Although the relationships between extraversion and positive affect and between neuroticism and negative affect are powerful, the patterns of causality accounting for them are unknown. We employed an experimental methodology to manipulate state extraversion and state neuroticism to determine their causal status. In Experiment 1, state extraversion was manipulated by instructing participants to act extraverted and introverted during two different discussions with two other participants. Participants reported more positive affect when instructed to act extraverted than when instructed to act introverted, and this finding was supported by the ratings of observers. In Experiment 2, state neuroticism was manipulated by instructing participants to act neurotic and emotionally stable during two different discussions. Participants reported more negative affect when instructed to act neurotic than when instructed to act stable. Thus, at least some part of the extraversion-positive affect and neuroticism-negative affect relationships may be accounted for by the causal influences of extraverted behavior and neurotic behavior, respectively.