Unit Overview
Computational thinking & problem solving (including declarative, OO and AI)
CIE A-Level Computer ScienceΒ· 5 min read π n/a
1. Unit at a Glance
This unit follows a logical learning progression: we start with the four core pillars of computational thinking, then move to contrast two widely used problem-solving paradigms (object-oriented and declarative), before applying these core concepts to artificial intelligence problem solving. Each skill builds on the last: decomposition simplifies complex problems, abstraction removes unnecessary detail, pattern recognition speeds up solution design, and paradigms provide structured frameworks to implement working solutions.
This unit is split into the following sub-topics:
Problem decomposition
Learn how to break large, complex problems into smaller, independently solvable sub-problems.
β β± 3 min
Abstraction
Understand how to hide irrelevant details to focus on the core requirements of a problem.
β β β± 3 min
Pattern recognition
Identify repeating patterns across problems to reuse existing, proven solution approaches.
β β β± 3 min
Object-oriented problem solving
Model real-world problems using objects, classes, inheritance, and encapsulation principles.
β β β β± 4 min
Declarative problem solving
Contrast declarative (what to solve) and imperative (how to solve) approaches to problem modelling.
β β β β± 4 min
Artificial intelligence fundamentals
Learn core AI concepts including knowledge representation and intelligent agent design.
β β β β β± 5 min
AI search techniques
Explore uninformed and informed search methods for solving state-space AI problems.
β β β β β± 6 min
2. Common Pitfalls
Wrong move:
Confusing abstraction with just deleting information from a problem
Why:
Abstraction is intentional: it hides irrelevant details for a specific use case, not just random information
Correct move:
Always document which details you are hiding and why they are not needed for your solution
Wrong move:
Treating OOP and declarative approaches as interchangeable for all problem types
Why:
Each paradigm is optimized for specific problem structures, and using the wrong one leads to overcomplicated solutions
Correct move:
Evaluate the problem structure and requirements before selecting a problem-solving paradigm
Wrong move:
Thinking computational thinking only applies to coding problems
Why:
Computational thinking is a general problem-solving framework used across many disciplines beyond software development
Correct move:
Practice applying decomposition and abstraction to non-coding problems to build your intuition
3. Quick Reference Cheatsheet
Concept | Key Summary |
|---|---|
Problem Decomposition | Breaking a complex problem into smaller, independent solvable sub-problems |
Abstraction | Hiding unnecessary details to reduce complexity and focus on core problem requirements |
Pattern Recognition | Identifying shared characteristics across problems to reuse proven solution designs |
Object-Oriented Paradigm | Model problems around entities (objects) that combine state (data) and behavior (methods) |
Declarative Paradigm | Specify what problem to solve, not the step-by-step procedure to solve it |
Uninformed Search | AI search that uses no problem-specific heuristics to guide path selection |
Informed Search | AI search that uses heuristics to prioritize more promising solution paths |
What's Next
Start your learning with the foundational skill of problem decomposition, the first core pillar of computational thinking. Once you complete all sub-topics in this unit, you will be ready to apply these problem-solving principles to advanced algorithm design in the next unit of the syllabus.