Day 2

KEYNOTE TALKS THE ROAD AHEAD

Self-organization of Computation in Neural Systems

Sakya Dasgupta

"Can we create an algorithm that allows a biological organism to think or remember?" This is the question that Sakya Dasgupta begins his lecture on Self-organization of Computation in Neural Systems. The researcher indicates that to build intelligent machines that can learn and think like people, it is necessary to develop algorithms that can learn general conditions from limited examples. In addition, systems need to understand cause and effect, and relationships between different objects. "It is important that the machines understand the physical reality of the way we do it," says the researcher. The processes closest to artificial intelligence are memory and learning, and can be the gateway to building systems that think like us. These processes allow us to build a network of neurons whose properties can be studied or programmed. We can explore how these networks of neurons can be used to make computations. In this aspect, we try to explore how plasticity allows the assembly of neural networks and how this mechanism can be developed in an unsupervised manner. In this aspect, the researcher focuses on ho to prevent disruption of the network by a mechanism called competition, where new inputs to the network modifies the existing ones to regulate the neuronal network activity. Definitely, the regulation of cell assemblies is one of the most interesting ways of road ahead.

Molecular Mechanisms of Gene X Environment Interactions

Elisabeth Binder

Dr. Elisabeth Binder focuses on the effect of genetic and epigenetic factors on the development of mental illness. There are multiple factors involved in he development of mental disease. For example, in the case of depression, there is not just one depression, but a wide range of environmental and genetic factors that influence people in different ways. "In psychiatry, the recognition of abuse in childhood is a big step forward, but in recent years we have realized that pregnancy is of prominent importance," says Dr. Binder. "In pregnancy, the mother's high levels of cortisol can pass through the placenta to the child." One of the major systems for the development of mental illness is the stress hormone system. Exposure to stressful sources releases glucocorticoid into almost every organ of the body. "Stress has an effect on the whole body, which even though we are not aware of it. Our body begins to function differently," she says. In fact, she says, glutocorticoids carry genetic information that somehow encodes the source of the stress effect and has the potential to modify future responses to the same stress.

At the same time, Professor Binder's research team has found through the study of human cells in culture that there is a wide variety of genes that can predispose individuals to the impact of stressful effects. "Methylation plays an important role in future responses to stress," she highlight. For example, the FKBP5 gene is highly sensitive to methylation and glucocorticoids. Even if there is no change in quantity, there is a memory that the methylation process allows us to identify. The methylation mechanism appears to offer new insights into how stress mechanisms progressively affect the body's response and affect the mental life of individuals.

STUDENT TALKS

Decoding the brain

In the session on synaptic time Synaptic Time-Dependent Plasticity Increases Signal Transmission Speed (STDP) , Pau Aceituno from the Marx Planck Institute, showed us the application of STDP , among with spectral properties of large random graphs results, to understand the dynamics of neural networks. His work focuses on the characteristics of STDP to decrease latencies. In case of a high complicated neuronal network, his research demostrate how difference in network structures change the evolution of the latency.

Subsequently, Paulina Dabrowska reported on her research on the study of network mechanisms leading to patterns in monkey motor cortex, using a Utah array to record spiking activity, similarities in the spiking activity among the population counts but different average FR distribution for simulated and experimental results (macaque motor cortex activities). This discrepancie can be associated to the connectivity parameters used in her model, which are derived from different species and cortices . Another reason is that the resting state can be notably less homogeneous that current simulation. At the end she gave us her thoughts on how can this model can be improved by a successive adaptation of the model conectivitity to value specific to monkey motor cortex, until experimental and simulation statistics agree, highligthing the vital role of statistics in neuroscience.

UNCONFERENCE

Perhaps the unconference session was one of the spaces that generated the most expectation among the doctoral students who attended ENCODS. Fortunately, we managed to collect a good and interesting amount of proposals. For three hours we had the opportunity to discuss p-value, better methods of water maze analysis, science communication, and even the students were able to do some salsa and yoga. Off course it is worth to mention those in relaxing session anywhere. Who would believe it? Thanks to the participation of all the sessions, they were a success.

Experience the Music Paint Machine

Manic VR

Close The Loop

Presentation Skills

Many participants are joining Anna Elena Pepe to present their elevator pitch and get an expert feedback and valuable tips. An elevator pitch is a very condensed talk about your work, that aims at convincing people in the short time that you spend with someone while you ride on an elevator or even waiting in line for the bathroom.

Microglia!

IMPRESSIONS

Google Sites

Report abuse