Abstract
Squid use eight arms and two slender tentacles to capture prey. The muscular stalks of the tentacles are elongated approximately 80% in 20-40 ms towards the prey, which is adhered to the terminal clubs by arrays of suckers. Using a previously developed forward dynamics model of the extension of the tentacles of the squid Doryteuthis pealeii (formerly Loligo pealeii), we predict how spatial muscle-activation patterns result in a distribution of muscular power, muscle work, and kinetic and elastic energy along the tentacle. The simulated peak extension speed of the tentacles is remarkably insensitive to delays of activation along the stalk, as well as to random variations in the activation onset. A delay along the tentacle of 50% of the extension time has only a small effect on the peak extension velocity of the tentacle compared with a zero-delay pattern. A slight delay of the distal portion relative to the proximal has a small positive effect on peak extension velocity, whereas negative delays (delay reversed along stalk) always reduce extension performance. In addition, tentacular extension is relatively insensitive to superimposed random variations in the prescribed delays along the stalk. This holds in particular for small positive delays that are similar to delays predicted from measured axonal diameters of motor neurons. This robustness against variation in the activation distribution reduces the accuracy requirements of the neuronal control and is likely due to the non-linear mechanical properties of the muscular tissue in the tentacle.