Latin-American Workshop on Computational Neuroscience
November 22-24, 2017
Francisco Sotres Bayón, PhD
Short Bio:
Dr. Francisco Sotres-Bayon, studied Biology at Universidad Nacional Autónoma de Mexico (UNAM), then did Doctorate studies at New York University (NYU) with Joseph LeDoux and Postdoctoral work at University of Puerto Rico (UPR) with Gregory Quirk. Currently he runs his own laboratory as Research Associate Professor in the Instituto de Fisiología Celular at UNAM. He has studied the neurobiology of defensive responses to threats (fear) and its extinction. Currently his research focuses on the brain mechanisms that underlie the competition between threat-related behaviors and reward-related behaviors in rats. He has published 18 articles and two book chapters which have been cited more than 2,000 times. He has received several honors, including the Return Home Fellowship awarded by the International Brain Research Organization.
Title of the Talk: Understanding the brain mechanisms of learned and innate motivational conflict
Abstract: The ability to approach rewards and avoid threatening environmental stimuli iscritical for survival. The basolateral amygdala (BLA), ventral striatum (VS) and medial prefrontal cortex (mPFC) have been implicated in reward-seeking and threat-related behaviors. However, it is not clear whether these brain regions are necessary when an animal is challenged to instantly weight whether the chance to obtain a reward is worth facing a threat. To address this, we have developed three behavioral tasks where rats are faced with the conflict of approaching a reward (food) despite the drive to avoid a threat (shock). These three tasks are similar in that the behavioral conflict involves presenting simultaneously stimuli with opposing motivational value. But the tasks differ in that the types of motivational value of the stimuli can be acquired (learned) and/or innate (not learned): one task involves the competition of two learned stimuli (learned/learned), another involves one learned stimulus in conflict with an innate stimulus (learned/innate) and a last one involves the competition of two innate stimuli (innate/innate). Using local pharmacological inactivations, we found that: 1) BLA is necessary for threatseeking behavior, whereas VS is necessary for reward-seeking behavior in the learned/learned conflict task, 2) prelimbic cortex region of the mPFC is necessary for the interaction of reward and threat-seeking behaviors in the learned/innate task, and finally 3) VS is necessary for reward-seeking behavior in the innate/innate conflict task. Together, these preliminary results support the notion that BLA is involved in threat-seeking, VS in reward-seeking and mPFC may be involved in reward-threat interaction in motivation conflict. Undergoing experiments explore systematically each of these structures in each our three motivational conflict tasks. We hypothesize that both BLA and VS are necessary for threat and reward-seeking behavior regardless of whether the motivational stimuli are learned or innate, but notably the prefrontal cortex is exclusively necessary when motivational conflict involves at least one learned motivational stimuli (as in our recent study: Ramirez-Lugo et al., 2016). Understanding the brain mechanisms underlying the behavioral responses of learned and innate motivational conflict in rats could help treat emotional disorders characterized by deficits in emotional decision-making in humans.
Prof. Laurence Devillers
Short Bio:
Academic qualifications:
Present academic position:
Full Professor of Computer Science (Affective Computing and AI) at Paris-Sorbonne University (2011-present) Researcher at LIMSI-CNRS Distinguished Lecturer for ISCA 2017-2018
Research work:
Her research background is in affective computing, machine learning, speech recognition, signal analysis, spoken dialog system, evaluation and ethics. Since 2001, she leads a team at LIMSI-CNRS (Orsay) on “Affective and social dimensions in spoken interaction with robots” (5-10 researchers). She already directed 10 PHD thesis (+ 4 current PHDs) and 5 post-docs. She participates in BPI ROMEO2 project, which has the main goal of building a social humanoid robot. She leads the EU CHISTERA project JOKER: JOKe and Empathy of a Robot. She is involved in the Eurobotics Topic Groups on social and affective robotics. She is member of the CERNA (French national committee) on the ethics of the Research in Robotics and heads the Machine Learning/AI and Ethics WG. She is also involved in the Affective Computing Committee of the IEEE Global Initiative for Ethical Considerations in the Design of Autonomous Systems (2016-17). She also wrote a book for a wide audience on the regulation, the ethics and the interaction man-robot (Plon editor in France): “ Robots and Humans: myths, fantasms and reality (Des robots et des homes: mythes, fantasmes et réalité)” (2017).
