Reward-associated modulation of motor adaptation - An fMRI study
Poster No:
2060
Submission Type:
Abstract Submission
Authors:
Khushboo Punjabi1,2, Penelope Tilsley1,2, Bruno Nazarian3, Martin Weygant4, Fabrice Sarlegna5, Nicole Malfait6, Jean-Philippe Ranjeva1,2, Jan-Patrick Stellmann1,2,7
Institutions:
1Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging, Marseille, France, 2APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging, Marseille, France, 3Aix Marseille Univ, CNRS, Centre IRM-INT@CERIMED, Institut des Neurosciences de la Timone; UMR 7289, Marseille, Marseille, 4Charité – Universitätsmedizin Berlin, NeuroCure Clinical Research Center, Charité Campus Mitte, Berlin, Germany, 5Aix Marseille University, CNRS, ISM; UMR 7287, Marseille, France, 6Aix Marseille University, CNRS, INT; UMR 7289, Marseille, France, 7APHM, Timone University Hospital, Department of Neuroradiologie, Marseille, France
First Author:
Khushboo Punjabi
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Co-Author(s):
Penelope Tilsley
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Bruno Nazarian
Aix Marseille Univ, CNRS, Centre IRM-INT@CERIMED, Institut des Neurosciences de la Timone; UMR 7289
Marseille, Marseille
Aix Marseille Univ, CNRS, Centre IRM-INT@CERIMED, Institut des Neurosciences de la Timone; UMR 7289
Marseille, Marseille
Martin Weygant
Charité – Universitätsmedizin Berlin, NeuroCure Clinical Research Center, Charité Campus Mitte
Berlin, Germany
Charité – Universitätsmedizin Berlin, NeuroCure Clinical Research Center, Charité Campus Mitte
Berlin, Germany
Jean-Philippe Ranjeva
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging
Marseille, France|Marseille, France
Jan-Patrick Stellmann
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging|APHM, Timone University Hospital, Department of Neuroradiologie
Marseille, France|Marseille, France|Marseille, France
Aix Marseille University, CNRS, CRMBM; UMR 7339, Institut Marseille Imaging|APHM, Timone University Hospital, CEMEREM, Institut Marseille Imaging|APHM, Timone University Hospital, Department of Neuroradiologie
Marseille, France|Marseille, France|Marseille, France
Introduction:
Adapting to delays in visual feedback of movements is an increasingly common phenomenon with the increased use of digital devices[1]. Reward prediction error (RPE) at the end of movement has been shown to impact motor adaptation[2]. Classical theories of motor control implicate the basal ganglia (BG) to specialize in reward-based learning and timing motor commands, and the cerebellum (Cb) in sensory prediction error (SPE)-based learning, and their feedback interacting only at the cortex (Ctx) level[3,4]. However, there is evidence supporting the involvement of Cb in reward processing and direct sub-cortical connections between these two structures[5,6,7]. Motor adaptation studies in humans mostly address this topic from a behavioural and electrophysiological perspective. These methods cannot test the interaction between sub-cortical structures and cortex during adaptation. Using a block-event mixed paradigm with a 3T fMRI protocol, we look for the BG-Cb-Ctx activity patterns associated with the effect of reward during temporal adaptation.
Methods:
20 healthy adults (10 females; mean age 25.5 ± 2.3 yrs) performed a task where they intercepted a target moving from right to left on the screen using ballistic movements with a joystick. Outcome feedback was manipulated to study RPE such that half of the successful trials were rewarded with an auditory-visual explosion of the target and visual feedback was perturbed to study for SPE by introducing a gradually increasing lag between the joystick position and its visual feedback (cursor). Three types of trials: rewarded (R+) and non-rewarded hits (R-) and misses (M), and two types of blocks: with or without delay (Baseline) were defined (Fig.1A). The entire task consisted of two rounds of baseline blocks followed by delay blocks (Fig.1B). Kinematic variables were calculated from the cursor coordinates using custom R scripts. Temporal hand error (THE) was computed as the temporal difference between the centers of the target and the joystick crossing the interception zone (positive if joystick crossed the interception zone before the target, otherwise negative).
Participants practiced the task and underwent imaging using T2*-w EPI and 3D T1-w MP2RAGE at 3T (Siemens Prisma). Images were preprocessed using Presurfer and the fMRIprep pipeline[8]. The onsets of R+, R- and M trials; Baseline and Delay blocks; and nuisance regressors for motion, WM and CSF signals were entered in a GLM design in participants' native space. The following contrasts were computed for each participant: R- vs M for differentiating success from failure, R+ vs R- for reward prediction, and R+ vs R- in delay vs baseline for the effect of reward in sensorimotor adaptation. These contrast maps were normalized to MNI space and used for their respective one-sample t-test 2nd level analysis with Nilearn. Group-level contrasts were visually inspected for differences in activation cluster peak locations.
