TASK REVISION: A NEGLECTED FORM OF WORK PERFORMANCE.
研究提出并测量了任务修订（纠正错误程序或角色期望的行为），发现目标设定会抑制该行为，而替代方案的显著性和监督责任则促进它，对理解工作绩效有启示。
This study addressed the problem of task revision, a virtually unresearched issue in the work performance literature. We defined task revision as action taken to correct a faulty procedure, an inaccurate job description, or a role expectation that is dysfunctional for an organization. Two experiments were constructed to measure task revision and test for its determinants. Results showed that goal setting inhibited task revision: instructions to your best were superior to a specific goal. Facilitators of task revision were the salience of alternatives and being in a supervisory position with accountability pressures. We discuss the implications of these results in terms of the functions of counter-role behavior for organizations and the need to broaden the construct of work performance. Work performance has long occupied a central role in organizational research. It was the primary issue during the beginnings of industrial psychology (e.g., Munsterberg, 1913), of major concern during the human relations movement (e.g., Likert, 1961), and at center stage as empirical research blossomed in organizational behavior (e.g., Vroom, 1964). Now voluminous, the literature on work performance ranges from the extensive study of organizational and social factors that influence work behavior to the analysis of cognitive processes underlying task effort. In recent years, the most common form of performance research has translated the issue into a cognitive question. The aim has not only been to find conditions under which people will work harder, but to explain the mental processes underlying task activity. For example, early versions of need theory, which emphasized stages and growth in human desires, have gradually given way to expectancy models in which valued outcomes are a part of a cognitive calculus hypothesized to precede behavior. Likewise, reinforcement theory, with its emphasis on external incentives and behavioral learning, has been overtaken by goal-setting approaches in which salient outcomes serve as mental targets for behavior. Though we do not yet know exactly how individuals process information in performance contexts,