SCM model registry

The model registry exposes a catalogue of peer-reviewed structural causal models (SCMs) across scientific domains. These models encode causal structure — nodes, edges, and constraints — that power intervention validation, counterfactual tracing, and integrity checking throughout the Wu-Weism platform.

Both endpoints in this group are public. You do not need an Authorization header to read the model registry.

List all SCM models

GET /api/scm/models

Returns all publicly available SCMs in the registry.

Response fields

success

boolean

required

true when the request succeeds.

models

object[]

required

Array of model summary objects.

Show model object properties

modelKey

string

Unique identifier for the model. Use this key when calling other SCM endpoints.

version

string

Semantic version of the model (e.g., "1.0.0").

domain

string

Scientific domain the model covers. See domain values below.

description

string

Human-readable summary of the model’s scope and intended use.

nodes

string[]

Variable names modelled as nodes in the causal graph.

edges

object[]

Directed causal edges between nodes, each with from and to properties.

constraints

object

Structural constraints and assumptions encoded in the model.

Example

curl --request GET \
  --url https://wuweism.com/api/scm/models

const response = await fetch('https://wuweism.com/api/scm/models');
const { models } = await response.json();

import requests

response = requests.get('https://wuweism.com/api/scm/models')
models = response.json()['models']

{
  "success": true,
  "models": [
    {
      "modelKey": "neural_topology_v1",
      "version": "1.0.0",
      "domain": "neuroscience",
      "description": "Causal model of structural brain connectivity and its relationship to cognitive outcomes, encoding known confounders including age, stress exposure, and genetic factors.",
      "nodes": ["cortisol_level", "stress_exposure", "hippocampal_volume", "age", "neuroplasticity"],
      "edges": [
        { "from": "stress_exposure", "to": "cortisol_level" },
        { "from": "cortisol_level", "to": "hippocampal_volume" },
        { "from": "age", "to": "hippocampal_volume" }
      ],
      "constraints": {
        "acyclic": true,
        "latentConfounders": ["genetic_risk"]
      }
    },
    {
      "modelKey": "alignment_bias_scm",
      "version": "1.0.0",
      "domain": "alignment",
      "description": "Structural causal model capturing how training data composition and RLHF feedback loops causally influence model alignment and emergent biases.",
      "nodes": ["training_data_distribution", "rlhf_reward_signal", "alignment_score", "emergent_bias"],
      "edges": [
        { "from": "training_data_distribution", "to": "alignment_score" },
        { "from": "rlhf_reward_signal", "to": "alignment_score" },
        { "from": "alignment_score", "to": "emergent_bias" }
      ],
      "constraints": {
        "acyclic": true,
        "latentConfounders": []
      }
    },
    {
      "modelKey": "hot_rosenthal_v1",
      "version": "1.0.0",
      "domain": "consciousness",
      "description": "Higher-order thought (HOT) causal model based on Rosenthal's representationalist framework, encoding causal relationships between first-order states, higher-order representations, and conscious experience.",
      "nodes": ["first_order_state", "higher_order_representation", "conscious_experience", "attention"],
      "edges": [
        { "from": "first_order_state", "to": "higher_order_representation" },
        { "from": "higher_order_representation", "to": "conscious_experience" },
        { "from": "attention", "to": "higher_order_representation" }
      ],
      "constraints": {
        "acyclic": true,
        "latentConfounders": ["global_workspace_access"]
      }
    }
  ]
}

Get a specific SCM model

GET /api/scm/models/{modelKey}

Returns the full specification for a single SCM, including its complete node list, edge set, constraints, and metadata.

Path parameters

modelKey

string

required

The unique model identifier. See the table below for known model keys.

Known model keys

`modelKey`	Domain	Description
`alignment_bias_scm`	`alignment`	Training data and RLHF effects on model alignment and emergent bias
`hot_rosenthal_v1`	`consciousness`	Higher-order thought causal model (Rosenthal framework)
`neural_topology_v1`	`neuroscience`	Structural brain connectivity and cognitive outcomes
`neural_dynamics_v1`	`theoretical_neuroscience`	Dynamic causal modelling of neural population activity
`iml_epistemology_v1`	`iml`	Epistemic causal structure of interpretable machine learning explanations

Domain values

The domain field uses one of the following values: alignment · consciousness · neuroscience · theoretical_neuroscience · iml · education · legal · ecology · evolutionary_biology · cognitive_psychology · scaling_laws · physics · abstract

Response fields

success

boolean

required

true when the request succeeds.

model

object

required

Full model specification.

Hide model properties

modelKey

string

Unique model identifier.

version

string

Semantic version string.

domain

string

Scientific domain.

description

string

Human-readable description of the model.

nodes

string[]

All variable names modelled as nodes in the directed acyclic graph.

edges

object[]

Directed causal edges. Each edge has from (string) and to (string).

constraints

object

Structural constraints, including acyclic (boolean) and latentConfounders (string[]).

Example

curl --request GET \
  --url https://wuweism.com/api/scm/models/neural_topology_v1

const response = await fetch('https://wuweism.com/api/scm/models/neural_topology_v1');
const { model } = await response.json();

import requests

response = requests.get('https://wuweism.com/api/scm/models/neural_topology_v1')
model = response.json()['model']

{
  "success": true,
  "model": {
    "modelKey": "neural_topology_v1",
    "version": "1.0.0",
    "domain": "neuroscience",
    "description": "Causal model of structural brain connectivity and its relationship to cognitive outcomes, encoding known confounders including age, stress exposure, and genetic factors.",
    "nodes": [
      "cortisol_level",
      "stress_exposure",
      "hippocampal_volume",
      "age",
      "neuroplasticity",
      "genetic_risk"
    ],
    "edges": [
      { "from": "stress_exposure", "to": "cortisol_level" },
      { "from": "cortisol_level", "to": "hippocampal_volume" },
      { "from": "age", "to": "hippocampal_volume" },
      { "from": "age", "to": "cortisol_level" },
      { "from": "neuroplasticity", "to": "hippocampal_volume" }
    ],
    "constraints": {
      "acyclic": true,
      "latentConfounders": ["genetic_risk"]
    }
  }
}

{
  "success": false,
  "error": "Model not found"
}

​List all SCM models

​Response fields

​Example

​Get a specific SCM model

​Path parameters

​Known model keys

​Domain values

​Response fields

​Example

List all SCM models

Response fields

Example

Get a specific SCM model

Path parameters

Known model keys

Domain values

Response fields

Example