Participants practiced the task and underwent imaging using T2*-w EPI and 3D T1-w MP2RAGE at 3T (Siemens Prisma). Images were preprocessed using Presurfer and the fMRIprep pipeline[8]. The onsets of R+, R- and M trials; Baseline and Delay blocks; and nuisance regressors for motion, WM and CSF signals were entered in a GLM design in participants' native space. The following contrasts were computed for each participant: R- vs M for differentiating success from failure, R+ vs R- for reward prediction, and R+ vs R- in delay vs baseline for the effect of reward in sensorimotor adaptation. These contrast maps were normalized to MNI space and used for their respective one-sample t-test 2nd level analysis with Nilearn. Group-level contrasts were visually inspected for differences in activation cluster peak locations.
Results:
THE, RT and movement end time didn't differ significantly between blocks, but RT and Movement end time show a moderate correlation with THE (Fig.1C). Cb cortex, Caudate (Cau) and putamen(Put) - involved in object identity and value association activate along with insula, cingulate cortex(CC), med. sup. frontal gyrus (FG) and amygdala in distinguishing feedback of success from failure (Fig.2A). R+ trials elicit activity in the dentate nucleus, Cau, and pallidum activate with Sup. FG and sup. parietal lobule, CC and SMA- areas associated with perceiving reward and motor planning (Fig.2B). For reward and adaptation coupling (Fig.2C), activations in the Cb cortex, Put, PCu, SFG and MFG, dorsal posterior CC and pre-SMA suggest reward and adaptation feedback integration via the motor planning areas[9].
Conclusions:
In addition to the hypothesized sub-cortical BG-Cb co-activations, consistent involvement of the salience and central executive network regions indicates an increased cognitive dependence in successful coping with temporal adaptation[10].
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Motor Behavior:
Motor Planning and Execution
Motor Behavior Other 1
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Subcortical Structures
Keywords:
Basal Ganglia
Cerebellum
FUNCTIONAL MRI
Other - Reward, Motor adaptation
1|2Indicates the priority used for review
Provide references using author date format
[2] J. Izawa and R. Shadmehr, “Learning from sensory and reward prediction errors during motor adaptation,” PLoS Comput Biol, vol. 7, no. 3, 2011, doi: 10.1371/journal.pcbi.1002012.
[3] A. Klaus, J. Alves da Silva, and R. M. Costa, “What, If, and When to Move: Basal Ganglia Circuits and Self-Paced Action Initiation,” Annu Rev Neurosci, vol. 42, no. 1, pp. 459–483, Jul. 2019, doi: 10.1146/annurev-neuro-072116-031033.
[4] K. Doya, “Complementary roles of basal ganglia and cerebellum in learning and motor control,” Curr Opin Neurobiol, vol. 10, no. 6, pp. 732–739, Dec. 2000, doi: 10.1016/S0959-4388(00)00153-7.
[5] D. Kostadinov and M. Häusser, “Reward signals in the cerebellum: Origins, targets, and functional implications,” Neuron, vol. 110, no. 8. Cell Press, pp. 1290–1303, Apr. 20, 2022. doi: 10.1016/j.neuron.2022.02.015.
[6] E. A. Pelzer et al., “Cerebellar networks with basal ganglia: feasibility for tracking cerebello-pallidal and subthalamo-cerebellar projections in the human brain,” European Journal of Neuroscience, vol. 38, no. 8, pp. 3106–3114, Oct. 2013, doi: 10.1111/EJN.12314.
[7] A. C. Bostan, R. P. Dum, and P. L. Strick, “Functional Anatomy of Basal Ganglia Circuits with the Cerebral Cortex and the Cerebellum,” Prog Neurol Surg, vol. 33, pp. 50–61, 2018, doi: 10.1159/000480748.
[8] O. Esteban et al., “fMRIPrep: a robust preprocessing pipeline for functional MRI,” Nature Methods 2018 16:1, vol. 16, no. 1, pp. 111–116, Dec. 2018, doi: 10.1038/s41592-018-0235-4.
[9] E. M. Gordon et al. “A somato-cognitive action network alternates with effector regions in motor cortex,” Nature, vol. 617, pp. 351–359, 2023. doi: 10.1038/s41586-023-05964-2.
[10] E. C. Ceislik et.al., “Success versus failure in cognitive control: meta-analytic evidence from neuroimaging studies on error processing,” bioRxiv, May 2023, doi: 10.1101/2023.05.10.540136.