Midazolam induces amnesia in a simple, one-trial, maze-learning task in young chicks

We report a simple, one-trial, learning paradigm which we have developed for use in young chicks. Chicks were separated from their brood mates and placed in a small isolation chamber. A 'T' corridor, or maze, connected the isolation chamber to the brood space, allowing the chick to escape isolation stress and rejoin the brood. When the chick successfully negotiated the corridor, the latency to perform this task was recorded. On a subsequent trial, any improvement in the speed of performance was reasoned to reflect the chick's memory of the task. Undrugged chicks always showed significant improvement in task latency if they were replaced in the maze 3 hours after a successful escape, suggesting that they had remembered the task. Chicks given midazolam (0.1 or 0.3 mg/kg, IP), a benzodiazepine, before the first escape, showed no improvement on their second escape. Improved performance was seen, however, if a second injection of midazolam was given before the second escape, suggesting a state-dependent effect.