Jacob S Prince

Jacob S Prince

PhD Student @ Harvard Psychology

Research Projects

An integrated theory of category-selective regions: evidence from deep neural networks

During my internship with Dr. Talia Konkle at Harvard in Summer 2018, I began a project on the representational structure of category-preferring regions in visual cortex. I tested competing theories of category information in the ventral visual pathway using a set of deep neural networks (DNNs) trained to perform face, scene, and object categorization. Comparing their representational match with human FFA and PPA led us to propose a theory that reconciles the modular and distributed hypotheses of information processing in the ventral stream: that these neural ROIs reflect different facets of a shared feature space supporting a wide range of downstream tasks.

Prince, JS., Konkle, T., (2020). An integrated theory of category-selective regions: evidence from deep neural networks. (Paper in prep.)

GLMsingle: a turnkey solution for accurate single-trial fMRI estimates

In my current role as a Research Associate with Dr. Michael Tarr at CMU, I have worked to enhance the quality of two massive fMRI datasets (BOLD5000 and NSD) measuring neural responses to rich naturalistic scenes. In close collaboration with Dr. Kendrick Kay (Univ. Minnesota), I have found that applying voxel-wise HRF optimization, data-driven denoising techniques, and ridge regression in combination can significantly improve the reliability of beta estimates within-subject, and importantly, boost representational consistency across datasets.

Prince, JS., Pyles, JA., Tarr, MJ., Kay, KN., (2020). GLMsingle: a turnkey solution for accurate single-trial fMRI estimates. (Paper in prep.)

Covert metrics of conscious visual perception: Pupil, microsaccade and blink dynamics

As an undergraduate at Yale, I worked with Dr. Hal Blumenfeld on characterizing the neural and psychophysical responses associated with visual perception. One challenge of perceptual tasks is their frequent reliance on behavioral reports, which can conflate brain activity associated with perception with that of reporting. I led the development of a report-free experimental paradigm that relies on classification of pupillary and eye metrics to predict whether a faint stimulus has been perceived. I conducted a behavioral and pupillometry study, and my analysis of this large dataset revealed that eye metrics (pupil diameter, microsaccade and blink rate) differed as a function of behavioral report (specifically, I found pupillary dilation and saccadic suppression during perceived trials). I treated this as a training dataset for a machine learning pipeline, and developed a classifier that accurately predicted perception on a trial-by-trial basis.

Prince, JS. & Ding, Z...Blumenfeld, H., (2020). Covert metrics of conscious visual perception: Pupil, microsaccade and blink dynamics. (Paper in prep.)