Assessing and improving military personnels metacognitive ability
Abstract
The goal of the current proposal is to develop tools to both assess and improve people metacognitive ability. Preliminary data suggest that existing techniques for measuring metacognitive ability are deficient in that they depend on the difficulty of the task or overall level of confidence. The majority of the measures of metacognitive ability increase trivially as the task becomes easier, which makes it difficult to separate improvements in metacognition from improvements in task performance. Even more problematic, most measures are biased in such away that individuals with a tendency to give high confidence would be assessed to have higher levels of metacognition, regardless of whether their ability is actually superior.We have recently developed an improved, empirically-validated measure of metacognitive ability based on a formal model of confidence generation. To test the formal model, Aim 1 will organize a very large database of existing confidence studies. Researchers will be invited to share their data for the purposes of testing measures of metacognition. So far, over 145 datasets with a total of almost 4 million trials have already been assembled. Aim 2 will use the data from this database to corroborate the new measure of metacognition. Finally, using the newly established measure of metacognition, Aim 3 will focus on the development of tools to improve metacognitive ability.The current proposal could serve as the basis for the development of both a standardized assessment of the metacognitive abilities of new Navy recruits and an intervention for improving the metacognitive ability for current Navy personnel. Such assessment and training will significantly improve decision-making capabilities in both human-human and human-machineteams. Indeed, these interventions will ensure that Navy personnel are able to both properly form and interpret confidence ratings communicated by others and thus enable effective decision making by both humans and machines
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 11, 2020
- Source ID
- N000142012622
Entities
People
- Dobromir Rahnev
Organizations
- Georgia Tech Research Corporation
- Office of Naval Research
- United States Navy