In experiments which require the production of binary sequences by subjects, it has been found repeatedly that the interpretation of experimental results depends critically upon the nature of nonrandom sequential patterning among successive responses. Examples from the literature are discussed, where an experiment was originally misinterpreted because inter-response patterning was not considered. This patterning has become an object of study in itself. It has been suggested (Ross & Levy, 1958) that the length of the sequence of responses is an important determinant of the nature of the interresponse patterning. Differences between subjects in the type of patterns produced have also been reported. In addition, the nature of patterning has been found to vary depending upon the instructions given and the experimental procedures used. The present study was designed to evaluate the importance of these three variables (sequence length, subjects, and instructions).