{
  "version": 1,
  "date": "2026-10-22",
  "status": "declared; full-model execution on the existing research host pending",
  "question": "Does the pinned released cat visual-cortex model, run with its released spontaneous recipe (param_spont) on the existing research host, produce finite spontaneous statistics, and how do its 24 means compare descriptively with the printed Figure 4 labels?",
  "model": {
    "name": "LSV1M v1.0",
    "commit": "04bf11b72ff5f78e66f28e037016e8e33b09c12e",
    "parameters": "param_spont/defaults",
    "entrypoint": "create_experiments_spont",
    "changes": "No changes to packaged upstream model files. The wrapper adds timing receipts and defers heavy plotting until raw data is saved.",
    "recipe_difference": "The released param and param_spont trees differ in their self-references, their recorder definitions, and the position of the unchanged native_nest: False entry within the cell block of l4_cortex_inh, l23_cortex_exc and l23_cortex_inh (same key and value). Network, input and simulation values are otherwise identical. recipe-check-009 described the difference as recorders only."
  },
  "source_commits": [
    {
      "directory": "mozaik",
      "commit": "f5c09a0eccc5786daf78881d883bbce586a00e6d"
    },
    {
      "directory": "PyNN",
      "commit": "981853dd567cefb3e337accc5c13e1c8c99385ad"
    },
    {
      "directory": "imagen",
      "commit": "a4af248e161b896471d820f9f39e9374d99b1c69"
    },
    {
      "directory": "nest-checkout",
      "commit": "41892a50f95ddf50966e2371151b207d098e814a"
    },
    {
      "directory": "stepcurrent",
      "commit": "fe91c0345a602d564a74ebe0ea2c66d9a4157005"
    }
  ],
  "runtime": {
    "python": "3.9.20",
    "packages": {
      "Cython": "0.29.36",
      "Jinja2": "3.1.6",
      "MarkupSafe": "3.0.4",
      "MorphIO": "3.3.6",
      "Pillow": "9.1.0",
      "PyNN": "0.11.0",
      "PyWavelets": "1.4.1",
      "Pygments": "2.21.0",
      "Sphinx": "7.4.7",
      "alabaster": "0.7.16",
      "attrs": "26.1.0",
      "babel": "2.18.0",
      "certifi": "2026.7.22",
      "charset-normalizer": "2.0.12",
      "cycler": "0.12.1",
      "docutils": "0.21.2",
      "elephant": "0.11.1",
      "execnet": "2.1.2",
      "fonttools": "4.60.2",
      "future": "0.18.2",
      "idna": "3.20",
      "imageio": "2.19.3",
      "imagen": "1.0",
      "imagesize": "1.5.0",
      "importlib_metadata": "8.7.1",
      "iniconfig": "2.1.0",
      "interval": "1.0.0",
      "junitparser": "2.4.3",
      "kiwisolver": "1.4.7",
      "lazyarray": "0.5.2",
      "llvmlite": "0.38.1",
      "matplotlib": "3.5.1",
      "mozaik": "0.1.0",
      "mpi4py": "3.1.6",
      "neo": "0.12.0",
      "nest-simulator": "3.4",
      "networkx": "3.2.1",
      "numba": "0.55.2",
      "numpy": "1.22.3",
      "numpyencoder": "0.3.0",
      "packaging": "26.3",
      "param": "1.5.1",
      "parameters": "0.2.1",
      "pip": "24.2",
      "pluggy": "1.6.0",
      "psutil": "5.9.0",
      "py": "1.11.0",
      "pyparsing": "3.3.3",
      "pytest": "7.1.1",
      "pytest-forked": "1.6.0",
      "pytest-timeout": "2.1.0",
      "pytest-xdist": "2.5.0",
      "python-dateutil": "2.9.0.post0",
      "quantities": "0.14.1",
      "requests": "2.32.5",
      "scikit-image": "0.19.3",
      "scipy": "1.8.0",
      "setuptools": "65.5.0",
      "six": "1.17.0",
      "snowballstemmer": "3.1.1",
      "sphinxcontrib-applehelp": "2.0.0",
      "sphinxcontrib-devhelp": "2.0.0",
      "sphinxcontrib-htmlhelp": "2.1.0",
      "sphinxcontrib-jsmath": "1.0.1",
      "sphinxcontrib-qthelp": "2.0.0",
