---
title: "Bojana Grujičić"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "Bojana Grujičić"
aliases: ["Bojana Grujicic","B. Grujičić","Grujicic"]
first_seen: "2026-07-12T00:00:00.000Z"
writer_model: "claude-opus-4-8"
connects_to: ["representational similarity analysis (RSA)","mechanistic explanation","deep convolutional neural networks","philosophy of neuroscience","underdetermination"]
seek_code_commit: "649b1a4"
---


Philosopher of neuroscience (Max Planck School of Cognition / Humboldt-Universität
zu Berlin / UCL Science and Technology Studies at the time of publication) whose
2024 *Synthese* paper "Deep convolutional neural networks are not mechanistic
explanations of object recognition" argues that representational similarity
analysis cannot license treating a deep network as a mechanistic explanation of
biological vision. She is the anchor author of the vault's RSA-underdetermination
thread: a single peer-reviewed paper whose thesis — that different similarity
measures pick out different mechanisms with no arbitration between them — grounds
several claim-notes and the substrate-laundering observation.

## References
- [[claim-representational-similarity-underdetermines-mechanism]] — her load-bearing thesis: RSA underdetermines shared mechanism
- [[claim-similarity-measures-pick-out-different-representational-vehicles]] — the mechanism, cashed out via correlation vs Euclidean distance
- [[claim-grujicic-frames-dcnn-critique-on-3m-not-marr-levels]] — her actual apparatus is Craver & Kaplan's 3M, not [[entity-david-marr|Marr]]'s levels
- [[observation-substrate-laundering-across-marr-levels]] — the vault's cross-level reading her argument grounds
- [[question-verify-grujicic-2024-rsa-underdetermination-dcnn]] — the verification thread on her paper