Reconstruction differences rival individual differences in the fly compass connectome and break trained circuit models

Md Ali Ashraf Yad; Nazifa Tabassum Nabila

cs.NEOct 21, 2026

Reconstruction differences rival individual differences in the fly compass connectome and break trained circuit models

Authors: Md Ali Ashraf Yad, Nazifa Tabassum Nabila

Organizations: Wrocław University of Science and Technology · Comilla University

Abstract

Neural network models are increasingly constructed from synapse-resolution connectomes, and the same circuit is now available in reconstructions from other animals and pipelines. We do not know whether such models are dependent on the reconstruction they are built from, and whether connectome differences are biological or measurement errors. We studied the Drosophila head-direction circuit in nine versions: four flies (male CNS, hemibrain, FlyWire, BANC) and five variants of the same fly varying only by thresholds, proofreading or synapse detection. Using wiring-based correspondences validated by anatomical labels, we place all differences on one scale. A change of synapse detection within one fly changed the circuit more than halfway to the smallest between-fly difference on four distance measures; proofreading differences were smaller but depended on the metric, and left and right halves of one fly differed no more than random halves. Networks trained on each wiring to hold a cued heading developed near-perfect ring attractors, but their error increased from below 1 degree to 10-28 degrees when all weights changed by 1%. Training on each wiring did not transfer to any other connectome, including other versions of the same fly; refitting thirty cell-type gains, we repaired filtering-level differences but remaining error depended on wiring distance. Training with noise lowered precision and gave only partial transfer. In imaging data, the fly compass drifted gradually in darkness. Connectome-constrained models should report the reconstruction they were built on and be tested for robustness across reconstructions.

Figures & tables

Figure 1: Study design. ( A ) The compass circuit and model ports are: cue to ER, angular velocity to PEN, readout from EPG. ( B ) Nine wiring versions: four flies from three groups (bold) and five same-fly variants, in which the variants differ in filtering, proofreading or detection of synapses. ( C ) Experiments: Distance measurements, Training on each wiring, Transplant through neuron maps, Cell-type gains.

Figure 2: Connectome differences on one scale. ( A ) Wiring-map distance by source of difference; grey band, noise floor (0.022–0.032). ( B ) Hold error of male CNS models after artificial rewiring ( Yad, 2026b ) against distance from the original (five networks per point); bands mark same-fly measurement and between-fly differences. The degree-preserving curve continues to the 50% shuffle (0.42, 18.2 ∘ ), off the plotted range.

Figure 3: Near-perfect but knife-edge ring attractors. ( A, B ) EPG activity of pilot networks 95 steps after the cue in the top two principal components, coloured by cued heading (same scale): round ring for real wiring, distorted loop for maximally shuffled wiring. ( C ) Slowest Jacobian multiplier of every trained Stage 4 network; dashed line, 1. ( D ) Hold error after scaling all weights by $-2$ to $+2\%$ (per-neuron gains, median of seeds 0–1).

Figure 4: No transfer; noise trades precision for robustness. ( A ) Error of clean per-neuron networks tested on each wiring (rows: trained on; median of trained seeds). ( B ) Exploratory noise sweep: own-wiring error and error with all weights scaled by 0.95 or 1.05, by training-noise level. Chance, dotted line. ( C ) Transplant error for male CNS $\leftrightarrow$ hemibrain and for pairs with FlyWire or BANC, with own-wiring error; dotted line, chance.

Figure 5: Transfer depends on distance, not on the source of the difference. ( A ) Transplant error against source–target distance, zero-shot (crosses) and after refitting about 30 cell-type gains (dots); dashed line, median own-wiring error over the plotted pairs. ( B ) Median error by source; percentages, transplants below 30 ∘ after refitting. Exploratory: 21 networks on seven versions, noise level 0.5; three refits did not complete.

Figure 6: The fly compass degrades gradually. ( A ) Mean squared change of bump-heading offset against lag (15 flies, all frames; shading, 95% CI over flies). ( B ) Diffusion coefficient from $\mathrm{MSD}=2Dt+2\sigma^{2}$ fits to no-turning runs of at least 5 s (9 flies); bars, 95% CI. Data: DANDI 000289 ( Basnak et al., 2025 ) .

Version	Source	Neurons	Connections
male CNS v1.0	FlyWire Codex (filtered)	434	27,246
male CNS, neuPrint	neuPrint	434	33,783
male CNS v0.9	Codex	434	27,246
hemibrain v1.2.1	neuPrint	409	28,942
FlyWire FAFB v783	Codex, Princeton synapses, filtered	433	16,741
FAFB unfiltered	Codex unfiltered file, $\geq$ 3 synapses	433	23,340

Table 1: The nine wiring versions. Head-direction circuit (EPG, PEG, PEN,

\Delta

7, ER1–6), at least three synapses per neuron pair.

