A FIELD EXPERIMENT WAS CONDUCTED TO TEST THE RELATIONSHIP BETWEEN PREVIOUS COMMITMENTS TO A THEFT VICTIM AND A BYSTANDER'S WILLINGNESS TO STOP THE THIEF. A POCKET CALCULATOR WAS STOLEN IN FRONT OF BYSTANDERS. IN ONE CIRCUMSTANCE, THE BYSTANDER WAS PREVIOUSLY COMMITTED TO WATCH THE VICTIM'S CALCULATOR (FIRSTPARTY COMMITMENT). IN A SECOND CIRCUMSTANCE, THE BYSTANDER MADE A COMMITMENT TO THE VICTIM TO WATCH THE BELONGINGS OF ANOTHER STUDENT WHO EARLIER HAD BEEN SEATED NEARBY (SECOND-PARTY COMMITMENT). THERE WAS ALSO A THIRD, NONCOMMITMENT GROUP. A TOTAL OF 67 PERCENT OF THE SUBJECTS INTERVENED TO KEEP THE CALCULATOR FROM BEING STOLEN. THEY WERE: 100 PERCENT OF THE SUBJECTS IN THE FIRST-PARTY COMMITMENT GROUP, 60 PERCENT IN THE SECOND-PARTY COMMITMENT GROUP, AND 40 PERCENT IN THE NO-COMMITMENT GROUP. THE DIRECT, FIRST-PARTY COMMITMENTS TO THE VICTIM OF A THEFT LED TO GREATER BYSTANDER INTERVENTION. THE TWO-PARTY SITUATION WAS APPARENTLY MORE COGNITIVELY COMPLEX AND CONFUSING FOR BYSTANDERS TRYING TO UNDERSTAND THEIR RIGHTS AND RESPONSIBILITIES, THAN FIRST-PARTY COMMITMENTS, WHICH WERE LESS AMBIGUOUS AND MORE EASILY UNDERSTOOD AS REQUIRING ACTION. REFERENCES, AND TABULAR DATA ARE INCLUDED. (AUTHOR ABSTRACT MODIFIED)