Separate Neural Substrates in the Human Cerebellum
for Sensory-motor Adaptation of Reactive and of
Scanning Voluntary Saccades
N. Alahyane, V. Fonteille, C. Urquizar, R Salemme, N. Nighoghossian, Denis
Pelisson, C. Tilikete

To cite this version:
N. Alahyane, V. Fonteille, C. Urquizar, R Salemme, N. Nighoghossian, et al.. Separate Neural Substrates in the Human Cerebellum for Sensory-motor Adaptation of Reactive and of Scanning Voluntary
Saccades. The Cerebellum, 2008, 7 (4), pp.595-601. �10.1007/s12311-008-0065-5�. �hal-02197091�

HAL Id: hal-02197091
https://hal.science/hal-02197091
Submitted on 30 Jul 2019

HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.

Separate Neural Substrates in the Human Cerebellum
for Sensory-motor Adaptation of Reactive and of
Scanning Voluntary Saccades
N. Alahyane, V. Fonteille, C. Urquizar, : Salemme, N. Nighoghossian, Denis
Pelisson, C. Tilikete

To cite this version:
N. Alahyane, V. Fonteille, C. Urquizar, : Salemme, N. Nighoghossian, et al.. Separate Neural Substrates in the Human Cerebellum for Sensory-motor Adaptation of Reactive and of Scanning Voluntary
Saccades. The Cerebellum, Taylor
Francis Ltd., 2008, 7 (4), pp.595-601. �10.1007/s12311-008-0065-5�. �hal-02197091�

HAL Id: hal-02197091
https://hal.archives-ouvertes.fr/hal-02197091
Submitted on 30 Jul 2019

HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.

Separate Neural Substrates in the Human Cerebellum
for Sensory-motor Adaptation of Reactive and of Scanning
Voluntary Saccades
N. Alahyane & V. Fonteille & C. Urquizar & R. Salemme &
N. Nighoghossian & D. Pelisson & C. Tilikete

Abstract Sensory-motor adaptation processes are critically
involved in maintaining accurate motor behavior throughout life. Yet their underlying neural substrates and taskdependency bases are still poorly understood. We address
these issues here by studying adaptation of saccadic eye
movements, a well-established model of sensory-motor
plasticity. The cerebellum plays a major role in saccadic
adaptation but it has not yet been investigated whether this
role can account for the known specificity of adaptation to

the saccade type (e.g., reactive versus voluntary). Two
patients with focal lesions in different parts of the cerebellum
were tested using the double-step target paradigm. Each
patient was submitted to two separate sessions: one for
reactive saccades (RS) triggered by the sudden appearance of
a visual target and the second for scanning voluntary
saccades (SVS) performed when exploring a more complex
scene. We found that a medial cerebellar lesion impaired
adaptation of reactive—but not of voluntary—saccades,

N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
D. Pelisson : C. Tilikete
INSERM, U864, Espace et Action,
Bron 69500, France

N. Nighoghossian
Hospices Civils de Lyon, Cerebrovascular Unit,
Hôpital Neurologique,
Lyon, France

N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
N. Nighoghossian : D. Pelisson : C. Tilikete
Université de Lyon,
Lyon 69003, France
N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
N. Nighoghossian : D. Pelisson : C. Tilikete
Université Lyon 1, Biologie Humaine,
Lyon 69003, France
N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
D. Pelisson : C. Tilikete
Hospices Civils de Lyon, Mouvement et Handicap,
Hôpital Neurologique,
Lyon 69003, France
N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
D. Pelisson : C. Tilikete
IFR19, Institut Fédératif des Neurosciences de Lyon,
Lyon 69003, France
N. Alahyane : V. Fonteille : C. Urquizar : R. Salemme :
D. Pelisson : C. Tilikete
IFR23, Institut Fédératif de Recherche sur le Handicap,
Bron 69500, France

N. Nighoghossian
Creatis UMR CNRS 5515 INSERM U630, UCB Lyon I,
Lyon, France

D. Pelisson (*)
“Espace et Action” U864, INSERM, University Lyon 1,
16 avenue du doyen Lépine,
69676 Bron cedex, France
e-mail: pelisson@lyon.inserm.fr

Present address:
N. Alahyane
Centre for Neuroscience Studies, Queen’s University,
Botterell Hall, Room 234,
Kingston, ON, Canada K7L 3N6
Present address:
V. Fonteille
Inserm, U742, Université Pierre et Marie Curie,
Paris 75005, France

whereas a lateral lesion affected adaptation of scanning
voluntary saccades, but not of reactive saccades. These
findings provide the first evidence of an involvement of the
lateral cerebellum in saccadic adaptation, and extend the
demonstrated role of the cerebellum in RS adaptation to
adaptation of SVS. The double dissociation of adaptive abilities is also consistent with our previous hypothesis of the
involvement in saccadic adaptation of partially separated
cerebellar areas specific to the reactive or voluntary task
(Alahyane et al. Brain Res 1135:107–121 (2007)).

