cerebellum and nonmotor function - xurxo mariño...cerebellum and nonmotor function peter l....

ANRV379-NE32-17 ARI 10 May 2009 9:35

Cerebellum and NonmotorFunctionPeter L. Strick,1,2,3 Richard P. Dum,2,3

and Julie A. Fiez2,4

1Veterans Affairs Medical Center, Pittsburgh, Pennsylvania 15261; 2Center for the NeuralBasis of Cognition, 3Systems Neuroscience Institute and the Department of Neurobiology,4Learning Research and Development Center and Department of Psychology,University of Pittsburgh, Pittsburgh, Pennsylvania 15260; email: [email protected]

Annu. Rev. Neurosci. 2009. 32:413–34

The Annual Review of Neuroscience is online atneuro.annualreviews.org

This article’s doi:10.1146/annurev.neuro.31.060407.125606

Copyright c© 2009 by Annual Reviews.All rights reserved

0147-006X/09/0721-0413$20.00

Key Words

prefrontal cortex, parietal cortex, dentate nucleus, brain mapping

AbstractDoes the cerebellum influence nonmotor behavior? Recent anatomicalstudies demonstrate that the output of the cerebellum targets multiplenonmotor areas in the prefrontal and posterior parietal cortex, as wellas the cortical motor areas. The projections to different cortical areasoriginate from distinct output channels within the cerebellar nuclei.The cerebral cortical area that is the main target of each output channelis a major source of input to the channel. Thus, a closed-loop circuit rep-resents the major architectural unit of cerebro-cerebellar interactions.The outputs of these loops provide the cerebellum with the anatomicalsubstrate to influence the control of movement and cognition. Neu-roimaging and neuropsychological data supply compelling support forthis view. The range of tasks associated with cerebellar activation isremarkable and includes tasks designed to assess attention, executivecontrol, language, working memory, learning, pain, emotion, and ad-diction. These data, along with the revelations about cerebro-cerebellarcircuitry, provide a new framework for exploring the contribution of thecerebellum to diverse aspects of behavior.

413

Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

M1: primary motorcortex

Contents

CEREBELLAR OUTPUTCHANNELS. . . . . . . . . . . . . . . . . . . . . . 416

TOPOGRAPHY OF FUNCTIONWITHIN THE DENTATE . . . . . . . 419

MOLECULAR GRADIENTSIN THE DENTATE . . . . . . . . . . . . . . 419

HUMAN DENTATE . . . . . . . . . . . . . . . . 420MACRO-ARCHITECTURE OF

CEREBRO-CEREBELLARLOOPS . . . . . . . . . . . . . . . . . . . . . . . . . . . 420WHAT IS THE FULL EXTENT

OF THE CEREBELLARINFLUENCE OVERNONMOTOR AREAS OFTHE CEREBRAL CORTEX? . . 422

CEREBELLO-LIMBIC CIRCUIT . . . 422WHAT IS THE FUNCTIONAL

IMPACT OF NONMOTORCEREBRO-CEREBELLARLOOPS? . . . . . . . . . . . . . . . . . . . . . . . . . . 423

HOW MIGHT THE CEREBELLUMINFLUENCE COGNITIONAND AFFECT? . . . . . . . . . . . . . . . . . . . 425

TOPOGRAPHICALINFORMATION CAN BE USEDTO ADVANCE THEORY . . . . . . . . 426

SUMMARY ANDCONCLUSIONS . . . . . . . . . . . . . . . . . 429

“In biology, if seeking to understand function,it is usually a good idea to study structure.”(Crick & Koch 2005, p. 6)

This statement by Crick & Koch is espe-cially applicable to the functions of the cere-bellum. We believe that important insightsinto cerebellar function can be gained froman anatomical analysis of cerebellar output.Indeed, recent results indicate that a majorcomponent of cerebellar output targets non-motor areas of the cerebral cortex. This fea-ture of cerebellar structure suggests the cere-bellum influences not only the generation and

control of movement but also nonmotor aspectsof behavior.

In this chapter we review new evidenceabout the areas of the cerebral cortex that arethe target of cerebellar output. We describethe functional map that has recently been dis-covered within one of the major output nu-clei of the cerebellum, the dentate nucleus. Wepresent evidence that the fundamental unit ofcerebro-cerebellar operations is a closed-loopcircuit. Finally, we provide clear examples fromimaging and lesion studies that the cerebellumhas a nonmotor function.

The cerebellum is massively interconnectedwith the cerebral cortex. The classical view ofthese interconnections is that the cerebellumreceives information from widespread corticalareas, including portions of the frontal, pari-etal, temporal, and occipital lobes (Figure 1)(Glickstein et al. 1985, Schmahmann 1996).This information was then thought to be fun-neled through cerebellar circuits where it ulti-mately converged on the ventrolateral nucleusof the thalamus (e.g., Allen & Tsukahara 1974,Brooks & Thach 1981). The ventrolateral nu-cleus was believed to project to a single corti-cal area, the primary motor cortex (M1). Thus,cerebellar connections with the cerebral cortexwere viewed as a means of collecting informa-tion from widespread regions of the cerebralcortex. The cerebellum was thought to per-form a sensorimotor transformation on its in-puts and convey the results to M1 for the gen-eration and control of movement. According tothis view, cerebellar output was entirely withinthe domain of motor control, and abnormal ac-tivity in this circuit would lead to purely motordeficits.

Our continuing analysis of cerebellar out-put and function has led us to challenge thisview (e.g., Akkal et al. 2007; Ben-Yehudah &Fiez 2008; Ben-Yehudah et al. 2007; Cloweret al. 2001, 2005; Dum & Strick 2003; Fiez1996; Hoover & Strick 1999; Kelly & Strick2003; Middleton & Strick 1994, 1996a, b, 2001;Schell & Strick 1984). It is now clear that ef-ferents from the cerebellar nuclei project to

414 Strick · Dum · Fiez

Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

multiple subdivisions of the ventrolateral tha-lamus (for a review, see Percheron et al. 1996),which, in turn, project to a myriad of cortical ar-eas, including regions of frontal, prefrontal, andposterior parietal cortex ( Jones 1985). Thus,the outputs from the cerebellum influence morewidespread regions of the cerebral cortex thanpreviously recognized. This change in per-spective is important because it provides theanatomical substrate for the output of the cere-bellum to influence nonmotor as well as motorareas of the cerebral cortex. As a consequence,abnormal activity in these circuits could leadnot only to motor deficits but also to cognitive,attentional, and affective impairments. Below,we provide additional support for this new per-spective based on the analysis of focal lesions ofthe cerebellum and activation of the cerebellumduring nonmotor tasks.

Prior neuroanatomical approaches for ex-amining cerebro-cerebellar circuits have beenhindered by a number of technical limitations.Chief among these is the multisynaptic natureof these pathways and the general inability ofconventional tracers to label more than the di-rect inputs and outputs of an area. To overcomethese and other problems, we developed the useof neurotropic viruses as transneuronal tracersin the central nervous system of primates (forreferences and a review, see Kelly & Strick 2000,2003; Strick & Card 1992). Selected strains ofvirus move transneuronally in either the retro-grade or anterograde direction (Kelly & Strick2003, Zemanick et al. 1991). Thus, one can ex-amine either the inputs to or the outputs froma site. The viruses we use as tracers move fromneuron to neuron exclusively at synapses, andthe transneuronal transport occurs in a time-dependent fashion. By careful adjustment of thesurvival time after a virus injection, we wereable to study neural circuits composed of twoor even three synaptically connected neurons.We used virus tracing to examine cerebello-thalamocortical pathways to a wide variety ofcortical areas (Akkal et al. 2007; Clower et al.2001, 2005; Hoover & Strick 1999; Kelly &Strick 2003; Lynch et al. 1994; Middleton &Strick 1994, 1996a, b, 2001) (Figure 2).

PO

CA

CgS

PS Ai

AsIP

PO

LuSTS

LS

CS

IO

Or 9

Brodmann Area

Num

ber o

f lab

eled

cel

ls/m

m

8 22 27 20 21 A153

2

412 6 19 24c 23 In7 1718

2

4

6

8

12

14

16

18

20

10

46 1 24 SII

Figure 1Origin of projections from the cerebral cortex to the cerebellum. (Top) Therelative density of corticopontine neurons is indicated by the dots on lateral andmedial views of the macaque brain. (Bottom) Histogram of relative density ofcorticopontine cells in different cytoarchitectonic areas of the monkey. Ai, As,inferior and superior limbs of arcuate sulcus, respectively; CA, calcarine fissure;CgS, cingulate sulcus; CS, central sulcus; IP, intraparietal sulcus; LS, lateralsulcus; Lu, lunate sulcus; IO, inferior occipital sulcus; PO, parietal-occipitalsulcus; PS, principal sulcus; STS, superior temporal sulcus. Adapted fromGlickstein et al. 1985, published in The Journal of Comparative Neurology,Vol. 235, No. 3, 1985, pp. 343–59. Copyright c© 1985. Alan R. Liss, Inc.Reprinted with permission of John Wiley & Sons, Inc.

www.annualreviews.org • Cerebellum and Nonmotor Function 415

Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Pre-SMA

SMAarm

9l

9m

AIP

46d

M1face

PMvarm

PMdarm

M1arm

M1leg

PS

FEF

ST

LS

CgS

10 mm

Pre-PMd

12

TE

Lu

IP

AS

CS

7b

7a

46v

Figure 2Targets of cerebellar output. Red labels indicate areas of the cerebral cortex that are the target of cerebellaroutput. Blue labels indicate areas that are not the target of cerebellar output. These areas are indicated onlateral and medial views of the cebus monkey brain. The numbers refer to cytoarchitectonic areas. AIP,anterior intraparietal area; AS, arcuate sulcus; CgS, cingulate sulcus; FEF, frontal eye field; IP, intraparietalsulcus; LS, lateral sulcus; Lu, lunate sulcus; M1, face, arm, and leg areas of the primary motor cortex; PMdarm, arm area of the dorsal premotor area; PMv arm, arm area of the ventral premotor area; PrePMd,predorsal premotor area; PreSMA, presupplementary motor area; PS, principal sulcus; SMA arm, arm areaof the supplementary motor area; ST, superior temporal sulcus; TE, area of inferotemporal cortex.