      "sphinxcontrib-serializinghtml": "2.0.0",
      "tifffile": "2024.8.30",
      "tomli": "2.5.0",
      "tqdm": "4.70.1",
      "urllib3": "1.26.20",
      "wheel": "0.38.4",
      "zipp": "3.23.1"
    },
    "nest": "3.4",
    "mpi_ranks": 1,
    "threads_per_rank": 16,
    "virtual_processes": 16,
    "environment": {
      "OMP_NUM_THREADS": "16",
      "OMP_WAIT_POLICY": "PASSIVE",
      "GOMP_SPINCOUNT": "0",
      "OPENBLAS_NUM_THREADS": "1",
      "MKL_NUM_THREADS": "1",
      "NUMEXPR_NUM_THREADS": "1",
      "MPLBACKEND": "Agg",
      "PYTHONHASHSEED": "0",
      "PYTHONDONTWRITEBYTECODE": "1",
      "OMPI_MCA_mpi_yield_when_idle": "1"
    }
  },
  "protocol": {
    "trial": 0,
    "replicates": 1,
    "pynn_seed": 5,
    "mozaik_effective_seed": 513,
    "released_mpi_seed_field": 1023,
    "seed_note": "mpi_seed is required by the model schema but not forwarded as mozaik_seed by the controller. Preserve released fields; the pinned framework defaults to 513.",
    "time_step_ms": 0.1,
    "analog_sample_period_ms": 1.0,
    "measured_window_ms": 40320.0,
    "bin_ms": 10.0,
    "bin_count": 4032,
    "background_luminance_cd_m2": 50.0,
    "workflow_derived_total_ms": 40770.0,
    "workflow_intervals_ms": [
      150,
      40320,
      150,
      150
    ],
    "workflow_note": "Blank reset before the measured InternalStimulus; measured segment; blank reset before the terminal blank; terminal blank. PyNN 0.11.0 adds one min_delay (0.1 ms) to its first run (pyNN/nest/simulator.py, _State.run), so NEST ends at 40,770.1 ms. The feasibility-010 probe logged requests of 150, 70, 150 and 150 ms and ended at 520.1 ms; the receipt wrapper was smoke-tested in checks/.",
    "workflow_run_calls_ms": [
      150.0,
      40320.0,
      150.0,
      150.0
    ],
    "nest_biological_time_after_calls_ms": [
      150.1,
      40470.1,
      40620.1,
      40770.1
    ]
  },
  "compatibility": {
    "self_connections": true,
    "decision": "Retain released True: preserves the previous unsuppressed diagonal weight rule in inspected ancestry. Exact archived runtime still unknown.",
    "change_commit": "c8261569a858ebbd3b91621b2d863d8059c89f37",
    "parent_commit": "d2ea60ce38f824d12ff0886e051ab3ad73939079"
  },
  "measurement": {
    "populations": [
      "V1_Exc_L4",
      "V1_Inh_L4",
      "V1_Exc_L2/3",
      "V1_Inh_L2/3"
    ],
    "metrics": [
      "firing_rate_hz",
      "isi_cv",
      "pairwise_correlation",
      "voltage_mV",
      "exc_conductance_nS",
      "inh_conductance_nS"
    ],
    "summaries": [
      "mean",
      "sd"
    ],
    "analysis_comparisons": 48,
    "atol": 1e-10,
    "rtol": 1e-12,
    "figure_source": "model/visualization_functions.py",
    "recording_recipe": "Released param_spont recorders: 12,128 / 3,467 / 12,176 / 3,487 spike-recorded and 400 / 399 / 400 / 398 analog-recorded neurons (L4 exc / L4 inh / L2/3 exc / L2/3 inh), equal to the paper S5 Table spontaneous counts (recipe-check-009).",
    "analysis_command": "See README.md; validate_datastore.py (protocol-007) then compare_reference.py",
    "secondary": {
      "name": "protocol-006 sample, spike-based measures",
      "neurons": {
        "V1_Exc_L4": 836,
        "V1_Inh_L4": 152,
        "V1_Exc_L2/3": 856,
        "V1_Inh_L2/3": 173
      },
      "definition": "Neurons spike-recorded by both the released param/ and param_spont/ recorders on this realization, frozen before the run in secondary/param-spike-subset.json.",
      "metrics": [
        "firing_rate_hz",
        "isi_cv",
        "pairwise_correlation"
      ],
