RE: mosquito project (was: Introduction)

["Gary McCallister" <mccallis@xxxxxxxxxxxxx>]

I received this e-mail as finals were beginning and I still haven't found time
to read the Logomecium paper.  I don't think I am  a good enough programmer to
add much to the discussion.  However, I am a pretty good mosquitologist and
general biologist, especially when it comes to invertebrates).

I might first note that the mosquito hunter does use random, expecially at the
barrier.  An odd phenomenon is that the "mopsquito" doesn't "find" it's way
around the barrier at all, but just keeps beating on the barrier randomly until
it works it's way around.  In some cases it suddenly breaks through the barrier
as if ti had discovered a hole in the screen.  It does this because the 
distance
of each move is also random, and if the distance is greater than the thickness
of the barrier, it will jump clear across.  That could be changed by putting a
limit on the random distance of each move.  

However, this very much mimics an insects behavior at a window screen when it
is attracted to the light or dark inside.  It has no capacity for thinking and
so will literally beat itself to death, randomly.

This next lengthy comment probably does not change much in the programming
discussion.  However, mosquitoes are technically not attracted to carbon
dioxide.  CO2 is simply an exciter.  And it isn't CO2 itself, it is a change in
the ambient concentration.  So a resting mosquito probably does not respond to
CO2 until other cues alert it too.  But when it does so, it becomes excited by 
a
change in the concentration.  (If the concentration remains high, the mosquito
becomes bored and settles down.)  

Since any CO2 plume is bound to be mixed and whirled by obstacles
cross-currents and potential source mobility the mosquito presumable senses a
continuously shifting, changing concentration.  When this occurs, the mosquito
remains agitated, moves "towards" the last high concentration until the []
drops, then it goes into "random" direction and distance until a new increased
plume is discovered.  

This is highly inefficient of time, but highly efficient in end result.  The
msoquito will almost invariably arrive at a source of CO2, even though it may
not be the source that it originally responded too.  However, it so much more
complex than that.  Mosquitoes can tell one source of CO2 from another and each
species has it's favorite sources.  Thus Culex would just as soon feed on a
bird, but will settle for a human if it is running low on fuel.  

In fact, host seeking is a multisensory affair.  It has been shown for other
insects that they respond not just to one attraction.  They may need to be
aroused by light (or dark as the case may be), may seek by chemical attractant,
may choose between sources based upon other chemical cues (mostly unknown at
this time), further refine their search based upon  color and movement, and 
make
a final selection based upon temperature and subtle chemical cues.  Most of 
this
is not known for mosquitoes (and most other insects as well).  

Now that's a computer model to be built.  :-)






Gary McCallister

http://www.onemanmormonbluesband.com
semi-serious and quasi-comic music for the non-disciminating consumer