“…We also excluded studies that reported only team- or organizational-level data because we were concerned with rewards' effects on individual performance. Considering our definition of creativity as involving original and useful production, we excluded studies or study conditions that (a) used self-reported measures of creativity because their validity as measures of performance is not well established (Kurtzberg, 2005); (b) used measures of creative performance that were not consistent with our definition of creativity as involving novelty, such as tests of convergent thinking (e.g., the Remote Association Test) and insight problems (e.g., the Maier coat rack problem or the Duncker candle task) because these measures have close-ended, single solutions; (c) used measures of creativity such as tests of closure (e.g., the Gestalt Completion Test), projective tests of creativity (e.g., the House–Tree–Person test), and word association tests because these measures do not involve any consideration of the aesthetic or practical value of the responses; (d) provided behavioral modeling or creativity training because, as pointed out by an anonymous reviewer, these studies may provide information that renders tasks noncreative (i.e., Cepe-Thomas, 1982; Glover, 1973; Glover & Sautter, 1977; Mendelson, 1973; and one study condition in Hennessey et al, 1989, and Gallman, 1973); or (e) gave highly descriptive feedback (e.g., specific behavioral information) to participants on their prior or current performance on the same type of task they were evaluated on or one that provided instructions that may have simplified the task (e.g., specific information about how creativity is evaluated), because, as pointed out by an anonymous reviewer, such information may simplify the task and render the output noncreative (i.e., Chambers, Goldman, & Kovesdy, 1977; Fallon & Goetz, 1975; Glover, 1973, 1979; Goetz, 1981; Kratochwill, Rush, & Kratochwill, 1979; Lane, Friedman, Goetz, & Pinkston, 1982; Rosen, 1979). Lastly, we excluded studies that did not report enough information to calculate an effect size ( k = 16).…”