Feeling guilty: little effect on false confession rate

Abstract

ABSTRACT In two experiments, we tested the hypothesis that guilt feelings would elevate the probability of making a false confession. In Experiment 1 (N = 146), a confederate induced guilt feelings by asking participants to cheat on a task. The experimenter then falsely accused participants of having pressed a forbidden key, causing a computer crash. In Experiment 2 (N = 108), a confederate was punished every time participants could not answer a quiz question. The confederate later cheated in a game and asked participants to take the blame. In Experiment 1, 100 participants (68.5%) falsely confessed to pressing the key. In Experiment 2, 39 participants (36.1%) falsely confessed to cheating. Guilt manipulations had no effect on false confession rates. When exploring the effect of guilt feelings, five of eight tests were statistically non-significant. As yet, there is insufficient evidence to argue that guilt feelings are a major determinant of false confessions.