The effects of distance and age on the accuracy of estimating perpetrator gender, age, height, and weight by eyewitnesses

Abstract

ABSTRACT Descriptions of perpetrators given by eyewitnesses are important in criminal cases, but the accuracy of eyewitnesses is often low. Research suggests that increased distance lowers accuracy of some descriptions and children and older adults tend to be the least accurate. To investigate the effects of distance and age on descriptive accuracy simultaneously, we presented 1588 participants with four separate live targets at distances between 5 and 110 m. After each viewing, they (numerically) estimated the target’s gender, age, height, and weight. We investigated high accuracy (±2 units) and serious errors (±10 units) of age, height, and weight estimates whereas a correct/incorrect gender estimate was categorized as high accuracy/serious error. We found that the likelihood of high accuracy at 5-meters was 95–99% for gender, 24–53% for age, 13–38% for height, 11–30% for weight, and at 110-meters it was 86–97% for gender, 15–23% for age, 13–25% for height, 12–28% for weight. The likelihood of serious errors at 5-meters was 1–5% for gender, 1–14% for age, 3–43% for height, 10–54% for weight, and at 110-meters it was 3–14% for gender, 11–38% for age, 11–53% for height, 17–56% for weight. Increased distances decreased accuracy and especially young children performed overall worse than adults.