      "method": "metrics.spike_metrics with the primary window and bin policy; positions and IDs checked against the frozen file.",
      "status": "Descriptive only; no upstream parity; not a pass criterion.",
      "command": "secondary_subset.py after validate_datastore.py"
    }
  },
  "acceptance": {
    "execution": "Exit 0; marker raw_data_saved true and explosion_detected false; four PyNN run steps requesting 150, 40,320, 150 and 150 ms, with NEST times 150.1, 40,470.1, 40,620.1 and 40,770.1 ms, 16 threads and 16 virtual processes; raw datastore retained. This is not a science pass.",
    "analysis": "Four populations, full 40320 ms window, 4032 bins, 48 finite agreeing numerical comparisons, raw input hashes unchanged.",
    "reference": "24 means; report signed differences and whether within half a unit of the printed last digit. Descriptive rounding comparison only.",
    "biological_equivalence_margin": null,
    "full_paper_reproduction_pass": null,
    "missing": [
      "exact historical runtime",
      "full-precision reference arrays and sample counts",
      "independent biological validation"
    ],
    "open_reference_issue": "Population-weighted rounded inhibitory conductance ~4.09 nS versus manuscript prose 3.56 nS; unresolved.",
    "execution_basis": "Run receipts after each PyNN run step; if the receipts are missing, Mozaik log lines plus NEST final status in the marker.",
    "stop_rule": "Every attempt is retained and reported. A failed or timed-out simulation may be repeated unchanged under protocol-007 only after a documented infrastructure failure (host reboot, loss of swap, full disk, out-of-memory kill or external stop), at most two attempts in total. Any other rerun, or any changed setting, is a new versioned protocol. The unchanged analysis programs may be rerun on the same saved datastore with different resource limits; any change to analysis code is a new version."
  },
  "compute": {
    "preflight": {
      "min_cpus": 4,
      "min_total_gib": 7.5,
      "min_available_gib": 6.5,
      "min_swap_total_gib": 60,
      "min_swap_free_gib": 55,
      "min_disk_free_gib": 20,
      "idle": "no other Felisyn simulation or analysis process; protocol-007 lock free",
      "identity": "package manifest, 37 model hashes, five source commits with clean trees"
    },
    "gates_note": "Engineering gates for the existing host, set from feasibility-010: a 0.52 s probe of this network peaked at 7.5 GiB memory plus 24.0 GiB swap. Not measured minimum requirements.",
    "host": "Existing DigitalOcean Regular droplet: 4 vCPU (1 thread per core), ~7.8 GiB RAM, Ubuntu 24.04; temporary 64 GiB swap file not in fstab.",
    "simulation_timeout_seconds": 259200,
    "timeout_grace_seconds": 120,
    "unit": {
      "runner": "started by root with systemd-run; runs as the unprivileged research user",
      "MemoryHigh": "7000M",
      "RuntimeMaxSec": 270000,
      "OOMPolicy": "continue"
    },
    "expected_duration_note": "Projection from feasibility-010: about 12.6 h excluding unmeasured steps. Not measured: retrieving, converting and saving 40 s of recordings under swap (with 70-150 ms segments the probe spent 13-15 min per presentation outside what Mozaik counts as simulation time), and any change in speed later in the run. The 72 h timeout is a stop rule, not a forecast: on the existing host a longer limit costs nothing, while a timeout forfeits the run.",
    "cost_note": "No new resources. The existing host is billed as before, whatever its load.",
    "authorization": "Owner instruction 22 Oct 2026: use the existing server; no new servers or provider accounts."
