Strikes and startles of northern pike (Esox lucius): a comparison of muscle activity and kinematics between S-start behaviors.

S-starts are a major class of fast-start behaviors that serve diverse locomotor functions in fishes, playing roles in both feeding strike and escape startle events. While movement patterns are similar during strike and startle, their motor control mechanisms have not been compared. To investigate heterogeneity in S-start responses and to test the hypothesis that S-starts are generated by the same patterns of muscle activity regardless of the behavioral context in which they function, we examined kinematic and muscle activity patterns of northern pike (Esox lucius) performing feeding and escape S-starts. Movements were recorded with high-speed video (250 Hz). Muscle activity was recorded from seven electrodes, one in the left adductor mandibulae and bilaterally in the anterior, midbody and posterior epaxial white muscle. Although S-shaped movements are produced in both feeding and escape, kinematics and electromyogram (EMG) patterns differ. Stage 1 (pre-propulsive movement) is significantly slower and more variable during feeding strikes and involves caudal bending with less rostral movement than recorded for startle behaviors. Correspondingly, there is strong caudal muscle activity prior to rostral activity during strikes, whereas in startles caudal muscle activity had near simultaneous onset with contralateral rostral activity. Onset of jaw muscle activity occurred significantly after the onset of axial muscle activity during feeding strikes. By contrast, during startles, jaw activity onset was nearly simultaneous with the onset of axial muscle activity. Stage 2 kinematics generally did not differ between the strike and startle; however, EMGs indicate that stage 2 movements are generated by different patterns of muscle activity for the two behaviors. Although strikes and startles are similar in their propulsive performance, they appear to be initiated and driven by fundamentally different motor control mechanisms. We suggest that S-start startle behavior is mediated by a simple system of descending reticulospinal input to spinal neurons while the S-start strike involves a more complex neural circuit, allowing greater modulation of stage 1 movements while maintaining high stage 2 performance.