Date of Award

Spring 2012

Document Type


Degree Name

Doctor of Philosophy (PhD)



Committee Director

Ivan K. Ash

Committee Member

Bryan E. Porter

Committee Member

Elaine M. Justice


Recent hindsight bias research suggests that modern Cognitive Reconstruction theories that model hindsight effects as non-unitary phenomena potentially confound their findings by not differentiating between judgment tasks. This experiment tests a non-unitary approach of modeling hindsight effects that predicts confidence ratings and outcome likelihood judgments to be independent tasks, governed by differing cognitive processes and susceptible to unique patterns of hindsight bias. Predictions specify that sense-making theories accurately account for hindsight bias effects for outcome likelihood ratings and expectation based adjustment models accurately account for "I would have known that!" hindsight bias effects for confidence ratings. Utilizing a within-subjects, narrative text paradigm, the proposed non-unitary approach was tested by investigating whether the effects of outcome congruency on hindsight bias results were moderated by the type of judgment task. Participants read stories, rated their confidence in predicting the outcome or the likelihood of possible outcomes, given either expected or unexpected story outcomes, and then asked to recall their ratings. Results supported the predictions of the proposed non-unitary approach with confidence ratings and outcome likelihood judgments producing opposite patterns of hindsight bias effects. Theoretical implications, study limitations and future research directions were also discussed.