CEREBELLAR OUTPUTCHANNELSIn our initial studies, we injected virus intophysiologically defined portions of M1 and setthe survival time to label second-order neuronsin the deep cerebellar nuclei (Hoover & Strick1999). In general, cerebellar projections to M1originate largely from neurons in the dentatenucleus (75%), although a smaller componentalso originates from interpositus (25%). Ourdescription focuses on the organization ofthe dentate. The dentate nucleus is a com-plex three-dimensional structure (Figure 3).Therefore, we created an unfolded map of thenucleus to display observations from different

experiments in a common framework(Figure 4) (Dum & Strick 2003).

Virus transport following injections into thearm representation of M1 labeled a compactcluster of neurons in the dorsal portion of thedentate at mid-rostro-caudal levels (Figure 2;Figure 3, far right panel; Figure 4, top centerpanel). Virus transport from the leg representa-tion of M1 labeled neurons in the rostral pole ofthe dorsal dentate (Figure 2; 4, top left panel),whereas virus transport from the face represen-tation labeled neurons at caudal levels of thedorsal dentate (Figures 2; Figure 4, top rightpanel). Clearly, each cortical area receives inputfrom a spatially separate set of neurons in the


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

D

M1 mm

PMvarm

Area 46Area 9lArea 9m M1arm

Figure 3Output channels in the dentate. The dots on representative coronal sections show the location of dentateneurons that project to a specific area of the cerebral cortex in the cebus monkey. The cortical target isindicated above each section. [Abbreviations are according to Figure 2 (M1, primary motor cortex; PMv,ventral premotor area)]. D, dorsal; M, medial. Adapted from Middleton & Strick 1996b.

dentate, which we have termed an output chan-nel (Middleton & Strick 1997). The rostral tocaudal sequence of output channels to the leg,arm, and face representations in M1 (Figure 4,top panels; Figure 5) corresponds well with thesomatotopic organization of the dentate pre-viously proposed on the basis of physiologicalstudies (e.g., Allen et al. 1978, Stanton 1980,Rispal-Padel et al. 1982, Asanuma et al. 1983,Thach et al. 1993).

The region of the dentate that contains neu-rons that project to M1 occupies only 30% ofthe nucleus (Dum & Strick 2003, Hoover &Strick 1999). This implies that a substantial por-tion of the dentate projects to cortical targetsother than M1. To test this proposal and definethe cortical targets of the unlabeled regions ofthe dentate, we injected virus into selected pre-motor, prefrontal, and posterior parietal areasof the cortex (Figure 2).

Virus transport from the arm representa-tions of the ventral premotor area (PMv) andthe supplementary motor area (SMA) demon-strated that both cortical areas are the target ofcerebellar output (Figure 2) (Akkal et al. 2007,Middleton & Strick 1997). The output channelsto these premotor areas are located in the sameregion of the dentate that contains the outputchannel to arm M1 (Figures 3, 5). We havespeculated that the clustering of output chan-nels to M1 and the premotor areas in the dorsalregion of the dentate creates a motor domain

within the nucleus (Figure 5) (Dum & Strick2003). Furthermore, the output channels to thearm representations of these motor areas appearto be in register within the dentate. This raisesthe possibility that the nucleus contains a sin-gle integrated map of the body within the motordomain.

Virus transport following injections intoprefrontal cortex revealed that some subfieldsare the target of dentate output, whereas othersare not (Middleton & Strick 1994, 2001). Den-tate output channels project to areas 9m, 9l, and46d, but not to areas 12 and 46v (Figures 2–5).Importantly, the extent of the dentate that is oc-cupied by an output channel to a specific area ofprefrontal cortex is comparable to that occupiedby an output channel to a cortical motor area(Figure 4). Thus, it is likely that the signal fromthe dentate to prefrontal cortex is as importantas its signal to one of the cortical motor areas.In addition, dentate output channels to areas ofprefrontal cortex are located in a different re-gion of the nucleus than the output channels tothe cortical motor areas. The output channelsto prefrontal cortex are clustered together ina ventral region of the nucleus that is entirelyoutside the motor domain. The output chan-nels to prefrontal cortex are also rostral to theoutput channel that targets the frontal eye field(Lynch et al. 1994).

Although the presupplementary motor area(PreSMA) has traditionally been included with


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

M1 leg (Jo17) M1 arm (Jo19) M1 face (Jo18)

Do

rsa

lD

ors

al

Area 46 (F11R) Area 9L (F11L) PreSMA (DA7L)

5 mm

13–1610–128–94–7

7–1264–52–3

13–1610–127–9

3–421

2–6

11–13107–93–6

10–1197–83–6

Do

rsa

lV

en

tra

lV

en

tra

lD

ors

al

Figure 4Unfolded maps of the dentate: output channels to different areas of the cerebralcortex in the cebus monkey. (Top panels) Somatotopic organization of outputchannels to leg, arm, and face M1 in the dorsal dentate. (Bottom panels) Ventrallocation of output channels to prefrontal cortex. The key below each diagramindicates the density of neurons in bins through the nucleus. Rostral is to theleft. [Abbreviations are according to Figure 1 (M1, primary motor cortex;PreSMA, presupplementary motor area)]. Adapted from Dum & Strick 2003(which includes a detailed description of the unfolding method).

the motor areas of the frontal lobe, we arguethat it should be considered a region of pre-frontal cortex (for reviews, see Akkal et al. 2007,Picard & Strick 2001). In support of this pro-posal, virus transport from the PreSMA labeledan output channel in the ventral dentate wherethe output channels to areas 9 and 46 are lo-cated (Figure 2; Figure 4, bottom; Figure 5).

This result illustrates that the topographic ar-rangement of output channels in the dentatedoes not mirror the arrangement of their tar-gets in the cerebral cortex. For example, thePreSMA is adjacent to the SMA on the me-dial surface of the hemisphere (Figure 2), butthe output channels to the two cortical areas arespatially separated from one another in the den-tate (Figure 5). Thus, the topographic arrange-ment of output channels in the dentate appearsto reflect functional relationships between cor-tical areas rather than the spatial relationshipsamong them.

Virus transport from regions of posteriorparietal cortex demonstrated that some of itssubfields are also the target of output channelslocated in the dentate (Figures 2, 5) (Cloweret al. 2001, 2005). For example, area 7b, whichin the cebus monkey is located laterally in theintraparietal sulcus, is the target of an outputchannel located ventrally in the caudal pole ofthe dentate (Figure 5). A second region of pos-terior parietal cortex, the anterior intraparietalarea (AIP), receives a focal projection from asmall cluster of neurons that is located dorsallyin the dentate at mid-rostro-caudal levels. Inaddition, the AIP receives a broadly distributedprojection from neurons that are scattered indentate regions that contain output channels toM1, the PMv, and perhaps other premotor ar-eas. This creates a unique situation in whichAIP may receive a sample of the dentate outputthat is streaming to motor areas in the frontallobe, as well as input from its own separate out-put channel. There is additional preliminary ev-idence that regions of the medial and lateralbanks of the intraparietal sulcus are the targetof cerebellar output (Ugolini et al. 2006). How-ever, area 7a, which is located on the inferiorparietal lobule (Figure 2), does not receive sub-stantial input from the dentate or other cerebel-lar nuclei (Clower et al. 2001). Currently, the in-formation about cerebellar projections to areasin posterior parietal cortex is complex and in-complete. It is clear, however, that several areasin this cortical region are the target of outputchannels from the ventral dentate.


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

TOPOGRAPHY OF FUNCTIONWITHIN THE DENTATE

We coalesced the unfolded maps from individ-ual experiments into a single summary diagramwhere the average location of each output chan-nel is indicated (Figure 5). This summary dia-gram emphasizes several notable features aboutthe topographic organization of the dentate. Asizeable portion of the nucleus projects to partsof the prefrontal and posterior parietal cortex.The output channels to prefrontal and posteriorparietal areas are clustered in a ventral and cau-dal region of the nucleus. Consequently, theseoutput channels are spatially segregated fromthose in the dorsal dentate that target motor ar-eas of the cortex. Thus, the dentate appears tobe spatially subdivided into separate motor andnonmotor domains that focus on functionallydistinct cortical systems. Another feature em-phasized by the summary diagram is that thecortical targets for large portions of the dentateremain to be determined.

