Stand a microphone too close to its own speaker and you already know what happens: a faint hum swells into an ear-splitting screech in about a second. Nothing changed the rules. Nobody turned a knob. The sound that came out of the speaker went back into the microphone, came out louder, went back in louder still, and the loop did the rest. That howl is not a glitch — it’s the purest demonstration of the single most important idea in systems thinking, and the subject of this entire course.
In the last course you learned to ask “and then what?” — to follow a decision past its obvious first result into the world’s reaction. That habit, second-order thinking, told you the later effects exist. This course answers the next question: why do those effects so often blow up, fizzle out, or wobble back and forth instead of just… happening once and stopping? The answer, almost every time, is a feedback loop.
What a feedback loop actually is
A feedback loop is any arrangement where a system’s output comes back around to become part of its own input. The screech is a loop. So is the bank account where this year’s interest earns next year’s interest. So is the thermostat that switches your heater off once the room is warm and back on once it cools. In every case, the thing the system produces feeds back and influences what it produces next — so the system is, in a real sense, reacting to itself.
That’s what systems thinking is: the discipline of looking at the loops between things instead of the things in isolation. A pile of parts behaves predictably. The same parts wired into a loop can do something no single part would ever do alone — explode, self-correct, or oscillate forever. The behaviour lives in the wiring, not the pieces.
The one-sentence version
A feedback loop is when a system’s output loops back to become its own input — so the system responds to itself, and that self-response is where runaway growth, stubborn stability, and endless oscillation all come from.
There are only two kinds (and one troublemaker)
Here’s the whole course in miniature. Every feedback loop is one of two types:
- A reinforcing loop (also called amplifying or positive) feeds an effect back so it gets bigger each time around. More leads to more. The screech, compound interest, a viral video, a bank run, a snowball rolling downhill. Reinforcing loops are why small things become enormous.
- A balancing loop (also called stabilizing or negative) feeds an effect back so it gets smaller, pulling the system toward some target. A thermostat holding a temperature, your body sweating to cool down, prices rising until buyers back off. Balancing loops are why most things stay roughly put.
And then there’s the troublemaker that makes both misbehave: the delay. When feedback arrives late — when the system can’t see the result of its last action yet — a balancing loop overshoots its target and starts to oscillate. It’s why the shower scalds you a few seconds after you crank the hot tap, why supply chains lurch between glut and shortage, why economies boom and bust. Delay is the difference between a system that settles and one that hunts back and forth forever.
You don’t have to take that on faith. Here’s the machine itself — flip between a reinforcing and a balancing loop, push the strength up, and then, on the balancing loop, add a delay and watch a calm glide turn into an oscillation:
Run the loop
One machine, three behaviours
Pick a loop type, set its strength, and — for a balancing loop — add a delay. Watch how a reinforcing loop runs away, a balancing loop glides to its goal, and a delay makes that same loop overshoot and oscillate.
A reinforcing loop feeds on itself: 20 compounds to about 3673 — roughly 183.7× the start — and just keeps climbing. Nothing here pulls it back; the output is its own input.
That single chart is the spine of everything ahead. Spend ten seconds dragging the sliders now; the rest of the course is just learning to read what you’re seeing.
Before you read — take a guess
A microphone picks up a sound, the speaker plays it louder, the mic picks THAT up, and a screech builds in a second. What kind of loop is this?
Why loops break ordinary intuition
Most of us reason in straight lines: a small cause makes a small effect, a big cause makes a big effect, and twice the input gives twice the output. That linear intuition is the default setting of the human brain, and it is exactly wrong for systems with loops.
A reinforcing loop turns a whisper into a screech — a tiny input becomes an overwhelming output. A balancing loop swallows a big shove and barely moves — a huge input produces almost no lasting change. And a delayed loop produces an effect that arrives after you’ve stopped looking, often pointing the opposite way from what you expected. None of that is intuitive if you think in lines. All of it is obvious once you think in loops.
The pitfall this whole course fights
The number-one systems-thinking mistake is assuming the world is linear — that effects are proportional to causes and arrive right away. They aren’t and they don’t. Loops make small things explode and big things vanish; delays make effects show up late and backwards. Linear intuition will fool you every single time there’s a loop in play.
How this connects to what you already know
This isn’t a fresh start — it’s the level below second-order thinking. When you asked “and then what?” and found that a consequence compounded (the bank run that fed on itself) or cancelled out (the price that rose until demand fell), you were looking at a feedback loop without naming it. Second-order thinking said the reaction matters. Feedback loops tell you what shape that reaction takes — runaway, self-correcting, or oscillating — which is exactly what lets you predict it instead of just bracing for it.
The same goes for incentives: a perverse incentive that spirals out of control (more cheating makes cheating more normal, which makes more cheating) is a reinforcing loop wearing a business suit. You’ve been meeting loops the whole time. Now you get the engineering.
The map of the course
Six teaching lessons, then one exam you can’t undo:
- Stocks & Flows — the vocabulary of every system: the bathtub model of things that accumulate (stocks) and the rates that fill or drain them (flows). You can’t talk about loops until you can talk about this.
- Reinforcing Loops — amplification: compounding, viral growth, bank runs, vicious and virtuous cycles, and why “more leads to more” produces explosions and collapses alike.
- Balancing Loops — stabilization: thermostats, body temperature, predator and prey, markets clearing — the loops that resist change and hold a system at a target.
- Delays & Oscillation — the troublemaker: why a late signal makes a stabiliser overshoot and ring, from the scalding shower to the supply-chain bullwhip.
- Leverage Points — where to push: the places in a system where a small, well-aimed intervention changes everything, and why the obvious place is usually the weak one.
Then a Final Exam — graded, one question at a time, one-way: once you answer, it locks. No back button, no retries.
How to use this course
One rule does most of the work: guess before you peek. Commit to an answer in your head before you reveal any explanation. The small sting of being wrong is what makes a model stick — and systems thinking is full of answers that feel right (think in lines!) and turn out backwards. The exercises are the lesson; the prose just sets them up.
Next up: stocks and flows — because before you can understand a loop, you have to understand the thing the loop is looping around.