Principles of brain architecture Jordan K Matelsky

Format:

Book

Thesis/Dissertation

Author/Creator:

Matelsky, Jordan K., author.

Contributor:

University of Pennsylvania. Bioengineering., degree granting institution.

Language:

English

Subjects (All):

Neurosciences.

Philosophy of science.

0317.

0402.

0800.

Local Subjects:

Neurosciences.

Philosophy of science.

0317.

0402.

0800.

Genre:

Academic theses

Physical Description:

1 online resource (118 pages)

Contained In:

Dissertations Abstracts International 87-12B

Place of Publication:

Ann Arbor : ProQuest Dissertations and Theses, 2026

Language Note:

English

Summary:

Brains are physical systems whose architecture shapes what neural circuits can compute, yet our measurements across scales and levels of abstraction remain jagged. Some areas of neuroscience have extraordinary detail. Others remain inscrutable, despite monumental efforts. When in general we try to understand how systems work, we often want to isolate a subsystem and study it in detail, but the brain's architecture is so complex and interconnected that this is rarely possible: in many cases we do not even have enough information to know where we can make valid mechanistic "dissections" to isolate simpler subsystems for study. This dissertation asks how incomplete structural data can still support defensible abstractions of brain architecture under that constraint: where a dissection cut is scientifically admissible, which regularities are stable enough to model, and which representations make those regularities visible.I approach that problem by making neural structure more observable, more compressible, and more comparable. I show how dense volumetric measurements become scientifically more useful when paired with reconstruction pipelines, graph abstractions, and the contactome as a representation of physical opportunity rather than only realized synapses. I then use developmental lineage, subgraph decomposition, and data-driven motif discovery to compress high-dimensional connectomes into candidate mesoscale structure, and use comparative analysis across individuals, developmental stages, and species to separate conserved architectural scaffolds from idiosyncratic details. Across these studies, I find that lineage provides predictive information about connectivity, motif-based representations preserve useful local and global structure, important connectome "scaffolds" are often conserved across individuals, and contact-based representations recover structural opportunity that the synaptic graph alone discards.Our results will be a set of defensible techniques, findings, and theoretical frameworks to make neural architecture measurable, comparable, and interpretable. Together, these studies shift explanation away from single open-ended stories and toward testable organizational principles linking neural structure, development, and signaling across species, scales, and time.

Notes:

Source: Dissertations Abstracts International, Volume: 87-12, Section: B.

Advisors: Körding, Konrad P. Committee members: Ungar, Lyle H.; Issadore, David A.; Schapiro, Anna C.; Wester, Brock A.

Ph.D. University of Pennsylvania 2026

Vendor supplied data

Local Notes:

School code: 0175

ISBN:

9798247973843

Access Restriction:

Restricted for use by site license