---
id: "20260723-0204-what-is-the-primary"
title: "Attwell & Laughlin 2001: the primary grey-matter signaling energy-budget split"
type: "capture"
status: "promoted"
promoted_to: ["[[claim-attwell-laughlin-2001-grey-matter-signaling-energy-split]]  # 2001 four-way split 47/34/13/3; ~81% signaling-dominant sum folded in","[[claim-howarth-2012-revised-brain-energy-budget-lowers-action-potential-share]]  # 2012 revision: AP 47%→21%, postsynaptic 34%→50%, total ~81%→~71%"]
not_promoted: ["The 47%+34%=~81% arithmetic sum (capture Claim 2) — not a distinct atomic claim; it is arithmetic over two figures from the same abstract sentence, so folded into the 2001 split note rather than given its own file.","Cerebellar-cortex 2012 split (resting 54% / postsynaptic 22% / AP 17%; Purkinje cells 15% of cerebellar signaling energy) — region-specific tangent, no verbatim exact_quote captured; left as a further lead. Region-specific budgets could earn their own note later but no kept claim rests on this.","Laughlin KITP conference-slide rescaling ('~25% goes to non-signaling maintenance' + the total-vs-signaling rescaling interpretation) — carries the capture's own [unverified-quant] flag, weak transport (extract_pdf tls: unverified), and is not load-bearing for any kept claim; left in inbox, not routed as a question per intake discipline (minor, nothing rests on it).","The '145 mL/100g grey matter/h per extra AP/neuron/s' oxygen-consumption figure — distinct quantitative claim seen only in a WebFetch summary, not re-verified against a second Tier 1–2 source this run; a lead, not a kept claim, and not load-bearing here.","The 2001 paper's downstream '≤15% of neurons simultaneously active' sparse-coding implication — a distinct mechanism-level claim, not chased to primary full text this run; links to [[claim-barlow-1961-efficient-coding-removes-sensory-redundancy]] territory. Left as a lead.","Person entity candidates (David Attwell, Simon B. Laughlin, Clare Howarth, Padraig Gleeson) and concept candidates (energy budget for signaling, action potential, postsynaptic potential, sparse coding) — no hubs built this run (bias against the flood). Attwell now anchors 2+ neuroenergetics notes and a possible MOC is forming; both logged to 00-meta/seek-flags.md as noticings rather than built here."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-07-23T00:00:00.000Z"
provenance: "batch run 2026-07-23; web research via WebSearch + WebFetch + extract_pdf"
derived_from: []
tags: ["neuroscience","brain-energy-budget","Attwell-Laughlin","action-potentials","synaptic-transmission","metabolic-constraints"]
source_tier_summary: "Tier 1 (PubMed-mirrored primary abstracts) for the core quantitative split; Tier 1-with-caveat (co-author's own conference slides, weak transport) for corroborating structural context"
---


**Topic question:** What is the primary energy-budget split in Attwell & Laughlin 2001, and does signaling (action potentials ~47% + postsynaptic effects ~34%) total ~81% of grey-matter signaling energy?

**Short answer the claims below support:** Yes, confirmed for the 2001 estimate specifically. The paper's own abstract states action potentials consume 47% and postsynaptic effects of glutamate consume 34% of the grey-matter signaling energy budget; these are the two largest of four named components (resting potential 13%, glutamate recycling 3%). 47 + 34 = 81, so signaling's two dominant components do total ~81% of the modeled signaling energy budget. However, a 2012 follow-up paper by an overlapping author (Attwell) substantially revised these proportions downward for the action-potential term — a caveat that matters for anyone citing the 2001 split as current.

---

## Claim: Attwell & Laughlin (2001) modeled grey-matter signaling energy as split 47% action potentials / 34% postsynaptic glutamate effects / 13% resting potential / 3% glutamate recycling

**Claim type:** Quantitative (specific percentages from a named study).

The paper, "An Energy Budget for Signaling in the Grey Matter of the Brain" (Attwell D, Laughlin SB, *Journal of Cerebral Blood Flow & Metabolism* 21(10):1133-45, 2001), used anatomic and physiologic data to model energy expenditure on components of excitatory (glutamatergic) signaling in rodent cortical grey matter. The paper's own abstract, reproduced verbatim on PubMed, states the four-way split directly.

