Functional organization of the human brain

Prix Jean Nicod 2023 - Cycle de conférences

Le Prix Jean Nicod vise à promouvoir les recherches philosophiques se rapportant à la cognition et à faire connaître en France les travaux réalisés à l’étranger dans ce domaine.
Le prix 2023 est décerné à Nancy Kanwisher, titulaire de la chaire Walter A. Rosenblith de neurosciences cognitives au département des sciences cérébrales et cognitives du Massachusetts Institute of Technology et chercheuse au McGovern Institute for Brain Research.
  Nancy Kanwisher © Illustration Marie Lods
Nancy Kanwisher © Illustration Marie Lods

L’esprit humain, son organisation, sa nature, ses relations avec le corps et avec le monde sont depuis toujours parmi les thèmes centraux de la philosophie. La psychologie contemporaine elle-même a pris naissance au sein de la philosophie. Elle s’est émancipée, mais l’émergence des sciences cognitives consacre d’une certaine façon le retour de la philosophie dans ce champ de recherche. Les développements de l’informatique et des neurosciences, en jetant une nouvelle lumière sur les phénomènes mentaux, ont eu pour effet de relancer le débat philosophique. La « philosophie de l’esprit » est ainsi plus florissante que jamais. Ce retour n’a rien d’une régression, car la philosophie dont il est question est en phase avec la recherche scientifique, informée par elle et en constante interaction avec elle.

Les Conférences Jean-Nicod visent à promouvoir les recherches philosophiques se rapportant à la cognition et à faire connaître en France les travaux réalisés à l’étranger dans ce domaine. Le conférencier présente ses recherches au cours d’un cycle de conférences qu’il rassemble ensuite en un livre.

Cycle de conférences (en anglais) de Nancy Kanwisher, titulaire de la chaire Walter A. Rosenblith de neurosciences cognitives au département des sciences cérébrales et cognitives du Massachusetts Institute of Technology et chercheuse au McGovern Institute for Brain Research, lauréate du prix Jean Nicod 2023.


Modularity of Mind and Brain and the Case of the FFA
Presentation of the Jean Nicod Prize and cocktail reception after the lecture
Thursday 7th December - 2pm - Salle des Actes, 45 rue d’Ulm 75005 Paris

Is the human mind structured, and if so what is that structure? This classic question, long the purview of philosophers, became scientifically tractable first as neurologists studied the sometimes specific cognitive deficits that resulted from brain damage, later as cognitive and developmental psychologists traced dissociations in the cognitive abilities of infants and children, and most recently as cognitive neuroscientists measured neural responses from the human brain. In this lecture I consider the case of face perception, charting the many lines of evidence that specialized neural machinery in the fusiform gyrus plays a specific and causal role in the perception and recognition of faces. The advent of fMRI in the 1990s for the first time enabled us to find and characterize the fusiform face area (FFA) noninvasively in normal subjects, and hence to study it in detail. Because the precise location of this region varies across individuals, we first localize it functionally in each participant individually, and then measure its response in a wide variety of conditions. We found that the FFA is a parade case of functional specificity, responding twice as strongly to faces as to any other stimuli. The response of this region is correlated with awareness of a face in binocular rivalry, modulated by spatial and object-based attention, and selectively increased when people closed their eyes and simply imagine faces. Electrical stimulation of the FFA produces a face percept, demonstrating the selective causal role of this region in face perception. Tsao and Freiwald further showed that macaques have similar face-selective patches, and then used the powerful tools available in animals to map out the sequence of representations across these face patches and their connectivity. The upshot of this story is that at least one patch of the human brain is extremely specific for the single mental function of face perception. But that conclusion raises an obvious question, tackled in the next lecture.

