AP Computer Science Principles · lesson 1 of 4 · 20 min
What is computational thinking?
Computational thinking is a problem-solving approach that borrows techniques from computer science and applies them to any domain. It is not about programming — it is about thinking clearly and systematically. The College Board identifies four pillars: decomposition , pattern recognition , abstraction , and algorithm design . When you decompose a problem you break it into smaller, more manageable sub-problems. When you recognise patterns you notice similarities between problems you have already solved. When you abstract you strip away irrelevant detail and focus on what matters. When you design an algorithm you create a precise, step-by-step solution. Real-world example
Planning a school event: decompose into venue, catering, invitations, and schedule. Recognise that invitations follow the same pattern every year. Abstract away the specific food choices when planning the timeline. Write an algorithm (checklist) for the day-of logistics.
Decomposition in depth
Decomposition means identifying the distinct parts of a problem so each part can be solved independently. A well-decomposed problem has sub-problems that are cohesive (each does one thing) and loosely coupled (changes to one do not break others). Consider building a weather app. The top-level problem decomposes into: fetch weather data, parse the response, display current conditions, display a 5-day forecast, and handle errors. Each of those can be decomposed further — fetching data involves choosing an API, constructing a URL, sending an HTTP request, and handling the response. On the AP exam, decomposition questions often ask you to identify which sub-problem a given code segment addresses, or to explain why breaking a problem into procedures makes it easier to test and maintain.
Abstraction in depth
Abstraction hides complexity behind a simpler interface. When you call
Math.sqrt(16)
you do not need to know the Newton–Raphson algorithm running underneath — the function abstracts that detail away. Good abstraction lets you think at a higher level.
There are two kinds of abstraction the AP exam tests:
procedural abstraction
(wrapping code in a named procedure) and
data abstraction
(representing complex information with a simpler structure like a list or object).
Procedural abstraction example
Instead of writing the same five lines to validate an email address in ten places, you write one isValidEmail(address) procedure and call it wherever needed. If the validation rules change, you update one place.
Algorithm design
An algorithm is a finite, unambiguous sequence of steps that solves a problem. Every algorithm must eventually terminate and must produce a correct result for all valid inputs. The AP exam uses a block-based pseudocode — you will never be asked to write Python or Java, but you must be able to read and trace pseudocode.
- PROCEDURE greet(name)
- DISPLAY(“Hello, ” + name)
- END PROCEDURE
greet(“Alice”) → displays “Hello, Alice” The AP pseudocode uses DISPLAY, INPUT, IF/ELSE, REPEAT UNTIL, FOR EACH, and PROCEDURE. Memorise these keywords — they appear in every exam.
Knowledge check
- Question of
- Hide hint
- Show hint
✓ Correct! ✕ Not quite
- Try again
- See results
- Next question
- Check answer
Back to AP Computer Science Principles
Next lesson: Binary numbers and data representation