pyWitness

pyWitness analyzes recognition memory and eyewitness identification data, providing tools for ROC and confidence accuracy characteristic (CAC) analysis, signal-detection-based model fitting, simulated data generation, and power analysis.

Key Features:

• ROC and CAC analysis: Performs receiver operating characteristic (ROC) and confidence accuracy characteristic (CAC) analyses for recognition memory experiments.
• Signal-detection-based model fitting: Fits signal-detection models to eyewitness identification and recognition memory data.
• Statistical comparisons: Enables statistical comparisons of model fits and empirical measures.
• Simulated data generation: Generates simulated datasets for hypothesis testing and model validation.
• Power analysis: Conducts power analyses to estimate required sample sizes for experimental designs.
• R interoperability (reticulate): Integrates with the R environment via the reticulate package to enable combined Python–R workflows.
• Open-science reproducibility: Documents algorithms, fits, and methods to support reproducible analyses.

Scientific Applications:

• Eyewitness identification in legal contexts: Analyzes identification performance and decision metrics relevant to forensic and legal settings.
• Memory accuracy and confidence judgments: Evaluates relationships between recognition accuracy and confidence in eyewitness reports.
• Experimental design and hypothesis testing: Uses simulation and power analysis to inform sample size and test theoretical predictions.
• Recognition memory research in cognitive psychology and forensic science: Supports empirical and modeling studies of recognition memory processes applied to forensic questions.

Methodology:

Performs ROC and CAC analyses, statistical comparisons, signal-detection-based model fitting, simulated data generation, and power analyses.