MOLECULAR GRADIENTSIN THE DENTATE

The division of the dentate into separatemotor and nonmotor domains is reinforcedby underlying molecular gradients within thenucleus (Akkal et al. 2007, Dum et al. 2002,Fortin et al. 1998, Pimenta et al. 2001). Fortinet al. (1998) reported that immuno-stainingfor two calcium binding proteins, calretininand parvalbumin, is greatest in ventral regionsof the squirrel monkey dentate. A monoclonalantibody, 8B3, which recognizes a chondroitinsulfate proteoglycan on subpopulations ofneurons, also differentially stains the dentatein cebus monkeys and macaques (Akkal et al.2007, Dum et al. 2002, Pimenta et al. 2001).Immunoreactivity for 8B3 is most intense inventral regions of the dentate that projectto prefrontal and posterior parietal areas ofcortex. In contrast, antibody staining is leastintense in the dorsal regions of the nucleusthat project to the cortical motor areas. Theseobservations suggest that 8B3 recognizes a

Dentate

1 mm

D

C

Do

rsa

lD

ors

al

Ve

ntr

al

M1face

M1arm

PMvarm

SMAarm

46d Pre-SMA

9

M1leg

MOTOR

NON-MOTOR

?

7b

FEF

Figure 5Summary map of dentate topography. The lettering on the unfolded mapindicates the cortical target of different output channels in the cebus monkey.The location of different output channels divides the dentate into motor andnonmotor domains. Staining for monoclonal antibody 8B3 is most intense inthe nonmotor domain. The dashed line marks the limits of intense staining forthis antibody. The designation of the region marked by “?” is unclear.[Abbreviations as in Figure 2 (FEF, frontal eye field; M1, primary motorcortex; PMv, ventral premotor area; Pre-SMA, presupplementary motor area;SMA, supplementary motor area)]. Adapted from Dum & Strick 2003 andAkkal et al. 2007.

significant portion of the nonmotor domainwithin the dentate. Our measurements indicatethat approximately 40% of the nucleus isintensely stained by 8B3. This analysis doesnot include the caudal portion of the dentate,


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

(marked by a “?” in Figure 5) because thisregion does not stain intensely for 8B3 and wehave yet to determine its cortical target. How-ever, based on its location, we suspect that thiscaudal region projects to a nonmotor area ofthe cerebral cortex. If this is the case, then thenonmotor domain of the dentate may representas much as 50% of the nucleus in the cebusmonkey.

HUMAN DENTATE

It has long been recognized that the humandentate is composed of a dorsal, microgyricportion and a ventral, macrogyric portion (forreferences and illustration, see Voogd 2003).Compared with the microgyric dentate, themacrogyric dentate is reported to (a) developlater, (b) have smaller cells, (c) display a selec-tive vulnerability in cases of neocerebellar at-rophy, and (d) have a higher iron content. Thislast observation suggests that molecular gradi-ents may exist within the human dentate as theydo in the monkey dentate; however, this possi-bility remains to be tested. Comparative studiessuggest that the dentate has expanded in greatapes and humans relative to the other cerebel-lar nuclei (Matano et al. 1985). Furthermore,most of this increase appears to be due to anexpansion in the relative size of the ventral halfof the dentate (Matano 2001). This observa-tion implies that the nonmotor functions of thedentate grow in importance in great apes andhumans.

MACRO-ARCHITECTURE OFCEREBRO-CEREBELLAR LOOPS

The cortical areas that are the target of cere-bellar output also project via the pons tothe cerebellar cortex (Glickstein et al. 1985,Schmahmann 1996). This observation suggeststhat cerebro-cerebellar connections may forma closed-loop circuit. We have tested this con-cept for a representative motor area (the armarea of M1) and a nonmotor area (area 46 inthe prefrontal cortex) (Kelly & Strick 2003). Inessence, we examined whether a specific region

of the cerebellar cortex both receives input fromand projects to the same area of the cerebralcortex.

We used retrograde transneuronal transportof rabies virus to define the Purkinje cells incerebellar cortex that project to M1 or to area46. The arm area of M1 receives input fromPurkinje cells located mainly in lobules IV–VIof the cerebellar cortex (Figure 6, left panel).In contrast, area 46 receives input from Purk-inje cells located mainly in Crus II of the an-siform lobule (Figure 6, right panel). We sawno evidence of overlap between the two sys-tems. Thus, the two areas of the cerebral cortexare the target of output from Purkinje cells thatare located in separate regions of the cerebel-lar cortex. Clearly, the separation of motor andnonmotor functions seen in the dentate nucleusextends to the level of the cerebellar cortex.

In separate experiments, we used antero-grade transneuronal transport of a herpes virusto define the granule cells in cerebellar cortexthat receive input from M1 or from area 46.The arm area of M1 projects to granule cellslocated mainly in lobules IV–VI, whereas area46 projects to granule cells mainly in Crus II.Again, each cerebral cortical area projects togranule cells that are located in a separate regionof the cerebellar cortex. Moreover, these find-ings indicate that the regions of the cerebellarcortex that receive input from M1 are the sameas those that project to M1. Similarly, the re-gions of the cerebellar cortex that receive inputfrom area 46 are the same as those that projectto area 46. Thus, M1 and area 46 form sepa-rate, closed-loop circuits with different regionsof the cerebellar cortex (Figure 7). Altogether,these observations suggest that multiple closed-loop circuits represent a fundamental architec-tural feature of cerebro-cerebellar interactions.

There are a number of important functionalimplications to these results. They suggest thatthe cerebellar cortex is not functionally homo-geneous. Instead, our results imply that cere-bellar cortex contains localized regions that areinterconnected with specific motor or nonmo-tor areas of the cerebral cortex. In fact, we haveargued that the map of function in the cerebellar


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Crus Ia

a

p

p

HVIII

I

F.pr.

F.ppd.

M1 Area 46

II

III

IV

V

VI

VII

VIII

IX

X

I

II

III

IV

V

VI

VII

VIII

IX

X

F.pr.

F.pr.

F.ppd.

F.ppd.

F.pr.

F.ps.

HVIIB

HVIIB F.apm.

F.ps.

F.in.cr.

S. int.cr.2

Crus II

Figure 6Regions of cerebellar cortex that project to areas of cerebral cortex. The black dots on the flattened surfacemaps of the cerebellar cortex indicate the location of Purkinje cells that project to the arm area of M1 (leftpanel ) or to area 46 (right panel ) in the cebus monkey. The Purkinje cells that project to M1 are located inlobules that are separate from those that project to area 46. Nomenclature and abbreviations are according toLarsell (1970). Adapted from Kelly & Strick 2003.

cortex is likely to be as rich and complex as thatin the cerebral cortex (Kelly & Strick 2003). Asa consequence, global dysfunction of the cere-bellar cortex can cause wide-ranging effects onbehavior (e.g., Schmahmann 2004). However,localized dysfunction of a portion of the cere-bellar cortex can lead to more limited deficits,which may be motor or nonmotor depending

on the specific site of the cerebellar abnormal-ity (e.g., Allen & Courchesne 2003, Fiez et al.1992, Gottwald et al. 2004, Schmahmann &Sherman 1998). Thus, precisely defining thelocation of a lesion, a site of activation, or arecording site is as important for studies of thecerebellum as it is for studies of the cerebralcortex.


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

M1

THM1

PNM1

TH46

PN46

DNM1

CBMM1

DN46

CBM46

46

Figure 7Closed-loop circuits link the cerebellum with thecerebral cortex. We illustrate two topographicallyseparate closed-loop circuits. One interconnects thecerebellum with M1 and the other interconnects thecerebellum with area 46. In each loop, the corticalarea projects to a specific site in the pontine nuclei(PN), which then innervates a distinct region of thecerebellar cortex (CBM). Similarly, a portion of thedentate nucleus (DN) projects to a distinct region ofthe thalamus, which then innervates a specificcortical area. Note that the cortical area, which isthe major source of input to a circuit, is the majortarget of output from the circuit. CBM, cerebellarcortex; DN, dentate; PN, pontine nucleus; TH,subdivisions of the thalamus.

WHAT IS THE FULL EXTENT OFTHE CEREBELLAR INFLUENCEOVER NONMOTOR AREAS OFTHE CEREBRAL CORTEX?

As noted above, the cortical targets for substan-tial portions of the dentate remain unidenti-fied. In addition, fastigial and interpositus nu-clei send efferents to the thalamus (Asanumaet al. 1983, Batton et al. 1977, Kalil 1981,Stanton 1980), and the cortical targets of thesedeep nuclei remain to be fully determined. Theclosed-loop architecture described above en-ables us to make some predictions about addi-tional cortical targets of cerebellar output (Dum& Strick 2003, Kelly & Strick 2003, Middleton& Strick 1998). If closed-loop circuits reflecta general rule, then all of the areas of cere-bral cortex that project to the cerebellum arethe target of cerebellar output. In addition tothe cortical areas we have already studied, the

cerebellum receives input from a wide varietyof higher-order, nonmotor areas. This includesareas of extrastriate cortex, posterior parietalcortex, cingulate cortex, and the parahippocam-pal gyrus on the medial surface of the hemi-sphere (Figure 1) (Brodal 1978; Glickstein et al.1985; Leichnetz et al. 1984; Schmahmann &Pandya 1991, 1993, 1997; Vilensky & vanHoesen 1981; Wiesendanger et al. 1979). Ifsome or all of these areas turn out to be cere-bellar targets, then the full extent of cerebellarinfluence over nonmotor areas of the cerebralcortex is remarkable and much larger than pre-viously suspected.