Appendix figures & tables15 assets

Supplementary material from the paper’s appendix.

Appendix

Wiring	Gains	Seed	40 steps	80 steps	200 steps
real	per neuron	0	0.68	1.26	4.40
real	per neuron	1	0.57	1.39	5.31
real	per neuron	2	88.11	88.60	89.99 (failed)
real	cell type	0	1.23	3.97	17.51
real	cell type	1	1.11	3.69	17.96
real	cell type	2	88.11	88.60	89.99 (failed)

Appendix

Table S1: Pilot hold errors (degrees; test sets of 40, 80 and 200 steps).

Wiring	Gains	Seed	Err.	PC2	H1	$\lambda_{1}$	$\lambda_{2}$	Ring	Return
real	neuron	0	4.4	1.000	1.34	1.000	0.986	yes	yes
real, $\Delta$ 7 silenced	neuron	0	82.5	1.000	0.51	0.975	0.949	yes	no
real	neuron	1	5.3	1.000	1.18	1.000	0.997	yes	yes
real, $\Delta$ 7 silenced	neuron	1	79.8	1.000	0.22	0.995	0.978	no	no
real	cell type	0	17.3	0.999	0.97	0.998	0.981	yes	yes
real, $\Delta$ 7 silenced	cell type	0	46.3	0.998	0.72	1.028	0.982	yes	yes

Appendix

Table S2: Pilot mechanism measures. Error at 200 steps (degrees), variance in the top two principal components of EPG activity, top persistence of the one-dimensional homology feature, the two largest Jacobian multipliers, and the two H4 criteria. Failed networks (seed 2 of the real wiring) are omitted.

Wiring	Gains	Seed	$\times$ 1	$\times$ 0.98	$\times$ 0.99	$\times$ 1.01	$\times$ 1.02
real	neuron	0	0.69	33.6	14.3	11.3	21.8
real	neuron	1	0.56	48.9	26.0	18.4	36.0
real	cell type	0	1.22	39.3	18.4	12.7	27.3
real	cell type	1	1.11	30.9	11.8	9.1	16.0
25% shuffle	neuron	0	4.39	53.9	29.3	26.0	44.4
25% shuffle	neuron	1	3.69	66.0	39.2	30.7	53.3

Appendix

Table S3: Pilot weight-scale test. Hold error (degrees, 40 steps) after multiplying all weights by the given factor, without retraining.

Source	Kind	Comparison	Distance
Filtering	same fly	male CNS Codex vs. neuPrint	0.022
	same fly	FlyWire filtered vs. unfiltered	0.038
Proofreading	same fly	male CNS v0.9 vs. v1.0	0.000
	same fly	BANC v626 vs. v888	0.104
Synapse detector	same fly	FlyWire Princeton vs. Buhmann	0.090
	same fly	FlyWire unfiltered vs. Buhmann	0.084

Appendix

Table S4: Wiring-map distances by source of difference , with the noise floor and the left–right control (random halves, five repeats; SD of the edge-removal control 0.003).

Fly	Cell type	Method	Agreement	Shuffled
male CNS	EPG	MDS of connectivity profiles	0.99	0.26–0.31
male CNS	PEN_a	anchored to EPG	1.00	0.36–0.49
male CNS	PEN_b	anchored to EPG	1.00	0.33–0.40
male CNS	PEG	anchored to EPG	1.00	0.30–0.38
hemibrain	EPG	MDS of connectivity profiles	0.99	0.25–0.30
hemibrain	PEN_a	anchored to EPG	1.00	0.31–0.35

Appendix

Table S5: Validation of compass positions inferred from wiring against glomerulus labels (circular agreement; shuffled labels as control).

Map	Cell type	Exact match
hemibrain $\rightarrow$ male CNS	EPG	100%
hemibrain $\rightarrow$ male CNS	PEN_a	100%
hemibrain $\rightarrow$ male CNS	PEN_b	100%
hemibrain $\rightarrow$ male CNS	PEG	83%
male CNS $\rightarrow$ hemibrain	EPG	100%
male CNS $\rightarrow$ hemibrain	PEN_a	100%

Appendix

Table S6: Exact glomerulus match of the neuron maps between the two labelled flies (target

\rightarrow

source).

Source	Target	Alignment $r$	Mirrored	Neurons used
male CNS	hemibrain	0.804	no	99%
male CNS	FlyWire	0.572	no	96%
male CNS	BANC	0.515	no	94%
hemibrain	male CNS	0.804	no	94%
hemibrain	FlyWire	0.570	yes	93%
hemibrain	BANC	0.528	no	93%

Appendix

Table S7: Neuron maps for all ordered fly pairs : correlation of log connectivity with used frame, with or without mirror frame, and fraction of source that is used.

