# Neuronal detection of social actions directs collective escape behaviour

- **DOI:** [10.1038/s41586-026-11041-1](https://doi.org/10.1038/s41586-026-11041-1)
- **Rating:** Sound (3.3 out of 5)
- **Verdict:** An integrative study identifies social-offset neural responses and matched withdrawal behaviour, but its causal neural framing exceeds the non-interventional evidence.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-09-29
- **Evaluation page:** https://www.nabu.science/paper/10.1038%2Fs41586-026-11041-1

## Key claims

1. Visual neurons in the midbrain optic tectum and thalamus increased their activity when virtual conspecifics escaped
   - Support: Well-supported
   - Type: Primary finding
   - Basis: within-session imaging comparisons and nested analyses directly identify escape-associated neural responses
   - Location: Abstract; Fig. 4c
2. fish retreated from virtual schools that escaped or disappeared, but only those swimming with biological burst-and-glide motion
   - Support: Well-supported
   - Type: Secondary
   - Basis: closed-loop assays include no-action and linear-motion controls with consistent withdrawal effects
   - Location: Abstract; Fig. 5l–o
3. collective escape from danger can be mediated by an individual’s visual perception of other escaping animals
   - Support: Supported
   - Type: Secondary
   - Basis: behavioural assays support visual social mediation, though they do not establish a specific neural mediator
   - Location: Abstract; Figs. 2–3
4. Neural encoding of this rapid social offset allows fish to infer danger from social information alone
   - Support: Tentative
   - Type: Secondary
   - Basis: separate tethered imaging and freely moving assays do not establish the neural population’s causal behavioural role
   - Location: Abstract; Discussion

## Strengths

- **Convergent evidence across experimental levels** Natural group behaviour, closed-loop virtual-conspecific assays, and two-photon imaging converge on biologically patterned social offset as a cue associated with withdrawal. The sequence is developed across Figs. 1–5.
  - Dimension: contribution
  - Location: Results, Figs. 1–5
- **Matched motion and action controls** Biological burst-and-glide motion is compared with matched linear motion, while escape, disappearance, no-action, and looming conditions separate social offset from generic visual change.
  - Dimension: methodological rigour
  - Location: Figs. 4–5; Extended Data Figs. 6–8
- **Transparent analysis and null reporting** The paper reports mixed-effects treatment of nested imaging data, bootstrapped chance baselines, motor-confound checks, and a chance-level escape-versus-disappearance decoder rather than suppressing the null result.
  - Dimension: methodological rigour
  - Location: Fig. 5e–i; Extended Data Figs. 6r–s, 7i, 8i–k; Methods, Statistics

## Limitations

- **Neural causation is not directly tested** Imaging identifies stimulus-responsive populations in tethered fish, while behaviour is measured in separate freely moving assays. No perturbation establishes that social-offset neurons direct, enable, or are required for withdrawal.
  - Dimension: methodological rigour
  - Location: Title; Abstract; Figs. 4–5; Discussion
- **Causal wording exceeds demonstrated linkage** The Title’s “directs” and the Abstract’s statement that neural encoding “allows” inference present a functional neural mechanism more strongly than the non-interventional evidence supports.
  - Dimension: reporting
  - Location: Title; Abstract
- **Generalization remains model-specific** The neural and behavioural mechanism is tested in one laboratory species and controlled virtual environments. Cross-species conservation and ecological transfer are proposed in the Discussion but not directly examined.
  - Dimension: application pathway
  - Location: Discussion, pp. 7–8

## Reliability checks

Caveats (4 of 4 checks).

- **Internal coherence:** Noted. The numerical and experimental narrative is generally coherent, but causal neural framing exceeds the direct linkage established between tethered imaging and freely moving behaviour. Large neuron counts relative to fish and session counts are mitigated by reported mixed-effects models.
- **Research conduct:** Clean. Animal approval, absence of competing interests, and named data and code locations are declared. No text-grounded research-conduct concern was identified.
- **Reference integrity:** Clean. 12 of 12 checkable references verified
  - 12 references in manuscript
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
