Face identification proficiency test designed using item response theory

Area of Science:

• Cognitive Psychology
• Forensic Science
• Psychometrics

Background:

• Accurate face identification is crucial in applied settings, particularly for forensic examiners.
• Existing proficiency tests have limitations, such as static item sets preventing repeated valid administration.
• Developing reliable and adaptable face identification proficiency measures is essential.

Purpose of the Study:

• To introduce and evaluate the Triad Identity Matching (TIM) test for measuring face identification proficiency.
• To utilize item response theory (IRT) for calibrating test items and assessing participant ability.
• To establish a flexible framework for proficiency testing across diverse ability levels.

Main Methods:

• Development of the Triad Identity Matching (TIM) test using face-image triads.
• Application of item response theory (IRT) to analyze test item difficulty and participant performance.
• Validation through simulations and correlation with existing face-recognition tests.

Main Results:

• The TIM test demonstrated a wide range of accuracy and spanned various difficulty levels, as shown by IRT modeling.
• IRT-based item metrics allowed for the creation of reliable proficiency subsets.
• The IRT model effectively evaluated participant ability across different sessions and populations.

Conclusions:

• The TIM test provides a valid, reliable, and flexible tool for assessing face identification proficiency.
• The developed IRT model enables accurate ability estimation and can be adapted for various user groups.
• This framework supports the development of standardized proficiency assessments for face identification specialists and individuals with face-processing challenges.