Operant-conditioning boxes are widely used in animal training, allowing researchers to shape specific behaviors through reinforcements and/or punishments. Commercially available devices are expensive and run with proprietary software and hardware, hampering adaptations for the specific needs of an experiment. Therefore, many low-cost and open-source devices have recently been developed, but there are still few options for studying auditory behaviors. To overcome this problem, we developed a device based on a computer and an Arduino Mega 2560 board, named OBAT (Operant Box for Auditory Tasks), designed to present two different auditory stimuli to small primates. It has three modules: sound delivery, response bars, and reward system. We estimate that OBAT is at least 4-10 times cheaper than commercially available operant-conditioning boxes. Data from a behavioral pilot test ensured that the device can be used to train a marmoset in an auditory discrimination task. In addition, despite its low cost, accuracy tests showed that the OBAT operates with a high temporal precision. All schematics and software source code are available so that other groups can easily replicate the experiment or adapt the device to their own needs.
Keywords: Arduino; Auditory processing; Discriminative learning; Operant-conditioning box.