Title of the Talk: Towards social and affective relations with a robot : Joke and Empathy of a Robot/ECA
Abstract: Talk during social interactions naturally involves the exchange of propositional content but also and perhaps more importantly the expression of interpersonal relationships, as well as displays of emotion, affect, interest, etc. in order to provide a companion-machine (robot or ECA) with the skills to create and maintain a long term social relationship through verbal and non verbal language interaction. Such social interaction requires that the robot has the ability to represent and understand some complex human social behavior. It is not straightforward to design a robot with such abilities. Social interactions require social intelligence and ‘understanding’ (for planning ahead and dealing with new circumstances) and employ theory of mind for inferring the cognitive states of another person. This talk will review modeling human emotions using speech and language and will discuss how to use these models in advanced dialogues employing complex social behaviors such as Joke and Empathy.
Prof. Dr. Sen Cheng
Short Bio: Sen Cheng is Professor of Computational Neuroscience in the Institute for Neural Computation at the Ruhr University Bochum, co-chair of the Mercator Research Group Structure of Memory, and speaker of the Research Department of Neuroscience. The goal of his research is to understand the cognitive and neural mechanisms of episodic memory and spatial representations. To this end, he mostly uses computational methods, including spiking neural networks, cognitive modeling and machine learning. He has published more than 20 peer-reviewed articles in the field of learning and memory.
Title of the Talk: Intrinsic sequences in the hippocampus for spatial navigation and memory storage
Abstract: The hippocampus in the mammalian brain is known for two seemingly disparate things: episodic memory in humans and spatial representation in rodents and other species. However, it is not understood how the hippocampus implements these two functions and what they have in common. In this talk, I will present a new theory in which the hippocampus stores neural sequences that represent episodic memories. I will present new modeling results that show how such sequences might be stored in the hippcampal circuit. Sequential activity also plays an important role in spatial behaviors. Place cells are (re-)activated in a sequential order that reflects the sequence of the animal’s prior locations or the upcoming trajectory, even when the animal is not moving. We can account for various types of sequential activity and their differences within a single model. I conclude that the key function of hippocampus is to generate intrinsic sequences that are used for episodic memory and spatial navigation.
Vinicius Rosa Cota
Short Bio:Vinícius R. Cota is associate professor in the Department of Electrical Engineering of Federal University of São João Del-Rei (UFSJ) in the field of Neuroengineering. He received his bachelor degree of Electrical Engineer from Federal University of Minas Gerais in 2002 and received his PhD in Bioinformatics from the same institution in 2007. During his doctorate, Dr. Cota studied the neurodynamics of forebrain areas in Epilepsy, aiming at EEG feature extraction for seizure prediction and the usage of electrical stimulation for seizure suppression. During his post doctorate in the International Neuroscience Institute of Natal (Rio Grande do Norte, Brazil), he investigated causal relations between states of the sleep-wake cycle and memory consolidation according to Hebbian postulates. At UFSJ, he currently researches both neurobiological and technological aspects of neuroengineering as a means to treat epilepsy and related comorbidities. With collaborators, he was the developed a nonstandard, temporally unstructured low frequency pattern of therapeutic brain electrical stimulation, of which he holds a patent. He is also the founder and leader of the Laboratory of Neuroengineering and Neuroscience (LINNce) at UFSJ.
Title of the Talk: Toward a neural prosthesis to the epileptic brain by tackling neural hypersinchronism
Abstract: It is now well established that neural function emerges from the network activity of interacting brain units, rather than that of isolated substrates. Interaction, by its turn, relies on a multitude of physiological mechanisms responsible for joining brain areas into synchronized oscillatory activity. By this token, disturbances in neural synchronization, either the lack or excess thereof, can lead to disease. Of particular interest, epilepsy, a neurological disorder of major importance worldwide, can be understood as a dysfunction of hypersynchronism. Our research group has successfully developed and tested a novel neurostimulation strategy to disrupt excessive neural synchronization as a means to treat refractory epilepsy and related neurological disorders: a temporally unstructured electrical stimulation with a low mean frequency, or NPS (after non-periodic stimulation). When applied to the amygdala of rats, NPS displayed robust anticonvulsant, antiepileptogenic, and anxiolytic effects, without jeopardizing memory, background anxiety, and motor functions. Mechanistic investigation using in vivo electrophysiology and functional imaging suggests both local circuit desynchronization and recruitment of extra-limbic areas with a modulatory / inhibitory effect on the limbic system underlie the therapeutic effect of NPS. We now pursue in-depth investigation on the relationship between temporal structure of electrical stimulation and synchronism in the brain.
Guillermo Cecchi, PhD
Short Bio: Guillermo Cecchi received an education in Physics (MSc, University of La Plata, Argentina), Physics and Biology (PhD, The Rockefeller University), and Imaging in Psychiatry (Postdoctoral Fellow, Cornell University). He has been interested in diverse aspects of theoretical biology, including Brownian transport, molecular computation, spike reliability in neurons, song production and representation in songbirds, statistics of natural images and visual perception, statistics of natural language, and brain imaging. In 2001 he joined IBM Research to work on computational approaches to brain function. In recent years, Dr. Cecchi has pioneered the use of a computational linguistics approach to quantify psychiatric conditions from short speech samples, applying it successfully to conditions as diverse as schizophrenia, mania, prodromal psychosis, and drug and alcohol intake.
