Version history

1 version on record. Newest first; the live version sits at the top with a live indicator.

  1. Live 7012074336b4
    5/17/2026, 4:35:28 PM
    Content snapshot
    {
      "kind": "infographic",
      "prompt": "Small numbers (single-digit to low-tens) of mouse cortical pyramidal neurons are causally sufficient to drive perceptual decisions, with the most extreme number coming from V1 'pattern-completion' neurons whose activation recalls a larger ensemble.",
      "provider": "other",
      "raw_fields": {
        "papers": [
          {
            "n": null,
            "doi": "10.7554/eLife.58889",
            "value": "~14",
            "method": "2p holographic optogenetics + 2p Ca imaging",
            "metric": "minimum neurons sufficient for perceptual detection",
            "n_analyzed": null,
            "ci_or_error": "saturates at ~37",
            "text_access": "abstract_only",
            "n_definition": "L2/3 pyramidal neurons activated per ensemble",
            "scope_region": "primary somatosensory cortex (barrel)",
            "study_system": "awake mouse barrel cortex",
            "taxonomic_level": "broad excitatory cell category",
            "scope_population": "L2/3 pyramidal neurons",
            "value_source_sentence": "By precisely titrating the number of neurons stimulated, we demonstrate that the lower bound for perception of cortical activity is ~14 pyramidal neurons.",
            "experimental_conditions": "all-optical 2p activation + detection task"
          },
          {
            "n": null,
            "doi": "10.1016/j.cell.2019.05.045",
            "value": "2",
            "method": "2p holographic optogenetics + Ca imaging",
            "metric": "pattern-completion neurons sufficient to bias visual behavior",
            "n_analyzed": null,
            "ci_or_error": null,
            "text_access": "abstract_only",
            "n_definition": "pattern-completion neurons activated to recall an ensemble",
            "scope_region": "primary visual cortex",
            "study_system": "awake mouse V1",
            "taxonomic_level": "fine functional subtype",
            "scope_population": "pattern-completion neurons within ensembles",
            "value_source_sentence": "activation of only two pattern completion neurons from behaviorally relevant ensembles improved performance, by reliably recalling the whole ensemble.",
            "experimental_conditions": "visual go/no-go discrimination"
          }
        ],
        "audit_issues": [
          {
            "dimension": "scope_region",
            "description": "S1 barrel cortex (Dalgleish/eLife 58889) vs. V1 (Carrillo-Reid/Cell 2019).",
            "entries_affected": [
              "10.7554/eLife.58889",
              "10.1016/j.cell.2019.05.045"
            ]
          },
          {
            "dimension": "scope_population",
            "description": "Targeting differs: random L2/3 pyramidal neurons (Dalgleish) vs. pre-identified pattern-completion neurons within a behaviourally-relevant ensemble (Carrillo-Reid). Therefore the '~14 vs. 2' difference reflects target-selection sophistication, not a baseline biological threshold.",
            "entries_affected": [
              "10.7554/eLife.58889",
              "10.1016/j.cell.2019.05.045"
            ]
          },
          {
            "dimension": "metric_definition",
            "description": "Dalgleish's '~14' is the minimum count sufficient to drive detection; Carrillo-Reid's '2' is the number of pattern-completion neurons whose activation biases an existing visual discrimination — a different operational definition.",
            "entries_affected": [
              "10.7554/eLife.58889",
              "10.1016/j.cell.2019.05.045"
            ]
          }
        ],
        "audit_verdict": "CAVEAT",
        "comparison_id": "perceptual-threshold-cortical-activation",
        "comparison_name": "Minimum number of mouse cortical neurons whose holographic activation drives behavior",
        "comparison_type": "convergent evidence",
        "what_it_reveals": "Small numbers (single-digit to low-tens) of mouse cortical pyramidal neurons are causally sufficient to drive perceptual decisions, with the most extreme number coming from V1 'pattern-completion' neurons whose activation recalls a larger ensemble.",
        "homogeneity_check": {
          "caveats": [
            "Cell 2019 uses V1 'pattern-completion' neurons; eLife 2020 uses randomly targeted L2/3 pyramidal neurons in S1 — the metric is conceptually compared but not directly equivalent (selective vs random targeting).",
            "Behavioral tasks differ: visual go/no-go (Cell 2019) vs perceptual detection (eLife 2020)."
          ],
          "n_definition_uniform": "false",
          "scope_region_uniform": "false",
          "taxonomic_level_uniform": "false",
          "scope_population_uniform": "false"
        },
        "suggested_plot_type": "forest plot",
        "mandatory_caption_caveats": [
          "The ~14 vs. 2 difference is largely driven by neuron-selection strategy (random L2/3 vs. selected pattern-completion neurons), not by a difference in baseline cortical sensitivity.",
          "Tasks differ (perceptual detection vs. visual go/no-go discrimination); cortical areas differ (S1 vs. V1)."
        ]
      },
      "section_id": "section_11",
      "source_url": "https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence/blob/79ce062d54a924ce05953ec90aa9d26044d2b48f/evidence/section_11_evidence_package.json",
      "target_ref": "wiki_page:computationalreviewrecurrence-11-pattern-completion",
      "review_repo": "ComputationalReviewRecurrence",
      "section_ref": "wiki_page:computationalreviewrecurrence-11-pattern-completion",
      "source_path": "evidence/section_11_evidence_package.json",
      "source_refs": [
        "paper:paper-d2118dd5013a",
        "paper:paper-e37e0b6bad65"
      ],
      "section_title": "11. Physiological signature III — pattern completion, replay, and sequence generation as recurrent-circuit read-outs in mouse cortex",
      "source_policy": {
        "mode": "public_source_pointer_with_short_context",
        "notes": [
          "Local review repositories are read-only inputs.",
          "SciDEX stores paper metadata, structured evidence, file pointers, and short citation contexts; it does not copy full review prose."
        ],
        "source_commit_sha": "79ce062d54a924ce05953ec90aa9d26044d2b48f",
        "source_repository_url": "https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence"
      },
      "generation_status": "complete",
      "review_bundle_ref": "analysis_bundle:ab-d9c479db9be9",
      "origin_url": "https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence/blob/79ce062d54a924ce05953ec90aa9d26044d2b48f/evidence/section_11_evidence_package.json",
      "commit_sha": "79ce062d54a924ce05953ec90aa9d26044d2b48f",
      "created_by": "persona-jerome-lecoq-gbo-neuroscience",
      "repository_url": "https://github.com/AllenNeuralDynamics/ComputationalReviewRecurrence"
    }