A plethora of laboratory studies have shown that including the study of worked examples during problem-solving practice improves learning (Sweller, 1999; Sweller & Cooper, 1985). While most worked-example research focuses on the use of correct examples, recent work suggests that asking children to explain a combination of correct and incorrect examples can be even more effective (e.g., Siegler & Chen, 2008). The combination of correct and incorrect examples has been shown to lead to improvements in both conceptual understanding and procedural skill in Algebra compared with procedural practice alone (Booth et al., in revision). It seems clear that both types of support are necessary, but what if extra support for knowledge construction is achieved through other types of innovative classroom practice? In that case, would it still be optimal to provide a combination of correct and incorrect examples, or would providing incorrect examples alone suffice for improving student learning? In the present study, the researchers test the contribution of correct vs. incorrect examples in the context of such support for knowledge construction--guided problem-solving practice with the Cognitive Tutor, a self-paced intelligent tutor system which provides students with feedback and hints as they practice (Koedinger, Anderson, Hadley, & Mark, 1997). Sixty-four eighth-grade Algebra I students (29 females, 35 males) at a mid-western middle school participated. This study was the first to test whether a combination of correct and incorrect examples is more beneficial than incorrect examples alone. The study's findings suggest that receiving incorrect examples can be beneficial regardless of whether it is paired with correct examples. This finding is especially important to note because when examples are used in classrooms and in textbooks, they are most frequently correctly solved examples. One figure is appended.