Some have challenged the concept that a ma-jor component of cerebellar output targets cog-nitive areas of the cerebral cortex. For example,Glickstein & Doron (2008) argue that the re-gions of prefrontal cortex that are the target ofdentate output are concerned with the controlof eye movements. They point out that a por-tion of area 46 in the cebus monkey is includedin a “prefrontal eye field” (Lynch & Tian 2006).However, our injection sites into area 46 displaylittle or no overlap with the prefrontal eye field(compare figure 2 in Middleton & Strick 2001with figure 2 in Tian & Lynch 1996). Our injec-tion sites into areas 9l and 9m also display littleor no overlap with known eye fields. Further-more, the output channels to prefrontal cortexare located in dentate regions that are largelyrostral to the output channel that targets thefrontal eye field (Figure 5). Indeed, the outputchannel to the frontal eye field is located in thecaudal pole of the dentate (Lynch et al. 1994).This region of the dentate is likely to lie out-side of the nonmotor domain because it doesnot stain intensely for 8B3. Thus, there is con-siderable evidence that the cerebellar output toprefrontal cortex is distinct from the cerebellarcontrol of eye movements.

CEREBELLO-LIMBIC CIRCUIT

In discussing the neural substrate for a cere-bellar influence over nonmotor functions, itis important to note the longstanding notion


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

that the cerebellum is interconnected with thelimbic system. Cerebellar stimulation can alterlimbic function and elicit behaviors like shamrage, predatory attack, grooming, and eating(e.g., Berntson et al. 1973, Reis et al. 1973,Zanchetti & Zoccolini 1954). Cerebellar le-sions can tame aggressive monkeys independentof gross motor abnormalities (Berman 1997,Peters & Monjan 1971). Classic electrophysio-logical evidence suggests that cerebellar stim-ulation, especially in portions of the fastigialnucleus and associated regions of vermal cortex,can evoke responses at limbic sites, includingthe cingulate cortex and amygdala (e.g., Anandet al. 1959, Snider & Maiti 1976).

The major weakness in the cerebello-limbichypothesis is the absence of a clear anatomicalsubstrate that links the output of the cerebel-lum, and especially the fastigial nucleus, withlimbic sites such as the amygdala. Althoughneuroanatomical evidence indicates that thedeep cerebellar nuclei are interconnected withthe hypothalamus (Haines et al. 1990), theseconnections do not appear sufficient to medi-ate all of the behavioral effects evoked by cere-bellar stimulation. Thus, the circuits that linkthe output of the cerebellar nuclei with regionsof the limbic system need to be explored usingmodern neuroanatomical methods.

WHAT IS THE FUNCTIONALIMPACT OF NONMOTORCEREBRO-CEREBELLAR LOOPS?

Anatomical evidence that the cerebellum exertsa significant influence over nonmotor corticalareas is complemented by neuroimaging andneuropsychological data, which indicate thisinfluence is functionally important. Some of thefirst physiological data in humans came froma landmark neuroimaging study published byPetersen and colleagues (1989). This studyused positron emission tomography (PET)imaging to identify localized changes inblood-flow associated with a hierarchy ofsimple language tasks. Each level of the taskhierarchy was compared with the adjacent

level to successively isolate areas involved inthe perception, production, and high-levellinguistic analysis of single words. The findingsconfirmed the importance of brain regions thathad previously been implicated in perceptual,motor, and language processing. For instance,bilateral changes in blood flow that localizedto the paravermal regions of the cerebellumwere found when subjects read aloud orrepeated words, as compared with when theypassively viewed or listened to words; suchactivity during speech production was easilyreconciled with evidence that damage to thecerebellum can produce motoric disturbancesin speech (Holmes 1939). At the same time,there were several surprises, including the factthat a region in the right lateral cerebellarhemisphere showed greater blood flow whensubjects performed a verb generation task (say-ing aloud an appropriate use for each noun),as compared with simply reading or repeatingnouns. The response location differed fromthe paravermal activation observed duringspeech production and led the authors to favora cognitive interpretation; however, they ac-knowledged that they were “unable to assign aspecific candidate set of computations requiredby our generate task that might be related tocerebellar activation” (Petersen et al. 1989).

Since the Petersen et al. (1989) report, func-tional neuroimaging has become a widely em-ployed method, with reports of cerebellar acti-vation now commonplace. The range of tasksassociated with activation in the cerebellum isdauntingly large, and includes tasks designedto probe the neural basis of attention, executivecontrol, language, working memory, learning,pain, emotion, and addiction (for referencesand reviews, see Glickstein 2007, Timmann &Daum 2007). In considering this accrued data,one essential question is whether the identifiedcerebellar activations are indicative of impor-tant nonmotor functions or somehow inciden-tal to cognition (Glickstein 2007). Although thetasks used in neuroimaging are often describedin nonmotor terms (e.g., as a language task or anemotion task), they include motor components


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

(e.g., subjects may respond with a keypress oneach trial, or be asked to maintain their gazeon presented stimuli). Thus, activation differ-ences could reflect differences in motor plan-ning or execution, which might occur as a taskbecomes more difficult and response times in-crease. If this is the case, then increased cerebel-lar engagement might alter motor output (e.g.,keypresses could be emitted with more force)without impacting the cognitive aspects of taskperformance (e.g., the particular responses thatare selected).

Principals of hemodynamic activation maydiffer between the cerebral and cerebellar cor-tices, and this also makes it difficult to interpretthe functional significance of changes in cere-bellar activation. In the cerebral cortex, hemo-dynamic responses are thought to reflect a com-plex integration and indirect measure of theactive synaptic circuitry within a particular cor-tical area. Although there is still much to beunderstood, a rough correspondence can oftenbe made between the functional characteristicsof a particular cortical area as determined bysingle-unit recording studies in nonhuman pri-mates, as compared with what has been learnedthrough functional neuroimaging studies in hu-mans (Logothetis 2008). The same correspon-dences between neuronal activity and changesin the hemodynamic state may not be true in thecerebellar cortex, which has a general cytoar-chitecture that is clearly distinctive from that ofthe cerebral cortex. For instance, the sole out-put neurons of the cerebellum (Purkinje cells)are inhibitory, and the principal excitatory driveonto Purkinje cells comes from one system thatprovides a relatively tonic but weak influence,and another system that exerts a more phasicbut extremely powerful influence.

Convergent data from human lesion studiesspeak to the functional relevance of the cerebel-lar activation observed in human neuroimagingstudies. Schmahmann & Sherman (1998) wereamong the first to tackle this question througha clinical review of patients with cerebellardamage. They concluded that these patientsexhibited a diverse set of symptoms, whichthey termed the cerebellar cognitive affective

syndrome. Subsequent studies further doc-umented the nonmotor effects of cerebellardamage, collectively providing evidence forimpairments on standardized and experimentalmeasures of attention and executive control,procedural memory, working memory, lan-guage, and visual-spatial processing (Timmann& Daum 2007). At the same time, a close ex-amination of this neuropsychological literaturereveals a high degree of inconsistency (Burk2007, Frank et al. 2007, Glickstein 2007).

The factors that account for the variabilitybetween neuropsychological studies have yetto be determined. False positive findings mightarise from failures to consider motor confoundsthat could affect cognitive task performance,and from the inclusion of participants whosedeficits might reflect noncerebellar sourcesof dysfunction (Burk 2007, Frank et al. 2007,Glickstein 2007). False negative findings canalso arise from several sources (Bellebaum &Daum 2007). The locus of cerebellar damageis likely to be a particularly important factor.Most of the studies have collapsed acrosssubjects with different lesion sites within thecerebellum. Although this permits group-basedstatistical comparisons, it ignores evidence offunctional topography within the cerebellum.As a consequence, the findings from a studyare likely to be influenced by the particular mixof lesion sites represented within the subjectgroup.

The results reviewed in earlier sections ofthis chapter suggest that the map of functionin the cerebellar cortex is likely to be as richand complex as that in the cerebral cortex. Inthis regard, we have previously questioned theuse of descriptors such as cerebellar patient,which tend to de-emphasize the importance ofthe functional topography of the cerebellar cor-tex (Kelly & Strick 2003). For example, lesionsthat damage cerebellar tissue, which is inter-connected with area 46, are unlikely to causemotor deficits, just as lesions that damage cere-bellar tissue, which is interconnected with M1,are unlikely to result in cognitive deficits. Thus,precisely defining the location of a lesion, siteof activation, or recording site is as important


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

for studies of the cerebellum as it is for studiesof the cerebral cortex.

Finally, although most of the neuropsycho-logical research has involved adult participants,it is important to acknowledge evidence thatthe cerebellum may be especially importantfor normal cognitive development. Within theneurosurgical literature, it has been recognizedthat children with a history of posterior fossatumors are at risk for a variety of intellec-tual, emotional, and educational impairments(Cantelmi et al. 2008). It was thought that thesedisturbances were caused by radiation exposurethat occurred as part of the tumor managementstrategy. However, more recent evidence sug-gests that these impairments are a side effectof the incidental cerebellar damage that fre-quently occurs during the tumor resection. Adevelopmental perspective may also help to ex-plain the associations that have been made be-tween neuroanatomical and neurophysiologicalabnormalities in the cerebellum and disorderssuch as autism (Amaral et al. 2008), schizophre-nia (Andreasen & Pierson 2008), and dyslexia(Nicolson et al. 2001).