Source	Comparison	Matching	$n$	Map	Weights	Jaccard	Shares
filtering	male CNS vs male CNS neuPrint	neuron ID	152	0.022	0.003	0.121	0.013
filtering	FlyWire vs FlyWire unfilt.	neuron ID	155	0.038	0.061	0.223	0.023
proofreading	male CNS v0.9 vs male CNS	neuron ID	152	0.000	0.000	0.000	0.000
proofreading	BANC v626 vs BANC	neuron ID	140	0.104	0.082	0.132	0.049
synapse detector	FlyWire vs FlyWire Buhmann	neuron ID	155	0.090	0.358	0.257	0.072
synapse detector	FlyWire unfilt. vs FlyWire Buhmann	neuron ID	155	0.084	0.304	0.336	0.065

Appendix

Table S8: Robustness of the decomposition to the distance measure. Angular wiring-map distance (main text) and three standard measures: neuron-level input-fraction dissimilarity (Weights) and Jaccard distance between edge sets (both on direction-tuned neurons, matched by identity within a fly and by rank maps between flies), and cell-type pathway-share distance (Shares). Spearman correlation with the map distance across the 12 comparisons: Weights 0.93, Jaccard 0.77, Shares 0.93 (all

p\leq 0.004

Trained on	male CNS	hemibrain	FlyWire	BANC
male CNS	1.2	89.4	90.7	89.2
hemibrain	87.9	0.9	86.2	90.7
FlyWire	91.1	91.7	0.7	88.6
BANC	89.3	88.5	83.9	1.1

Appendix

Table S9: Stage 4 transplants, cell-type gains (clean training, 3–5 networks per row).

Trained on	male CNS	hemibrain	FlyWire	BANC
male CNS	0.7	82.0	91.1	88.0
hemibrain	82.2	0.6	88.2	90.2
FlyWire	91.0	91.2	0.5	89.6
BANC	87.3	89.1	90.1	0.5

Appendix

Table S10: Stage 4 transplants, per-neuron gains (clean training, 3–5 networks per row).

Level	Trained on	male CNS	hemibrain	FlyWire	BANC
0	male CNS	0.7	78.7	89.0	88.1
	hemibrain	77.6	0.7	89.4	90.3
	FlyWire	90.6	89.4	0.5	90.0
	BANC	87.4	87.1	87.3	0.4
0.25	male CNS	7.0	78.5	87.8	88.1
	hemibrain	57.0	17.9	88.7	87.5

Appendix

Table S11: Noise sweep : Transplant matrices at each training-noise level (per-neuron gains, three seeds per row; two failed clean networks excluded).

Family	Test	Statistic	$p$	$p_{\mathrm{Holm}}$
H1 transplant (pre.)	median transplant ratio R, celltype gains	R = 87	–	–
H1 transplant (pre.)	median transplant ratio R, neuron gains	R = 162	–	–
H2 level (pre.)	neuron-level R $>$ cell-type R (paired Wilcoxon, 12 pairs)	median 162 vs 87	2.4 $\times 10^{-4}$	7.3 $\times 10^{-4}$
H3 distance (pre.)	$\rho$ (wiring distance, R), celltype gains	$\rho$ = -0.33	0.85	0.85
H3 distance (pre.)	$\rho$ (wiring distance, R), neuron gains	$\rho$ = +0.25	0.21	0.42
H4 mechanism (pre.)	banc celltype: networks meeting ring + fast return	3/5	–	–

Appendix

Table S12: All tests. Prespecified tests (pre.) and exploratory tests (Expl.), with uncorrected

p

and Holm-corrected

p

within the prespecified family.

Fly	Gains	$n$	H4	$\lambda_{1}$ med.	$\lambda_{1}$ range	Radius var.	$\|$ Winding $\|=1$	Err. 200
BANC	cell type	5	3	1.000	0.998–1.009	0.21	5/5	12.0
BANC	per neuron	5	5	1.001	0.991–1.001	0.15	5/5	2.0
FlyWire	cell type	5	4	1.003	1.000–1.044	0.21	4/5	8.7
FlyWire	per neuron	5	5	1.001	1.000–1.003	0.21	5/5	3.5
hemibrain	cell type	4	3	1.014	1.008–1.021	0.30	4/4	12.7
hemibrain	per neuron	4	4	1.001	1.000–1.002	0.27	4/4	4.2

Appendix

Table S13: Mechanism measures of the Stage 4 networks (trained networks only): H4 count, slowest Jacobian multiplier (median and range), radius variation and winding number of the trajectory in the top two principal components (exploratory), and hold error at 200 steps (degrees).

