Fixation, the difficulty in adapting cognitive processes to meet task changes, has received little attention in sequence learning literature. This is surprising given the degree to which humans must continuously adapt to changing sequential structures. In this experiment, we investigated whether learning a sequence was impeded by preceding practice with sequences involving either a random, simple or complex structure. Learning of a novel complex sequence in a serial reaction time (SRT) task was compared to a control group which did not receive previous practice. Across 12 SRT blocks, those who previously practiced complex and random sequences demonstrated less overall RT improvement than the control group. To distinguish sequence-dependent learning from general practice effects, RT performance was compared between block 12 and block 13, where a random sequence was introduced. Previous practice with random and complex sequences was detrimental for sequence-dependent learning as compared to the control group while it was benefitted by previous practice with the simple sequence. These findings indicate that fixation can result from previous exposure to randomized or complex sequence structures. Prior practice with a random sequence encourages fixation on stimulus-oriented planning processes which subsequently precludes use of introduced sequence information. Prior practice with a sequence structure promotes response-oriented planning and sequence complexity determines the extent to which this planning becomes sequence-specific. Following simple sequences, response-oriented planning can be readily adapted to the acquisition of more elaborate sequence structures. However, complex sequences require increased sequence specificity and so response-oriented planning processes become more fixated to learned sequence structures and consequently there is less flexibility to acquire a novel complex sequence.