A conditioning-based task comprising a modified version of an active avoidance task reported previously (Miyakawa et al., 2019) was used to monitor tinnitus-associated behaviors. A two-compartment shuttle box was separated by a barrier with a narrow open door. The shuttle box was equipped with an electrical grid floor connected to a shock generator. During the active-avoidance training period (1–2 weeks), the mice learned to associate passing through the doorway with sound. Nine sounds were chosen at random from the following stimulus set: five pure tones (4, 8, 16, 20, and 32 kHz) and four octave-band noises (4–8, 8–16, 10–20, and 16–32 kHz), played at three intensities each (40, 50, and 60 dB). Each trial began with a holding period of 5–40 s, during which a mouse stayed on one side of the two-compartment shuttle box. At the end of the holding period, a sound was played. If the mouse passed through the doorway within 8 s of sound onset, a new trial was automatically initiated. If the mouse failed to pass through the doorway within the time window, a mild foot shock (0.2 mA for 0.5 s) was delivered once every 7 s until it passed through the doorway. When a mouse achieved a 70 % success rate – successful passing through the doorway within 8 s of sound onset, it was transferred to a testing session. The average number of times that a mouse passed through the doorway during silent probe trials for 3 days represented the likelihood that the mouse was experiencing tinnitus. After the training and last testing phases, two additional testing phases were performed 10 days after the mice received unilateral noise trauma and 10 days after cortical surface stimulation via the graphene electrode array.
Member
