Event Abstract

Back to Event

Simulation of plasticity damage in the cerebellar cortex during cerebellum-driven tasks

Alice Geminiani1*, Claudia Casellato1, Egidio D‘Angelo2, 3 and Alessandra Pedrocchi1
  • 1 Politecnico di Milano, Department of Electronics, Information and Bioengineering, Italy
  • 2 University of Pavia, Department of Brain and Behavioral Sciences, Italy
  • 3 Istituto Neurologico IRCCS Fondazione C. Mondino , Brain Connectivity Center, Italy

Introduction The role of the cerebellum in motor learning has been demonstrated through neurophysiological studies [1] and computational models [2]. The organized modular circuit and neural plasticity at different sites are essential to achieve adaptation and memory of cerebellar motor commands during specific tasks. In particular, plasticity in the cerebellar cortex is the main mechanism driving learning [3]: Long Term Potentiation (LTP) and Depression (LTD) change the activity of Purkinje Cells (PCs) that modulate the cerebellar output. Animal experiments have shown that a damage to cortical LTD prevents from adaptation in cerebellum-driven tasks, like Eye-Blinking Conditioning (EBC) and Vestibulo-Ocular Reflex (VOR) [4]. During EBC, after the repeated paired presentation of a Conditioned Stimulus (CS) and an Unconditioned Stimulus (US), the subject learns to produce a Conditioned Response (CR) anticipating US [5]. During VOR, when the subject is looking at a target and the head is moving, a compensatory eye motion keeps the gaze fixed on the target [6]. In both tasks, the cerebellum controls the output motor command modulation, thanks to synaptic plasticity. Therefore, a damage to cortical LTD rate results in impaired learning. Altered cortical LTD has been associated with human pathologies [7]; however, the main studies report data from animals, because it is difficult to isolate the contribution of single damages in patients. Therefore, a new tool to investigate pathological conditions is necessary, to fill the gap between animal and human studies and provide deeper insight into cerebellar diseases. A promising solution is to modify realistic computational models of human neural circuits and then embed them into closed-loop simulations to reproduce misbehaviours. In the current study, we used this approach to investigate LTD rate damage in the cerebellum during EBC and VOR, showing the promising role of computational neuroscience in understanding pathologies. Materials and Methods Simulations were performed through a realistic Spiking Neural Network (SNN) representing a cerebellar microcomplex, with physiological topology and connection ratios [1]. To simulate EBC and VOR, we provided two input signals (state information and noxious stimulus/error), which were coded by specific neural populations and depended on the protocol (Fig. 1). The output motor command produced a time-event (CR) when overcoming a threshold during EBC, while during VOR the signal drove continuously the eye motion. The network included specific learning rules at three cortical and nuclear plasticity sites. The healthy model was able to achieve fast adaptation due to cortical plasticity and slow consolidation of motor capabilities thanks to nuclear plasticity [3]. The learning rule parameters were optimized through a Genetic Algorithm [8, 9]. LTD rate damage was reproduced by decreasing the learning rule parameter regulating cortical LTD, from 10% to 70% of reduction. Each simulation consisted of 100 repetitions: • EBC: 10 blocks of 10 trials with 9 CS-US and 1 CS-alone repetitions; CS lasted 350 ms; US lasted 100 ms and co-terminated with CS. • VOR: in each trial, the target was fixed while the head was moving for 1 s according to a sinusoidal trajectory with amplitude 5° and frequency 1 Hz. To evaluate the outcome, we computed the %CR on blocks of 10 trials for EBC and the Root Mean Square gaze Error (RMSE) for VOR; the gaze error was the difference between the desired and actual eye position movement. Results During EBC the main effect of LTD rate reduction was a delayed acquisition of motor performance (Fig. 2A). In fact, when decreasing LTD rate by 10% to 50%, the model achieved the same %CR as in healthy conditions during the late acquisition blocks, but learning was slower. However, in case of severe damage (70% decrease), conditioning did not occur: there was not a significant improvement of %CR throughout the trials, with maximum 10% CR. This case matched the results of experiments on mice with LTD rate damage, where the %CR did not significantly increase at the end of the protocol [7]. During VOR simulations, for damages up to 50%, the RMSE decreased throughout the acquisition trials, by tuning the output motor command, without any significant delay in learning. In particular, for an intermediate damage, the RMSE decreased as fast as in healthy conditions but the residual value was higher (Fig. 2B). This result differs from the outcome of EBC simulations where conditioning occurred in late acquisition even for an intermediate damage. On the other hand, when LTD rate impairment was higher, the RMSE did not meaningfully decrease during the task, demonstrating that learning was completely compromised, like in EBC. This case reproduced the same lack of adaptation showed in mice during VOR in the reference study [4]. Discussion and Conclusion The present study provides a successful example on the use of realistic models of brain areas to simulate pathologies. We evaluated the behavioural effects of a damage to cortical LTD during two different cerebellum-driven tasks. The model was able to reproduce the same lack of motor learning as in animal experiments; then, by associating low-level impairment and misbehavior, we could predict the damage amount in the reference studies. Moreover, we could achieve a deeper insight into the role of cortical LTD. In fact, the delayed conditioning during EBC simulations for damages up to 50% proved that this plasticity mechanism is fundamental for fast learning. However, in case of more complex tasks like VOR, cortical LTD is crucial not merely for a time association, but also for a continuous tuning in amplitude and phase of the output. Further analysis on the low-level neural activity and synaptic weights would shed light on the underlying altered mechanisms. In particular, it could help testing compensatory mechanisms and suggesting possible treatments. Then, we could use the same approach to simulate different pathologies, involving damages to neural populations, such as silencing PCs to simulate cortical degeneration [10]. Finally, the investigation of pathologies through a computational model could suggest new elements to be introduced in the model. This will increase the realism and the reliability of simulations, allowing to use the model as a clinical tool for faster diagnosis and treatment evaluation.

