory guided 161. Peters, B., Kaiser, J., Rahm, B., and Bledowski, C. (2021). Object-based
action. Nat. Neurosci. 22, 477–483. https://doi.org/10.1038/s41593-018- attention prioritizes working memory contents at a theta rhythm. J. Exp.
0335-6. Psychol. Gen. 150, 1250–1256. https://doi.org/10.1037/xge0000994.

151. Boettcher, S.E.P., Gresch, D., Nobre, A.C., and van Ede, F. (2021). 162. Siegel, M., Warden, M.R., and Miller, E.K. (2009). Phase-dependent
Output planning at the input stage in visual working memory. Sci. Adv. neuronal coding of objects in short-term memory. Proc. Natl. Acad.
7, eabe8212. https://doi.org/10.1126/sciadv.abe8212. Sci. USA 106, 21341–21346. https://doi.org/10.1073/pnas.0908193106.

152. van Ede, F., Board, A.G., and Nobre, A.C. (2020). Goal-directed and stim- 163. Bahramisharif, A., Jensen, O., Jacobs, J., and Lisman, J. (2018). Serial
ulus-driven selection of internal representations. Proc. Natl. Acad. Sci. representation of items during working memory maintenance at letter-se-
USA 117, 24590–24598. https://doi.org/10.1073/pnas.2013432117. lective cortical sites. PLoS Biol. 16, e2003805. https://doi.org/10.1371/
journal.pbio.2003805.
153. van Ede, F., Deden, J., and Nobre, A.C. (2021). Looking ahead in working
memory to guide sequential behavior. Curr. Biol. 31, R779–R780. https:// 164. Stokes, M.G. (2015). ‘‘Activity-silent’’ working memory in prefrontal cor-
doi.org/10.1016/j.cub.2021.04.063. tex: a dynamic coding framework. Trends Cogn. Sci. 19, 394–405.
https://doi.org/10.1016/j.tics.2015.05.004.
154. Miniussi, C., Wilding, E.L., Coull, J.T., and Nobre, A.C. (1999). Orienting
attention in time. Modulation of brain potentials. Brain 122, 1507–1518. 165. Baldauf, D., and Deubel, H. (2010). Attentional landscapes in reaching
https://doi.org/10.1093/brain/122.8.1507. and grasping. Vision Res. 50, 999–1013. https://doi.org/10.1016/j.
visres.2010.02.008.
155. Inagaki, H.K., Chen, S., Daie, K., Finkelstein, A., Fontolan, L., Romani, S.,
and Svoboda, K. (2022). Neural algorithms and circuits for motor plan- 166. Heuer, A., Ohl, S., and Rolfs, M. (2020). Memory for action: a functional
ning. Annu. Rev. Neurosci. 45, 249–271. https://doi.org/10.1146/an- view of selection in visual working memory. Vis. Cogn. 28, 388–400.
nurev-neuro-092021-121730. https://doi.org/10.1080/13506285.2020.1764156.

156. van Ede, F., Chekroud, S.R., and Nobre, A.C. (2019). Human gaze tracks 167. Tatler, B.W., and Land, M.F. (2015). Everyday visual attention. In The
attentional focusing in memorized visual space. Nat. Hum. Behav. 3, Handbook of Attention, pp. 391–421. https://doi.org/10.7551/mitpress/
462–470. https://doi.org/10.1038/s41562-019-0549-y. 10033.003.0020.

986 Neuron 111, April 5, 2023