**Sourcing floor check:** Quantitative claim (specific percentages) — requires Tier 1-2. Met: the exact quote is the study's own abstract text, reproduced by PubMed (NLM), the standard indexing mirror of the primary publication's own words. Independently corroborated by a Google Scholar bibliographic lookup confirming the same PMID/DOI/citation, and by the paper's co-author's own later conference slides (see Further leads) reproducing a consistent structural breakdown.

| Field | Value |
|---|---|
| source_url | https://pubmed.ncbi.nlm.nih.gov/11598490/ |
| source_author | David Attwell, Simon B. Laughlin |
| source_date | 2001-10 |
| source_tier | 1 |
| exact_quote | "Action potentials and postsynaptic effects of glutamate are predicted to consume much of the energy (47% and 34%, respectively)" [with resting potential 13% and glutamate recycling 3% given in the same abstract] |
| doi | 10.1097/00004647-200110000-00001 |
| journal_landing_url | https://journals.sagepub.com/doi/10.1097/00004647-200110000-00001 (resolves; full text paywalled, abstract confirmed via PubMed mirror) |

---

## Claim: Action potentials (47%) and postsynaptic effects (34%) together account for ~81% of the modeled grey-matter signaling energy budget in the 2001 estimate

**Claim type:** Quantitative (arithmetic sum of two Tier-1-sourced figures from the same source).

47% + 34% = 81%. Both addends are drawn from the same directly-quoted abstract sentence above (Claim 1), naming the same denominator (energy budget for signaling in the grey matter of the brain — the paper's own title and stated scope). No additional primary-source hop is needed for the sum itself since it is simple arithmetic over two already-Tier-1-sourced numbers with a shared, explicitly stated denominator. This directly confirms the numeric premise in the topic question: signaling's two largest modeled components, action potentials and postsynaptic glutamate effects, dominate the 2001 grey-matter signaling energy budget, with the remaining ~16-19% split between resting-potential maintenance (13%) and glutamate recycling (3%) — leaving a small (~3%) unaccounted residual, consistent with rounding across four modeled categories.

| Field | Value |
|---|---|
| source_url | https://pubmed.ncbi.nlm.nih.gov/11598490/ |
| source_author | David Attwell, Simon B. Laughlin |
| source_date | 2001-10 |
| source_tier | 1 |
| exact_quote | (same abstract quote as Claim 1; 81% is a direct sum of the two quoted percentages, not a separately quoted figure) |
| derivation_note | 47 + 34 = 81; arithmetic, not an independently sourced statistic |

---

## Claim: A 2012 follow-up study by an overlapping author team (Howarth, Gleeson, Attwell) substantially revised the 2001 split, lowering the action-potential share and raising the postsynaptic share

**Claim type:** Quantitative (specific percentages from a named follow-up study) + historical (supersession of an earlier estimate).

Howarth C, Gleeson P, Attwell D, "Updated Energy Budgets for Neural Computation in the Neocortex and Cerebellum," *Journal of Cerebral Blood Flow & Metabolism* (2012), re-modeled the same signaling-energy budget after new data showed mammalian action potentials are more energy-efficient than the 2001 model assumed. For cerebral cortex, the revised split is: postsynaptic glutamate receptors 50%, action potentials 21%, resting potentials 20%, presynaptic transmitter release 5%, transmitter recycling 4%. Under this revision, action potentials + postsynaptic effects sum to 71%, not 81% — a materially different total from the 2001 estimate, driven almost entirely by the action-potential term falling from 47% to 21%. This means the "~81%" figure in the topic question is accurate as a description of the 2001 paper specifically, but is not the current best estimate from the same research group eleven years later.

**Sourcing floor check:** Quantitative claim — Tier 1-2 required. Met: exact quote is the 2012 paper's own abstract, reproduced via PubMed mirror, same standard as Claim 1.

| Field | Value |
|---|---|
| source_url | https://pubmed.ncbi.nlm.nih.gov/22434069/ |
| source_author | Clare Howarth, Padraig Gleeson, David Attwell |
| source_date | 2012 |
| source_tier | 1 |
| exact_quote | "most signaling energy (50%) is used on postsynaptic glutamate receptors, 21% is used on action potentials, 20% on resting potentials, 5% on presynaptic transmitter release, and 4% on transmitter recycling" |
| context_quote | estimates "need reevaluating following recent work demonstrating that action potentials in mammalian neurons are much more energy efficient than was previously thought" |
| doi_landing_url | https://journals.sagepub.com/doi/10.1038/jcbfm.2012.35 (resolves; full text paywalled, abstract confirmed via PubMed mirror) |

