Vestibular short latency responses to pulsed linear acceleration in unanesthetized animals

Linear acceleration transients were used to elicit vestibular compound action potentials in non-invasively prepared, unanesthetized animals for the first time (chicks, Gallus domesticus, n=33). Responses were composed of a series of up to 8 dominant peaks occurring within 8 msec of the stimulus. Response amplitudes for 1.0 g stimulus ranged from 1 to 10 μV. A late, slow, triphasic, anesthesia-labile component was identified as a dominant response feature in unanesthetized animals. Amplitudes increased and latencies decreased as stimulus intensity was increased (MANOVA P < 0.05). Linear regression slope ranges were: amplitudes = 1.0-5.0 μV/g; latencies = - 300 to -1100 μ sec/g. Thresholds for single polarity stimuli (0.035 ± 0.022 g, n = 11) were significanly lower than those of alternating polarity (0.074 ± 0.028 g, n = 18, P < 0.001). Bilateral labyrinthectomy eliminated responses whereas bilateral extirpation of cochleae did not significantly change response thresholds. Intense acoustic masking (100/104 dB SL) produced no effect in 2 animals, but did produce small to moderate effects on response amplitudes in 7 others. Changes were attributed to effects on vestibular end organs. Results of unilateral labyrinth blockade (tetrodotoxin) suggest that P1 and N1 preferentially reflect ipsilateral eight nerve compound action potentials whereas components beyond approximately 2 msec reflect activity from vestibular neurons that depend on both labyrinths. The results demonstrate that short latency vestibular compound action potentials can be measured in unanesthetized, non-invasively prepared animals.