4 months after a stroke affecting the caudal paramedian
part of the right cerebellar hemisphere (infarct of the medial
branch of the postero-inferior cerebellar artery). Note that
the available imaging data did not allow us to delineate
very clearly the lesions’ boundaries. However, it is obvious
that the two lesions were clearly separate as they involved
the lateral part of anterior lobe in P1 and the medial part of
posterior lobe in P2. At the time of recording, a right-sided
limb ataxia (P1) or axial ataxia (P2) were still present, but
there was no ocular nystagmus.

Keywords Sensory-motor . Adaptation . Cerebellum .
Saccades . Plasticity . Neuropathology

Experimental Set-up

Introduction
The cerebellum is known as the main neural center for
cellular plasticity underlying long-term adaptation of motor
behavior, but its precise role is still intensively debated [2–
5]. In particular, whether the task specificity of sensorymotor adaptation [5–9] relies on the selective involvement
of different adaptable cerebellar circuits is still unknown.
We addressed this issue using saccadic adaptation—a
progressive adjustment of the amplitude of saccadic eye
movements to a systematic perturbation of target position—
as a key model of sensory-motor adaptation [8]. Previous
works suggested that adaptation of reactive saccades (RS)
elicited by the sudden presentation of a novel visual target
critically depends on the medial part of the posterior (MP)
cerebellum [10–13]. In contrast, adaptation of scanning
voluntary saccades (SVS) has been much less studied,
although these endogenously generated saccades are predominant in daily life. Based on the pattern of adaptation
transfer between RS and SVS in healthy humans, we
recently suggested that SVS and RS adaptations rely both
on a common brainstem site and on two separate, but partly
overlapping, cerebellar sites [1]. We provide here direct
evidence for this hypothesis by reporting on saccadic
adaptation abilities in two volunteers with isolated cerebellar
lesion (Fig. 1a). Indeed, this case study disclosed two
opposite patterns of adaptation deficits for reactive and
voluntary saccades depending on the lesioned cerebellar site.

Materials and Methods
Patients
P1, a 40-year-old man, was tested 10 days after a stroke
located at the intermediate level of the rostral right
hemisphere of the cerebellum (infarct of antero-inferior
cerebellar artery). P2, a 61-year-old man, was tested

The set-up and protocols were approved by the local ethics
committee (CCPPRB-Lyon B) and have been described in
detail in a previous paper [1]. Visual targets (6 mm black dots
on a gray background) were presented on a video screen by
means of a 142-Hz visual stimulus generation system (CRS,
Cambridge, UK). A helmet-mounted EyeLink II eye-tracker
(SR Research, Mississauga, Canada) was used to record 2-D
binocular eye movements (500 Hz sampling, accuracy better
than 0.1°). Real-time monitoring of eye position allowed us
to modify the visual display during primary saccades within
10 ms of a pre-defined event trigger (i.e., eye velocity
exceeding a 100°/s threshold).
Procedure
After providing written informed consent, each patient
underwent RS and SVS adaptations in separate sessions. A
single target (RS) or a set of targets (SVS) shifted backward
during each primary saccade (Fig. 1b,c), in order to induce
an adaptive gain reduction of both leftward and rightward
saccades. As a consequence of saccadic suppression [14,
15], such target shifts were unnoticed by the patients.
Adaptation after-effect was measured during blocks of
unperturbed trials, i.e., without intra-saccadic target shift,
before (‘pre’) and after (‘post’) completing five blocks of
the adaptation run (‘ada1’ to ‘ada5’).
RS session The adaptation of RS was tested in this session.
The first adaptation blocks comprised 48 trials each (24
leftward saccades randomly intermingled with 24 rightward
saccades). Each adaptation trial (Fig. 1b) started with the
presentation of a central fixation point (FP) for a pseudorandom period (1,300 or 1,500 or 1,700 ms). Then, FP was
replaced by a target (T1) located randomly at 8° to the right
or to the left (±8°). During the ensuing horizontal RS, T1
shifted back to T1′ by 25% of the initial target displacement
for blocks ada1, ada2, and ada3 (144 trials), and by 40% for
blocks ada4 and ada5 (96 trials). Finally, T1′ was switched
off 800 ms after T1 onset, and FP re-appeared 1,200 ms
later for the next trial. Pre-adaptation and post-adaptation

