evidence-accumulation-selector

To choose between DDM and LBA models for choice response-time data, evaluate your experimental design and research objectives. The evidence accumulation selector uses a decision tree to match your data structure to the appropriate model for valid parameter estimation.

Use EZ-diffusion instead of the full DDM for cognitive modeling when working with limited trials per condition. It serves as a statistically valid alternative for two-choice tasks where the full DDM parameter estimation might fail due to insufficient trial counts.

Evidence accumulation modeling requires specific trial count thresholds to ensure valid parameter estimation. The selector enforces these trial count limits and distributional assumptions to prevent misapplying models like DDM or LBA to inadequate datasets.

Interpreting drift rates and boundary separation in response-time models involves decomposing latent cognitive processes. Drift rate indicates information processing speed, while boundary separation reflects response caution, guided by expert parameter interpretation logic.

For experimental designs with more than two response alternatives, race models are typically recommended over standard DDM. The model selection logic evaluates your specific choice-RT task structure to determine if a race model is necessary.

Yes, research literacy is required to select the correct cognitive model for your dataset. Understanding foundational experimental design principles and statistical assumptions ensures you can properly navigate the decision tree and interpret the resulting model parameters.