Learn HSSM

Use this section for guided material that builds understanding through complete examples. If you are new to HSSM, follow the first path in order; the remaining paths assume that you can already fit and inspect a basic model.

Start with a complete workflow

1. Install HSSM.
2. Fit and check a first DDM.
3. Fit and check one useful regression.
4. Add hierarchical structure.
5. See one analysis from data to interpretation.

The optional scenic route preserves the comprehensive tour of model families, priors, regressions, hierarchies, comparison, custom likelihoods, and low-level PyMC. It is a reference journey, not a required step in the first-user path.

Learn a model family

• Hierarchical DDM regressions connect an experimental design to parameter formulas.
• Choice-only models cover decisions without response times.
• Poisson race models introduce a multi-accumulator model.
• HMMs with DDM emissions model regime changes across trials.
• Attentional DDMs connect fixation covariates to evidence accumulation.

Learn reinforcement-learning models

Start with RLSSM basics, then choose the route that matches your data:

• choice-only RLSSMs;
• custom RLSSMs with ssms.rl;
• restless learners; or
• registering a custom RLSSM.

Understand modeling choices

• Coming from HDDM maps familiar HDDM concepts to HSSM.
• Likelihood kinds in HSSM explains the available likelihood kinds.
• Centered and non-centered parameterizations establish the basic trade-off.
• Random-slope prior diagnostics show how parameterization decisions affect real models.

Ready to solve a specific problem? Continue to the how-to guide index. For exact signatures and supported helpers, use the reference index.