2Being able to replicate real experiments with computational simulations is a unique opportunity 3 to refine and validate models with experimental data and redesign the experiments based on 4 simulations. However, since it is technically demanding to model all components of an experiment, 5 traditional approaches to modeling reduce the experimental setups as much as possible. In this 6 1 Allegra Mascaro, Falotico, Petkoski et al.
Towards closed-loop experiments and simulationsstudy, our goal is to replicate all the relevant features of an experiment on motor control and 7 motor rehabilitation after stroke. To this aim, we propose an approach that allows continuous 8 integration of new experimental data into a computational modeling framework. First, results show 9 that we could reproduce experimental object displacement with high accuracy via the simulated 10 embodiment in the virtual world by feeding a spinal cord model with experimental registration of the 11 cortical activity. Second, by using computational models of multiple granularities, our preliminary 12 results show the possibility of simulating several features of the brain after stroke, from the 13 local alteration in neuronal activity to long-range connectivity remodeling. Finally, strategies 14 are proposed to merge the two pipelines. We further suggest that additional models could be 15 integrated into the framework thanks to the versatility of the proposed approach, thus allowing 16 many researchers to achieve continuously improved experimental design. 17 18 response of the environment itself, in that the output of the brain is relevant only if it has the ability to 19 impact the future and hence the input the brain receives. This "closed-loop" can be simulated in a virtual 20 world, where simulated experiments reproduce actions (output from the brain) that have consequences 21 (future input to the brain) (Zrenner et al., 2016). To the aim of reproducing in silico the complexity of real 22 experiments, different levels of modeling shall be integrated. However, since modeling all components of an 23 experiment is very difficult, traditional approaches of computational neuroscience reduce the experimental 24 setups as much as possible. An "Embodied brain" (or "task dynamics", see Zrenner et al. (2016)) approach Allegra Mascaro, Falotico, Petkoski et al.
Towards closed-loop experiments and simulationscould overcome these limits by associating the modelled brain activity with the generation of behavior 25 within a virtual or real environment, i.e. an entailment between an output of the brain and a feedback 26 signal into the brain (Tessadori et al., 2012; DeMarse et al., 2001; Reger et al., 2000). The experimenter 27 can interfere with the flow of information between the neural system and environment on the one hand 28 and the state and transition dynamics of the environment on the other. Closing the loop can be performed 29 effectively by (i) validating the models on experimental data, and (ii) designing new experiments based on 30 the hypotheses f...