Quantifying cognitive processes in the human brain using measures of dependence

The question of localizing cognition in the human brain is an old and difficult one. Particularly challenging is to understand the fascinating human ability to recognize and identify faces. The exquisite capacity to perceive facial features has been explained by the activity of neurons particularly responsive to faces, found in the fusiform gyrus and the anterior part of the superior temporal sulcus. This dissertation hypothesizes and demonstrates that it is possible to detect automatically the recognition of faces solely from processed electroencephalograms (EEG) in an online fashion with high temporal resolution using measures of statistical dependence applied on steady-state visual evoked potentials (ssVEPs).;EEG recordings are first modeled as an indexed family of random variables belonging to a stochastic process. Measures of dependence exploit bivariate distributions among pairwise channel recordings, and is a more realistic approach to quantify the joint spatio-temporal data distribution than previous methods just working with the marginal distributions, since the latter implicitly assume statistical independence between measurements. Standard and novel dependence measures were applied to estimate dependence within the filtered current source density (CSD) data. Based on previous and recent literature, the analysis included measures of (i) linear and monotone correlation (Pearson's r, Spearman's rho and Kendall's tau), (ii) synchrony (using phase-locking statistics), (iii) mutual information (using k-nearest neighbors), and (iv) entropy (permutation and approximate entropy). Novel approaches to quantify dependence are proposed using the concepts of generalized association (GMA and TGMA) and weighted-permutation entropy (WPE).;Dependencies between channel locations were assessed for two separate conditions elicited by distinct pictures flickering at a rate of 17.5 Hz. Filter settings were chosen to minimize the distortion produced by bandpassing parameters on dependence estimation. A dynamic graph visualizing the dependence evolution in time was generated for each condition and dependence measure. Several concepts from graph theory were adapted to analyze the resulting graphs and identify the active recording sites. Measures of centrality were particularly useful in determining the main channels involved in the cognitive response and a connected components analysis was employed to study in depth the network structure.;A classification framework based on information theoretical concepts is further developed by computing a similarity measure between two matrices storing the dependence information. This measure is then used to determine whether or not two matrices share the same condition. Besides, statistical analysis was performed for automated stimuli classification on six participants using the Kolmogorov-Smirnov test. Results show active regions in the occipito-parietal part of the brain for both conditions with a greater dependency between occipital and inferotemporal sites for the face stimulus. This aligns with previous evidence suggesting re-entrant organization of the ventral visual system, showing heightened re-entry when viewing meaningful or salient stimuli. Further research should investigate whether the communication pattern observed in this study is direct or enabled via one or more intermediate sources.