Response surface methodology was used to optimize the processing parameters of the fiber extraction from sorghum bran. The studied independent factors were ethanol/bran ratio, time, temperature, and number of treatment cycles. A three-level four-variable Box-Behnken design (BBD) was used to establish the optimum conditions of extraction. The results showed that the experimental data could be fitted to a second-order polynomial equation using multiple regression analysis and the model was highly significant (P<0.0001). The optimum extraction conditions were 11.8 mL·g−1 of ethanol/bran ratio, 60 min, 65°C, and 7 extraction cycles. The experimental yield was 35.52% which is close to the value predicted by the BBD model (34.36%). By applying the two combinations of factors generated by the path of steepest descent, the first combination (12 mL/g, 60 min, 65°C, and 8 cycles) allowed a yield of 35.50%, while the second (11 mL/g, 70 min, 55°C, and 8 cycles) exhibited a yield of 39.90% which is higher than that from the BBD model (P<0.05). Compared to the first combination generated by the path of steepest descent, the BBD model conditions were economical with small number of cycles and low ethanol/bran ratio.