What Other Mental Functions get Their Own Private Patch of Real Estate in the Brain?
Friday 8th December - 10am - Salle des Actes, 45 rue d’Ulm 75005 Paris

Having established that at least one region of the cortex is highly specialized for a particular mental function, we set off in search of others. Over the next few years we identified the extrastriate body area (EBA), which responds very selectively to images of bodies and body parts, and the parahippocampal place area (PPA), which responds selectively to images of places. Newer data-driven methods enabled us to identify two surprising new functional selectivities in the cortex : for music in auditory cortex, and for visually presented food in high-level visual cortex. Other lines of work have identified cortical regions implicated in yet higher-level perceptual functions, including one selectively responsive during the perception of third-party social interactions (containing information about whether those interactions were competitive or cooperative), and another broadly implicated in intuitive physical reasoning. But what about abstract, uniquely human cognitive functions ? Rebecca Saxe showed that a region in the right temporo-parietal junction was extremely selectively engaged in "theory of mind", or thinking about what other people are thinking, a mental function already shown to have a distinctive developmental trajectory and a selective deficit in autism, and now apparently a private patch of brain as well. Evelina Fedorenko then turned to the long-standing and contentious question of the specificity of brain regions for language. Using the individual-subject functional ROI approach, Fedorenko showed that in fact the brain regions for language are remarkably specifically engaged in language per se, with almost no response during these other high-level mental processes, from arithmetic to music to working memory and cognitive control. Thus, language and thought are not the same thing in the brain. The work described in this lecture fills out the picture of the human brain as containing a large number of regions that are highly specialized for particular mental functions, providing us with an initial rough sketch of the human mind.

How Does All this Functional Organization Arise over Development?
Tuesday 12 December - 10am - Salle Jaurès, 29 rue d’Ulm, 75005 Paris

As described in the first two lectures, the last 25 years of research in human cognitive neuroscience have given us a glorious new picture of the functional organization of the human cortex, with dozens of regions, each specialized for a particular mental function, all present in approximately the same location in every normal person. It is impossible not to wonder how the precursors of all of this intricate and systematic structure arose over evolution, and how the structure of the cortex gets built in the life of each individual. Although these are among the hardest questions to answer about the human brain, tantalizing clues are beginning to emerge. Here I review recent findings from my collaborations with Rebecca Saxe and Heather Kosakowski, who developed methods for functionally scanning awake human infants. We have found that the FFA, PPA, and EBA are all present, in adultlike locations, and with approximately adultlike selectivity, by 6 months of age, and music selective responses appear to be present in the auditory cortex of sleeping one-month-old infants. The early appearance of these selective responses limit the total amount of experience available to instruct the development of these regions. But how do these regions "know" where to arise in the cortex ? Multiple lines of evidence suggest that early-developing long-range connections from one cortical region to another play a central role in determining the cortical location where each selectivity arises. Indeed, the location of visual word form area, which responds selectively to words and letterstrings only after children are taught to read, can be predicted from patterns of cortical connectivity in the same children before they learn to read. Currently contested theories appeal to bottom-up developmental processes that construct perceptual processing systems from the sensoria inward, versus top-down processes that invoke not just the perceptual properties of the stimulus, but its significance to the infant.

How and Why?
Thursday 14 December - 10am - Salle Jaurès, 29 rue d’Ulm, 75005 Paris

Understanding specific brain regions therefore requires understanding how they perform their specific computations. Until recently we did not have plausible computational models of many of these functions. But all that has suddenly changed with the explosion of recent successes with artificial neural networks. In this lecture, I sketch the manyfold implications of these advances in AI for the organization and function of the human brain. First, artificial neural networks (ANNs) now succeed at many tasks similar to those conducted in specialized brain regions, from face recognition to speech perception and language processing. These ANNs thus provide the first computationally precise hypotheses for how these functions might work in the brain. Further, to a remarkable degree, responses in visual, auditory, and language cortex are well predicted by ANNs optimized for visual, auditory, and language tasks, respectively. For example, we can now predict with astonishing accuracy exactly how strongly the FFA, PPA, and EBA will respond to a novel image. But beyond providing testable and computationally explicit hypotheses for the computations conducted in the brain, these ANN models can inform not just how the brain works but why it works the way it does. With Katharina Dobs, we found that that a network trained on both face and object recognition spontaneously segregated itself into two separate systems, without any built-in priors to do so, suggesting that the statistics of experience may suffice, without face-specific innate predispositions, for the brain to achieve the functional organization it does. The stunning convergence between the organization and function of the brain and completely different and nonbiological ANNs optimized for similar tasks, is transforming cognitive science and neuroscience, from a focus on describing phenomena of the mind and brain and their underlying mechanisms, to a deeply theoretical enterprise of asking (and sometimes even answering) why they work the way they do.

Mis à jour le 20/10/2023