In summary, the neuroimaging and neu-ropsychological literatures provide compelling,although not conclusive, evidence that the hu-man cerebellum has important nonmotor func-tions. The precise nature of these functionsremains poorly understood. There are still rel-atively few neuroimaging and neuropsycholog-ical studies that have specifically targeted thecerebellum for a priori investigation. Most ofthe neurophysiological work in nonhuman pri-mates has utilized motor tasks that do not di-rectly map onto the cognitive and affective tasksthat have been employed in human research,and a large number of assumptions must bemade about the coupling between neuronal fir-ing and the neurovasculature in the cerebel-lum. Despite these limitations, a number ofperspectives have been proposed that help toexplain how the cerebellum may exert an im-portant influence over human cognition andaffect. In the sections below, we briefly reviewthree such perspectives and discuss the role thatanatomical information may play in advancing

our understanding of the normal motor func-tions of the cerebellum.

HOW MIGHT THE CEREBELLUMINFLUENCE COGNITION ANDAFFECT?

Speculations about the nonmotor functions ofthe cerebellum often begin by noting that thecytoarchitectonic structure of the cerebellumis uniform. This fact supports the overarchingassumption that there are computational prin-ciples that apply to both the motor and nonmo-tor functions of the cerebellum. Accordingly,researchers have often turned to the motor lit-erature for insights, as explained below.

One perspective is that cerebellar activa-tion often reflects its fundamental role in tim-ing. The timing-related view has its roots inlong-established clinical observations that mo-tor coordination can be severely disrupted bycerebellar injury (Holmes 1939). Modern le-sion studies have shown that patients withcerebellar damage have difficulties accuratelyproducing and perceiving time intervals (Ivry& Keele 1989), and functional imaging stud-ies have found effector-independent activationin the lateral cerebellum when subjects areasked to produce a complex rhythmic sequence(Bengtsson et al. 2005). Anatomically, it hasbeen suggested that recurrent networks withinthe cerebellum permit fine discriminationsamong different spatiotemporal patterns of in-put to the cerebellar cortex (de Solages et al.2008). With the loss of precise timing informa-tion and control, motor commands and internalcognitive states may no longer be appropriatelyselected and sequenced at a fine level. Thus,motorically, an individual may become less co-ordinated, and, cognitively, they may exhibitproblems with task-shifting and other forms ofexecutive control.

A second perspective is that cerebellaractivation reflects the use of sensorimotor im-agery, such as imagined speech. In the simplestversion of this hypothesis, representations andprocesses that would be engaged during actualmovement are co-opted to provide internal


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

representations that assist cognition. On themotor front, several studies have shown thatthe cerebellum is active when subjects imaginemaking a movement, such as a finger tap(Hanakawa et al. 2008). Similarly, the cerebel-lum is active when subjects are instructed toimagine producing simple speech utterances(Ackermann et al. 1998). These findings raisethe possibility that the cerebellum will berecruited whenever subjects engage in innerspeech—intuitively, the use of one’s internal(imagined) voice to represent, maintain,and organize task-relevant information andconscious thoughts. Humans exhibit a strongtendency to engage in verbal coding andrecoding, and thus internal speech-based rep-resentations may be important for a wide arrayof tasks that do not explicitly require speechor language processing. Beyond inner speechper se, recent findings suggest that conceptualknowledge of the world may rest, in part, uponinternally driven activation of perceptual andmotor representations (embodied cognition)(Barsalou 1999). For instance, when subjectsview action words that are associated with themovements of particular body parts (e.g., kick,throw, and lick), somatotopically organizedactivation can be found in sensorimotor areasof the cerebral cortex (Hauk et al. 2004), andactivation of a pain-related network has beenfound when individuals experience social rejec-tion (Eisenberger & Lieberman 2004). Corticalareas associated with a putative mirror neuronsystem have garnered most of the attention inthis area of research. However, there is someevidence that the cerebellum may be importantas well (Fuentes & Bastian 2007, Leslie 2004).

A third view is that the cerebellum is alearning machine that supports the adaptiveplasticity needed for the emergence of skilledbehavior. This general idea has been captured inmodels of motor control, which typically con-tain three basic elements: (a) internal modelsthat either predict the sensory input that shouldoccur as a consequence of a motor output (for-ward models) or that predict the movementsnecessary to achieve a goal (inverse models),(b) a comparison process that detects errors

in predicted versus actual outcomes, and (c) alearning process that uses error informationto adaptively modify internal models so thatmovement execution can be fast and accurate.There is broad agreement that the cerebellumis crucial for the adaptive control of many dif-ferent types of motor behavior, although as ofyet there is little consensus on the nature of thecomputations it performs (e.g., which types ofinternal models are used, and whether they aresimply passed from the cerebral cortex to thecerebellum or computed within the cerebellum)(Wolpert et al. 1998). Internal representationsare also central to cognition. Consequently,it may be just as important to ensure thealignment of different types of cognitive rep-resentations as it is to ensure the alignment ofsensory and motor representations. Cerebellarprocessing may help to adaptively modifyinternal representations so that the desiredgoals of cognition can be achieved in a skilled,and error-free, manner (Ito 2008).

Support for all three of these viewpoints canbe found within the existing literature. This ispartly because the ideas are to some extent in-terrelated, and thus they are not mutually exclu-sive. For instance, an internal model can be seenas a form of sensorimotor imagery (Fuentes &Bastian 2007), and the accurate detection of anerror may rely on the coincident timing of in-put from two representational maps (de Solageset al. 2008). It is also true, however, that re-searchers have yet to evaluate these perspectiveswith a high degree of anatomical and theoreti-cal specificity. In the section below, we return tothe issue of topographic organization within thecerebellum, and we consider how such informa-tion may be used to advance our understandingof its nonmotor functions.

TOPOGRAPHICALINFORMATION CAN BE USEDTO ADVANCE THEORY

The benefits that should accrue from carefulconsideration of anatomical detail within thecerebellum are predicated on evidence that con-nections between the cerebellum and nonmotor


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

regions of the cerebral cortex are organizedinto anatomically and functionally dissociableloops. Unfortunately, topographic informationhas thus far played a modest role in functionalinterpretations of cerebellar activity. Too often,neuroimaging researchers simply refer to an ac-tivation focus as in the cerebellum, and thoseusing neuropsychological methods analyze be-havioral results without regard to the specificlocation of cerebellar damage.

The widespread use of the Talairach &Tournoux atlas system (Talairach & Tournoux1988) accounts for some of the inattention toneuroanatomical detail. This atlas has provideda standard convention for reporting neu-roimaging results, but it was developed for thecerebral cortex. The stereotactic procedureswere not validated for the cerebellum, and theatlas does not even provide labels for its majorhemispheric fissures. Newly developed tools,such as an atlas for the human cerebellum(Diedrichson 2006, Dimitrova et al. 2006,Makris et al. 2005, Schmahmann et al. 1999),can help to surmount the limitations of theTalairach & Tournoux atlas. However, someof the inattention to anatomical detail reflectsa deeper issue. Namely, the correspondencebetween anatomical landmarks and functionalareas is poorly understood within the cerebel-lum as compared with the cerebral cortex.

As information accrues about the correspon-dence between gross anatomical structure inthe cerebellum and specific cortico-cerebellarloops, neuroimaging researchers will be in abetter position to leverage knowledge about theprecise locus of activation or site of damage tomake inferences about function.

In principle, it is possible to make progresson delineating the functional topography of thecerebellum using neuroimaging data alone. Forinstance, in the motor domain, the loci ofactivation associated with lip, tongue, hand,and foot movements have been examined (e.g.,Grodd et al. 2001). The results indicate thatactivation foci in the cerebellum are somato-topically organized, as predicted on the basis ofprior findings from nonhuman research. To il-lustrate how this approach could be extended

to the cognitive domain, consider the typicalfoci of activation reported in studies of read-ing and verbal working memory. Turkeltaubet al. (2002) used a voxelwise meta-analysis pro-cedure to determine which regions of activationare consistently found when subjects read alouda visually presented word, as compared with asimple baseline task (e.g., view a fixation cross).Across a set of 11 studies, they found evidencefor reliable changes in the right and left par-avermal cerebellar cortex. Chein et al. (2002)used a similar meta-analysis procedure to de-termine which regions of activation are con-sistently found when subjects perform a verbalworking memory task (e.g., a two-back task withvisually presented stimuli), as compared with acontrol task with similar perceptual-motor de-mands (e.g., a one-back task with visually pre-sented stimuli). Across a set of 30 studies, theyfound evidence for reliable changes in the rightlateral cerebellar cortex. The average foci ofcerebellar activation found in these two meta-analyses are separated by more than 22 mm (seeFigure 8), well beyond the typical 15+ mmrange often associated with the functional dif-ferentiation of cortical foci (Xiong et al. 2000).Thus, it is likely that cerebellar activation focifound in the reading and verbal working mem-ory meta-analyses reflect the engagement ofdifferent cortico-cerebellar loops.

