Loading...
Analysis and Comparison of Different Approaches to Testing Representational Models for Brain Activity Patterns
Mirzazadeh, Pouneh | 2021
308
Viewed
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 53819 (02)
- University: Sharif University of Technology
- Department: Mathematical Sciences
- Advisor(s): Sharifitabar, Mohsen; Nili, Hamed
- Abstract:
- The representation concept links the information processed by the brain back to the world and enables us to understand what the brain does at a functional level.Representational models specify how activity patterns in the population of neurons relate to sensory stimuli, motor responses, or cognitive processes. In an experimental context, representational models can be defined as hypotheses about the distribution of activity profiles across experimental conditions. Three different methods are currently being used to test such hypotheses: encoding analysis, pattern component modeling (PCM), and representational similarity analysis (RSA). All three evaluate the second moment of the distribution of activity profiles, which determines the representational geometry, and thus how well any feature can be decoded from population activity.In this research, we go deeper into RSA method that first quantifies the representational
geometry by calculating a dissimilarity measure for all pairs of conditions and then compares the estimated representational dissimilarities to those predicted by the model. Comparing these two dissimilarities can be done by evaluating the likelihood of estimated dissimilarities under the model, so we assume a Gaussian distribution on the dissimilarities and present an analytical expression for the mean and (co)variance of unbiased estimators of Euclidean and Mahalanobis distances. We then use derivatives of likelihood in the iterative algorithms to estimate the model’s
scaling factor to maximize the likelihood - Keywords:
- Coding ; Decoding Algorithm ; Representational Similarity Analysis ; Representational Models ; Pattern Components Modeling ; Brain Representations