Available online at www.sciencedirect.

com Current Opinion in

ScienceDirect Neurobiology

Attention control in the primate brain

Rober Boshra1 and Sabine Kastner1,2,a

Abstract decision making, action selection). In the visual domain,

Attention is fundamental to all cognition. In the primate brain, it attentional selection can be based on spatial location,
is implemented by a large-scale network that consists of areas feature, or object information (e.g. looking for a person
spanning across all major lobes, also including subcortical with an orange hat standing at a traffic light). Even though
regions. Classical attention accounts assume that control over a typical selection takes only a few hundred milliseconds,
the selection process in this network is exerted by ‘top-down’ the network in the primate brain that is engaged during
mechanisms in the fronto-parietal cortex that influence sensory this process encompasses a multitude of areas and is
representations via feedback signals. More recent studies distributed across all major lobes as well as subcortical
have expanded this view of attentional control. In this review, structures including the thalamus, basal ganglia, and the
we will start from a traditional top-down account of attention superior colliculus [1e5]. It has been a longstanding goal
control, and then discuss more recent findings on feature- of the field of cognitive neuroscience to identify mecha-
based attention, thalamic influences, temporal network dy- nisms deployed across this distributed network that ac-
namics, and behavioral dynamics that collectively lead to count for the efficiency of the selection process. We will
substantial modifications. We outline how the different refer to these broadly as ‘attentional control mechanisms’.
emerging accounts can be reconciled and integrated into a Of note, these are different from control mechanisms
unified theory. related to why an item is selected, which involve moti-
vational, emotional, rule-based, and other cognitive pro-
Addresses cesses that are beyond the scope of this review [6e8].
1
Princeton Neuroscience Institute, Princeton University, Princeton, NJ,
08544, USA
2
Department of Psychology, Princeton University, Princeton, NJ, The first evidence that the attention network contained
08544, USA regions exerting critical control functions came from
neuropsychological studies in human patients showing
Corresponding authors: Boshra, Rober ([email protected]); that unilateral lesions, especially of higher-order cortex,
Kastner, Sabine ([email protected])
(Boshra R.)
may cause impairments in directing attention to
a
Dedicated to the memory of Dr. Leslie G. Ungerleider (1946–2020), contralateral visual space. This syndrome is known as
an amazing scientist, mentor, and friend. visuospatial hemineglect and, in severe cases, may lead
to complete disregard of the affected space, not only
with respect to the external space, but also in reference
Current Opinion in Neurobiology 2022, 76:102605
to the body or internal representations of memory sys-
This review comes from a themed issue on Systems Neuroscience tems [9,10]. For example, patients may eat from only
Edited by Joshua Johansen and Laura L Colgin half of their plate, dress half of their body, or recall
For complete overview of the section, please refer the article collection - landmarks from ipsi-, but not contra-lesional cognitive
Systems Neuroscience maps. Visuospatial neglect may also follow unilateral
Available online 15 July 2022 lesions at different sites including the basal ganglia and
https://doi.org/10.1016/j.conb.2022.102605
thalamus [11,12], thereby underlining the distributed
nature of possible control functions across the network.
0959-4388/© 2022 Elsevier Ltd. All rights reserved.
However, emphasis on the most frequently observed
sites, that is, the inferior parietal lobule and superior
Keywords temporal cortex [13,14], have contributed to attention
Spatial attention, Feature-based attention, Pulvinar, Rhythmic attention,
control models that center on fronto-parietal ‘top-down’
Fronto-parietal network, Attention control.
control mechanisms that operate by providing feedback
to sensory cortices (extensively reviewed in [15]).
Introduction
The term attention refers to a set of mechanisms aimed at In this review, we will discuss the changing views of
selectively prioritizing the neural representations that are attention control mechanisms that have emerged in
most relevant to one’s current behavioral goals. Attention recent years. We will primarily focus on studies using
mechanisms are highly flexible and provide the foundation intracranial electrophysiology in non-human primates,
on which most other cognition can operate (e.g. memory, but will also make connections to the human literature.

www.sciencedirect.com Current Opinion in Neurobiology 2022, 76:102605

2 Systems Neuroscience

We will take the traditional model of attentional top- processing in downstream areas. The fronto-parietal
down control of fronto-parietal over visual cortex during activation during directed attention encompasses a
space-based selection as a starting point and introduce multitude of topographic areas that have a preferential
three more recent control accounts along with evidence in representation of the contralateral visual hemifield
their support: advances in feature-based control, thalamic (Figure 1a top, left; [17]). Accordingly, spatial attention
control, and control of temporal dynamics in network to the contralateral hemifield evokes a stronger signal
interactions. We propose that the control mechanisms than to the ipsilateral hemifield, thus generating a
associated with each account are not mutually exclusive contralateral spatial biasing signal in each topographic
but may rather synergistically interact and complement area (Figure 1a top, right; [18]). The sum of biasing
one another. Together, the view that emerges from these signals across all fronto-parietal areas is similar in
different lines of work is that attention control is magnitude across the two hemispheres suggesting a
distributed across multiple network nodes exhibiting balanced spatial attention control system in which visual
specific control functions over space, objects, and time. space within a hemifield is largely controlled by the
contralateral hemisphere (interhemispheric competition
Fronto-parietal control of the selection account; [18e20]. The large-scale attention network
process that has been observed in humans is generally conserved
Neuroimaging studies have provided a detailed account in non-human primates, including key regions in frontal
of the functional neuroanatomy of the attention network. cortex, particularly the frontal eye fields (FEF) and
When human subjects are attending to a location in visual adjacent areas in dorso- and ventrolateral prefrontal
space in anticipation of a target stimulus, activity in cortex (PFC; [21]), as well as in posterior parietal cortex,
frontal and parietal cortex is sustained relative to activity particularly the lateral intraparietal area (LIP) and adja-
in visual cortex, reflecting the cognitive operations of the cent parietal regions such as V6/V6A, and the ventral
task, but not sensory processing (Figure 1a bottom; [16]). intraparietal area (VIP) [22e24].
Such results are in line with a top-down model of spatial
attention control, in which the fronto-parietal network Important evidence that frontal cortex is a source of top-
generates feedback signals that influence visual down feedback signals comes from studies demonstrating
Figure 1