Title of the Talk: Speech-based automated diagnosis and prognosis of neuropsychiatric disorders
Abstract: We describe recent preliminary studies demonstrating that computational analysis of spoken and written language can provide for highly accurate diagnostics and prognosis over a wide variety of psychiatric and neurologic conditions, including psychosis, drug abuse, Parkinson’s, and Alzheimer’s. These results are based on the mathematical formalization of psychiatric qualitative knowledge related to the characterization of the conditions (e.g., derailment and literality in psychosis) and drug effects (e.g., increased intimacy/affection with the use of the recreational drug ecstasy), as well as novel linguistic feature extraction approaches, including the application of statistical network theory. We will also discuss the implications for mental health (and possibly, computer science) of a systematic application of this methodology, extended to include similar readily available behavioral data such as voice and video.
Prof. Fernando Cendes
Short Bio: After graduating in medicine and completing a neurology residency at UNICAMP, Brazil, Fernando Cendes did a postdoctoral fellowship on EEG, neuroimaging, and epilepsy at McGill University–Montreal Neurological Institute and Hospital, Montreal, QC, Canada from 1991–1997 and received a Ph.D. in Neuroscience from McGill University in 1997. He is currently Full Professor of Neurology and Coordinator of the epilepsy surgery program at the Department of Neurology, Medical School of the University of Campinas (UNICAMP), Brazil. He is a member of the Diagnostic Methods Commission of the International League Against Epilepsy and Associate Editor of Epilepsy. His current research interests include neuroimaging and the medical and surgical treatment of epilepsy.
Title of the Talk: Imaging normal and abnormal brain plasticity
Abstract: Successful behavior throughout the evolution of species depends on the capacity to adapt to the environment. Practice elicits use-dependent plasticity and may result in encoding adaptive strategies for subsequent recall (Hummel & Cohen 2005). Thus, learning leads to plastic changes in the cerebral cortex that store the learned strategies and make them available to optimize future behavior (Pascual-Leone et al., 2005). Brain plasticity has been defined as any enduring change in cortical properties either morphological or functional in response to environmental stimuli (Hummel & Cohen 2005). Different forms of brain plasticity have been studied in the literature in both animal models and humans. In addition to learning and memory, neural plasticity participates in the process of functional recovery that follows brain lesions such as stroke. There is also “bad” plasticity leading to progressive structural and functional deterioration of the CNS, which occurs, for example, in epilepsy and neurodegenerative disorders, such as Alzheimer’s disease and Parkinson’s disease. In the past, brain plasticity was thought to be a property only of the developing brain, but it has been shown to occur in adults (Doyon & Benali 2005). Plastic changes can occur both at the level of the synapse and the axon, resulting in changes to connections that can have behavioral implications. Perhaps the most striking example of brain plasticity is the process of memory. Studies of retrograde amnesia in memory impaired patients, neuroimaging with healthy volunteers, and experimental animals have shown that the recall of acquired memories is initially dependent on the hippocampus, but over time there is a gradual increase in dependency upon extra-hippocampal regions, such as neocortex (Clarke et al. 2009; Squire and Bayley, 2007; Wiltgen et al., 2004). Synaptic plasticity (Zhao et al. 2007), hippocampal neurogenesis (Kitamura et al. 2009), and brain-derived neurotrophic factor (BDNF) (Bekinschtein et al. 2007) play a key role in memory processes in adult rodents. The hippocampus is the most affected structure in mesial temporal lobe epilepsy (MTLE). MTLE is the most common form of partial epilepsy, characterized by atrophy of mesial temporal structures, mossy fiber sprouting, spontaneous recurrent seizures and cognitive deficits, in particular memory. In fact, much what is now known about memory in humans comes from observations and studies in patients with MTLE (Scoville and Milner, 1957). Studies have reported clinical, cognitive, and imaging deterioration over time in patients with MTLE. Experimental and pathologic data are concordant with the fact that MTLE is a progressive disorder and that the progressive neuronal loss or gray matter atrophy may not be restricted to the hippocampus. This “bad” plasticity is associated with poorer seizure control and a longer duration of epilepsy (Coan et al. 2009) and may recover, at least in part, after cessation of seizures, for example, after successful surgical treatment (Yasuda et al. 2009). Functional MRI (fMRI) can add important information about the cerebral dysfunctions related to seizures and medication, including those affecting memory and language. Memory encoding fMRI paradigms that activate language areas in both hemispheres have been used for investigating the network plasticity that occurs in epilepsy. A fMRI study showed that a complex network including parietal and frontal cortices are involved in verbal memory encoding and retrieval tasks in normal controls and patients with MTLE-HS (Alessio et al. 2013). However, the extension of these activations is more intense and widespread, particularly in the frontal lobes, in patients with left MTLE-HS, suggesting a functional reorganization of verbal memory processing due to the failure of the left hippocampal network system, or perhaps a bilateral limbic dysfunction. Of particular interest are the recent connectivity and functional network studies. Normal brain function depends on a complex interaction of structural and functional networks. An organized baseline brain function has been defined in imaging studies as the “default mode network,” which is normally suspended or deactivated during specific tasks. Disruptions in these networks in fMRI studies appear to be associated with cognitive and behavioral impairments seen in different disorders such as Autism, Alzheimer’s Disease, epilepsy, and others.
