nucleation

LIT-650: Conscious birds

4 min read

Properties4
StatusActive — read in full 2026-10-03 (NOTE-488); worth reading as a short, current statement of the case for consciousness in birds and of what each of three theories (global neuronal workspace, recurrent processing, IIT) would need to find in an avian brain. Its strongest evidence is second-hand: crow nidopallium caudolaterale neurons that track the reported percept rather than the stimulus (Nieder et al. 2020). Its sharpest argument is that the bird's prefrontal analogue is nuclear, not laminar, so conscious report does not need cortical layering. The authors say the findings "do not allow for a strong conclusion".
Tagsconsciousness, neuroscience, cognition, self
Cited as “paper” byNOTE-488 — Conscious birds
Cited as “source” byTHEORY-100 — A neural correlate of conscious report does not require laminar cortex: in crows it is carried by the nidopallium caudolaterale, a nuclear pallial region whose neurons follow the reported percept rather than the stimulus

Graph View

Gianmarco Maldarelli and Onur Güntürkün (2025), Philosophical Transactions of the Royal Society B 380(1939):20240308, theme issue “Evolutionary functions of consciousness” — DOI-10.1098/rstb.2024.0308

Key takeaways

  • Report without cortex. In crows trained to report whether a near-threshold stimulus appeared, single neurons in the nidopallium caudolaterale (NCL), the functional analogue of mammalian prefrontal cortex, fired with the report, whether or not a stimulus was shown, and were silent on “not seen” reports. A later study found separate NCL populations coding reported presence and reported absence. The NCL is nuclear, not layered: the cortex-like layers and columns found in birds (Stacho et al. 2020) are confined to sensory pallium. So, if birds are conscious, a correlate of conscious report does not need laminar architecture.
  • Theories as checklists for a different brain. The pigeon forebrain connectome has a few densely interlinked hubs (NCL and four others) joined to local modules, the arrangement global neuronal workspace theory (GNWT) requires, so GNWT’s anatomical requirements are “largely met”. Recurrent connectivity in the visual pallium and NCL meets recurrent processing theory’s (RPT’s) anatomy, though recurrence was never switched off experimentally. IIT’s requirements are unmet “(yet)”: nobody has measured the perturbational complexity index (PCI) in birds or found the grid-like microcolumn lattice IIT is linked to in mammals.
  • The mark test underrates birds. Only magpies (contested by a failed replication) and Indian house crows pass it. Pigeons and roosters fail it, but pigeons under food competition, and roosters warning of a predator, treat their mirror image differently from a conspecific. The authors read self-awareness as graded, multidimensional and fitted to each species’ ecology.

Standing in the record

Filed on 2026-10-03 at the owner’s request, with Newen & Montemayor (LIT-626) from the same theme issue, for the record’s holdings on animal consciousness and its theories. No anthology topic holds it, and it carries no instruction for machine-learning practice.

It is the record’s first work on consciousness in a non-mammalian brain and its first direct use of GNWT and RPT on an animal. It connects as follows.

  • Butlin et al. (LIT-056) derive “indicator properties” from the same theories (RPT-1–2, GWT-1–4) and apply them to AI systems. This paper makes the same theory-heavy move on a bird, without using their list. The parallel is instructive: the paper finds the two functionalist theories’ requirements met by a brain built differently from ours, which is what a computational functionalist would expect. That is my reading, not the authors’ claim.
  • THEORY-023 says AI consciousness cannot currently be settled. Its promote_when asks for a comparative programme across animals in which a biological property predicts where consciousness markers fall. Avian evidence of this kind is that programme’s raw material. The NCL result counts against cortical lamination as such a property. It says nothing either way about life or metabolism, the properties Seth’s biological naturalism (LIT-135) turns on.
  • IIT 3.0 (LIT-445) and Barrett et al. (LIT-185). The paper proposes PCI as “a good substitute” for Φ in birds. Barrett et al. argue that PCI is a proxy, not an approximation, capturing at most two of IIT’s six axioms. So even a positive avian PCI would test IIT less directly than this paper implies.
  • Newen & Montemayor (LIT-626), in the same issue, read GNWT as needing prefrontal cortex and argue that basic consciousness is subcortical. This paper reads GNWT as a connectome property that a nuclear pallium can meet. The two disagree about what GNWT requires. Neither cites the other. The two also use the mirror test differently: Newen & Montemayor count magpies among the species that pass it, while this paper notes a failed replication and doubts the test.
  • Veit (LIT-105) argues for a gradualist, evolutionary approach to animal consciousness. This paper ends on the same gradualism (“not … an ‘all-or-nothing’ cognitive function”) and the same adaptationist inference. It is consistent with Veit’s approach but offers no account of why consciousness evolved, only evidence of where it is.

No relation is declared. It tests no theory the record holds, it compares itself against nothing the record holds, and it argues for no account of consciousness that another entry excludes.

Backlinks