Reinforced Correct Answers to Next Stage Problems Produced the Highest Stage Performance in Traditional Nonliterates Found in the World

Thirty-nine nonliterate Nepalese adults were given 2 stage-based isolation-of-variables instruments: the thatched roof problem and laundry problem. The thatched roof problem was very similar to the laundry instrument, just differing in content. The thatched roof task was used as the training instrument and the laundry instrument was used as the transfer task instrument. The participants practiced on the thatched roof instrument. With the transfer task instrument, correct answers were reinforced with money. From the beginning of the measured stage of performance in the transfer instrument, the M stage 9.77 (SD 1.48) increased to M stage 10.72 (SD 1.45) at the end of the transfer task training and testing, t(38) 16.7013, p .00000. This is roughly 1 stage from pretest to posttest. Also, the frequency of people performing at the lower stages (Primary Stage 8 and Concrete Stage 9) decreased at posttest. The frequency of people at the higher stages (Abstract Stage 10 to Metasystematic Stage 13) increased at posttest. This showed that training with reinforcement had a positive effect on increasing stage performance from pretest to posttest. This finding strongly suggests that all testing should include repeated presentation of very similar items and that reinforcement needs to be used for correct answers. Otherwise there is the risk of underestimating what tasks people can successfully complete and what their stage of performance is.