Topographic distinctions between the acti-vation patterns associated with two tasks canbe used to constrain functional interpretations.For instance, behavioral data clearly indicatethat an articulatory rehearsal (or inner speech)strategy is commonly used to help main-tain verbal information in working memory(Baddeley 2003). Functional interpretations ofcerebellar activation during verbal workingmemory have assumed that this activation re-flected the use of the cerebellum to supportarticulatory rehearsal, perhaps through someform of motor speech imagery (Desmond &Fiez 1998). However, a simple motor imageryaccount fails to explain why the regions of thecerebellum that are engaged during articulatoryrehearsal (e.g., during verbal working memorytasks) are not the same as those engaged during


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

–100 –90

VIIIAVIIIB

VIIB VIIIAVIIIAVIIIA

VIVIVIVI

VV

VV

IVIV

IIIIII

IXIX

x: ± 10 mm x: ± 30 mm x: ± 30 mm

z-co

ordi

nate

(mm

)y-coordinate (mm)

VIIIB VIIIBVIIIB

VIIB VIIBVIIB

Cr IICr IICr II Cr IICr IICr II

Cr ICr ICr I Cr ICr ICr I

–80 –70 –60

–60

–50

–40

–30

–20

–10

0

–50 –40 –30 –100 –90 –80 –70 –60 –50 –40 –30 –100 –90 –80 –70 –60 –50 –40 –30

VIIIAVIIIA

VIVI

VV

VIIIBVIIIBVIIBVIIB

Cr IICr IICr II

Cr ICr ICr I

Figure 8Topography of cerebellar activation reported in neuroimaging studies of reading, verbal working memory, cue-elicited craving, andpain processing. Foci are plotted on sagittal sections using the atlas sections and procedures developed by Schmahmann et al. (1999);left-lateralized foci are plotted using filled symbols and right-lateralized foci are plotted using open symbols. Vermal and bilateralparavermal foci ( filled and open stars; left section) were reported in a meta-review of single-word reading (Turkeltaub et al. 2002), whereasa right-lateral focus (open diamond, middle section) was reported in a meta-review of verbal working memory (Chein et al. 2002). Bilateralresponses have been found in studies of cue-elicited craving (triangles; Olbrich et al. 2006, Smolka et al. 2006, Wilson 2008, Xiao et al.2006) and studies involving the presentation of pain-related visual stimuli (squares; Jackson et al. 2005, Moriguchi et al. 2007, Oginoet al. 2007, Singer et al. 2004, Ushida et al. 2008). The average locations of the responses are shown in the middle section, whereas theindividual foci reported across studies are shown in the right section.

actual speech (e.g., during simple word readingtasks). Instead, a more complex account seemsto be necessary. For instance, the differencesin activation loci may reflect differences inthe use of articulatory versus prearticulatoryrepresentations (Ackermann et al. 2007), in-ternal versus external guidance of movementplanning (Debaere et al. 2003), or the pro-duction of well-learned versus novel articu-latory sequences (Ben-Yehudah & Fiez 2008,Rauschecker et al. 2008).

Outside of the motor and cognitive do-mains, cerebellar activation is frequently re-ported in task comparisons that involve affec-tive processing. These include tasks used instudies of drug addiction and reward process-ing (Thoma et al. 2008), emotion appraisal andregulation (Schutter & van Honk 2005), andpain processing (Peyron et al. 2000). Such stud-ies usually emphasize cerebral cortical areas,such as orbitofrontal cortex, ventromedial pre-frontal cortex, and the amygdala. However, in-dividuals with cerebellar damage can exhibit af-fective dysfunction (Schmahmann & Sherman

1998), and thus greater consideration of thefunctional contributions of the cerebellum maybe warranted.

Careful consideration of the activation lociassociated with different affective tasks may be auseful starting point for assessing the specificityof cerebellar involvement in emotional process-ing. To illustrate this point, we identified twogroups of studies in which participants simplyviewed images of affectively arousing stimuli,as compared with control conditions involvingneutral stimuli. In one group, drug-dependentindividuals viewed images containing drug-related stimuli (e.g., nicotine-dependent indi-viduals saw an image of their hands holdinga cigarette). Such stimuli can elicit significantincreases in subjective craving for the drug(Wilson et al. 2005). In another group ofstudies, participants viewed images of injuredbody parts, or body parts experiencing painfulstimulation (e.g., a picture of a hand beingpierced by a needle). Such stimuli can elicita subjective sense of pain, which may be as-sociated with feelings of empathy toward an


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

injured person (Ushida et al. 2008). The fociof lateral cerebellar activation reported acrossthese two groups of studies are shown inFigure 8.

Within each group, the activation foci tendto cluster together. This is encouraging becausethe studies were selected to be as homogenousas possible to optimize the likelihood that asimilar network of brain regions would be re-cruited during the active and control tasks. Atthe same time, the typical locus of activationseems to differ across the two groups of stud-ies. This observation suggests that the cere-bellar activation is not simply a reflection ofthe motor imagery that is automatically trig-gered by the visual stimuli, because both typesof stimuli are likely to engage similar motor ef-fectors (e.g., hand and arm movements asso-ciated with smoking a cigarette, or withdraw-ing from a painful stimulation). If these resultsare confirmed, they would provide additionalsupport for the concept that separate cerebro-cerebellar circuits are engaged in response tothe visual stimuli that induce internal states as-sociated with craving versus pain.

SUMMARY AND CONCLUSIONS

The dominant view of cerebellar function overthe past century has been that it is concernedwith the coordination and control of motor ac-tivity (Brooks & Thach 1981). It is now appar-ent that a significant portion of the output ofthe cerebellum projects to nonmotor areas ofthe cerebral cortex, including regions of pre-frontal and posterior parietal cortex. Thus, theanatomical substrate exists for cerebellar outputto influence the cognitive and visuospatial com-putations performed in prefrontal and posterior

parietal cortex (Clower et al. 2001, 2005; Mid-dleton & Strick 2001). As a corollary, abnormal-ities in cerebellar structure and function havethe potential to produce multiple motor andnonmotor deficits.

The output to nonmotor areas of the cere-bral cortex originates specifically from a ventralportion of the dentate. This nonmotor region ofthe dentate is recognized by several molecularmarkers. Several authors have argued that ven-tral dentate and related regions of the cerebel-lar hemispheres are selectively enlarged in greatapes and humans (Leiner et al. 1991, Matano2001). Indeed, the enlargement of the ventraldentate in humans is thought to parallel theenlargement of prefrontal cortex. These obser-vations have led to the proposal that the den-tate’s participation in nonmotor functions maybe especially prominent in humans (e.g., Leineret al. 1991, Schmahmann & Sherman 1998).

Anatomical evidence that the cerebellumexerts an influence over nonmotor regionsof the cerebral cortex is complemented bydata from neuroimaging and neuropsychol-ogy. These methodologies have provided com-pelling evidence that the cerebellum plays afunctionally important role in human cogni-tion and affect. The theoretical perspectivesthat have been used to explain how and whythe cerebellum contributes to nonmotor taskshave drawn heavily upon the motor literature.Although these theoretical perspectives are use-ful, they are not computationally well-specified.As a result, they only loosely constrain the inter-pretation of cerebellar function in specific taskcontexts. Greater attention to anatomical infor-mation should help to significantly advance cur-rent understanding of cerebellar involvement innonmotor function.

DISCLOSURE STATEMENT

The authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

This work was supported in part by funds from the Office of Research and Development, Med-ical Research Service, Department of Veterans Affairs, and U.S. Public Health Service grantsR01NS24328, R01MH56661, P40RR018604 (P.L.S.), and RO1MH59256, R01DA18910 ( J.A.F.).


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

LITERATURE CITED

Ackermann H, Mathiak K, Riecker A. 2007. The contribution of the cerebellum to speech production andspeech perception: clinical and functional imaging data. Cerebellum 6:202–13

Ackermann H, Wildgruber D, Daum I, Grodd W. 1998. Does the cerebellum contribute to cognitive aspectsof speech production? A functional magnetic resonance imaging (fMRI) study in humans. Neurosci. Lett.247:187–90

Akkal D, Dum RP, Strick PL. 2007. Supplementary motor area and presupplementary motor area: targets ofbasal ganglia and cerebellar output. J. Neurosci. 27:10659–73

Allen G, Courchesne E. 2003. Differential effects of developmental cerebellar abnormality on cognitive andmotor functions in the cerebellum: an fMRI study of autism. Am. J. Psychiatry 160:262–73

Allen GI, Gilbert PF, Yin TC. 1978. Convergence of cerebral inputs onto dentate neurons in monkey.Exp. Brain Res. 32:151–70

Allen GI, Tsukahara N. 1974. Cerebrocerebellar communication systems. Physiol. Rev. 54:957–1006Amaral DG, Schumann CM, Nordahl CW. 2008. Neuroanatomy of autism. Trends Neurosci. 31:137–45Anand BK, Malhotra CL, Singh B, Dua S. 1959. Cerebellar projections to limbic system. J. Neurophysiol.

22:451–57Andreasen NC, Pierson R. 2008. The role of the cerebellum in schizophrenia. Biol. Psychiatry 64:81–88Asanuma C, Thach WT, Jones EG. 1983. Distribution of cerebellar terminations and their relation to other

afferent terminations in the ventral lateral thalamic region of the monkey. Brain Res. 286:237–65Baddeley A. 2003. Working memory: looking back and looking forward. Nat. Rev. Neurosci. 4:829–39Barsalou LW. 1999. Perceptual symbol systems. Behav. Brain Funct. 22:577–609Batton RR, Jayaraman AI, Ruggiero D, Carpenter MB. 1977. Fastigial efferent projections in the monkey: an

autoradiographic study. J. Comp. Neurol. 174:281–306Bellebaum C, Daum I. 2007. Cerebellar involvement in executive control. Cerebellum 6:184–92Ben-Yehudah G, Fiez JA. 2008. Impact of cerebellar lesions on reading and phonological processing. Ann. N.

