Serendip is an independent site partnering with faculty at multiple colleges and universities around the world. Happy exploring!

Emergence 2009: Final Projects, III

Emergence

Biology 361 = Computer Science 361
Bryn Mawr College, Spring 2009

 

Back to index of models

 

 

Download/view: DecisionMaking.nlogo

WHAT IS IT?
This model shows the relation between age and decision-making. Older turtles tend to make better decisions and avoid what is bad or unsafe. Health goes down when bad decisions are made. Turtles can reach immortality if they can learn to make the right decisions 100% of the times before they die from making several bad decisions.
HOW IT WORKS

NUMTURTLES slider determines how many turtles are going to be there at the beginning.

NUMHATCH determines the number of turtles born every 100 ticks.

COLORTURTLEBYAGE switch colors the turtles based on their age if it is on - white corresponding to the youngest turtles, and dark magenta corresponding to the oldest turtles. If the switch is off, it colors the turtles based on their health - white being the healthiest and dark violet being the unhealthiest.

If OLDDEATH? switch is turned on, it means that turtles will die if they reach a certain old age. If it is turned off, then turtles will not die from old age; they will only die from poor health in that case.

The YOUNGLEARNFAST? switch determines whether the young turtles learn faster or the older ones. When a turtle loses health by landing on an unsafe patch, its probability to avoid unsafe patches increases and based on that switch, the probabilty increases much more for younger ones than the older ones, or vice versa.

CREATEZONES creates unsafe areas based on the selections on the switches zone1, zone2, zone3 and zone4.

SETUP creates turtles based on the number specified in NUMTURTLES. It creates a range of turtles - an equal number of each possible age. It also distributes the turtles randomly in the world.

GO makes the turtles move around the world randomly, avoiding unsafe patches based on their probability of avoidance, and getting older over time.

The different monitors show how many turtles of each age, health or avoidance probability are currently present in the world.

The plot of Health Vs Age shows the health of turtles with a certain age, and what their avoidance probaility is as well, which is represented by the color of the dots (white being turtles with low probabillity and the darker ones having a higher probability).

HOW TO USE IT
Turn the switch of whichever zones you want. Click on CREATEZONES to see the chosen zones. Select the number of turtles and the number to hatch, along with the correct switches for COLORTURTLEBYAGE, OLDDREATH? and YOUNGLEARNFAST. Then click Setup to have the choices made by the user show up in the world. Press GO (single) to see what happens one step at a time, or GO (forever) to see the commands being executed continuously.
THINGS TO NOTICE

Do more turtles reach immortality (avoidance of 100%) if young turtles learn faster, or if older turtles learn faster? Why?

Which zone or combination of zones seem to get more turtles reach immortality faster?

What is generally the age and health of the turtles that reach 100% avoidance level? Is there a trend?

Are there generally more older turtles or younger ones at any time in the world?

THINGS TO TRY

Try out different combinations of NUMTURTLES and zones with OLDDEATH? turned off and NUMHATCH set to 0. What is the health level of most of the turtles that reach immortality at the end?

Notice the color of the dots in the Health Vs Age plot. The darker dots represent turtles with high avoidance level, and the lighter ones are turtles with low avoidance level.

Play with the YOUNGLEARNFAST? button. Do more turtles reach high avoidance level if this is switched on or off?

EXTENDING THE MODEL
It would be interesting to have the turtles color in a way that will represent both the age and haelth together.
RELATED MODELS
My previous model, PercentVisited_barriers shows how turtles learn to find and avoid unsafe patches that are distributed randomly around the world.
CREDITS AND REFERENCES
Thanks to Professor Paul Grobstein for helping to make the idea for this model more concrete and fellow student Evan Raskin for his help and expertise with Netlogo.

 

Models created using NetLogo.

Comments

Post new comment

The content of this field is kept private and will not be shown publicly.
To prevent automated spam submissions leave this field empty.
3 + 4 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.