| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDITORIAL FOCUS
; Perritt et al. 1994). Despite this preference, humans have an extraordinarily wide facial range. When studying human facial evolution, the primary question is thus probably not why the average but why the variety.
A similar issue arises in motor-pattern generation, in which a stereotyped motor pattern is often implicitly assumed to exist for each motor task. The task's motor pattern is then defined by averaging repeated performances of it, and variation around this mean often is ascribed to preparation damage or uncontrolled variables. Because animals perform multiple tasks, this organization would require multiple neural networks (1 per task) or multi-tasking networks that are modified to produce multiple outputs. Both solutions entail additional nervous system complexity with presumed additional metabolic cost (Laughlin 2001). An alternative (not necessarily exclusive) is to not produce task-optimized outputs but to produce the same wide range of outputs regardless of the task being performed. For each task, a subset of these outputs will likely perform it at least reasonably well with this "just-good-enough" inefficiency being "cheaper" than having multiple networks, or complex, multi-tasking networks, that do each task perfectly.
The article in this issue (Horn et al., p. 157–180) from the labs of Vlad Brezina and Klaude Weiss on Aplysia feeding highlights this issue. Aplysia feed on algae with a wide variety of sizes and shapes, and one might imagine Aplysia optimize their feeding movements for each alga type. However, Horn et al. show that Aplysia feeding motor-pattern parameters (cycle period, neuron burst durations) are highly variable both in vivo and in vitro. Remarkably, this variability remains even when Aplysia eat a uniform strip of sea weed (an invariant task). Both the feeding muscles (Weiss et al. 1993; Brezina et al. 2000) and the feeding neural network (Sweedler et al. 2002; Koh et al. 2003) are highly modulated, and thus the variability could arise from history-dependent effects. However, when these effects are minimized by comparing only successive cycles, much of the variability remains. The authors use motor-nerve stimulation to show that the variability is also not due to randomness in muscle response, and most of the variability therefore originates in the motor-pattern-generating network itself. Many Aplysia bites will consequently be suboptimal, and some may result in no food intake whatsoever. Aplysia thus apparently uses a just-good-enough strategy in which, provided most bites result in some food intake, it doesn't matter whether they are perfect bites.
One could argue that Aplysia uses this strategy because its nervous system has insufficient neurons to calculate alga-specific optimum-feeding movements. However, all nervous systems are metabolically expensive, and there is thus presumably always evolutionary pressure to generate behavior with the smallest possible number of neurons. The need to compromise between always generating optimal behaviors and the cost of maintaining the neurons this would require thus likely applies to all animals. Just-good-enough strategies are most useful in situations in which a range of related tasks are present in the environment (and thus producing a range of motor outputs means that a fairly good result will fairly often occur regardless of the task at hand) and in which the cost of failures is low. Large "random" variability is therefore unlikely in locomotory motor patterns. For instance, although in terrestrial locomotion substrates are often highly variable and large cycle-by-cycle variations are therefore required to well match step and substrate, this matching cannot be done with random cycle-by-cycle variation because the cost of failure (falling) is probably prohibitive. Alternatively, in swimming and flying the media are relatively uniform, there is hence really only one task and thus no obvious advantage to producing cycle-by-cycle variations.
Mastication is another story, however, as we eat foodstuffs with a wide variety of physical properties, and even suboptimal chewing would provide some food processing without serious deleterious effects. Only recently have techniques been developed that allow complete description of human chewing movements (Buschang et al. 2000), and whether chewing is optimized as a function of foodstuff is unclear. However, even when great care is taken to regularize chewing (a single food type—gum—chewed on only 1 side) and only the 10 most similar chews (of 50-60 total) are analyzed, humans show considerable movement parameter variability (SDs approximately ±30%) (Buschang et al. 2000). This work focused on the means of the parameters rather than their variation, and the variability could thus arise from irregularly spaced bouts of regular chewing, slow variation in long chewing bouts, or Aplysia-like chew-by-chew random variation. Nonetheless, these data suggest that, for some motor patterns, humans may also use a just-good-enough strategy. However, as Horn et al. well demonstrate, proving this will require recognizing that, in some cases, the most functionally important component of a system's output is not its mean but its variability and range.
Neuroscience Program, Department of Biological Sciences, Ohio University, Athens, Ohio 45701
Address reprint requests and other correspondence to: S. L. Hooper (E-mail: hooper{at}ohio.edu).
REFERENCES
Brezina V, Orekhova IV, and Weiss KR. Optimization of rhythmic behaviors by modulation of the neuromuscular transform. J Neurophysiol 83: 260–279, 2000.
Buschang PH, Hayasaki H, and Throckmorton GS. Quantification of human chewing-cycle kinematics. Arch Oral Biol 45: 461–474, 2000.[CrossRef][ISI][Medline]
Horn CC, Zhurov Y, Orekhova IV, Proekt A, Kupfermann I, Weiss KR, and Brezina V. Cycle-to-cycle variability of neuromuscular activity in Aplysia feeding behavior. J Neurophysiol 91: 157–180, 2004.
Koh HY, Vilim FS, Jing J, and Weiss KR. Two neuropeptides colocalized in a command-like neuron use distinct mechanisms to enhance its fast synaptic connection. J Neurophysiol 90: 2074–2079, 2003.
Langlois JH and Roggmann LA. Attractive faces are only average. Psychol Sci 1: 115–121, 1990.
Laughin SB. Energy as a constraint on the coding and processing of sensory information. Curr Opin Neurobiol 11: 475–480, 2001.[CrossRef][ISI][Medline]
Perrett DI, May KA, and Yoshikawa S. Facial shape and judgements of female attractiveness. Nature 368: 239–242, 1994.[CrossRef][Medline]
Sweedler JV, Li L, Rubakhin SS, Alexeeva V, Dembrow NC, Dowling O, Jing J, Weiss KR, and Vilim FS. Identification and characterization of the feeding circuit-activating peptides, a novel neuropeptide family of Aplysia. J Neurosci 22: 7797–7808, 2002.
Weiss KR, Brezina V, Cropper EC, Heierhorst J, Hooper SL, Probst WC, Rosen SC, Vilim FS, and Kupfermann I. Physiology and biochemistry of peptidergic cotransmission in Aplysia. J Physiol 87: 141–151, 1993.
This article has been cited by other articles:
|
|
 |
|
  C. S. Lum, Y. Zhurov, E. C. Cropper, K. R. Weiss, and V. Brezina Variability of Swallowing Performance in Intact, Freely Feeding Aplysia J Neurophysiol, October 1, 2005; 94(4): 2427 - 2446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Oprisan, A. A. Prinz, and C. C. Canavier Phase Resetting and Phase Locking in Hybrid Circuits of One Model and One Biological Neuron Biophys. J., October 1, 2004; 87(4): 2283 - 2298. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