---

## Further leads

- Howarth, Gleeson & Attwell 2012 also modeled cerebellar cortex separately, finding a very different split there: resting potentials 54%, postsynaptic receptors 22%, action potentials only 17%, with Purkinje cells using just 15% of total cerebellar signaling energy — worth its own note on region-specific energy budgets. (Source: PubMed abstract, https://pubmed.ncbi.nlm.nih.gov/22434069/, Tier 1.)
- Simon Laughlin's own conference slide deck ("Energy and the Designs of Brains," KITP UCSB, undated) reproduces a structurally consistent but differently-partitioned breakdown explicitly attributed to "Attwell & Laughlin, J Cereb Blood Flow Metabolism 2001": post-synaptic current 25%, resting potentials (neurons+glia) 10%, action potentials 35%, maintenance 25%, pre-synaptic 5% — of *total* grey matter energy (i.e., including non-signaling maintenance), rather than of signaling energy alone. Rescaling by excluding the 25% maintenance term reproduces figures close to the 47/34/13% signaling-only split, suggesting internal consistency between the two framings, but this specific rescaling and the "~25% goes to non-signaling maintenance" claim were not independently confirmed against the paper's full text this run. Source URL: https://online.kitp.ucsb.edu/online/brain08/laughlin/pdf/Laughlin_Brain_KITP.pdf — fetched via extract_pdf; provenance recorded `tls: "unverified"`, so per the safety spec this should not be treated as sole support for any load-bearing claim. Marked **[unverified-quant — needs primary]** for the specific "25% maintenance" figure and the rescaling interpretation.
- The 2001 paper's stated downstream implication — that grey-matter energy constraints favor sparse/distributed neural codes with "≤15% of neurons simultaneously active" — is a distinct mechanism-level claim worth its own note; not chased down to primary full text this run.
- The 2001 paper also reportedly estimates that "an increase in activity of 1 action potential/cortical neuron/s will raise oxygen consumption by 145 mL/100 g grey matter/h" (seen in a WebFetch summary of the PubMed abstract) — a distinct quantitative claim, not yet independently re-verified against a second Tier 1-2 source this run.
- Full primary-text access to both papers (Sage Journals) was blocked by paywall/anti-bot (HTTP 403) during this run; a library-mediated fetch of the actual PDF (rather than the abstract) would let a future pass quote the results/discussion sections directly rather than relying on the abstract alone.
- Possible connection to [[claim-brain-inference-bound-like-ai-at-system-level]] and [[claim-synaptic-plasticity-cheap-fraction-of-transmission-energy]] — both discuss brain energy allocation; a promoted claim-note from this capture should check for overlap/complementarity rather than duplication.

## Safety flags

None fired this session. One fetched source (Simon Laughlin's KITP conference-slide PDF) recorded `tls: "unverified"` in its extract_pdf provenance, which per the safety spec earns elevated suspicion — its content was reviewed and contains no addressed-to-AI language, override language, claimed authority, tier self-assignment, file-system instructions, credential requests, or urgency framing. It reads as an ordinary academic conference slide deck. No safety-log entry required beyond this note; it is not used as sole support for any load-bearing (core) claim above, consistent with the weak-transport rule.

## Entity candidates

- David Attwell — person — co-author of both the 2001 and 2012 papers; UCL neuroscientist, central figure in brain energy-budget modeling.
- Simon B. Laughlin — person — co-author of the 2001 paper; Cambridge zoologist known for efficient-coding and energy-cost-of-information work; his KITP slides are a secondary primary-author account of the same budget.
- Clare Howarth — person — lead author of the 2012 revision paper.
- Padraig Gleeson — person — co-author of the 2012 revision paper.
- energy budget for signaling (grey matter) — concept — the specific quantitative framework this capture is about; candidate for its own concept/MOC note distinct from any single claim.
- action potential — concept — one of the two dominant cost components in both the 2001 and 2012 budgets; likely already has or warrants its own note given how load-bearing it is across neuroscience captures.
- postsynaptic potential / glutamate receptor signaling — concept — the other dominant cost component; candidate for its own note.
- neural coding efficiency / sparse coding — concept — the downstream implication the 2001 paper draws from its energy budget (≤15% of neurons active); links to [[claim-barlow-1961-efficient-coding-removes-sensory-redundancy]] territory.