Latency of prediction-error / deviance signals in mouse V1

Timing and qualitative form of mouse V1 prediction-error responses vary across paradigms (sequence vs. oddball) and may distinguish 'omission-error' vs. 'substitution-error' channels — informative for which version of predictive coding best fits the data.

{
  "papers": [
    {
      "n": null,
      "doi": "10.1093/cercor/bhad163",
      "value": "100-150",
      "method": "extracellular recording + MbTDR",
      "metric": "late suppression window (ms)",
      "n_analyzed": null,
      "ci_or_error": null,
      "text_access": "abstract_only",
      "n_definition": "mice (count not in abstract)",
      "scope_region": "V1",
      "study_system": "mouse V1",
      "taxonomic_level": "broad category",
      "scope_population": "all V1 units pooled",
      "value_source_sentence": "Neural responses to expected stimuli were suppressed in a late window (100-150 ms) after stimulus onset following training, whereas responses to novel stimuli were not.",
      "experimental_conditions": "sequence learning, awake"
    },
    {
      "n": null,
      "doi": "10.1093/cercor/bhad215",
      "value": "150-230",
      "method": "16-channel LFP + MUA",
      "metric": "deviance onset latency (ms) in supragranular V1",
      "n_analyzed": null,
      "ci_or_error": null,
      "text_access": "abstract_only",
      "n_definition": "mice (count not in abstract)",
      "scope_region": "V1 L2/3",
      "study_system": "mouse V1",
      "taxonomic_level": "broad category",
      "scope_population": "all V1 units pooled",
      "value_source_sentence": "Multiunit activity and current source density profiles showed that although basic adaptation to redundant stimuli was present early (50 ms) in layer 4 responses, DD emerged later (150-230 ms) in supragranular layers (L2/3).",
      "experimental_conditions": "visual oddball, awake"
    },
    {
      "n": null,
      "doi": "10.1101/2024.01.20.576433",
      "value": "increased at expected stimulus time",
      "method": "two-photon calcium imaging",
      "metric": "omission-evoked activity",
      "n_analyzed": null,
      "ci_or_error": null,
      "text_access": "abstract_only",
      "n_definition": "mice (count not in abstract)",
      "scope_region": "V1 L2/3",
      "study_system": "mouse V1 L2/3",
      "taxonomic_level": "subcategory",
      "scope_population": "L2/3 pyramidal cells",
      "value_source_sentence": "We find increased neural activity at the time an expected, but omitted, stimulus would have occurred but no significant prediction error responses following an unexpected stimulus substitution.",
      "experimental_conditions": "multi-day sequence learning, awake"
    }
  ],
  "audit_issues": [
    {
      "dimension": "study_system",
      "description": "Rows mix paradigms (sequence learning vs. visual oddball vs. multi-day sequence learning with omission probes) and modalities (extracellular + MbTDR vs. LFP+MUA vs. two-photon Ca²⁺).",
      "entries_affected": [
        "10.1093/cercor/bhad163",
        "10.1093/cercor/bhad215",
        "10.1101/2024.01.20.576433"
      ]
    },
    {
      "dimension": "metric_definition",
      "description": "Reported timings reflect different events: 'late suppression window' (100–150 ms, sequence-learning suppression of expected stimuli), 'deviance-onset latency' (150–230 ms, oddball deviant detection), and 'omission-evoked activity' (qualitative). Not all rows share the same time-zero or response definition.",
      "entries_affected": [
        "10.1093/cercor/bhad163",
        "10.1093/cercor/bhad215",
        "10.1101/2024.01.20.576433"
      ]
    },
    {
      "dimension": "scope_population",
      "description": "Row 3 is L2/3 pyramidal cells only; rows 1–2 pool all V1 units.",
      "entries_affected": [
        "10.1101/2024.01.20.576433"
      ]
    }
  ],
  "audit_verdict": "CAVEAT",
  "comparison_id": "mouse_v1_deviant_timing",
  "comparison_name": "Latency of prediction-error / deviance signals in mouse V1",
  "comparison_type": "cross-study conflict",
  "what_it_reveals": "Timing and qualitative form of mouse V1 prediction-error responses vary across paradigms (sequence vs. oddball) and may distinguish 'omission-error' vs. 'substitution-error' channels — informative for which version of predictive coding best fits the data.",
  "homogeneity_check": {
    "caveats": [
      "Different paradigms (sequence learning vs. oddball)",
      "Population definition differs (L2/3-only vs. all units)",
      "Methods range from MUA/LFP to two-photon imaging"
    ],
    "n_definition_uniform": "false",
    "scope_region_uniform": "true",
    "taxonomic_level_uniform": "false",
    "scope_population_uniform": "false"
  },
  "suggested_plot_type": "timeline",
  "mandatory_caption_caveats": [
    "Paradigms differ (sequence-learning suppression vs. oddball deviance vs. omission probes); the reported timings are not all measured from the same stimulus event.",
    "Population definition differs (all V1 units vs. L2/3-only)."
  ]
}

[
  "paper:paper-805e7b60808f",
  "paper:paper-c0f0ae0b432b",
  "paper:paper-d6acd7459360"
]

{
  "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"
}