Neurostatistics: applications, challenges and expectations.

ABSTRACT

Brain function and its relations to cognition and behavior can be elucidated only by the use of various complementary methods. Over the past 20 years, we have been studying the brain mechanisms underlying spatial processes using different methods, including the recording of single cell activity in behaving monkeys, functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) in human subjects, all performing the same tasks. These methods provide partially overlapping perspectives, resulting in a gain in knowledge beyond the province of the individual method. A common aspect in this endeavor is the statistical analysis of the data acquired by different methods, especially regarding the encoding of information in unitary elements (single cell activity in neurophysiology, blood oxygenation level-dependent (BOLD) activation of voxels in fMRI, magnetic field strength in MEG) and the decoding of information from ensembles. In this paper we illustrate the various approaches, their data analysis and possible applications to medicine in the context of operations in space.