Grant: Discovery Grants Program - Individual
McGill University
$44,000.00
Cortical representation of visual information: dynamic, naturalistic stimulation, neuroimaging, and topological analysis
EF 2023-24
Résumé
This proposed five year program of research is a continuation of my ongoing program of research concerned with cortical visual representation, and seeks to leverage our recent developments in naturalistic stimulus design, MRI hardware, and analysis to test models of cortical visual representation. Specifically, we advance the perspective that dynamic, complex, naturalistic stimulation can provide an ecologically-valid view on cortical visual representations, and at times substantially different from what is predicted based on highly controlled, synthetic, or static stimuli. Our data collection approach utilizes very high-resolution fMRI using a custom device that we have developed, and we are developing an analytic approach that facilitates interpretation of complex data by using methods in algebraic topology. We propose a transformative and complimentary method of assessing representational "fit" using topological data analysis. In the first Aim of this program, we leverage professional-quality 3-D animation sets to develop controlled dynamic, complex naturalistic movies. This process allows us to not only experimentally manipulate a naturalistic scene, but also to extract statistics about both the depth features of the scene (depth map, optic flow, etc), but also image features of the movie too. Using this novel approach allows us to make inference about the visual representations of natural 3-D objects from the neuroimaging data, and particularly the dissociable representation of binocular disparity and depth. In the second Aim, we seek to first determine the statistical properties of 3-D objects-what are the fundamental 3-D shape primitives? Using an approach from spectral manifold analysis, we will evaluate 11,000 natural laser-scanned objects and infer both topological similarities and also the implicit 3-D shape features that they express. In the second part of this theme, we will evaluate how these data-driven 3-D shape primitives relate to cortical object responses. Finally, in Aim 3 we address the challenge of causal testing of mechanistic cortical models. Using our recent development of an MRI device that allows for arbitrary stimulation of the brain using Transcranial Magnetic Stimulation (TMS) while simultaneously imaging the whole brain, we will directly compare cortical models and cortical responses to assess the extent to which computational models actually represent the way the brain processes visual information. The transformative features of this research program are the utilization of controlled, complex dynamic stimulation to test models of cortical visual representation measured with high-resolution (1mm) functional MRI, the use of topological tools for both stimulus characterization and assessing representational geometry, and the use of causal testing with neuroimaging for direct evaluation of mechanistic models of cortical function.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2023-05840
Name nom
Farivar-Mohseni, Reza
Department département
Ophthalmology & Visual Sciences
Organization id
61
Institution établissement
McGill University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
44000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1502
Committee name en
Biological Systems and Functions
Committee name fr
Systèmes et fonctions biologiques
Area of application code
1102
Area of application group en
Health, education and social services
Area of application group fr
Santé publique, éducation et services sociaux
Area of application en
Biomedical engineering
Area of application fr
Génie biomédical
Research subject code
2950
Research subject group en
Applied mathematics
Research subject group fr
Applied mathematics
Research subject en
Mathématiques appliquées
Research subject fr
Mathématiques appliquées
Application title
Cortical representation of visual information: dynamic, naturalistic stimulation, neuroimaging, and topological analysis
Keyword
big data; complex data; cortico-cortical communication; manifold analysis; Multivariate analysis; Neuroimaging; topological data analysis; Transcranial magnetic stimulation; Visual representation
Application summary
This proposed five year program of research is a continuation of my ongoing program of research concerned with cortical visual representation, and seeks to leverage our recent developments in naturalistic stimulus design, MRI hardware, and analysis to test models of cortical visual representation. Specifically, we advance the perspective that dynamic, complex, naturalistic stimulation can provide an ecologically-valid view on cortical visual representations, and at times substantially different from what is predicted based on highly controlled, synthetic, or static stimuli. Our data collection approach utilizes very high-resolution fMRI using a custom device that we have developed, and we are developing an analytic approach that facilitates interpretation of complex data by using methods in algebraic topology. We propose a transformative and complimentary method of assessing representational "fit" using topological data analysis. In the first Aim of this program, we leverage professional-quality 3-D animation sets to develop controlled dynamic, complex naturalistic movies. This process allows us to not only experimentally manipulate a naturalistic scene, but also to extract statistics about both the depth features of the scene (depth map, optic flow, etc), but also image features of the movie too. Using this novel approach allows us to make inference about the visual representations of natural 3-D objects from the neuroimaging data, and particularly the dissociable representation of binocular disparity and depth. In the second Aim, we seek to first determine the statistical properties of 3-D objects-what are the fundamental 3-D shape primitives? Using an approach from spectral manifold analysis, we will evaluate 11,000 natural laser-scanned objects and infer both topological similarities and also the implicit 3-D shape features that they express. In the second part of this theme, we will evaluate how these data-driven 3-D shape primitives relate to cortical object responses. Finally, in Aim 3 we address the challenge of causal testing of mechanistic cortical models. Using our recent development of an MRI device that allows for arbitrary stimulation of the brain using Transcranial Magnetic Stimulation (TMS) while simultaneously imaging the whole brain, we will directly compare cortical models and cortical responses to assess the extent to which computational models actually represent the way the brain processes visual information. The transformative features of this research program are the utilization of controlled, complex dynamic stimulation to test models of cortical visual representation measured with high-resolution (1mm) functional MRI, the use of topological tools for both stimulus characterization and assessing representational geometry, and the use of causal testing with neuroimaging for direct evaluation of mechanistic models of cortical function.