The four moves
"Computational thinking" is an intimidating name for four moves your child already makes in pieces. Naming them helps you spot and strengthen them:
- Decomposition — break a big problem into smaller, doable parts. ("Clean your room" becomes: bed, then floor, then desk.)
- Pattern recognition — notice what repeats, so you can reuse a solution instead of starting over. (Every level of this game has a trap right before the exit.)
- Abstraction — ignore the details that don't matter and keep the ones that do. (A subway map isn't geographically accurate; it keeps only what helps you ride.)
- Algorithms — write clear, ordered steps that always work, precisely enough that someone else could follow them exactly.
The through-line is precision: thinking clearly enough that a machine — or another person — could follow you without guessing. That's the skill, and it long predates computers.
Why it matters more in the age of AI
AI tools do what you describe. Describe a task vaguely and you get vague output; break it down and state it precisely and you get something useful. Computational thinking is exactly what lets a child direct a powerful tool and judge its answer — the same ability we explore in learning in the age of AI. It's not about competing with the machine; it's about being able to think clearly enough to command one.
Where children practice it here
Games turn out to be an unusually good gym for this, because a game's rules are precise and the feedback is instant. In our arcade, a child decomposes a broken program to find the one wrong line, spots the pattern behind a bouncing ball, abstracts a snake bot's strategy into a few numbers, and writes the algorithm a bot runs on. Same four moves, wearing a game's clothes. And you don't need our arcade — the section below works with a cardboard box and a marker.
Common questions
Do you need a computer to learn it?
No. A child practices it whenever they give precise directions, sort things by a rule, plan a recipe's steps, or explain a game's exact rules. Many good programs teach it "unplugged" first, then connect it to code.
Is it the same as coding?
No. Coding is one place to practice it, but the thinking is broader — it shows up in math, science, writing, and everyday planning. Coding just makes the thinking visible and testable.
What age can a child start?
The four moves can be nurtured very young through play, logic, and cause-and-effect. Connecting them to real code tends to click from around ages 10+.
How do I nurture it at home?
Ask "what are the steps?" and "what's the rule?" about ordinary things. Let your child give you exact instructions and follow them literally — including the funny failures — so they feel how precise a real instruction must be.
The "exact instructions" game
Have your child tell you exactly how to make a peanut-butter sandwich, and follow their words literally — if they say "put peanut butter on the bread," put the jar on the bag. The laughter is the lesson: they're discovering how precise an algorithm has to be. Then try the same precision on a game rule in The Glitch Arcade.
Open the Arcade — free Games as laboratories