Fronto-parietal top-down control of attention. a) Top: A spatial attention task activates fronto-parietal regions (color indicates p < 0.001 of contrast:
‘attend to periphery’ vs. ‘attend to fixation’) in humans including FEF and IPS of the contralateral hemisphere (right; adapted from [18]) that largely
overlaps with topographic maps of the fronto-parietal network (color wheel maps visual field angle to color overlay; left; [17]). Bottom: Directing attention to
a peripheral target (horizontal black lines) location activates FEF and V4 during an expectation period (grey bars) preceding target presentation (red bars).
V4’s activity is further increased by the visual stimuli, whereas FEF’s activity reflects the cognitive operation involved with the task (i.e. allocation of
attention in space) (adapted from [16]). b) Electrical microstimulation using currents below saccade-inducing thresholds in FEF induces attention-like
effects (red) in V4 neurons with receptive fields overlapping the saccade field. Black indicates control data when FEF is not stimulated (adapted from [27]).
c) A schematic of the macaque brain, the flow of sensory bottom-up (black arrows) signals, and the trajectory of control signals (red arrows) originating
from higher-order cortical areas in the fronto-parietal network that backpropagate to sensory, lower-order cortex.

Current Opinion in Neurobiology 2022, 76:102605 www.sciencedirect.com

 Attention control in the primate Boshra and Kastner 3

that electrical microstimulation of FEF can induce were abolished in FEF (Figure 2a, right), and behavioral
attention-like effects. First, stimulation of FEF can in- performance in a feature-based attention task was
crease the behavioral performance to discriminate a impaired [42]. These findings suggest that VPA is a
stimulus, similar to the effects of attentional allocation on source for top-down feedback signals related to feature-
discriminability [25,26]. And second, FEF stimulation based attention control that may propagate to other
leads to attention-like effects in downstream area V4 higher-order areas such as FEF [42] as well as visual
such as increasing neuronal firing rates (Figure 1b; [27]). cortex [43] to boost the processing of relevant stimuli
Causal manipulations in humans using transcranial mag- across the visual field.
netic stimulation have corroborated these findings by
showing qualitatively similar effects [28]. Further evi- Feature-based attention control is not confined to PFC.
dence comes from lesion studies of PFC demonstrating A temporal area in dorsal posterior inferotemporal cortex
that modulatory effects of attention are attenuated in (PITd), located on the inferior bank and lip of the su-
visual cortex [29]. Together, these and other lines of perior temporal sulcus (STS), anterior to MT
research have provided compelling evidence for a hier- (Figure 2b, top) has recently been implicated in atten-
archical top-down model (c.f. [30]), in which the allo- tion control function. PITd is strongly connected with
cation of attention to locations, features, and objects are prefrontal areas VPA, FEF, as well as LIP (Figure 2d,
controlled by a higher-order fronto-parietal network that [44,45]). This area has been found to be activated in
influences processing in visual cortex via feedback sig- monkeys performing a motion detection attention task
nals (Figure 1c). in a neuroimaging study [46], and was targeted for
recording studies that revealed strong attention modu-
New perspectives on feature-based control lation (Figure 2b, bottom). Further, microstimulation of
There is extensive evidence of non-spatial effects of PITd causally shifted the location at which attention
attention that enhance neural responses to behaviorally was allocated [47]. While PITd appears to be responsive
relevant stimulus features (e.g. color or motion; see to visual stimuli of different feature categories, neurons
[31e33]). Feature-based attention operates in parallel do not exhibit specific feature selectivity [47].
across the visual field, such that an attended feature re-
ceives a boost independent of its spatial location, PITd is not only connected with fronto-parietal cortex,
supporting a notion that feature- and space-based atten- but may also receive modulatory input from the supe-
tion mechanisms may be controlled by partially non- rior colliculus. In neuroimaging studies, inactivation of
overlapping networks [34e38]. In contrast, based on the superior colliculus (SC) during a motion discrimi-
human neuroimaging findings, the fronto-parietal network nation attention task [48] has been shown to reduce
is engaged during non-spatial selections of stimulus fea- activation in the vicinity of PITd, a region termed fSTS
tures [33] or objects [39,40], characterizing it as a ‘domain- due to its location at the floor of STS (Figure 2c, top;
general’ controller with little functional specialization. Bogadhi et al., 2019). Recording studies showed a
However, the limitations of imaging methods make it un- reduction (but not abolishment) of attentional modu-
clear whether the underlying control functions associated lation effects as a consequence of SC inactivation
with space-, feature- and object-based attention are linked (Figure 2c, bottom; Bogadhi et al., 2021). While it is not
to specialized subpopulations of neurons within the clear whether fSTS and PITd are overlapping or
network or recruit common circuits. Recent studies in non- separate but adjacent areas, inactivation of this part of
human primates have shed light on this issue. cortex underlines its role in attention control by
inducing behavioral deficits akin to visuo-spatial
Several studies have probed the role of FEF and sur- hemineglect [48]; however, exact pathways affected
rounding prefrontal areas in the control of feature-based by perturbing this area remain unexplored and may be
attention. FEF neurons have been shown to exhibit best probed with systematic causal manipulation
activity that reflects the behavioral relevance of relevant studies of simultaneous multi-area recordings. Neurons
features in a visual search task [41]. However, FEF’s in this area were also responsive to different feature
neurons do not exhibit intrinsic feature selectivity. categories as well as objects. Although the exact rela-
Therefore, FEF may receive feature-relevant informa- tionship between PITd and fSTS remains to be
tion from a source area that represents not only a spatial determined, these studies show that this part of cortex
map, but also incorporates feature information. Neurons exhibits functional characteristics of an attention con-
in the ventral prearcuate gyrus (VPA), located lateral to trol area that may be specialized in feature- and object-
the principal sulcus and anterior to the arcuate sulcus based attention [46,48], and highlights afferent signals
(Figure 2a, left, inset), were found to be feature selec- from SC that may be routed to this temporal area
tive and were modulated by spatial attention [42]. VPA through pulvinar (Figure 2d). The anatomical pro-
exhibits sustained feature-selective responses that are jections between this temporal region and VPA
obtained at earlier response latencies than in FEF (Figure 2d, [45]) may also enable the flow of feature-
(Figure 2a, left). Inactivation studies of VPA showed based control signals between PFC and the feature
that feature- (but not space-) based attention effects representative areas close to PITd/fSTS.
www.sciencedirect.com Current Opinion in Neurobiology 2022, 76:102605
4 Systems Neuroscience

