Variables as Named Boxes in Memory

Programs need names for values because code has to refer to data without knowing the exact number in advance. Instead of writing 95 everywhere a test score matters, you write score, and the program looks up whatever is currently stored there. This makes code readable and lets the same instructions work with different data each time they run.

When you write int score = 0, two things happen. The declaration int score tells the program to set aside a labeled spot in memory sized for an int, and the = 0 immediately stores the value 0 in that spot. If you then write score = 10, the program does not find a new spot. It goes to the same labeled location and replaces 0 with 10. Write score = 25 next, and 10 is gone, replaced by 25. The box has one compartment. A new assignment does not add to it, it overwrites it.

This is why a variable is not a history log. People new to this sometimes picture score as remembering 0, then 10, then 25, like a list. It does not. At any instant, a variable names exactly one storage location holding exactly one value, and that value is whatever was assigned most recently. This idea matters immediately once you start evaluating expressions and tracing method calls, because every time you see a variable's name, you need to know precisely what single value sits there right now, not what it used to hold.