Figure 1
Image
Figure 2
Image

Acknowledgements

The work was supported by grants of the European Union (CEREBNET FP7-ITN238686, REALNET P7-ICT270434, Human Brain Project HBP-604102) and Regione Lombardia (HBP-Lombardia project).

References

References
1. D’Angelo E, Casali S. Seeking a unified framework for cerebellar function and dysfunction: from circuit operations to cognition. Front Neural Circuits (2013) 6(116):1-23. doi:10.3389/fncir.2012.00116.
2. Casellato C, Antonietti A, Garrido JA, et al. Adaptive robotic control driven by a versatile spiking cerebellar network. PLoS One (2014) 9(11):1-17. doi:10.1371/journal.pone.0112265.
3. D’Angelo E. The organization of plasticity in the cerebellar cortex: from synapses to control. Prog Brain Res (2014) 210:31-58. doi:10.1016/B978-0-444-63356-9.00002-9.
4. De Zeeuw CI, Hansel C, Bian F, et al. Expression of a protein kinase C inhibitor in Purkinje cells blocks cerebellar LTD and adaptation of the vestibulo-ocular reflex. Neuron (1998) 20:495-508. doi: 10.1016/S0896-6273(00)80990-3.
5. Bracha V, Zhao L, Irwin KB, Bloedel JR. The human cerebellum and associative learning: dissociation between the acquisition, retention and extinction of conditioned eyeblinks. Brain Res (2000) 860:87-94. doi:10.1016/S0006-8993(00)01995-8.
6. Boyden ES, Raymond JL. Active reversal of motor memories reveals rules governing memory encoding. Neuron (2003) 39(6):1031-1042. doi:10.1016/S0896-6273(03)00562-2.
7. Miyata M, Kishimoto Y, Tanaka M, et al. A Role for Myosin Va in Cerebellar Plasticity and Motor Learning: A Possible Mechanism Underlying Neurological Disorder in Myosin Va Disease. J Neurosci (2011) 31(16):6067-6078. doi:10.1523/JNEUROSCI.5651-10.2011.
8. Carlson KD, Nageswaran JM, Dutt N, Krichmar JL. An efficient automated parameter tuning framework for spiking neural networks. Front Neurosci (2014) 8(10):1-15. doi:10.3389/fnins.2014.00010.
9. Antonietti A, Casellato C, Garrido JA, et al. Spiking Neural Network With Distributed Plasticity Reproduces Cerebellar Learning in Eye Blink Conditioning Paradigms. IEEE Trans Biomed Eng (2016) 63(1):210-219. doi:10.1109/TBME.2015.2485301.
10. Geminiani A, Antonietti A, Casellato C, D’Angelo E, Pedrocchi A. A Computational Model of the Cerebellum to Simulate Cortical Degeneration During a Pavlovian Associative Paradigm. In: IFBME Proceedings on the XIV Mediterranean Conference on Medical and Biological Engineering and Computing (2016): 1063-1068. doi:10.1007/978-3-319-32703-7_210.

Keywords: Cerebellum, Learning, Long Term Depression, Spiking Neural network, Pathological model, Eye-Blinking Conditioning, vestibulo-ocular reflex

Conference: Neuroinformatics 2016, Reading, United Kingdom, 3 Sep - 4 Sep, 2016.

Presentation Type: Poster

Topic: Computational neuroscience

Citation: Geminiani A, Casellato C, D‘Angelo E and Pedrocchi A (2016). Simulation of plasticity damage in the cerebellar cortex during cerebellum-driven tasks. Front. Neuroinform. Conference Abstract: Neuroinformatics 2016. doi: 10.3389/conf.fninf.2016.20.00086

Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.

The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.

Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.

For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.

Received: 31 May 2016; Published Online: 18 Jul 2016.

* Correspondence: Dr. Alice Geminiani, Politecnico di Milano, Department of Electronics, Information and Bioengineering, Milano, 20133, Italy, alice.geminiani@unipv.it