Home Up StarLogo Online Resources



BACK to Content Areas

BACK to Links

Excerpt from my Final Document, "Using Emergent Design to Implement Computer-Enabled Learning Environments in K-12 Public Schools"( Copyright 2002 by Carol Caldwell-Edmonds, all rights reserved)

A Computer-enabled Demonstration of Emergent Levels in Complex systems: Simulation

Computer simulations demonstrate emergent levels by giving learners concrete examples to observe.   Mitchel Resnick directed the creation of one such simulation tool for exploring complex systems: a variation of Logo called StarLogo.  Instead of one turtle[1], StarLogo features the capability to program masses of turtles—hundreds or thousands—in order to model large systems of individual parts such as the molecules of a gas, a colony of ants, or traffic on a highway.  In his book, Turtles, Termites, and Traffic Jams, (2000) Resnick describes his work with students who developed a simulation based on automobile traffic.  On the screen is a simple city street scene and the Logo “turtles,” or cars in this incarnation, that are programmed to do very simple movements: any individual car decelerates if it is closer than one or two car-lengths to another car, otherwise it accelerates. In the program code, the command “random” appears in lines which determine the initial placement of cars on the street and the speed of the cars.  There is an option for including a police car with a radar trap in the median.  Running the simulation with the radar trap shows a traffic jam forming near it.  However, if the radar trap is not included, traffic jams still eventually form.  Further, in both simulations (with and without the radar trap), the point where the cars start crowding together moves backwards across the screen as the cars move forward.


Figure 2. a and b  Screenshots of the StarLogo Traffic Simulation Program

wpeE.jpg (43888 bytes)



Traffic Simulation in StarLogo after running several seconds shows the traffic jam “drifting backwards.”


In his book, Resnick describes how the behavior of the traffic jam—backward movement in relation to the cars’ direction—often surprised students who worked with him as he developed StarLogo.  Some students seemed to think that the jam should also move forward.  Resnick attributes their expectation to “slippage between levels” [of interaction in the simulation].  Students saw the jam as a collection of cars (correctly on one level, but not on the “jam” level) and then assumed the behavior of the emergent level (the traffic jam) would exhibit the same behavior as elements of another level (the cars moving forward). Resnick and Wilensky (1996) use this same example to illustrate how people can be confused by different levels of interactions and by the idea of “emergent levels.”

We are focusing on yet another meaning of levels, which we call the “emergent view” of levels.  Our focus is on levels that arise from interactions of objects at lower levels—like a traffic jam that emerged from the interactions among the cars.  These levels might seem similar to the part/whole levels: just as a year is made up of months, traffic jams are made up of cars.  But the jam/car relationship is different in some very important ways.  For one thing, the composition of the jam keeps changing; some cars leave the jam and other cars enter it.  Moreover, the jam arises from interactions among the cars; it is not just a simple accumulation of cars.   Months do not interact to form a year; they simply accumulate or add up.  A year can be viewed essentially as a long month.  But a traffic jam is not just a big car.  It is qualitatively different.  And that is what led to the high school students’ surprise: the jam behaved very differently from the cars, moving backwards while the cars within it moved forward.  (Resnick and Wilensky 1996)


Resnick also describes experiments students tried with the traffic program.  When the cars were all started at equal intervals with equal speeds, no traffic jams formed.  It was the randomness programmed into each individual car’s position and speed, meant to simulate reality, which eventually caused traffic jams to form. Emergent levels can arise by chance interactions as well as from external causes like radar traps.  The cars had no plan or intention to cause traffic jams, they were individually only decelerating or accelerating depending on their distance from the one car directly ahead.  In the simulation without the radar trap, chance interactions caused the traffic jam.


The programming environment that created this traffic simulation was designed for high school level students, and students were instrumental in its development.  Computer simulations enable complex ideas to become accessible to school age students through manipulation of concrete images on a screen. 

[1]Logo programming languages feature a turtle figure that moves or draws or calculates depending on commands typed one at a time or listed in full programs.  The turtle can be given any shape and background on the screen.