MCB 419: Homework

Individual Project #1

Inital draft (HW #7): due Tue Mar 10
Final version (Proj #1): due Tue Mar 17

This assignment is part 1 of your first individual project for the class. The project will span two weeks. This week you will select one of three possible project topics and develop a preliminary version of your NetLogo simulation (due Mar 10). Next week you will complete your simulation study and turn in the final version of your project write-up (due Mar 17). The three possible project topics are:

Topic A: Area-restricted search: modulation of behavior
Topic B: Multi-sensory informational cues
Topic C: Simulated evolution: genomic memory

This project will be more open-ended than your previous homework assignments. You will not be given a 'demo' version of the model. Instead, you will have to design and create your own model from scratch, based on brief descriptions on the 'Guidelines' tab. Select whichever project you find most interesting.

For this week, your main goal is to develop a working version of the NetLogo model. In the next assignment, you will will use this model to carry out some quantitative studies, and write up your results in the form of a short report.

Answer the questions found on the 'Assignment' tab and email your responses along with a copy of the netlogo file to mcb419@gmail.com with 'hw07' in the Subject line.

This assignment is due by 11:59PM (midnight) on Tue, Mar 10.

Design Guidelines

Topic A: Area-restricted search: modulation of behavior

In Lecture 9, we discussed area-restricted search (ARS) behavior. In C. elegans, we saw that dopamine was involved in modulating the frequency of high-angle turns, such that the worm's search behavior changed depending on how much time had elapsed since the worm had last been in contact with food. For this project, you will develop a NetLogo simulation of ARS illustrating how a slow modulatory signal can mediate changes in search behavior, and how ARS can improve foraging efficiency under certain conditions. Specifically, you will use your model to provide support for the following two hypotheses:

When food is clustered, an area-restricted search strategy can improve foraging efficiency relative to a non-modulated search strategy.
When food is randomly distributed, an area-restricted search strategy does not provide any benefit relative to a non-modulated search strategy.

Topic B: Multisensory informational cues

In Lecture 6, we discussed Dusenbery's idea that certain stimuli are 'causal' whereas others are 'informational'. In his paper, Dusenbery gave an example of how the scent and color of a flower have importance to a bee only because they are associated with nectar, which has nutritional value for the bee. (nectar is 'causal'; scent and color are 'informational'). For this project, you will develop a NetLogo simulation in which two different 'informational' sensory cues are associated with the presence of food. You will then design a controller that makes use of these informational cues. For example, you might create a simulation in which food is found only in bright, warm regions of the environment. You will use your model to:

Demonstrate that under appropriate conditions, the use of 'informational' cues can significantly improve foraging efficiency relative to design that ignores these cues.
Use your simulation to clarify the factors that define 'appropriate conditions' in the above statement.

Topic C: Simulated evolution: genomic memory

In Lecture 6, we discussed the idea that an organism's genome provides a memory of what designs and strategies have been most successful in previous generations. For this project, you will develop a NetLogo simulation in which certain parameters of the bot's foraging stragegy are encoded into a set of NetLogo variables that are treated as a genome. When the bot reproduces, it passes along this genetic information to its offspring, with variation. Using a foraging scenario similar to hw05, in which bots eat, grow and reproduce, you will allow your bots to 'evolve' over multiple generations. Use this scenario to provide support for the following hypotheses:

Over the course of multiple generations, the processes of variation and natural selection will tend to 'converge' on a controller design that is well-matched to the foraging task in this environment.
The foraging performance of the 'final' design (after convergence) will be significantly improved relative to that of the initial 'seed' population.

Assignment

You can access a copy of the assignment file HERE, or copy and paste from the text below.

=================================================================
MCB 419 Homework 7 (Spring 2009)

When you've finished answering all the questions, email a copy 
of this file (hw07.txt) along with your NetLogo file (hw07.nlogo)
to mcb419@gmail.com with 'hw07' in the Subject line. 

=================================================================


1. Which one of the three project topics did you select?



2. Provide a brief overview of your NetLogo simulation: 

   a) key features of the 'environment' that you created

   b) summary of your bot's sensory capabilities

   c) summary of your bot's motor capabilities

   d) overview of your bot's controller design

   e) Are there any special instructions for running the 
      simulation or manipulating the user-interface?


3. Did you encounter any technical problems or do you have any 
   questions that still need to be resolved before next week?



=================================================================
END OF THE ASSIGNMENT
=================================================================