Fig. 1. MRI scans and adaptation protocols. a FLAIR (patient
P1) and diffusion (patient P2)
images showing stroke lesions
(arrows) located at the intermediate level of the rostral right
hemisphere (P1) and at the
caudal paramedian part of the
right hemisphere (P2). b RS
adaptation for a rightward trial.
Peripheral target T1 turned on
simultaneously when fixation
point FP turned off (rightward
arrow), eliciting a primary saccade. During this saccade, T1
shifted backward to T1′ (leftward arrow). c SVS adaptation
for a rightward trial. Subjects
sequentially scanned targets
(from T0 to T3) at their own
pace. During each of the three
SVS, the whole display shifted
backward (leftward arrows)
until reaching the T0′′′–T3′′′
positions

trials were identical to adaptation trials, except that target
T1 did not shift but instead disappeared during primary
saccades. Each test block comprised 12 leftward saccades
randomly intermingled with 12 rightward saccades.
SVS session The adaptation of SVS was tested in this
second session. The adaptation blocks comprised 16 trials

of three SVS each. Each adaptation trial (Fig. 1c) started
with the presentation of FP at 8° to the right (leftward trial)
or left (rightward trial) of the screen center and 4° above.
After 1,600 ms, a circle surrounding FP was presented
simultaneously with targets T0 to T3 (T0 located below FP)
along the central horizontal axis with an 8° regular spacing.
The circle disappeared 500 ms later, signalling subjects to

perform a downward saccade to T0 followed by a sequence
of three horizontal SVS. During each of these SVS, the
entire set of targets was shifted horizontally backward by
25% of the horizontal target separation for blocks ada1,
ada2, and ada3 and by 40% for blocks ada4 and ada5. Preadaptation and post-adaptation trials differed from adaptation trials in two aspects. First, only three targets (FP, T0,
and T1) were presented: T0 at the center of the screen, FP
at 4° above T0, and T1 at −8° (left) or +8° (right) from T0.
Subjects thus performed a vertical saccade from FP to T0,
followed by a single horizontal SVS from T0 to T1. Second,
this display did not shift but instead disappeared during the
SVS. Each ‘pre’ and ‘post’ block was subdivided in two
sub-blocks of 12 rightward and 12 leftward trials.

Data Analysis
Signals from the two eyes were averaged to yield the
cyclopean eye position. After automatic detection and
visual check of all primary saccades, the horizontal saccade
gain (=ratio of eye displacement amplitude to target
eccentricity) was submitted to one-way ANOVAs with the
“trial block” factor (seven levels=‘pre’, ‘ada1’–‘ada5’, and
‘post’) separately for P1 and P2, for RS and SVS, and for
rightward and leftward saccades. Post hoc tests (LSD
Fisher) compared saccade gain during ‘ada’ blocks and
‘post’ blocks to the ‘pre’ block. Finally, the adaptation
after-effect [100*(‘pre’ gain−‘post’ gain)/‘pre’ gain] was
compared to the 95% confidence intervals computed in six
healthy subjects from our previous study [1].

Results
Table 1 summarizes the mean baseline gain of RS and SVS
measured during the pre-adaptation block. For statistical
analysis, these data were compared to control data collected
under the same conditions in a group of six healthy subjects
(age range 22–48 years, two women, four men; see [1]).
Saccade gains in patients were considered as abnormal (see
arrows in the table) if the means were outside the 95%
confidence intervals computed from the control mean gains.
These comparisons revealed that saccades in patients were
either normometric or hypometric, with the exception of a

very moderate gain increase for contralesional RS in P1.
Note that neither this pattern of dysmetria nor the betweensubjects variations of gain variability (standard deviation,
s.d.) closely paralleled the pattern of adaptation deficits
presented in the following text.
Time-course of Saccadic Gain Changes
Figure 2a shows the changes in saccade gain in the five
adaptation blocks and the post-adaptation block relative to
pre-adaptation. Each patient showed a specific pattern of
adaptation depending on the direction (ipsilesional versus
contralesional) and type (RS versus SVS) of saccade. For
the contralesional direction, both patients showed a decrease in RS and SVS gain in both adaptation and postadaptation blocks, consistent with the expected effect of
backward adaptation training. ANOVAs confirmed a significant effect of the “trial block” factor (F=8 to 41, P<0.001;
post hoc comparisons with the ‘pre’ block shown in
Fig. 2a). These gain reductions followed a time course
closely resembling that of healthy subjects [1] and revealed
that adaptation of contralesional saccades was preserved in
P1 and P2. For ipsilesional saccades, a different picture
emerged. P1 exhibited preserved adaptation abilities for RS,
as shown by the significant decrease of gain in adaptation
and post-adaptation blocks (ANOVA, F=23, P<0.001).
However, P1’s adaptation abilities for SVS were altered: the
gain change during adaptation (F = 7, P < 0.001) was
moderate and inconsistent between blocks, and it was not
maintained in post-adaptation (post hoc comparison with
‘pre’, P>0.05). P2 showed the opposite pattern: SVS
adaptation was fully preserved (F=10, P<0.001), whereas
RS adaptation was strongly impaired both during the
adaptation and post-adaptation phases (F=2, P>0.15).
Adaptation After-effect
We then compared the adaptation after-effect in P1 and P2
(Fig. 2b) to 95% confidence intervals computed from our
previous study in six healthy subjects [1]. For the contralesional direction, both RS and SVS after-effects were within
control confidence intervals, strengthening our conclusion
that saccadic adaptation was normal despite the cerebellar
lesions. In contrast, for the ipsilesional direction, adaptation
after-effects were significantly smaller for SVS in P1, and for

