Huh, funny. 40 years passed, but this is still so relatable

This reveals what we call NETWORK–CODE INVERSION:

- Novices use domain-specific networks to produce relatively unstructured, domain-general (not tuned to the task) codes.

- Experts repurpose domain-general control networks to encode highly structured, domain-specific knowledge.

9/n

3️⃣ WHERE do these optim codes live?

🧠 Experts rely more on
- working-memory systems
- navigation-related regions
- memory-retrieval networks

👁️ Novices rely more on
- early visual cortex
- face/object regions
- language areas

👉 A shift from domain-specific → domain-general control networks.

8/n

2️⃣ HOW are representations structured?

Using manifold dimensionality (Participation Ratio), we find lower-dimensional, more compressed neural codes in experts. And these compressed manifolds carry more task-relevant information.

👉 Experts pack more information into fewer dimensions.

7/n

WHAT is represented? -- brain

Brain–model RSA shows the same shift: both groups encode visual features, but only experts encode high-level, task-relevant structure.

In other words,

👉 expertise changes WHAT is represented: from low-level, surface features to high-level, relational structure.

4/n

1️⃣ WHAT is represented? -- behavior

Behavioral RSA shows that experts organize their value judgments around relational and goal-relevant structure. Visual similarity barely plays a role.

Novices, meanwhile, show much less structured preferences.

3/n

To test this, we built a stimulus set spanning low-level (visual similarity) to high-level (strategy, checkmate) structure, and turned these into RDMs capturing our theoretical models.

**Enter: Representational Similarity Analysis (RSA).**

2/n

Super excited to share a new preprint!

We asked a simple-but-big question:

What changes in the brain when someone becomes an expert?

Using chess ♟️ + fMRI 🧠 + representational geometry & dimensionality 📈, we ask:

1️⃣ WHAT information is encoded?
2️⃣ HOW is it structured?
3️⃣ WHERE is it expressed?

1/n

It was great reconnecting with friends and colleagues at #CCN2025 in Amsterdam and presenting our latest #expertise work.

We 👀 into how #chess experts represent the board, and how the content, structure, and location of these repr shift w/ expertise.⬇️

To determine the source of the signal, we ran a PPI analysis. The foveal ROI was functionally connected with both per V1 and LOC, but not other regions.

So there seems to be a local path within V1 and a route from higher-level visual areas—though not all info survives the trip.

Our MVPA results showed that foveal V1 indeed contained feedback about perceptually distinct categories (cars vs. bikes) but not finer semantic distinctions (male vs. female faces).

In contrast, higher-level areas like FFA and LOC robustly decoded those semantic details.

To tease apart low- and high-level info, we compared our data to two models capturing perceptual vs. semantic/categorical features.

If foveal V1 encodes perceptual details, we’d expect it to align more with TDANN’s predictions than with CLIP’s -- and vice-versa.

We recorded BOLD signal from various ROIs, including FFA, LOC, retinotopically defined “peripheral” and several "foveal" ROIs to capture potential feedback across varying spatial scales.

We used MVPA to see whether activation patterns reflected perceptual or categorical info.

We ran an fMRI experiment where participants performed a same/different task on two briefly presented peripheral images — faces (male/female) or vehicles (cars/motorbikes) — 7 degrees away from central fixation, ensuring the fovea wasn’t stimulated directly.