  },
  "model_files": {
    "LICENSE": "a26d3c2f8bd909a008bc7c7bb2427f16a18f2ae2ce9762228b0b80240c9e9c27",
    "README.md": "bae111703bc70b9950935f92639c7daf95f6279ceb97732198486200408a6053",
    "analysis_and_visualization.py": "24422016b54a084dd07aa8705ff81786d2377f2d1de0f344e90758f208c5d7de",
    "experiments.py": "f7695ce945a4115f2285896993c9230e047c3488888e4099538377f42560b292",
    "eye_path.pickle": "b125a9d152bb27aa34e37db2b0036b52858b738873538afe61a2fb7c064f0430",
    "image_naturelle_HIGHC.bmp": "c9ebccb7f186988cc5d7073db523a7af78c5abe138b1700edf41cf1598591403",
    "model.py": "55e1317df2fe5526e452693f13718c55369444b3f97481bb250b3189d4a3a9bc",
    "or_map_new_16x16": "e7c93ca03308a68c3818d3d91af8b26fd626c226d5c58c229fa317acaa206779",
    "param/SpatioTemporalFilterRetinaLGN_defaults": "5ef31eb2e0f0d6d2a5c8ae7d839b53a511dd923a10ac54d31076b08c0dab9bce",
    "param/defaults": "6fe8db805f00d96c7c4575faa1c2b6c7c2a3f6b9cd1cf8ae2c462b05fe75c230",
    "param/l23_cortex_exc": "92391389b750e4ffe68cd6a4a3c7a3d0a5b20f634b9aa2220ce6a3a9d5932a1c",
    "param/l23_cortex_inh": "85b1fd631bbc4af7d73599ed529048379d0d204e145343e0457f078e1c0d0de6",
    "param/l23_exc_rec": "263b58f203fd7f7169763bbdf635c62527a754d9940cacff0962f38b634c2d31",
    "param/l23_inh_rec": "a56b4a19ae553e9a809f04598971e6c899e4128237d0b942a675e697cfa84d95",
    "param/l4_cortex_exc": "12497632889d6d646e96a8fdd3c542b34db8f69dba2edaa51d038578ac95959e",
    "param/l4_cortex_inh": "bddb86fa833f3c34b9fede96cad7514f0540ca50b8adad75ac7b03a0325f1403",
    "param/l4_exc_rec": "263b58f203fd7f7169763bbdf635c62527a754d9940cacff0962f38b634c2d31",
    "param/l4_inh_rec": "a56b4a19ae553e9a809f04598971e6c899e4128237d0b942a675e697cfa84d95",
    "param/lgn_rec": "31328ff4b8603b66bfd0e736b403fd7840b4ab1dfe67d74c0c68c056a034b721",
    "param_spont/SpatioTemporalFilterRetinaLGN_defaults": "979212f6035e9abb9e41b4ae6a88b4f0327cfd5bdc58a5b49f958c125d6f8665",
    "param_spont/defaults": "5947743f5129eff48a1e6aab39f14cb0b2002e185b6051078bc4fc93a1b5372a",
    "param_spont/l23_cortex_exc": "6b08839188b6c818ae113f0346d3805a00d48cac583a31db75feb6e3214a7a74",
    "param_spont/l23_cortex_inh": "000570aa1be671c5941645ba8ed777c170619b7ae925d87cfcb8a331a8b2223d",
    "param_spont/l23_exc_rec": "a8fb8ac6261e77855397869c72928006a4b9be231f990c5bc24ec6fc4b546f07",
    "param_spont/l23_inh_rec": "a8fb8ac6261e77855397869c72928006a4b9be231f990c5bc24ec6fc4b546f07",
    "param_spont/l4_cortex_exc": "0efd315e9226e5121182ea6bf3155188ee874aebc17d1860f47350b0a55004af",
    "param_spont/l4_cortex_inh": "caf6a8046e186ce2612ef8225a4d9ee9be62d23c77f9ce59ea37f25f7300c983",
    "param_spont/l4_exc_rec": "201262fce8e7e35fc1b7dab7d1fa7136a3e40345770fb599aa945ffb7d19e39f",