Figure 2

Advances in feature-based attention control. a) Left: During a visual search task in macaques, VPA (location schematically presented in inset) exhibits
feature-based attention effects earlier than those observed in FEF. Right: Feature-based effects (red) observed in FEF (top) are abolished following
inactivation of VPA (bottom). Spatial attention effects (green) are not affected following inactivation (adapted from [42]). b) PITd is strongly activated during
a motion discrimination task (top; adapted from [46]) and its neurons exhibit considerable attentional modulation (bottom; adapted from [47]). c) Top:
Inactivation of SC induces a decrease in activation in an area at the floor of the STS during a motion detection task. Bottom: fSTS exhibits attention
modulation in its neuronal responses (left). This modulation is reduced, but not abolished following inactivation of SC (right; adapted from [48,49]). d) New
nodes discovered may factor into the control of feature-based attention. VPA may generate a behaviorally-relevant template for object and feature in-
formation to FEF (solid red arrow) and to V4 (dotted arrow). Relevant anatomical connectivity is denoted in red, highlighting projections between the PITd/
fSTS area (abbreviated PITd*) with LIP, FEF, and VPA (red lines). The SC causally influences temporal cortex in the PITd/fSTS area (dashed red line)
most likely by way of pulvinar (black lines).

Coordination of large-scale networks: periphery, integrate it with internal representations, and

Thalamic control drive motor output to generate behavior within just a
How can a large-scale network that consists of a multi- few hundred milliseconds? There is evidence that a
tude of nodes process information from the sensory central thalamic structure interconnected heavily with

Current Opinion in Neurobiology 2022, 76:102605 www.sciencedirect.com

 Attention control in the primate Boshra and Kastner 5

the cortex, the pulvinar, may play an important role in interconnected cortical areas and their projection zone
organizing the cortical attention network in time and in the pulvinar, while monkeys performed a spatial
aiding its processing efficiency [3,50]. Pulvinar is the attention task [58,59]. Focusing on the cue-target in-
largest nucleus in the primate thalamus and its expan- terval, which represents a pure cognitive state (i.e. the
sion during evolution scales with that of the primate allocation of attention at a cued location in expectation
neocortex [51e53]. It is almost exclusively inter- of target onset), during which the attention network is
connected with the cortical visual system through two prepared for target selection, neuronal spiking activity
well-established cortico-pulvinar pathways that parallel was not only enhanced, but also showed increased cor-
the canonical input-output relationships constituting relations in a common frequency band (alpha, low beta,
the visual processing hierarchy [54,55]: (i) a trans- <20 Hz) of the local field potential. The pulvinar
thalamic feedforward pathway that originates from layer appeared to have the strongest influence on this coor-
5 of a cortical area, loops through a dedicated pulvinar dinated activity in the cortical areas, whereas the
projection zone and terminates in layer 4 of an upstream cortical areas had little influence on these interactions
area, and (ii) a feedback pathway that originates from (Figure 3a; [58]). These functional thalamo-cortical
layer 6 of a cortical area and projects to its pulvinar interactions have been shown for extrastriate visual
projection zone. An important characteristic of the cortex [58], fronto-parietal cortex [59], and interactions
transthalamic feedforward pathway is that it indirectly between higher-order and extrastriate cortex [60], and
connects two cortical areas that share a direct cortico- may thus constitute a general principle of functional
cortical projection [56,57]. Based on this anatomical thalamo-cortical organization during attention. These
connectivity, pulvinar is in an ideal position to results are consistent with a role for pulvinar as ‘time-
coordinate functional interactions across a cortical keeper’, coordinating and optimizing functional in-
large-scale network [58]. This idea was probed in teractions across cortical networks to enhance the
studies that recorded simultaneously from two directly efficiency of signal processing between nodes.

Figure 3

Thalamic control and temporal dynamics of attention. a) Granger causality analysis shows pulvinar influences on cortical extrastriate areas V4 and
TEO (top) during an attention task when attention is directed at the RF (red) vs. away from the RF (blue). However, Granger causality analysis shows no
direct influences between the two extrastriate areas (bottom; adapted from [58]), suggesting that pulvinar initially organizes the cortical attention network in
time. b) Left: Detection behavior (as represented by hit-rate [HR] in a modified Egly-Driver task) is predicted by the phase of neural theta rhythms in three
major nodes involved in attention control: LIP, FEF, and pulvinar. Using a sine fit (black), HR is higher in one phase of theta (termed ‘good theta’) than in the
opposing phase (termed ‘poor theta’; adapted from [59,73]). Right: Functional connectivity between nodes in the attention network is modulated by neural
theta phase. Spike-LFP coupling between pulvinar and LIP in monkeys has two broad states during the cue-target delay interval of an attention task. In the
‘good theta’ phase, pulvinar spikes are coupled to LIP LFP in alpha frequencies (top), while LIP spikes are coupled to pulvinar LFP during the ‘poor theta’
(bottom; adapted from [59]). c) Schematic of connectivity between pulvinar and cortical regions involved in attention function. For cortico-cortical con-
nections (dotted black lines) there exists an indirect route through pulvinar (red lines). The directionality of the connectivity evolves dynamically over time,
organized by a theta rhythm; however, interactions beyond LIP, mediodorsal pulvinar, and FEF (panel B) during this rhythmic reweighing remain unexplored.
Further, whether similar temporal dynamics apply to areas engaged in feature-based attention is an open question (Box 1).

