http://friam.383.s1.nabble.com/Tiny-brained-bees-solve-a-complex-mathematical-problem-Queen-Mary-University-of-London-tp5763398p5764663.html
Sounds amazing to me, but for quite a different reason. I remember when
they were first doing the tracking of individual bees (back when I could still
claim to be studying animal behavior professionally). One of the most startling
findings of the old research was that bees almost never flew optimal paths. In
fact, you could watch streams of bees fly very unideal paths for hours,
sometimes days. This past research either followed bees foraging at large, or
gave them a single, temporary super-food-source, which the experimenter would
move around every so often. Bees would routinely fly 3-4 times farther than
they had to, in very convoluted ways, (presumably) because they were using
landmarks to get to the food source.
So what is different here?
Skimming the article: This research placed four artificial "flowers" in a 8.7
meter by 7.3 meter greenhouse. The flowers never moved, and always provided
food. They introduced the four flowers one at a time, so that if bees always
visited the flowers in the order they were introduced, they would fly in a
maximally inefficient order (making a 5 pointed star, with the hive entrance as
one of the points). Flying a circle around the greenhouse would visit all the
flowers much more efficiently. From what I can tell, they did not track the
bees in the air, but only noted the sequence in which they visited the hives.
-- On the positive side, the used computer simulation (but not, so far as I can
tell, agent based modeling) to determine what their null hypothesis should be.
On the negative side, their model did not seem to include natural constraints
that they later mention as being well established in the literature (e.g.,
that, in a given foraging bout, bees sometimes re-visit the same flower they
were just at, but rarely go back to flowers visited earlier). -- That said, it
is seems clear that: 1) bees used optimal paths fairly often (i.e.,
flying in a clockwise or counterclockwise circle), and 2) in the small number
of trials they had, bees never flew in the least optimal pattern (the five
pointed star). However, 3) bees flew in some sort of suboptimal pattern about
40% of the time.
Other thoughts: 1) The use of the term "trapline" is
weird, awkward, and distracting. It has antiquated significance, and I'm not
sure why they don't just talk about "optimal" foraging paths. 2) I suspect
there are several ways to make a more realistic computer simulation that would
produce similar foraging patterns with semi-random movements. While it is
admirable that they used a model to define null expectations, the simplicity of
their model is a bit suspect. 3) Future studies should create
configurations in which the optimal foraging path is not a circle around the
outside of the room. There are just too many ways to get a circle that don't
involve dynamic optimization of flight patterns.
Anyway, those are
thoughts from a first pass. If the experiment was more sophisticated, I would
judge it as definitely worthy of some Friamer throwing a little of their ABM
mojo at it. As it is, who knows? Low cost, low payoff. Either way, it seems
worth keeping track of where this research leads in the future.
Eric
On Mon, Nov 22, 2010 01:09 PM,
Russ Abbott
<[hidden email]> wrote:
Sounds amazing, but it's not quite as significant as the press
release makes it out to be. Here's the abstract from <a href="<a
href="http://www.journals.uchicago.edu/doi/abs/10.1086/657042"
onclick="window.open('http://www.journals.uchicago.edu/doi/abs/10.1086/657042');return false;">http://www.journals.uchicago.edu/doi/abs/10.1086/657042">The American Naturalist</a>. All the bees need do is keep track of the distance for any route and then through experimentation among different routes select the shortest found so far.
-- Russ Abbott
On Mon, Nov 22, 2010 at 9:00 AM, Owen Densmore
<owen@...> wrote:
GAs, Ant algorithms, now Bees!
<a href="http://www.qmul.ac.uk/media/news/items/se/38864.html" target="" onclick="window.open('http://www.qmul.ac.uk/media/news/items/se/38864.html');return false;">http://www.qmul.ac.uk/media/news/items/se/38864.html
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at <a href="http://www.friam.org" target="" onclick="window.open('http://www.friam.org');return false;">http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
Eric Charles
Professional Student and
Assistant Professor of Psychology
Penn State University
Altoona, PA 16601
Meets Fridays 9a-11:30 at cafe at St. John's College