    "param_spont/l4_inh_rec": "201262fce8e7e35fc1b7dab7d1fa7136a3e40345770fb599aa945ffb7d19e39f",
    "param_spont/lgn_rec": "31328ff4b8603b66bfd0e736b403fd7840b4ab1dfe67d74c0c68c056a034b721",
    "parameter_search_analysis.py": "3570e9ef1f9b581cd8ce7c6bae21b52087b60cc00bc65bd99a9305b24e59a9ef",
    "run.py": "50de45497c6d784066aaad66844769b68d11afb76ca0abf2df533e2ab6cd6172",
    "run_analysis.py": "cad6ba25ee4d9ed66807d09f642d19574d4b97a16089362eafbab7c5f3d30b0c",
    "run_parameter_search.py": "bdf775ece1ca5696fbf9a6c27c9df135eed31dc41b71593a30482ed1f068e2c9",
    "run_spont.py": "a7f59fbab3ce2ff0e56082dad9ae01367d9189fea1528405112a98825bc7d8ff",
    "run_stc.py": "697d0515a62b2be5c8d4ee1d62a186ded236e6fdbc62294d40e5d3ca6a4a7b15",
    "visualization_functions.py": "1f1bc47d44629c1a9ff81a7a59771217799305fe2cd33ff044c75db4c74eaabd"
  },
  "supersedes": {
    "protocol": "protocol-006",
    "relationship": "Versioned follow-up. Protocol-006 is unchanged and was not executed.",
    "changes": [
      "Recording recipe: released param_spont/defaults instead of param/defaults (recipe-check-009).",
      "Execution: 16 virtual processes as 1 MPI rank x 16 OpenMP threads with passive waiting, instead of 16 ranks x 1 thread (feasibility-010).",
      "Host: the existing 4-vCPU, ~7.8 GiB research host with a 64 GiB swap file, instead of a proposed 128 GiB rental; host-specific gates.",
      "Timeout: 72 h with a 120 s grace period, instead of 3 h with 60 s.",
      "Wrapper: adds NEST timer receipts after each PyNN run step and NEST final status; the model, workflow and seeds are unchanged.",
      "Analysis: same definitions, tolerances and comparisons; recordings exported as NumPy arrays instead of JSON lists to fit memory.",
      "Declared secondary analysis: spike-based measures on the protocol-006 sample of neurons that param_spont also records."
    ]
  },
  "decomposition": {
    "released_runs": "16 processes (model README: 16 processes on an EPYC 7302 with 128 GB RAM)",
    "declared_in_protocol_006": "16 MPI ranks x 1 thread",
    "used": "1 MPI rank x 16 threads",
    "evidence": "feasibility-010, tiny upstream model: equal virtual-process counts gave 0 of 2,208 differing spike trains; traces agreed within 2.2e-7 apart from the first recorded sample of some segments (136 of 144 signal groups held the same values assigned to different neurons; 4 voltage groups differed by up to 10 mV; cause not investigated); a 2-versus-16 control differed in 2,205 of 2,208. Full-network sheets: neuron positions and recorder selections under 1 x 16 equal recipe-check-009 under 16 x 1.",
    "limit": "Not verified bitwise at full scale; results are those of the 1 x 16 split."
  }
}
