Proceedings of the Annual Meeting of the Cognitive Science Society

Decades of feedback research have suggested that feedback is more effective in correcting errors than confirming the right responses. A study conducted by Eskreis-Winkler and Fishbach (2019) challenged this notion by showing that people learn less from feedback that indicates their answer is incorrect (failure feedback) than feedback that indicates their answer is correct (success feedback) even after incentivizing learning, manipulating response correctness, and controlling for background knowledge and mental inferences required for learning across conditions. Across two randomized experiments, we extended this work to investigate whether changing the focus of feedback from the self (“You answered this question correct/incorrect!”) to the task (“The answer was correct/incorrect!”) would reduce the difference between success and failure feedback. We replicated the previous study’s main finding that people learn less from failure feedback than success feedback. However, the focus of feedback message (task vs self) did not have the hypothesized effect. We suggest future research further investigate the impact of feedback focus using in-person experimental settings with more powerful designs and we recommend a set of motivational factors to investigate to determine how learning from failure feedback can be optimized. Keywords: learning; education; feedback; motivation; ego threat; replication