Subject: Feedback Loop and Metaphor

Posted by John Gunkler on 2/19/2006
In Reply To:Feedback Loop and Metaphor Posted by Richard Turnock on 2/19/2006

Message:

Richard,

I am afraid that I'm not yet comfortable with a Ferris Wheel as a metaphor for a feedback loop. While it captures the notion of things going around a circle and returning to the beginning, I don't see how it illustrates the key idea behind a feedback loop -- namely, the idea that what returns through the loop somehow affects what next is sent through it. You only have "feedback" when what you send out now somehow comes back to influence what you send out later.

I can think of some ways to salvage the Ferris Wheel, but (unfortunately) not as a metaphor. For example:

1. You could begin by setting up the situation of a fun place that has a Ferris Wheel (a fair, or Navy Pier in Chicago, or a midway) and ask students, "What is it about the Ferris Wheel that makes you get more interested in it?"
You may get answers like "The music" or "The lights shining on it" or "Its motion" or "Its huge size, sticking up above everything else" or "A friend who likes to ride it and gets excited" -- and, if you're really lucky, you may hear, "Crowds of excited kids getting on and off it." If you don't hear this last, you could prompt it by asking, "If you heard and saw a lot of other kids being excited by riding the Ferris Wheel, would that make more interested in riding it?"

2. Now, you set up the idea that you're interested in keeping track of the "popularity" of the Ferris Wheel and ask, "How would we know how popular the Ferris Wheel is?"
Expect to hear at least one person say, "By seeing (or counting) how many people ride it during the day."

3. So, the "stock" we're interested in is the number of people riding on the Ferris Wheel at any particular time -- which is a nice stock to track, since it only changes at the times the Ferris Wheel stops to load and unload passengers. And, of course, you could also keep track of the total number who ride during a set period (say, one day, or one afternoon.) And (see the next point), it might also be interesting to keep track of the number of people lined up to ride next.

4. Now you can talk about the effect the number of people riding right now has on the number who will line up to ride the next time. This gets interesting, because I think you'll find a threshold effect and two causal loops. That is, when only a few people are riding, there's not much impact on encouraging others to ride through the generation of excitement -- but there may be a positive effect due to noticing that the line isn't very long and "we can get right on!" In the middle range of number of riders, where the wheel is quite full every time (and riders are screaming with delight) but the line remains short enough that you can get right on, you will see a maximum positive impact on the number of riders getting in line. In the highest range, where the wheel is completely full but the queue is very long, you may see a negative effect on the number of people who get into line.

Maybe someone more imaginative than I can find a way to make the Ferris Wheel into a true metaphor for a feedback loop, but I suspect it will require doing things with the Ferris Wheel that don't naturally occur.

John Gunkler