www.sciencedirect.com Current Opinion in Neurobiology 2022, 76:102605

6 Systems Neuroscience

Pulvinar inactivation studies have corroborated these The ‘sampling’ state emphasizes visual processing
ideas by showing that, as a consequence, functional (associated with better behavioral performance) and is
interactions between cortical areas are weakened characterized by greater synchronization of gamma
[61,62], demonstrating that cortico-cortical informa- (>35 Hz) and beta (w15e30 Hz) activity within and
tion transmission is greatly compromised without between frontal and parietal cortex [73]. Specifically,
pulvinar’s influences. Further, during pulvinar inacti- spiking activity of visual neurons is exclusively correlated
vation, neuronal activity in visual cortex is greatly with gamma activity, which has often been observed
attenuated [62,63], indicating a loss of responsiveness during enhanced visual processing [79], while spiking
to visual input. Thus, it is possible that pulvinar activity of oculomotor neurons is exclusively correlated
controls the excitability of neurons to respond to with beta activity, which has been linked to the sup-
visual information from downstream areas, thereby pression of motor actions [80,81]. The pulvinar co-
playing a role of an ‘enabler’ of cortical function [64]. ordinates these cortico-cortical dynamics across the
attention network only during the sampling state in
Attention and the dimension of time: similar ways as described in the previous section, thereby
Control of temporal dynamics underlining its proposed function in sensory gating
According to classical attention theories [15,65e67] and (Figure 3b; [59]). The ‘shifting’ state (associated with
our subjective experience, neural processing appears to relatively worse performance) is characterized by a
be continuous during attentional deployment. However, release of suppression in the motor system observed as
recent studies have shown that attention function un- decreased motor-suppressive beta oscillations in LIP and
folds over time and is characterized by alternating pe- FEF [73], which may also affect SC. This is accompanied
riods of relatively enhanced or diminished visual by greater synchronization of alpha activity between
processing at the attended location, which are associ- parietal cortex and pulvinar (Figure 3b; [59]), which is
ated with respective better or worse behavioral perfor- hypothesized to temporarily interrupt the transthalamic
mance in detecting visual targets. Behavioral studies gating function, thereby leading to a relative suppression
have shown that these alternations occur at a rhythm in of visual processing. The latter state provides windows of
the theta range (w4e6 Hz; [68e71]). Importantly, opportunities to disengage from the presently attended
these behavioral rhythms have been found both in location and to direct attention to a new location or event
humans and monkeys, thus constituting an evolutionary in the environment. The rhythmic alternations of sam-
conserved and possibly fundamental property of atten- pling and shifting states during attention deployment
tion [59,68,72]. provide critical flexibility, not only to prioritize visual
processing at an attended location, but also to provide
The behavioral rhythms have been linked to neuronal opportunities to re-allocate attentional resources, if
theta rhythms in fronto-parietal cortex and the thal- behaviorally desirable, without locking them into a
amus. The phase of the neuronal theta rhythm in particular state for extended periods of time [82]. Thus,
relation to target appearance in a visual detection task even when attention is sustained at a spatial location, the
is predictive of behavioral outcome on a trial-by-trial underlying network interactions are highly dynamic,
basis in both humans and monkeys, thereby linking providing a neural model for the flexibility of a cognitive
behavior and network-level neuronal signals function (see [83] for a circuit model).
(Figure 3b; [72,73]). The theta rhythm across the
fronto-parietal-thalamic network appears to present a Conclusions: Attentional control 2.0
mechanism that temporally coordinates potentially Taken together, findings from the recent literature
conflicting sensory and motor functions within the introduce new functional nodes and mechanisms that
network. Specifically, the network contains both are implicated in attention control beyond a fronto-
neuronal populations for visual processing that gets parietal, top-down framework. Substantial evidence
boosted during attentional deployment and for shift- points to pulvinar as a critical node that is involved in
ing attention or executing eye movements to a new the active propagation of attention control signals. Pu-
target of interest [74e78]. Sampling and shifting tative temporal control areas discussed in this review
cannot occur at the same time and require temporal [84,85], as well as SC [86,87] may communicate with
coordination of the associated neuronal populations fronto-parietal nodes using pathways involving pulvinar.
and the wider circuits they are embedded in. Theta- This places pulvinar as a potential hub for control signals
dependent attentional rhythms appear to set up two to route through efficiently, either as a means to coor-
alternating states e for ‘sampling’ and ‘shifting’ e that dinate processing between control areas or to propagate
are characterized by complex temporal dynamics from control nodes to sensory areas that exhibit atten-
across fronto-parietal cortex and the thalamus. tional modulation.

Current Opinion in Neurobiology 2022, 76:102605 www.sciencedirect.com

 Attention control in the primate Boshra and Kastner 7

Increasing evidence indicates a dynamic unfolding of Although future studies need to elucidate the role of the
attention function and its exerted control over time. reported temporal control areas and how they relate to
The large-scale network with thalamus at its center may prefrontal feature-based attention signals originating in
dynamically alter functional connectivities across nodes VPA (Box 1), current evidence indicates that neurons in
at timescales similar to those obtained in behavioral PITd/fSTS exhibit feature- and space-based attention
studies, setting up alternating attentional states that are effects as well as object selectivity [47,49]. The finding of
characterized by complex temporal network dynamics. spatial and non-spatial effects in that region and its
We propose that these shifting functional connectivities location in the visual processing hierarchy along the object
enable behavioral strategies for efficient exploration of processing ventral visual stream suggest that PITd/fSTS
the environment as well as provide a means to alternate may act as a control node that integrates sensory infor-
between enhanced sensory processing and motor- mation relating to the basic form, color, and motion and
engaged shifts of attention. Specifically, we propose may be involved in the control of object-based attention.
that these dynamic interactions, which are coordinated
by theta rhythms, not only influence the fronto-parietal In summary, we propose that the non-canonical nodes of
attention network, but may also modulate the putative attention control discussed in this review are integral
temporal control areas, as well as putative orienting parts of one conjoined large-scale network, or constitute
signals originating in SC (see Box 1). subdivisions of smaller subnetworks that function in
tandem in a dynamic fashion, as attention function un-
folds over time.
Box 1. Open Questions

