Tuesday, March 22, 2022
Friday, March 18, 2022
Using quantitative and qualitative measures, we found a clear and reproducible pattern: Negative information had a much stronger effect on people’s attention, information processing, and behavior, consistent with the negativity bias found in other domains (Baumeister et al. 2001, Rozin and Royzman 2001). Qualitative comments accompanying the evaluators’ decisions to adjust their scores suggest that, as a result of exposures to critical information, evaluators devoted greater attention to evaluation-criteria-specific tasks, such as scrutinizing the proposal for critiques and weaknesses. In contrast, exposures to neutral and higher scores led to shorter comments, with a greater focus on non-criteria-specific aspects of evaluation, such as confidence in their judgment or achieving consistency with the other reviewers, which did not prompt additional information processing of the evaluation task at hand. Thus, provided with the opportunity to deliberate and influence each other, evaluators are more likely to focus on proposal weaknesses than strengths. This asymmetry suggests that reviewers are more concerned with false positives (i.e., type I errors) than false negatives (i.e., type II errors), as the furnishing of negative information weighs more heavily on reviewers’ decisions than positive information of comparative magnitude. This finding may help explain what many see as “conservatism bias” in funding novel projects, which has conjured slogans such as “conform and be funded” and “bias against novelty” (Nicholson and Ioannidis 2012, Boudreau et al. 2016). If the risk of proposals is associated with their weaknesses, then, relative to independent evaluations, postsharing evaluations favor more conservative projects. These decisions, in turn, directly shape the disruptiveness of innovation occurring at the knowledge frontier.