DstarM

DstarM estimates diffusion-model parameters for two-choice reaction time (RT) data while recovering nondecision time distributions without pre-specifying their form.

Key Features:

• D∗M method: Implements the D∗M method to estimate diffusion-model parameters without pre-specifying the distribution of nondecision times.
• Diffusion-model parameter estimation: Recovers decision-related parameters of two-choice diffusion models from RT data.
• Nonparametric nondecision-time estimation: Provides a second-step nonparametric estimate of the nondecision time distribution, capturing perceptual and motor processes.
• Bias reduction: Addresses bias introduced by the common assumption of uniformly distributed nondecision times.
• R package implementation: Available as an R package for computational analysis of two-choice RT data.
• Validation: Validated through extensive simulation studies demonstrating accurate retrieval of decision parameters and nondecision time distributions.

Scientific Applications:

• Cognitive psychology: Quantifies decision-making dynamics in two-choice tasks by separating decision and nondecision components.
• Neuroscience: Dissects perceptual and motor contributions to reaction time for studies of neural mechanisms of choice.
• Choice reaction task analysis: Improves parameter inference for experimental and modeling studies of two-choice RT tasks.

Methodology:

Uses the D∗M method to estimate diffusion-model parameters without specifying the nondecision time distribution, followed by a nonparametric estimation of the nondecision time distribution; validation employed extensive simulation studies.