Conflict of interest
Do thalamocortical interactions support and control other None declared.
cognitive functions? Higher-order thalamus, and particularly
pulvinar, has been shown to contribute to cognitive control. It is an
open question whether thalamocortical circuits support other Acknowledgments
cognitive functions in the primate brain, as suggested by recent This work was supported by grants from NEI (2R01EY017699; SK), NIMH
studies in rodents [88]. For example, working memory recruits (2R01MH064043, P50MH109429; SK), and NSERC (PDF-557604-
2021; RB).
overlapping mechanisms in PFC [89]. It will be an exciting venue
for future research to investigate whether similar thalamocortical
influences, possibly exerted by other nuclei such as the medi- References
odorsal nucleus, would subserve the control of working memory in Papers of particular interest, published within the period of review,
a manner similar to what has been found in attention function. have been highlighted as:
What is the role of temporal putative control areas in the large- * of special interest
scale primate attention network? The causal effects of PITd and * * of outstanding interest
fSTS perturbation on behavior and the puzzling functional simi-
larity to fronto-parietal control areas make temporal cortex a 1. Capotosto P, Tosoni A, Spadone S, Sestieri C, Perrucci MG,
fruitful target for future investigation using tasks that dissociate Romani GL, Della Penna S, Corbetta M: Anatomical segrega-
feature- from space-based attention effects, from a network tion of visual selection mechanisms in human parietal cortex.
perspective to probe its interactions with VPA, FEF, and LIP, and J Neurosci 2013, 33:6225–6229.
within the framework of rhythmic attention. 2. Caspari N, Janssens T, Mantini D, Vandenberghe R, Vanduffel W:
Do the temporal dynamics of attention factor into the control of Covert shifts of spatial attention in the macaque monkey.
feature-based or object-based attention? While recent work sug- J Neurosci 2015, 35:7695–7714.
gests that feature-based attention has rhythmic properties similar
3. Halassa MM, Kastner S: Thalamic functions in distributed
to spatial attention [90], it remains to be examined how that is cognitive control. Nat Neurosci 2017, 20:1669–1679.
reflected in underlying neural activity specific to feature-based
effects. That is particularly of interest in PITd/fSTS and VPA as 4. Kastner S, Ungerleider LG: Mechanisms of visual attention in
control areas that exhibit effects of both feature- and space- the human cortex. Annu Rev Neurosci 2000, 23:315–341.
based attention. 5. Krauzlis RJ, Lovejoy LP, Zénon A: Superior colliculus and
What is the role of the signals that pass from SC to temporal visual spatial attention. Annu Rev Neurosci 2013, 36:165–182.
cortex in attention function? The causal influence of SC on the
6. Womelsdorf T, Everling S: Long-range attention networks:
PITd/fSTS region suggests the existence of SC afferents, likely by circuit motifs underlying endogenously controlled stimulus
way of pulvinar, that modulate neuronal signals in temporal cortex. selection. Trends Neurosci 2015, 38:682–700.
However, the functional role of these signals is not clear, as SC
7. Inzlicht M, Bartholow BD, Hirsh JB: Emotional foundations of
inactivation only reduces attention effects in fSTS, but does not
cognitive control. Trends Cognit Sci 2015, 19:126–132.
abolish them. The ability to detect novel events is an orienting
function with localized cortical regions in humans [91–93] but not 8. Kouneiher F, Charron S, Koechlin E: Motivation and cognitive
in monkeys [94]. One hypothesis is that detection of novel events control in the human prefrontal cortex. Nat Neurosci 2009, 12:
939–945.
may be more closely linked to SC in monkeys. SC afferents in
fSTS may carry coarse signals that inform about the behavioral 9. Brain WR: Visual orientation with special reference to lesions
relevance of objects across the visual field that then propagate of the right cerebral hemisphere. Brain J Neurol 1941, 64:
from temporal cortex to other control areas either directly or 244–272.
through pulvino-cortical projections. 10. Critchley M: The parietal lobes. Williams & Wilkins; 1953.
11. Damasio AR, Damasio H, Chui HC: Neglect following damage
to frontal lobe or basal ganglia. Neuropsychologia 1980, 18:
123–132.

