The multivariate effect of task sequence, memory support, and state anxiety was investigated using a nonverbal concept acquisition task. Ninety-five Indian college students were randomly assigned to four treatment conditions resulting from the task sequences of 1) easy-to-hard and 2) hard-to-easy and from memory support versus nonmemory support. Results showed that subjects receiving the easy-to-hard sequence did significantly better on the performance test than did those receiving the hard-to-easy sequence. Also, the memory support treatment groups had significantly fewer errors on the performance test than did the nonmemory support groups. The instructional condition of easy-to-hard sequence plus memory support was the optimal treatment and provided manipulatable variables for applied usage. (Author)