Y. Acad. Sci. 1145:260–74Ben-Yehudah G, Guediche S, Fiez JA. 2007. Cerebellar contributions to verbal working memory: beyond

cognitive theory. Cerebellum 6:193–201Bengtsson SL, Ehrsson HH, Forssberg H, Ullen F. 2005. Effector-independent voluntary timing: behavioral

and neuroimaging evidence. Eur. J. Neurosci. 22:3255–65Berman AJ. 1997. Amelioration of agression: response to selective cerebellar lesions in the rhesus monkey.

Int. Rev. Neurobiol. 41:111–19Berntson GG, Potolicchio SJ Jr, Miller NE. 1973. Evidence for higher functions of the cerebellum: eating

and grooming elicited by cerebellar stimulation in cats. Proc. Natl. Acad. Sci. USA 70:2497–99Brodal P. 1978. The corticopontine projection in the rhesus monkey. Origin and principles of organization.

Brain 101:251–83Brooks VB, Thach WT. 1981. Cerebellar control of posture and movement. In Handbook of physiology, Section

1. The nervous system, ed. VB Brooks, Vol. 2, Motor Control, Part II, pp. 877–946. Bethesda: Am. Physiol.Soc.

Burk K. 2007. Cognition in hereditary ataxia. Cerebellum 6:280–86Cantelmi D, Scweizer TA, Cusimano MD. 2008. Role of the cerebellum in the neurocognitive sequelae of

treatment of tumors of the posterior fossa: an update. Lancet Oncol. 9:569–76Chein JM, Fissell K, Jacobs S, Fiez JA. 2002. Functional heterogeneity within Broca’s area during verbal

working memory. Physiol. Behav. 77:635–39Clower DM, Dum RP, Strick PL. 2005. Basal ganglia and cerebellar inputs to ‘AIP.’ Cereb. Cortex 15:913–20Clower DM, West RA, Lynch JC, Strick PL. 2001. The inferior parietal lobule is the target of output from

the superior colliculus, hippocampus and cerebellum. J. Neurosci. 21:6283–91Crick FC, Koch C. 2005. What is the function of the claustrum? Philos. Trans. R. Soc. Lond. B Biol. Sci.

360:1271–79de Solages C, Szapiro G, Brunel N, Hakim V, Isope P, et al. 2008. High-frequency organization and synchrony

of activity in the Purkinje cell layer of the cerebellum. Neuron 58:775–88


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Debaere F, Wenderoth N, Sunaert S, Van Hecke P, Swinnen SP. 2003. Internal vs external generation ofmovements: differential neural pathways involved in bimanual coordination performed in the presenceor absence of augmented visual feedback. Neuroimage 19:764–76

Desmond JE, Fiez JA. 1998. Neuroimaging studies of the cerebellum: language, learning and memory. TrendsCogn. Sci. 2:355–58

Diedrichsen J. 2006. A spatially unbiased atlas template of the human cerebellum. Neuroimage 33:127–38Dimitrova A, Zeljko D, Schwarze F, Maschke M, Gerwig M, et al. 2006. Probabilistic 3D MRI atlas of the

human cerebellar dentate/interposed nuclei. Neuroimage 30:12–25Dum RP, Li C, Strick PL. 2002. Motor and nonmotor domains in the monkey dentate. Ann. N. Y. Acad. Sci.

978:289–301Dum RP, Strick PL. 2003. An unfolded map of the cerebellar dentate nucleus and its projections to the cerebral

cortex. J. Neurophysiol. 89:634–39Eisenberger NI, Lieberman MD. 2004. Why rejection hurts: a common neural alarm system for physical and

social pain. Trends Cogn. Sci. 8:294–300Fiez J. 1996. Cerebellar contributions to cognition. Neuron 16:13–15Fiez JA, Petersen SE, Cheney MK, Raichle ME. 1992. Impaired nonmotor learning and error detection

associated with cerebellar damage. Brain 115:155–78Fortin M, Marchand R, Parent A. 1998. Calcium-binding proteins in primate cerebellum. Neurosci. Res. 30:155–

68Frank B, Schoch B, Richter S, Frings M, Karnath TO, Timmann D. 2007. Cerebellar lesion studies of cognitive

function in children and adolescents—limitations and negative findings. Cerebellum 6:242–53Fuentes CT, Bastian AM. 2007. ‘Motor cognition’—what is it and is the cerebellum involved? Cerebellum

6:232–36Glickstein M. 2007. What does the cerebellum really do? Curr. Biol. 17:R824–27Glickstein M, Doron K. 2008. Cerebellum: connections and functions. Cerebellum 7:589–94Glickstein M, May JG, Mercier BE. 1985. Corticopontine projection in the macaque: the distribution of

labeled cortical cells after large injections of horseradish peroxidase in the pontine nuclei. J. Comp. Neurol.235:343–59

Gottwald B, Wilde B, Mihajlovic Z, Mehdorn HM. 2004. Evidence for distinct cognitive deficits after focalcerebellar lesions. J. Neurol. Neurosurg. Psychiatry 75:1524–31

Grodd W, Hulsmann E, Lotze M, Wildgruber D, Erb M. 2001. Sensorimotor mapping of the human cere-bellum: fMRI evidence of somatotopic organization. Hum. Brain Mapp. 13:55–73

Haines DE, May PJ, Dietrichs E. 1990. Neuronal connections between the cerebellar nuclei and hypothalamusin Macaca fascicularis: cerebello-visceral circuits. J. Comp. Neurol. 299:106–22

Hanakawa T, Dimyan MA, Hallett M. 2008. Motor planning, imagery, and execution in a distributed motornetwork: a time-course study with functional MRI. Cereb. Cortex 18:2775–88

Hauk O, Johnsrude I, Pulvermuller F. 2004. Somatotopic representation of action words in human motor andpremotor cortex. Neuron 41:301–7

Holmes G. 1939. The cerebellum of man. Brain 62:1–30Hoover JE, Strick PL. 1999. The organization of cerebello- and pallido-thalamic projections to primary motor

cortex: an investigation employing retrograde transneuronal transport of herpes simplex virus type 1.J. Neurosci. 19:1446–63

Ito M. 2008. Control of mental activities by internal models in the cerebellum. Nat. Rev. Neurosci. 9:304–13Ivry RB, Keele SW. 1989. Timing functions of the cerebellum. J. Cogn. Neurosci. 1:136–52Jackson PL, Meltzoff AN, Decety J. 2005. How do we perceive the pain of others? A window into the neural

processes involved in empathy. Neuroimage 24:771–79Jones EG. 1985. The Thalamus. New York: Plenum. 935 pp.Kalil K. 1981. Projections of the cerebellar and dorsal column nuclei upon the thalamus of the rhesus monkey.

J. Comp. Neurol. 195:25–50Kelly RM, Strick PL. 2000. Rabies as a transneuronal tracer of circuits in the central nervous system.

J. Neurosci. Methods 103:63–71Kelly RM, Strick PL. 2003. Cerebellar loops with motor cortex and prefrontal cortex of a nonhuman primate.

J. Neurosci. 23:8432–44


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Larsell O. 1970. The Comparative Anatomy and Histology of the Cerebellum from Monotremes through Apes.Minneapolis: Univ. Minnesota Press. 269 pp

Leichnetz GR, Smith DJ, Spence RF. 1984. Cortical projections to the paramedian tegmental and basilar ponsin the monkey. J. Comp. Neurol. 228:388–408

Leiner HC, Leiner AL, Dow RS. 1991. The human cerebro-cerebellar system: its computing, cognitive, andlanguage skills. Behav. Brain Res. 44:113–28

Leslie KR. 2004. Functional imaging of face and hand imitation: towards a motor theory of empathy.Neuroimage 21:601–7

Logothetis NK. 2008. What we can do and what we cannot do with fMRI. Nature 453:869–78Lynch JC, Hoover JE, Strick PL. 1994. Input to the primate frontal eye field from the substantia nigra, superior

colliculus, and dentate nucleus demonstrated by transneuronal transport. Exp. Brain Res. 100:181–86Lynch JC, Tian JR. 2006. Cortico-cortical networks and cortico-subcortical loops for the higher control of

eye movements. Prog. Brain Res. 151:461–501Makris N, Schlerf JE, Hodge SM, Haselgrove C, Albaugh MD, et al. 2005. MRI-based surface-assisted

parcellation of human cerebellar cortex: an anatomically specified method with estimate of reliability.Neuroimage 25:1146–60

Matano S. 2001. Brief communication: proportions of the ventral half of the cerebellar dentate nucleus inhumans and great apes. Am. J. Phys. Anthro. 114:163–65

Matano S, Baron G, Stephan H, Frahm HD. 1985. Volume comparisons in the cerebellar complex of primates.II. Cerebellar nuclei. Folia Primatol. 44:182–203

Middleton FA, Strick PL. 1994. Anatomical evidence for cerebellar and basal ganglia involvement in highercognitive function. Science 266:458–61

Middleton FA, Strick PL. 1996a. The temporal lobe is a target of output from the basal ganglia. Proc. Natl.Acad. Sci. USA 93:8683–87

Middleton FA, Strick PL. 1996b. New concepts regarding the organization of basal ganglia and cerebellaroutput. In Integrative and Molecular Approach to Brain Function, ed. M Ito, Y Miyashita, pp. 253–71.Amsterdam: Elsevier Sci.

