The primary purpose of this study was to replicate and extend a study by Gravina, Austin, Schroedter, and Loewy (2008). A similar self-monitoring procedure, with the addition of self-monitoring accuracy training, was implemented to increase the percentage of observations in which participants worked in neutral postures. The accuracy training required the three participants to practice self-monitoring with the experimenter at least 20 times and to meet the criteria of 90% accuracy for the last 10 monitors. Feedback was delivered after each monitor. Two postures for each of the three participants were targeted by the intervention, and all six improved with an average effect size of 4.7 (range: 3.4 to 8.2) compared to Gravina et al., in which effect sizes averaged 1.9 (range: -0.1 to 3.2). In addition, participant self-monitoring was more accurate overall (77%) when compared to Gravina et al. (44%). (Contains 2 tables and 1 figure.)