Hypothesis	Prediction	Outcome
H1 transfer	Robust ( $R<2$ ) or fragile ( $R>5$ ) transfer between flies	Fragile: median $R=87$ (cell-type gains, registered) and 162 (per-neuron gains)
H2 parameter level	Neuron-level parameters transfer worse than cell-type parameters	Supported: $R$ larger at neuron level in 12 of 12 pairs, $p=2.4\times 10^{-4}$ (Holm $7.3\times 10^{-4}$ )
H3 distance	Transfer failure increases with wiring distance	Not supported ( $\rho=-0.33$ and $+0.25$ ); all transplants at chance (floor effect)
H4 mechanism	Real wirings form a ring attractor with fast return (at least 3 of 5 seeds)	Supported in all 8 fly $\times$ parameterization combinations; 31 of 36 networks
H5 drift	Descriptive comparison of model drift in darkness with the DANDI 000289 estimate; no directional prediction	Fly drift reported (darkness $D=0.047$ , bright cue 0.023, excess 0.024 rad 2 /s, $-0.010$ to 0.076); model comparison not made (deviation 8)

Appendix

Table S14: Preregistered hypotheses and outcomes.

True $D$	True $\sigma$	Fitted $D$ (mean)	SD	Bias	Fitted $\sigma$
0.002	0.1	0.0020	0.0002	+1.2%	0.100
0.002	0.3	0.0020	0.0007	+0.6%	0.300
0.01	0.1	0.0102	0.0012	+2.2%	0.099
0.01	0.3	0.0101	0.0021	+1.2%	0.302
0.04	0.1	0.0379	0.0060	-5.2%	0.102
0.04	0.3	0.0393	0.0070	-1.7%	0.300

Appendix

Table S15: Test of the drift-method on simulated data (20 replicates per condition).

Continuous Attractor Neural Networks

See all 9 papers

FlyHack: Stress-Testing Mechanistic Claims in a Connectome-Constrained Visual Model
Oct 9, 2026Aashish DhawanMechanistic Interpretability Visual Neuroscience
Department of Computer & Information Science & Engineering University of Florida
Sensorimotor behaviour from an untrained brain-and-nerve-cord connectome
Oct 24, 2026John R. Minnick, Jesus Gonzalez-Ferrer, Jinghui Geng +4Neural Processes Embodied Navigation
Department of Electrical and Computer Engineering, University of California, Santa Cruz, CA, USA. · UC Santa Cruz Genomics Institute, University of California, Santa Cruz, CA, USA. · Department of Biomolecular Engineering, University of California, Santa Cruz, CA, USA. +1
How much of fly walking is written in the wiring?
Sep 29, 2026Isabel Guan, Yuntian Zhao, Dingyuan Zhang +2Animal Behavior Analysis
The Hong Kong University of Science and Technology · ZENBOT · Nanyang Technological University +1

Oct 9, 2026J/K move · Enter open · S save

FlyHack: Stress-Testing Mechanistic Claims in a Connectome-Constrained Visual Model

Aashish Dhawan

Department of Computer & Information Science & Engineering University of Florida

Mechanistic Interpretability Visual Neuroscience

Open paper AskPDF arXiv

Connectome-constrained neural networks make biological components directly accessible to intervention, but their mechanistic conclusions can still depend on the comparison model, intervention, circuit representation and trained solution. We stress-test these dependencies in flyvis, a Drosophila motion-vision model, using its 50-model ensemble, task-matched causal CNN controls and replicated interventions. At the level of model behaviour, three-level reverse-phi reversal in flyvis is strongly period- and velocity-dependent and also appears in task-matched CNNs; neither family reproduces the fly's stationary-sawtooth response. Adversarial direction reversal requires a perturbation bound of 0.21--0.27 times stimulus contrast in flyvis versus 0.13--0.25 in CNNs, but the gap over a parameter-matched CNN largely closes for larger CNNs. At the level of mechanism, matching attribution to the intervention matters: for held-out-mean replacement, first-order edge attribution patching reaches pooled AUROC 0.970 and Spearman 0.872, versus 0.933 and 0.126 when its zero-aligned score is evaluated against the same ground truth. Circuit conclusions also vary with representation and ensemble member. Greedy cell-type sets meet a model-relative T4/T5 tuning criterion in 25 of 50 models, while T4/T5 output silencing reduces the tested optic-flow behaviour to at most two of eight clips in 40. In six models where the cell-type search first succeeds, the corresponding connection sets can be pruned to 13--31 added classes; under an additional two-sided response-amplitude check, only two of those six first-success cell-type sets still meet the tuning threshold. These results support a narrower use of connectome-constrained models: they make mechanistic hypotheses directly testable, but the resulting claims must be scoped to the model family, intervention, representation and trained solution.

Abstract

Continuous Attractor Neural Networks

Explore similar work

FlyHack: Stress-Testing Mechanistic Claims in a Connectome-Constrained Visual Model

Sensorimotor behaviour from an untrained brain-and-nerve-cord connectome

How much of fly walking is written in the wiring?