|
|
|
|
CONTRIBUTORS:
|
|
|
CONFERENCE NAME:
|
|
|
CONF. LOCATION:
|
Washington DC
|
|
CONFERENCE YEAR:
|
2005
|
|
PUB TYPE:
|
Conference Presentation
|
|
SUBJECT(S):
|
behavior, neuroscience, neurobiology, chaos theory, nonlinear forecasting
|
|
DISCIPLINE:
|
Biology
|
|
HTTP:
|
http://bjoern.brembs.net/download.php?view.33
|
|
LANGUAGE:
|
English
|
|
PUB ID:
|
103-426-002
(Last edited on
2006/04/13 05:24:11 GMT-6)
|
|
SPONSOR(S):
|
|
|
ABSTRACT:
Ever since the ancient Greeks, animals have been considered automata, or robots: if one only knew all their input variables, one could predict the motor output they would produce. This view carried the cybernetic approach to insect behavior more than 20 years ago and still today, there is an open debate whether apparent spontaneous behavior constitutes mere "random noise" in an otherwise strictly deterministic system, or whether there is genuine 'spontaneity' built into the system.
We analyze both the temporal structure of short pulses of torque (torque spikes), in tethered Drosophila and angular flight components of foraging honeybees. The fly's environment consists of a cylindrical panorama arranged to center the fly within the cylinder. The cylinder is either featureless (white) and motionless, or the rotational speed of the cylinder (with visual patterns) can be controlled by the fly's yaw torque. The bees are searching for their hive after displacement into an area lacking natural landmarks. Although trained to feed close to the hive, the bees, after being displaced, are still able to use a general landscape memory formed during orientation flights to find their way home. After circling and returning to the release site repeatedly, the bees eventually find landmarks and they switch to direct flights towards the hive.
Our analysis includes estimating fractal dimensions, computing log-survivorship curves, auto-correlations, power spectrum and Lévy exponents, as well as evaluating flight path geometry.
None of our analyses support the hypothesis that the temporal structure of spontaneous behavior corresponds to white noise. Rather, they suggest that spontaneous behavior is neither "random" nor strictly determined. Our data match current propositions that spontaneous behavior exhibits a degree of determinism sufficient to achieve goals (e.g. foraging and mating), paired with the necessary unpredictability to avoid predation and maximize search efficiency.
Therefore, it appears that the probabilistic nature of spontaneous behavior serves an evolutionarily conserved function.
|
|
|
|
STATISTICS
|
|
Click on # to view
|
|
Citations
|
|
0
|
|
References
|
|
0
|
|
Comments
|
|
0
|
|
Quality
|
|
0/0.00
|
|
Interest
|
|
0/0.00
|
|
View(er)s
|
|
2/264
|
|
|
|
|
|
|
| Prev |
Next |
|