Table 1 Mean (s.d.) baseline saccade gain measured in the pre-adaptation block

Patient P1
Patient P2

Ipsilesional RS

Ipsilesional SVS

Contralesional RS

Contralesional SVS

0.89 (0.11)
0.76 (0.13) ⇩

0.83 (0.06) ⇩
0.92 (0.03)

0.98 (0.06) ⇧
0.76 (0.09) ⇩

0.94 (0.12)
0.91 (0.03)

⇩ and ⇧: statistically significant decrease or increase (P<0.05) relative to controls’ 95% confidence intervals. see Ref. [1]

Fig. 2. Adaptation abilities of patients P1 and P2. a Mean gain
change of contralesional (top) and ipsilesional (bottom) saccades in
adaptation (‘ada1–ada5’) and post-adaptation (‘post’) trial blocks
relative to the ‘pre’ block. Negative values indicate a gain decrease
whereas positive values indicate a gain increase. *P<0.05, **P<0.01,
***P<0.001, ns P>0.05 (post hoc LSD Fisher tests comparing

saccade gains to the ‘pre’ saccade gain). b Adaptation after-effect:
mean percent gain change (100*(‘pre’−‘post’)/‘pre’) for RS (white
bar) and SVS (black bar) computed in patients. These values were
statistically compared to the means (filled squares) and 95%
confidence intervals (whiskers) computed in six healthy subjects from
our previous study [1]: *P<0.05, ns P>0.05

RS in P2 (and marginally larger for RS in P1) as compared
to controls. Thus, the reduced levels of ipsilesional saccade
adaptation described above (Fig. 2a) did not result from
idiosyncratic fluctuations but rather from the cerebellar
lesion itself.

Discussion
The double dissociation observed for ipsilesional saccades
reveals that medial (patient P2) and lateral (patient P1) parts
of the cerebellum selectively participate in RS and in SVS
adaptation, respectively. These findings confirm the contribution of the MP cerebellum to RS adaptation [10–13].
More importantly, they demonstrate for the first time that
cerebellar areas located more laterally than the MP
cerebellum also contribute to saccadic adaptation (SVS).
Remarkably, a role of the lateral cerebellum in saccade
control has so far remained elusive [16], notably because
the effects of lesions to this area are weak or absent. Thus,
the function of both saccade-related activity and oculomotor outputs of these lateral areas are still a mystery. The
present study reveals that one possible function is the
adaptive control of SVS, in line with the recent proposal of
a lateral cerebellum specialization for voluntary saccades
[17]. Unfortunately, for practical reasons, RS and SVS
adaptation could only be tested once in each patient,
preventing us from testing for the consistency of the
present findings. However, despite the known variability
of saccadic adaptation from session to session, the deficits
reported in the patients do likely result specifically from
their lesion because contralesional saccade gain always
adapted normally.
Our second original finding is that RS and SVS
adaptations rely upon separate neural substrates in the
cerebellum, which directly supports our predictions [1].
Recent studies in healthy human subjects have measured
the pattern of adaptation transfer either between RS and
SVS [1, 18] or from RS and SVS to hand pointing
movements [19] and to antisaccades (Cotti et al., submitted)
and have all proposed that RS and SVS adaptation rely on
separate mechanisms. The present study confirms this
proposal by revealing, in the human cerebellum, a neural
substrate of this functional separation between the two
types of saccadic adaptation. Further recordings in groups
of patients must be performed to more clearly define the
lateral and medial cerebellar areas involved respectively in
SVS and RS adaptation. High resolution 3-D imaging of
the cerebellum would thus be necessary to establish
detailed anatomo-functional correlations. Note also that
the contribution of these lateral and medial cerebellar areas
may not be the same with respect to the acquisition and/or
retention of adaptation, as suggested by the present

