sdtlu: An R package for the signal detection analysis of eyewitness lineup data.

Area of Science:

• Psychology
• Forensic Science
• Statistical Modeling

Background:

• Eyewitness lineups are crucial in criminal investigations.
• Signal Detection Theory (SDT) is increasingly used to analyze lineup data.
• Existing SDT analysis methods can be complex and time-consuming.

Purpose of the Study:

• Introduce the sdtlu R package for streamlined SDT analysis of lineup data.
• Provide automated functions for processing data, estimating parameters, and calculating performance measures.
• Facilitate model-based analysis without requiring Monte Carlo simulations.

Main Methods:

• Developed the sdtlu R package in the R programming language.
• Incorporated closed-form solutions for simultaneous and sequential lineups.
• Included support for show-ups and estimation of base rates in real-world lineups.

Main Results:

• sdtlu automates the processing of lineup data and SDT parameter estimation.
• The package computes model-based performance measures like AUC and diagnosticity.
• Uncertainty intervals are determined using bootstrapping, and parameters can be compared across datasets.

Conclusions:

• The sdtlu package offers a user-friendly and efficient tool for SDT analysis of eyewitness lineup data.
• It enables robust model-based analyses, enhancing the scientific rigor of lineup research.
• sdtlu supports various lineup types and provides comprehensive graphical outputs.