Emotional Fluency and the Discrepancy-Attribution Hypothesis

The purpose of this experiment was to investigate emotional fluency within the framework of the discrepancy-attribution hypothesis. A mismatch between expected and actual fluency results in a feeling of surprisingly good fit, which can lead people to misattribute fluency gained from stimulus manipulations to pastness. Experiment 1 replicated Whittlesea s (2002) finding that words completing high constraint sentence stems were judged to have been studied in a previous phase more often than words completing low constraint sentence stems, regardless of whether the words were actually studied or not. Experiment 1 also provided support for the emotional-fluency hypothesis; participants judged emotional words to be old more often than neutral words. In Experiment 2, participants judged whether they actually remembered target words or whether these words merely felt familiar. Participants claimed to remember emotional words more than neutral words, but there was no difference between emotional and neutral words in participants familiar responses. This finding suggests that fluency from emotional materials does not merely lead to a vague feeling of familiarity but rather to a strong belief that the item is actually remembered