observations. In any case, with this unique anatomofunctional organization, the cerebellum may not only
implement plasticity related to adaptation per se, but may
also process task-related signals to switch between different
adapted states, as previously suggested for limb motor
responses in animals [20, 21]. This advance in knowledge
of the selective and plastic functions housed in the
cerebellum will help our understanding of how the central
nervous system maintains, over the long term, proficient
visuo-motor interactions with our environment.
Acknowledgements Work performed in the “Mouvement et Handicap” IFNL platform (Lyon, France). Supported by “Hospices Civils
de Lyon” (HCL/P 2002.303), INSERM U864, CNRS (DP), and a
FRM fellowship (NA). Thanks to A. Farnè, N. Holmes, and B. White
for valuable comments.

References
1. Alahyane N, Salemme R, Urquizar C, Cotti J, Guillaume A,
Vercher J-L, Pélisson D (2007) Oculomotor plasticity: are
mechanisms of adaptation for reactive and voluntary saccades
separate. Brain Res 1135:107–121
2. Ito M (1984) The cerebellum and neural control. Raven, New York
3. Ito M (2006) Cerebellar circuitry as a neuronal machine. Prog
Neurobiol 78:272–303
4. Schmahmann JD (1997) The cerebellum and cognition. Academic,
San Diego
5. Bloedel JR (2004) Task-dependent role of the cerebellum in motor
learning. Prog Brain Res 143:319–329
6. Deubel H (1995) Separate adaptive mechanisms for the control of
reactive and volitional saccadic eye movements. Vis Res 35:
3529–3540
7. Krakauer JW, Ghilardi MF, Ghez C (1999) Independent learning
of internal models for kinematic and dynamic control of reaching.
Nat Neurosci 2:1026–1031
8. Hopp JJ, Fuchs AF (2004) The characteristics and neuronal substrate
of saccadic eye movement plasticity. Prog Neurobiol 72:27–53
9. Wainscott SK, Donchin O, Shadmehr R (2005) Internal models
and contextual cues: encoding serial order and direction of
movement. J Neurophysiol 93:786–800
10. Desmurget M, Pélisson D, Urquizar C, Prablanc C, Alexander
GE, Grafton ST (1998) Functional anatomy of saccadic adaptation
in humans. Nat Neurosci 1:524–528
11. Takagi M, Zee DS, Tamargo RJ (1998) Effects of lesions of the
oculomotor vermis on eye movements in primate: saccades. J
Neurophysiol 80:1911–1931
12. Barash S, Melikyan A, Sivakov A, Zhang M, Glickstein M, Thier
P (1999) Saccadic dysmetria and adaptation after lesions of the
cerebellar cortex. J Neurosci 19:10931–10939
13. Straube A, Deubel H, Ditterich J, Eggert T (2001) Cerebellar lesions
impair rapid saccade amplitude adaptation. Neurology 57:2105–
2108
14. Volkmann FC, Schick AM, Riggs LA (1968) Time course of visual
inhibition during voluntary saccades. J Opt Soc Am 58:562–569
15. Goodale MA, Pélisson D, Prablanc C (1986) Large adjustments in
visually guided reaching do not depend on vision of the hand or
perception of target displacement. Nature 320:748–750
16. Pélisson D, Goffart L, Guillaume A (2003) Control of saccadic
eye movements and combined eye/head gaze shifts by the medioposterior cerebellum. Prog Brain Res 42:69–89

17. Simo LS, Krisky CM, Sweeney JA (2005) Functional neuroanatomy of anticipatory behavior: dissociation between sensorydriven and memory-driven systems. Cereb Cortex 15:1982–
1991
18. Collins T, Dore-Mazars K (2006) Eye movement signals influence
perception: evidence from the adaptation of reactive and volitional
saccades. Vis Res 46:3659–3673
19. Cotti J, Guillaume A, Alahyane N, Pélisson D, Vercher JL (2007)
Adaptation of voluntary saccades, but not of reactive saccades,

transfers to hand pointing movements. J. Neurophysiol 98:602–
12
20. Lewis RF (2003) Context-dependent adaptation of visually-guided
arm movements and vestibular eye movements: role of the
cerebellum. Cerebellum 2:123–130
21. Shimansky Y, Wang JJ, Bauer RA, Bracha V, Bloedel JR (2004)
On-line compensation for perturbations of a reaching movement is
cerebellar dependent: support for the task dependency hypothesis.
Exp Brain Res 155:156–172