Two sources of bias explain errors in facial age estimation

Area of Science:

• Cognitive psychology
• Social psychology
• Forensic science

Background:

• Accurate age estimation is crucial for social interactions, age-restricted services, and law enforcement.
• Previous research indicates age judgments are error-prone, but the underlying mechanisms remain unclear.

Purpose of the Study:

• To systematically investigate biases in human age estimation.
• To identify and quantify the sources of error in age perception.
• To understand the influence of recent and normative face exposure on age judgments.

Main Methods:

• Utilized a large database of 3948 standardized passport images spanning diverse ages.
• Conducted three experiments with 84 participants across various age groups.
• Analyzed estimation errors to identify systematic biases in age perception.

Main Results:

• Average age estimation error was approximately 8 years.
• Identified two primary biases: assimilative serial dependency (bias towards preceding face) and general age stereotypes (younger faces appear older, older faces appear younger).
• These biases collectively explained approximately 95% of the variance in age estimates.

Conclusions:

• Age perception is significantly influenced by top-down factors, including recent perceptual experience and established normative representations of faces.
• Understanding these biases is key to improving the accuracy of age estimation in critical real-world contexts.
• Findings have implications for face perception research and practical applications requiring accurate age assessment.