Study tests when real connectome wiring beats shuffled nulls
Original title: When Does the Real Wiring Win? Testing connectome structure-function claims against unmeasured physiology
In the mushroom body, real Kenyon-cell wiring beats a shuffle by 0.09 under two protocols but loses by 0.09 when physiology is data-constrained.
The paper separates three ways of comparing real connectome wiring with shuffled or rewired nulls: physiology tuned on real wiring and transferred (P1), each wiring tuned on its own (P2), and P2 restricted to physiology consistent with independent data (P3). Using synthetic ground truth, placebo nulls and three Drosophila circuit models, some claims survive every protocol, including segregated reward and punishment dopamine and head-direction topography. In the mushroom body, real Kenyon-cell output wiring beats a cross-subtype identity shuffle by 0.09 under P1 and P2 but loses by 0.09 under P3. At the imaging-implied share of about 44% alpha'beta' cells, the real-wiring model fails to learn.
- P1 transfers physiology tuned on real wiring; P2 tunes each wiring separately; P3 constrains physiology to independent data.
- Mushroom-body real wiring beats the shuffle by 0.09 under P1 and P2, then loses by 0.09 under P3.
- 95% confidence intervals exclude zero, and held-out runs confirm the reversal.
- In flyvis, a mirrored connectome is within 0.07 median of the real error, but ranking depends on the fifty-fold checkpoint.
From the original article
1 min read at arxiv.orgConnectome-constrained models are often used to argue that a measured wiring diagram outperforms shuffled or rewired wiring, and hence that structure shapes function. Such comparisons can confound wiring with how the models are tuned: a connectome fixes who connects to whom, not the excitability, gains or plasticity rates the model also needs, and modellers often tune these until the real-wiring model works.
By Yeyuan Liu · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.