Middleton FA, Strick PL. 1997. Dentate output channels: motor and cognitive components. In Progressin Brain Research. 114: The Cerebellum from Structure to Control, ed. CI De Zeeuw, P Strata, J Voogd,pp. 55–66. Amsterdam: Elsevier

Middleton FA, Strick PL. 1998. Cerebellar output: motor and cognitive channels. Trends Cognit. Sci. 2:348–54Middleton FA, Strick PL. 2001. Cerebellar projections to the prefrontal cortex of the primate. J. Neurosci.

21:700–12Moriguchi Y, Decety J, Ohnishi T, Maeda M, Mori T, et al. 2007. Empathy and judging other’s pain: an fMRI

study of alexithymia. Cereb. Cortex 17:2223–34Nicolson RI, Fawcett AJ, Dean P. 2001. Developmental dyslexia: the cerebellar deficit hypothesis. Trends

Neurosci. 24:508–11Ogino Y, Nemoto H, Inue K, Saito S, Kakigi R, Goto F. 2007. Inner experience of pain: imagination of pain

while viewing images showing painful events forms subjective pain representation in human brain. Cereb.Cortex 17:1139–46

Olbrich HM, Valerius G, Pais C, Hagenbuch F, Ebert D, Juengling FD. 2006. Brain activation during cravingfor alcohol measured by positron emission tomography. Aust. N. Z. J. Psychiatry 40:171–78

Percheron G, Francois C, Talbi B, Yelnik J, Fenelon G. 1996. The primate motor thalamus. Brain Res. Rev.22:93–181

Peters M, Monjan AA. 1971. Behavior after cerebellar lesions in cats and monkeys. Physiol. Behav. 6:205–6Petersen SE, Fox PT, Posner MI, Mintun M, Raichle ME. 1989. Positron emission tomographic studies of

the processing of single words. J. Cogn. Neurosci. 1:153–70Peyron R, Laurent B, Garcia-Larrea L. 2000. Functional imaging of brain responses to pain. A review and

meta-analysis. Neurophysiol. Clin. 30:263–88Picard N, Strick PL. 2001. Imaging the premotor areas. Curr. Opin. Neurobiol. 11:663–72Pimenta AF, Strick PL, Levitt PR. 2001. A novel proteoglycan epitope is expressed in functionally discrete

patterns in primate cortical and subcortical regions. J. Comp. Neurol. 430:369–88


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Rauschecker AM, Pringle A, Watkins KE. 2008. Changes in neural activity associated with learning to articulatenovel auditory pseudowords by covert repetition. Hum. Brain Mapp. 29:1231–42

Reis DJ, Doba N, Nathan MA. 1973. Predatory attack, grooming, and consummatory behaviors evoked byelectrical stimulation of cat cerebellar nuclei. Science 182:845–47

Rispal-Padel L, Cicirata F, Pons C. 1982. Cerebellar nuclear topography of simple and synergistic movementsin the alert baboon (Papio papio). Exp. Brain Res. 47:365–80

Schell GR, Strick PL. 1984. The origin of thalamic inputs to the arcuate premotor and supplementary motorareas. J. Neurosci. 4:539–60

Schmahmann JD. 1996. From movement to thought: anatomic substrates of the cerebellar contribution tocognitive processing. Hum. Brain Mapp. 4:174–98

Schmahmann JD. 2004. Disorders of the cerebellum: ataxia, dysmetria of thought, and the cerebellar cognitiveaffective syndrome. J. Neuropsychiatry Clin. Neurosci. 16:367–78

Schmahmann JD, Doyon J, McDonald D, Holmes C, Lavoie K, et al. 1999. Three-dimensional MRI atlas ofthe human cerebellum in proportional stereotaxic space. Neuroimage 10:233–60

Schmahmann JD, Pandya DN. 1991. Projections to the basis pontis from the superior temporal sulcus andsuperior temporal region in the rhesus monkey. J. Comp. Neurol. 308:224–48

Schmahmann JD, Pandya DN. 1993. Prelunate, occipitotemporal, and parahippocampal projections to thebasis pontis in rhesus monkey. J. Comp. Neurol. 337:94–112

Schmahmann JD, Pandya DN. 1997. Anatomic organization of the basilar pontine projections from prefrontalcortices in rhesus monkey. J. Neurosci. 17:438–58

Schmahmann JD, Sherman JC. 1998. The cerebellar cognitive affective syndrome. Brain 121:561–79Schutter DJLG, van Honk J. 2005. The cerebellum on the rise in human emotion. Cerebellum 4:290–94Singer T, Seymour B, O’Doherty J, Kaube H, Dolan RJ, Frith CD. 2004. Empathy for pain involves the

affective but not the sensory components of pain. Science 303:1157–62Smolka MN, Buhler M, Klein S, Zimmermann U, Mann K, et al. 2006. Severity of nicotine dependence modu-

lates cue-induced brain activity in regions involved in motor preparation and imagery. Psychopharmacology184:577–88

Snider RS, Maiti A. 1976. Cerebellar contributions to the Papez circuit. J. Neurosci. Res. 2:133–46Stanton GB. 1980. Topographical organization of ascending cerebellar projections from the dentate and

interposed nuclei in Macaca mulatta: an anterograde degeneration study. J. Comp. Neurol. 190:699–731Strick PL, Card JP. 1992. Transneuronal mapping of neural circuits with alpha herpesviruses. In Experimental

Neuroanatomy: A Practical Approach, ed. JP Bolam, pp. 81–101. Oxford, UK: Oxford Univ. PressTalairach J, Tournoux P. 1988. Co-Planar Stereotaxic Atlas of the Human Brain: 3-Dimensional Proportional System:

An Approach to Cerebral Imaging. New York: Stuttgart. 122 pp.Thach WT, Perry JG, Kane SA, Goodkin HP. 1993. Cerebellar nuclei: rapid alternating movement, motor

somatotopy, and a mechanism for the control of muscle synergy. Rev. Neurol. 149:607–28Thoma P, Bellebaum C, Koch B, Schwarz M, Daum I. 2008. The cerebellum is involved in reward-based

reversal learning. Cerebellum 7:433–43Tian JR, Lynch JC. 1996. Corticocortical input to the smooth and saccadic eye movement subregions of the

frontal eye field in Cebus monkeys. J. Neurophysiol. 76:2754–71Timmann D, Daum I. 2007. Cerebellar contributions to cognitive functions: a progress report after two

decades of research. Cerebellum 6:159–62Turkeltaub PE, Eden GF, Jones KM, Zeffiro TA. 2002. Meta-analysis of the functional neuroanatomy of

single-word reading: method and validation. Neuroimage 16:765–80Ugolini G, Prevosto V, Graf W. 2006. Differences in cerebellar nuclear input to medial (MIP/VIPm) and

lateral (VIPl/LIPv) intraparietal areas, revealed by retrograde transneuronal transfer of rabies virus. Soc.Neurosci. Abstr. Viewer/Planner Progr. No. 242.19

Ushida T, Ikemoto T, Tanaka S, Shinozaki J, Taniguchi S, et al. 2008. Virtual needle pain stimuli activatescortical representation of emotions in normal volunteers. Neurosci. Lett. 439:7–12

Vilensky JA, van Hoesen GW. 1981. Corticopontine projections from the cingulate cortex in the rhesusmonkey. Brain Res. 205(2):391–95

Voogd J. 2003. The human cerebellum. J. Chem. Neuroanat. 26:243–52


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

ANRV379-NE32-17 ARI 10 May 2009 9:35

Wiesendanger R, Wiesendanger M, Ruegg DG. 1979. An anatomical investigation of the corticopontaineprojection in the primate (Macaca fascicularis and Saimiri sciureus)–II. The projection from frontal andparental association areas. Neuroscience 4:747–65

Wilson SJ. 2008. Self-focused versus other-focused cognitive strategies for coping with smoking cue-exposure: A functionalmagnetic resonance imaging study. Doc. diss., Univ. of Pittsburgh. 138 pp.

Wilson SJ, Sayette MA, Delgago MR, Fiez JA. 2005. Instructed smoking expectancy modulates cue-elicitedneural activity: a preliminary study. Nicotine Tob. Res. 7:637–45

Wolpert DM, Miall RC, Kawato M. 1998. Internal models in the cerebellum. Trends Cogn. Sci. 2:338–46Xiao Z, Lee T, Ahang JX, Wu Q, Wu R, et al. 2006. Thirsty heroin addicts show different fMRI activations

when exposed to water-related and drug-related cues. Drug Alcohol Depend. 83:157–62Xiong J, Rao S, Jerabek P, Zamarripa F, Woldorff M, et al. 2000. Intersubject variability in cortical activations

during a complex language task. Neuroimage 12:326–39Zanchetti A, Zoccolini A. 1954. Autonomic hypothalamic outbursts elicited by cerebellar stimulation.

J. Neurophysiol. 17:475–83Zemanick MC, Strick PL, Dix RD. 1991. Direction of transneuronal transport of herpes simplex virus 1 in

the primate motor system is strain-dependent. Proc. Natl. Acad. Sci. USA 88:8048–51


Ann

u. R

ev. N

euro

sci.

2009

.32:

413-

434.

Dow

nloa

ded

from

arj

ourn

als.

annu

alre

view

s.or

gby

Iow

a St

ate

Uni

vers

ity o

n 06

/26/

09. F

or p

erso

nal u

se o

nly.

cerebellum and nonmotor function - xurxo mariño...cerebellum and nonmotor function peter l....

Documents