• Guillermo Macbeth
  • Eugenia Razumiejczyk
Keywords: calibration, debiasing, underconfidence bias, verbal tasks


The underconfidence bias is defined as the underestimation of the subjective success in comparison with the objective or actual success achieved by an agent in a series of tasks. An experiment that dissolves the underconfidence bias in verbal tasks by means of two simple cognitive interventions is presented. The aim of the first debiasing intervention is to train the participants in verbal tasks; the aim of the second is to train them in calibration on verbal tasks. 210 university students participated in the experiment. Five experimental hypotheses related to the underconfidence bias debiasing were tested. Subjects were randomly assigned to a control group or to one of two experimental groups. In study phase one experimental group was trained in verbal tasks and the other in calibration on verbal tasks. In test phase a verbal aptitude test was administered to the three groups, followed by a subjective estimation of success task. Results show that the second cognitive intervention is more efficient than the first to achieve the debiasing of the underconfidence bias. It is suggested, in addition, that the training in specific tasks dissolves the calibration bias when it offers the occasion to train indirectly the cognitive processes of calibration.


Download data is not yet available.

Author Biographies

Guillermo Macbeth
Universidad del Salvador Buenos Aires Argentina
Eugenia Razumiejczyk
Universidad del Salvador Buenos Aires Argentina
How to Cite
Macbeth, G., & Razumiejczyk, E. (1). DEBIASING OVERCONFIDENCE BIAS IN VERBAL TASKS. Anales De Psicología / Annals of Psychology, 24(1), 143-149. Retrieved from
Basic Psychology