Shape memory alloy (SMA) actuators exhibit considerable hysteresis between the supply voltage (conventionally used in resistive heating) and strain characteristics of the SMA. Hence, it is not easy to control the strain of a thin-SMA wire, unless a model is developed that can match the actuator's nonlinearities for predicting the supply voltage required by the SMA system accurately. The work presented in this paper proposes the use of a black-box technique called the adaptive neurofuzzy inference system (ANFIS) to study the hysteretic behavior of SMAs. The input parameters for such an ANFIS model would be a physical variable at time t and at a time t + n, where n is a time shift. The present work studies the effect of a time shift on the actuator nonlinearities for two ANFIS models. One of the models studies the relationship between the desired displacement of an SMA and the supply voltage across the SMA, while the other model predicts the actual displacement of an SMA from the feedback temperature. A novel SMA–Constantan thermocouple records the feedback temperature.