Reducing objectivity bias from visual evidence with a ‘consider-the-opposite’ bias warning

Video is often regarded as a complete, objective account of events, and therefore we rarely question its validity. However, our interpretation of video can be biased by many factors. Therefore, we investigated whether we could reduce the overestimating of the objectivity of visual evidence used in police cases. To do that, we showed participants either camera footage of a possible crime, or a textual description of this video. Half of the participants were given a bias warning which asked them to explicitly generate an alternative explanation to their first impression of the evidence (‘consider-the-opposite’ strategy); the other half did not. We compared these four groups on how likely they thought it was that the suspects were guilty, their verdict (guilty/not guilty), and the confidence they had in their judgment. We expected that participants who saw the video would judge the probability of guilt of the suspects as higher than people who read the text, but found the opposite....Show moreVideo is often regarded as a complete, objective account of events, and therefore we rarely question its validity. However, our interpretation of video can be biased by many factors. Therefore, we investigated whether we could reduce the overestimating of the objectivity of visual evidence used in police cases. To do that, we showed participants either camera footage of a possible crime, or a textual description of this video. Half of the participants were given a bias warning which asked them to explicitly generate an alternative explanation to their first impression of the evidence (‘consider-the-opposite’ strategy); the other half did not. We compared these four groups on how likely they thought it was that the suspects were guilty, their verdict (guilty/not guilty), and the confidence they had in their judgment. We expected that participants who saw the video would judge the probability of guilt of the suspects as higher than people who read the text, but found the opposite. This could be due to certain aspects of the evidence being more salient in visual or textual form. We explored this using qualitative analysis of participants’ first impressions and alternative explanations of the evidence. Contrary to our expectations, the bias warning did not effectively reduce this difference in guilt judgments between people who saw the video and people who read the text. We did not find that individual differences moderated the effectiveness of the warning, but differences in perspective taking and the perceived objectivity of video did predict the estimated likelihood of guilt. We also did not find that confidence in judgment differed between the four groups. We discuss implications of these findings and give recommendations for future research to further explore how we regard certain visual and textual elements in evidence, and how individual differences play a role in our interpretation of evidence.Show less

Leiden University Student Repository

Documents

In Collections

Reducing objectivity bias from visual evidence with a ‘consider-the-opposite’ bias warning