Abstract
The cross-race effect (CRE) in eyewitness identification represents a persistent challenge for the criminal justice system, frequently leading to wrongful convictions. While behavioral disparities in cross-racial identification are well-documented, the underlying cognitive mechanisms, particularly under high-stress conditions, remain poorly understood. This study utilized advanced eye-tracking technology to investigate the cognitive load and gaze dynamics of eyewitnesses during simulated lineup tasks. A sample of 120 participants (evenly distributed between Black and White individuals) viewed a simulated crime video under either high-stress or low-stress conditions, followed by an 8-person photo lineup. Eye-tracking metrics, including fixation duration, fixation count, and task-evoked pupillary response (TEPR) as an index of cognitive load, were continuously recorded. Behavioral results confirmed a significant CRE, which was severely exacerbated by the high-stress manipulation. Eye-tracking analysis revealed that cross-racial trials yielded significantly greater pupil dilation and longer, more fragmented fixation durations on non-diagnostic facial features, indicating elevated cognitive load and inefficient visual scanning. These findings suggest that the cognitive demands of processing unfamiliar out-group racial features, combined with stress-induced working memory depletion, severely compromise the reliability of eyewitness decisions. This research provides objective, physiological evidence of the cognitive strain inherent in cross-racial identifications, offering critical insights for forensic profiling, judicial instructions, and eyewitness expert testimony.