www.sciencedirect.com Current Opinion in Neurobiology 2022, 76:102605

8 Systems Neuroscience

12. Karnath HO, Himmelbach M, Rorden C: The subcortical anat- 34. Hayden BY, Gallant JL: Combined effects of spatial and
omy of human spatial neglect: putamen, caudate nucleus and feature-based attention on responses of V4 neurons. Vis Res
pulvinar. Brain J Neurol 2002, 125:350–360. 2009, 49:1182–1187.
13. Karnath H-O, Ferber S, Himmelbach M: Spatial awareness is a 35. McAdams CJ, Maunsell JHR: Attention to both space and
function of the temporal not the posterior parietal lobe. Nature feature modulates neuronal responses in macaque area V4.
2001, 411:950–953. J Neurophysiol 2000, 83:1751–1755.
14. Mort DJ, Malhotra P, Mannan SK, Rorden C, Pambakian A, 36. Hayden BY, Gallant JL: Time course of attention reveals
Kennard C, Husain M: The anatomy of visual neglect. Brain J different mechanisms for spatial and feature-based attention
Neurol 2003, 126:1986–1997. in area V4. Neuron 2005, 47:637–643.
15. Nobre Anna C, Kastner Sabine. The Oxford handbook of atten- 37. Hopf J-M: Attention to features precedes attention to
tion. Oxford University Press; 2014. locations in visual search: evidence from electromagnetic
brain responses in humans. J Neurosci 2004, 24:1822–1832.
16. Kastner S, Pinsk MA, De Weerd P, Desimone R, Ungerleider LG:
Increased activity in human visual cortex during directed 38. Zhang W, Luck SJ: Feature-based attention modulates
attention in the absence of visual stimulation. Neuron 1999, feedforward visual processing. Nat Neurosci 2009, 12:24–25.
22:751–761.
39. Serences JT: Control of object-based attention in human
17. Silver MA, Kastner S: Topographic maps in human frontal and cortex. Cerebr Cortex 2004, 14:1346–1357.
parietal cortex. Trends Cognit Sci 2009, 13:488–495.
40. Baldauf D, Desimone R: Neural mechanisms of object-based
18. Szczepanski SM, Konen CS, Kastner S: Mechanisms of spatial attention. Science 2014, 344:424–427.
attention control in frontal and parietal cortex. J Neurosci
2010, 30:148–160. 41. Zhou H, Desimone R: Feature-based attention in the frontal
eye field and area V4 during visual search. Neuron 2011, 70:
19. Kinsbourne M: Hemi-neglect and hemisphere rivalry. Adv 1205–1217.
Neurol 1977, 18:41–49.
42. Bichot NP, Heard MT, DeGennaro EM, Desimone R: A source
20. Szczepanski SM, Pinsk MA, Douglas MM, Kastner S, * * for feature-based attention in the prefrontal cortex. Neuron
Saalmann YB: Functional and structural architecture of the 2015, 88:832–844.
human dorsal frontoparietal attention network. Proc Natl Acad This is the first report of VPA as a source for feature-based attention
Sci USA 2013, 110:15806–15811. effects. The authors recorded from multiple prefrontal regions and uti-
lized latency analyses to discover that feature-based effects were
21. Squire RF, Noudoost B, Schafer RJ, Moore T: Prefrontal con- observed earlier in VPA as compared to FEF. An additional VPA inacti-
tributions to visual selective attention. Annu Rev Neurosci vation experiment elegantly showed a causal effect on FEF signals,
2013, 36:451–466. conserving space-based attention but abolishing feature-based effects.
22. Bisley JW, Goldberg ME: Neuronal activity in the lateral intra- 43. Bichot NP, Xu R, Ghadooshahy A, Williams ML, Desimone R: The
parietal area and spatial attention. Science 2003, https:// * role of prefrontal cortex in the control of feature attention in
doi.org/10.1126/science.1077395. area V4. Nat Commun 2019, 10:5727.
Using reversible inactivation, this study examined the role of VPA in the
23. Cook EP, Maunsell JHR: Attentional modulation of behavioral
control of feature-based attention in V4. VPA inactivation decreased
performance and neuronal responses in middle temporal and
feature-based attention effects in V4 neurons and reduced the ability of
ventral intraparietal areas of macaque monkey. J Neurosci
the animals to locate the target during a visual search task.
2002, 22:1994–2004.
44. Sani I, McPherson BC, Stemmann H, Pestilli F, Freiwald WA:
24. Premereur E, Janssen P, Vanduffel W: Effector specificity in * Functionally defined white matter of the macaque monkey brain
macaque frontal and parietal cortex. J Neurosci 2015, 35:
reveals a dorso-ventral attention network. Elife 2019, 8, e40520.
3446–3459.
Using fMRI and high resolution ex-vivo diffusion MRI, the study showed
25. Cameron EL, Tai JC, Carrasco M: Covert attention affects the compelling evidence of anatomical connectivity between PITd and
psychometric function of contrast sensitivity. Vis Res 2002, fronto-parietal attention areas FEF and LIP. The study highlights a
42:949–967. potential interaction of the three areas during attention function.

26. Moore T, Fallah M: Microstimulation of the frontal eye field and 45. Schall J, Morel A, King D, Bullier J: Topography of visual cortex
its effects on covert spatial attention. J Neurophysiol 2004, 91: connections with frontal eye field in macaque: convergence
152–162. and segregation of processing streams. J Neurosci 1995, 15:
4464–4487.
27. Moore T, Armstrong KM: Selective gating of visual signals by
microstimulation of frontal cortex. Nature 2003, 421: 46. Stemmann H, Freiwald WA: Attentive motion discrimination
370 – 373. * recruits an area in inferotemporal cortex. J Neurosci 2016, 36:
11918–11928.
28. Ruff CC, Blankenburg F, Bjoertomt O, Bestmann S, Freeman E, Using whole brain fMRI in monkeys, PITd was found to be
Haynes J-D, Rees G, Josephs O, Deichmann R, Driver J: Concur- modulated during an attentive motion-discrimination task. Using several
rent TMS-fMRI and psychophysics reveal frontal influences on task variations, the authors identify the area to be strongly modulated by
human retinotopic visual cortex. Curr Biol 2006, 16:1479–1488. attention while not exhibiting signs of motion selectivity. These findings
identify PITd as a putative attention control node in a manner that ex-
29. Gregoriou GG, Rossi AF, Ungerleider LG, Desimone R: Lesions tends beyond classical views of fronto-parietal attention control.
of prefrontal cortex reduce attentional modulation of
neuronal responses and synchrony in V4. Nat Neurosci 2014, 47. Stemmann H, Freiwald WA: Evidence for an attentional priority
17:1003–1011. * * map in inferotemporal cortex. Proc Natl Acad Sci USA 2019,
116:23797–23805.
30. Awh E, Belopolsky AV, Theeuwes J: Top-down versus bottom- The study presents a thorough electrophysiological investigation of the
up attentional control: a failed theoretical dichotomy. Trends PITd region, identified by the same group using neuroimaging in an
Cognit Sci 2012, 16:437–443. earlier study. Using a combination of experiments, the authors show that
PITd responses fit the characteristics for an attention priority map
31. Maunsell JHR, Treue S: Feature-based attention in visual
including strong attentional modulation, spatial representation of
cortex. Trends Neurosci 2006, 29:317–322.
receptive fields across the visual field, visual responsiveness to multiple
32. Serences JT, Boynton GM: Feature-based attentional modu- feature types, and limited tuning to feature dimensions. Using electric
lations in the absence of direct visual stimulation. Neuron microstimulation, PITd was also causally linked to shifting the locus of
2007, 55:301–312. attention, emphasizing its role as a putative attention control node.

33. Greenberg AS, Esterman M, Wilson D, Serences JT, Yantis S: 48. Bogadhi AR, Bollimunta A, Leopold DA, Krauzlis RJ: Spatial
Control of spatial and feature-based attention in frontopar- * * attention deficits are causally linked to an area in macaque
ietal cortex. J Neurosci 2010, 30:14330–14339. temporal cortex. Curr Biol 2019, 29:726–736.e4.