Prof. Antonio C. Roque, PhD
Short Bio: Antonio C. Roque was born in Sao Paulo, Brazil in 1963. He received his BSc degree in Physics from the State University of Campinas, Brazil, in 1983, and his PhD degree in Computer Science and Artificial Intelligence from the University of Sussex, UK, in 1992. He joined the faculty of the Department of Physics of the University of Sao Paulo at Ribeirao Preto, Brazil, in 1993 where he is now Associate Professor. He founded and is the current coordinator of the Laboratory of Neural Systems (sisne.org), a pioneer laboratory in computational neuroscience in Brazil. His research interests are detailed computational modeling of neurons and brain structures and animal behavior. Throughout his career, he published more than fifty journal papers and supervised eighteen PhD theses. He is the creator and organizer of the Latin American School on Computational Neuroscience (LASCON). He is currently Principal Investigator and Technology Transfer Coordinator of the Research, Innovation and Dissemination Center for Neuromathematics (NeuroMat), a Sao Paulo Research Foundation (FAPESP) Center. He is also former member of the Board of Directors of the Organization for Computational Neuroscience (OCNS) and the Brazilian Society for Neuroscience and Behavior (SBNeC).
Title of the Talk: A spiking neural network with synaptic noise to model spontaneous cortical activity dynamics
Abstract: Networks of cortical neurons display spontaneous activity even in the absence of external stimuli. The spontaneous cortical network activity is often described as synchronized during slow-wave sleep and under certain anesthetics, and asynchronous during quiet wakefulness. The underlying mechanisms that control transitions between these cortical states are not completely known. Here, a spiking network model with synaptic noise is used to study this problem. The model is composed of a mixture of excitatory and inhibitory neurons belonging to different electrophysiological cortical cell classes. In the absence of synaptic noise, the network displays spontaneous collective oscillations that resemble alternating up and down states observed in synchronized cortical states. When synaptic noise is added, the network activity is characterized by intermittent transitions between asynchronous and collective oscillating states. Systematic analysis of firing rates, power spectra and voltage series shows that characteristics of these two states are similar to those of asynchronous and synchronized cortical states. The results suggest that synaptic noise may be an important mechanism underlying transitions between asynchronous and rhythmic states in cortical networks.
Dr. Pedro Schetatsky, PhD
Short Bio:
Title of the Talk: Electricity and the Brain: past, present and future
Abstract: Nothing more common place than the phrase "to understand the future is necessary to know the past. However, this fits neatly with the evolution of noninvasive neuromodulation, which was born along with the advent of electricity itself - a reflection of the Industrial Revolution in Europe and the United States - declined temporarily with the "era of modern pharmacotherapy" and emerged from the ashes as a promising alternative to it. In addition to refractoriness of psychoactive drugs, these are generally costly and induce side effects that limit their prolonged use, opening space for non-pharmacological treatment. With the help of emerging technology, different neuromodulatory modalities tend to expand their clinical and home use in the long term, and can even be monitored at a distance. In this practical aspect, the stimulation by continuous electric current is much more attractive than the Transcranial Magnetic Stimulation, the latter being more used at the research level. The clinical use of a practice depends on the research. But research also depends on clinical use. The history of electrical and magnetic stimulation advanced the practice before the theory. And you got it right in many ways! Thanks to the use of two specialty medical specialty have emerged, Psychiatry and Neurology, hand in hand until today. It is up to the new generation to increasingly improve these techniques, both in the methodology of stimulation and in the development of drugs (eg, selective serotonergic reuptake inhibitors) and practices (eg, physical activity) that potentiate brain-induced electrical plasticity.
The confirmation of other invited speakers will be done soon.