[Comp-neuro] Research Positions in Computational Neuroscience at OIST

Kenji Doya doya at oist.jp
Mon Sep 19 03:34:58 CEST 2011


Neural Computation Unit (PI: Kenji Doya) at Okinawa Institute of Science and Technology (OIST) is recruiting researchers and technical staff for the following four topics in its new research projects in Decision Making and Psychiatric Disorders:
1. Optical Imaging of Predictive Models in Cortical Microcircuits
2. Neuroinformatics for Decision Making Research
3. Machine Learning Approach to Diagnosis of Psychiatric Disorders
4. Modeling Dopamine and Serotonin Systems and Psychiatric Disorders

Please refer to the descriptions below for the details of each position. An applicant should have Ph.D., M.S., or a comparable degree in a relevant area. The contract is initially till March 2011 and renewable up to four years. Please send a CV and a brief description of research interest to:
nc-position-2011 at oist.jp
by October 10th, 2011. The search will continue until the positions are filled.

****
1. Optical Imaging of Predictive Models in Cortical Microcircuits
 
For perception in partially-observable environment and model-based decision making, prediction of the environmental state by combining previous percepts and the agent’s own action is critical. There have been a number of theoretical models of how cortical microcircuits implement predictive forward models and Bayesian inference, which deserve experimental testing. The researcher will work on two-photon imaging of mouse parietal/premotor/prefrontal cortex and/or analysis of massively parallel neural recording data to discover the neural circuit mechanism of "mental simulation".

2. Neuroinformatics for Decision Making Research
 
We are starting a new research program in “Brain Mechanisms for Prediction and Decision Making” with ten laboratories ranging from philosophy, psychology, neuroscience, psychiatry, machine learning and robotoics. They will produce a variety of rich experimental data including fMRI and PET images, multi-electrode recoding, optical imaging, and behavioral data from humans, primates, rodents, and fish. The role of the researcher or technical staff is to design, construct and maintain the database for efficient exchange of data between the collaborating labs and their public dissemination for further analysis and modeling. Development of software tools for data preprocessing, statistical data analysis, data-driven modeling and visualization is also an important research topic.
 
3. Machine Learning Approach to Diagnosis of Psychiatric Disorders
 
Our collaborating psychiatrists are collecting multi-dimensional data from depression and control subjects, including structural MRI, functional MRI under resting state and task performance, SNPs for serotonin/dopamine related genes, blood biomarkers, psychological tests, and clinical diagnosis. The researcher will develop a machine learning paradigm for the diagnosis and prediction of subtypes of depression. The first stage will take a supervised learning approach using doctors’ diagnoses, and the second stage will take an unsupervised learning approach to identify undocumented disease subtypes.

4. Modeling Dopamine and Serotonin Systems and Psychiatric Disorders

The monoaminergic systems including dopamine and serotonin play important roles in healthy behaviors and psychiatric disorders. Based on the neural and chemical recording and optogenetic manipulation data in awake behaving rodents, the researcher will construct a dynamic model of the interactions of the dopamine and serotonin systems with behaviors and the environment for the regulation of reinforcement and aversive learning under normal and pathological conditions.

****
Relevant links
Okinawa Institute of Science and Technology: http://www.oist.jp/
Neural Computation Unit: http://www.nc.irp.oist.jp/

----
Kenji Doya <doya at oist.jp>
Neural Computation Unit, Okinawa Institute of Science and Technology
1919-1 Tancha, Onna, Okinawa 904-0412, Japan
Phone: +81-98-966-8594; Fax: +81-98-966-2891
http://www.nc.irp.oist.jp/






More information about the Comp-neuro mailing list