Current Opinion in Neurobiology 2022, 76:102605 www.sciencedirect.com

 Attention control in the primate Boshra and Kastner 9

Using a combination of whole-brain neuroimaging and causal pertur- coherence between V4 and LIP in low frequencies, but had no local
bation, a region at the floor of STS was localized that showed greater effect on LFP activity in either V4 or LIP.
decreases in fMRI activity than FEF and LIP following SC inactivation.
Behaviorally, signs of hemi-spatial neglect were observed during an 62. Zhou H, Schafer RJ, Desimone R: Pulvinar-cortex interactions
attention task. Inactivation of this STS region also affected behavior, but in vision and attention. Neuron 2016, 89:209–220.
did not influence oculomotor function, emphasizing a degree of func-
63. Purushothaman G, Marion R, Li K, Casagrande VA: Gating and
tional specialization among the control nodes in the attention network.
control of primary visual cortex by pulvinar. Nat Neurosci
49. Bogadhi AR, Katz LN, Bollimunta A, Leopold DA, Krauzlis RJ: 2012, 15:905–912.
* Midbrain activity shapes high-level visual properties in the
64. Lakatos P, O’Connell MN, Barczak A: Pondering the pulvinar.
primate temporal cortex. Neuron 2021, 109:690–699. e5.
Neuron 2016, 89:5–7.
Bogadhi et al. investigated the electrophysiological properties of fSTS
neurons and the effect of SC inactivation during a motion detection 65. Petersen SE, Posner MI: The attention system of the human
task. fSTS neurons exhibited attentional modulation that was reduced brain: 20 Years after. Annu Rev Neurosci 2012, 35:73–89.
following SC inactivation. Neural activity in fSTS also exhibited effects
of event-detection and object-selectivity, both of which were reduced 66. Posner MI: Cognition and neural systems. Cognition 1981, 10:
following SC inactivation. While delineation between fSTS and PITd is 261–266.
not clear, these findings, in conjunction with previous neuroimaging
work on fSTS, highlight the involvement of a node in temporal cortex in 67. Treisman A: Features and objects: the fourteenth bartlett
attention control that is modulated by SC activity. memorial lecture. Q J Exp Psychol Sect A 1988, 40:201–237.

50. Kastner S, Fiebelkorn IC, Eradath MK: Dynamic pulvino-cortical 68. Fiebelkorn IC, Saalmann YB, Kastner S: Rhythmic sampling
interactions in the primate attention network. Curr Opin within and between objects despite sustained attention at a
Neurobiol 2020, 65:10–19. cued location. Curr Biol 2013, 23:2553–2558.

51. Baldwin MKL, Balaram P, Kaas JH: The evolution and functions 69. Landau AN, Fries P: Attention samples stimuli rhythmically.
of nuclei of the visual pulvinar in primates. J Comp Neurol Curr Biol 2012, 22:1000–1004.
2017, 525:3207–3226.
70. Song K, Meng M, Chen L, Zhou K, Luo H: Behavioral oscilla-
52. Chalfin BP, Cheung DT, Muniz JAPC, de Lima Silveira LC, tions in attention: rhythmic a pulses mediated through q
Finlay BL: Scaling of neuron number and volume of the band. J Neurosci 2014, 34:4837–4844.
pulvinar complex in new world primates: comparisons with
71. VanRullen R, Carlson T, Cavanagh P: The blinking spotlight of
humans, other primates, and mammals. J Comp Neurol 2007,
attention. Proc Natl Acad Sci USA 2007, 104:19204–19209.
504:265–274.
72. Helfrich RF, Fiebelkorn IC, Szczepanski SM, Lin JJ, Parvizi J,
53. Smaers JB, Gómez-Robles A, Parks AN, Sherwood CC: * * Knight RT, Kastner S: Neural mechanisms of sustained
Exceptional evolutionary expansion of prefrontal cortex in
attention are rhythmic. Neuron 2018, 99:854–865. e5.
great apes and humans. Curr Biol 2017, 27:714–720.
Using electrocorticography recordings obtained during two attention
54. Marion R, Li K, Purushothaman G, Jiang Y, Casagrande VA: tasks in humans, the study demonstrated that neural substrates of
Morphological and neurochemical comparisons between rhythmic attention, which have been shown in monkeys, are indeed
pulvinar and V1 projections to V2. J Comp Neurol 2013, 521: conserved in humans in states of apparent sustained attention to
813–832. cued locations. Cortical excitability, indexed by high-frequency band
activity, was found to be modulated by theta rhythms in the human
55. Markov NT, Vezoli J, Chameau P, Falchier A, Quilodran R, fronto-parietal network in both tasks, shaping behavioral
Huissoud C, Lamy C, Misery P, Giroud P, Ullman S, et al.: performance.
Anatomy of hierarchy: feedforward and feedback pathways
in macaque visual cortex. J Comp Neurol 2014, 522:225–259. 73. Fiebelkorn IC, Pinsk MA, Kastner S: A dynamic interplay within
* * the frontoparietal network underlies rhythmic spatial atten-
56. Adams MM, Hof PR, Gattass R, Webster MJ, Ungerleider LG: tion. Neuron 2018, 99:842–853.e8.
Visual cortical projections and chemoarchitecture of ma- Simultaneous recordings from monkey fronto-parietal areas LIP and
caque monkey pulvinar. J Comp Neurol 2000, 419:377–393. FEF reveal a distinct reweighing of functional connectivity between the
two areas during an attention task coinciding with the phases of
57. Shipp S: The functional logic of cortico–pulvinar connec- neuronal theta rhythms. Findings support that phase of neural theta
tions. Philos Trans R Soc Lond B Biol Sci 2003, 358:1605–1624. delineates two states: ‘good theta’ phase is dominated by LIP gamma-
band activity and characterized by enhanced sensory sampling asso-
58. Saalmann YB, Pinsk MA, Wang L, Li X, Kastner S: The pulvinar
ciated with high behavioral performance, while the ‘poor theta’ phase is
regulates information transmission between cortical areas
dominated by high FEF beta-band activity and is characterized by low
based on attention demands. Science 2012, 337:753–756.
behavioral performance.
59. Fiebelkorn IC, Pinsk MA, Kastner S: The mediodorsal pulvinar
* 74. Corbetta M, Akbudak E, Conturo TE, Snyder AZ, Ollinger JM,
coordinates the macaque fronto-parietal network during
Drury HA, Linenweber MR, Petersen SE, Raichle ME, Van
rhythmic spatial attention. Nat Commun 2019, 10:215.
Essen DC, et al.: A common network of functional areas for
Using simultaneous recordings from FEF, LIP, and mediodorsal pulvi-
attention and eye movements. Neuron 1998, 21:761–773.
nar, this study demonstrated the rhythmic reweighing of connectivities
between pulvinar and the fronto-parietal network during an attention 75. Fiebelkorn IC, Kastner S: Functional specialization in the
task. Increased behavioral performance aligned with a phase of neural attention network. Annu Rev Psychol 2020, 71:221–249.
theta (termed ‘good theta’) and corresponded to activity propagating
from pulvinar to cortex. Conversely, poor behavioral performance 76. Gregoriou GG, Gotts SJ, Desimone R: Cell-type-specific syn-
aligned with ‘poor theta’ phase and corresponded to neural activity chronization of neural activity in FEF with V4 during atten-
propagating from LIP to pulvinar. tion. Neuron 2012, 73:581–594.
60. Saalmann YB, Ly R, Pinsk MA, Kastner S: Pulvinar influences 77. Moore T, Fallah M: Control of eye movements and spatial
parietal delay activity and information transmission between attention. Proc Natl Acad Sci USA 2001, 98:1273–1276.
dorsal and ventral visual cortex in macaques. bioRxiv 2018,
https://doi.org/10.1101/405381. 78. Thompson KG: Neuronal basis of covert spatial attention in
the frontal eye field. J Neurosci 2005, 25:9479–9487.
61. Eradath MK, Pinsk MA, Kastner S: A causal role for the pulvinar
* in coordinating task-independent cortico–cortical in- 79. Fries P, Reynolds JH, Rorie AE, Desimone R: Modulation of
teractions. J Comp Neurol 2021, 529:3772–3784. oscillatory neuronal synchronization by selective visual
Using reversible inactivation and simultaneous electrophysiological attention. Science 2001, 291:1560–1563.
recordings from V4, LIP, and their projection zone in the dorsal part of
80. Pogosyan A, Gaynor LD, Eusebio A, Brown P: Boosting cortical
lateral pulvinar, this study shows the causal influence of pulvinar on
activity at beta-band frequencies slows movement in
cortico-cortical connectivity. Pulvinar inactivation reduced phase
humans. Curr Biol 2009, 19:1637–1641.

