Shaping and Chaining are both effective in training new behavior. Shaping, or the reinforcement of successive approximations to a target response, is used to develop novel responses or reestablish responses no longer occurring. Chaining is used to develop fixed sequences of responses. The combination of the two procedures was examined using twelve participants (8 men and 4 women) who were randomly assigned to one of three chaining comparisons in an individual organism design: Forward Chaining (FC) versus Backward Chaining (BC), FC versus Total Task Chaining (TTC), or BC versus TTC. A video tracking system recorded the coordinate position of a participant's hand in three-dimensional space, relative to the position of three targets randomly selected by the computer. Contingent upon contact with a shaping sphere around the current target in the chain component being trained, reinforcement in the form of a computer-emitted tone occurred. Upon completion of a training component, a different to e and points were presented as reinforcement. Reinforcement amount was equated between FC, BC, and TTC procedures, and a test phase requiring 5 repetitions of the sequence to demonstrate that learning took place. The following was found: (a) in the FC-BC comparison BC was generally more effective than FC, (b) in the FC - TTC comparison FC was generally more effective than TTC, but (c) in the BC - TTC comparison there were fewer differences between BC and TTC. The outcomes for the FC versus BC and the BC versus TTC comparisons are in contrast to the literature, and require further investigation. The automated chaining procedures used in this study show promise as a method to be used in a rehabilitative setting.