People have incredible faith in their visual experience, even in the often-uncertain context of legal decisions. Jurors find video evidence uniquely credible and persuasive, yet little is known about whether the manner in which people watch such materials affects their decisions. In my dissertation, “More than meets the eye: Building a visual attention-based model of intergroup legal punishment,” I proposed a model of legal punishment which uses visual attention to better understand when out-group bias is likely to emerge in legal judgments. Across 6 studies, I found that group identifications interact with visual attention to predict punishment for intergroup altercations.
Across the studies, I measured social group identification with extant groups (i.e. police, racial groups) or novel groups. I presented participants with videotaped intergroup altercations featuring an in-group and out-group actor engaged in physical conflict. Participants’ visual attention to the actors was measured, using covert eye-tracking, or manipulated via instructions. Finally, participants reported on the facts of the scene and assigned legal punishment to the actors.
Results demonstrated that the more that people selectively attended to an out-group actor, the greater the discrepancy in decisions among differently identified individuals. Only among individuals who attended frequently to the out-group actor did the expected pattern of out-group bias in legal punishment emerge. Further, selective attention polarized punishment by shifting the way people interpreted or recalled the facts of the scene. These discrepancies persisted even when participants were given an additional opportunity to watch the evidence. Indeed, patterns of visual attention persisted across viewings, suggesting a form of visual confirmation bias in out-group punishment.
Finally, this bias was attenuated by instructions to watch evidence relationally, equally attending to all aspects of the scene. White participants trained to watch a cross-race fight relationally assigned the Black actor significantly less jail and probation time than participants who watched naturally. This effect was qualified by a significant interaction between attention instructions and group identification: instructions to watch the scene relationally particularly reduced the punishment of participants strongly identified with their White in-group—those most likely to evince out-group bias in the control condition.
Together, these studies elaborate the process by which people can come to different conclusions about the same information. This work extends the literature on out-group bias, offering a comprehensive model of biased judgment that stems from the earliest stages of information processing. In the context of a legal system where minority offenders are disproportionately likely to face harsher punishment, understanding when such disparities emerge is the key to combatting them. Indeed, jury instructions to contain one’s racial biases have proven ineffective at eliminating disparity; my results suggest that an attention-based intervention could be a subtler, more effective way to achieve the same goal. Finally, these studies have implications for the role of video in the courtroom. The rules of evidence have not kept pace with evolving technology. By integrating established psychological knowledge of social and attentional processes with the needs and realities of the courtroom, we can more effectively combat bias and ensure justice.