www.sciencedirect.com Current Opinion in Neurobiology 2022, 76:102605

10 Systems Neuroscience

81. Zhang Y, Chen Y, Bressler SL, Ding M: Response preparation The study utilized simultaneous multi-area recordings and multivariate
and inhibition: the role of the cortical sensorimotor beta decoding techniques to show that in prefrontal cortex, both attention and
rhythm. Neuroscience 2008, 156:238–246. working memory utilize a similar mechanism of representing encoded
information relevant to task demands. The data, collected during a color
82. Fiebelkorn IC, Kastner S: A rhythmic theory of attention. matching task, investigated neural activity underlying 1) maintenance of
Trends Cognit Sci 2019, 23:87–101. information, and 2) attending to a cued location, and showed that neural
activity pertaining to both tasks share similar neural correlates in pre-
83. Aussel A, Fiebelkorn IC, Kastner S, Kopell NJ, Pittman-
frontal areas, but not in sensory areas such as V4, supporting the idea of
Polletta BR. In Interacting rhythms enhance sensitivity of target
a shared control resource that serves both cognitive functions.
detection in a fronto-parietal computational model of visual
attention; 2021, https://doi.org/10.1101/2021.02.18.431872. 90. Re D, Inbar M, Richter CG, Landau AN: Feature-based attention
* * samples stimuli rhythmically. Curr Biol 2019, 29:693–699.e4.
84. Kaas JH, Lyon DC: Pulvinar contributions to the dorsal and
Using a behavioral task composed of overlaying color clouds, this study
ventral streams of visual processing in primates. Brain Res
suggests that rhythmicity of attention does not only pertain to space-based
Rev 2007, 55:285–296.
selection, but also to feature-based selections. The authors show that
85. Yeterian EH, Pandya DN: Corticothalamic connections of the while attention rhythmically samples a single object at 8 Hz, sampling of
superior temporal sulcus in rhesus monkeys. Exp Brain Res features from two different objects takes place at 4 Hz per object.
1991, 83.
91. Vossel S, Geng JJ, Fink GR: Dorsal and ventral attention
86. Harting JK, Huerta MF, Frankfurter AJ, Strominger NL, Royce GJ: systems: distinct neural circuits but collaborative roles.
Ascending pathways from the monkey superior colliculus: an Neuroscientist 2014, 20:150–159.
autoradiographic analysis. J Comp Neurol 1980, 192:853–882.
92. Corbetta M, Patel G, Shulman GL: The reorienting system of
87. Rovó Z, Ulbert I, Acsády L: Drivers of the primate thalamus. the human brain: from environment to theory of mind. Neuron
J Neurosci 2012, 32:17894–17908. 2008, 58:306–324.

88. Nakajima M, Halassa MM: Thalamic control of functional 93. Corbetta M, Shulman GL: Spatial neglect and attention net-
cortical connectivity. Curr Opin Neurobiol 2017, 44:127–131. works. Annu Rev Neurosci 2011, 34:569–599.

89. Panichello MF, Buschman TJ: Shared mechanisms underlie 94. Patel GH, Yang D, Jamerson EC, Snyder LH, Corbetta M, Ferrera VP:
* * the control of working memory and attention. Nature 2021, Functional evolution of new and expanded attention networks in
592:601–605. humans. Proc Natl Acad Sci USA 2015, 112:9454–9459.

Current Opinion in Neurobiology 2022, 76:102605 www.sciencedirect.com