Matrices are usually taught as grids of numbers with odd multiplication rules. That's the hard way in. A matrix is an action on space: it moves every point, while keeping grid lines straight and evenly spaced.
- Try each preset, and read off the arrows.
- Choose "Flatten". The shaded square collapses, and the determinant goes to zero.
- Choose "Quarter turn". The dashed lines vanish. Why?
- Drag the arrows until they swap sides. What happens to the determinant?
Reading a matrix
The columns are just where the two basis arrows land. That fixes everything else, because any point is some of one arrow plus some of the other.
Determinant
The determinant is how much areas get scaled. If it's zero, space is flattened and you can't undo it, so the matrix has no inverse.
Eigenvectors
Most arrows get knocked off their line. But some special directions stay on their own line, and are only stretched. These are eigenvectors, and the stretch is the eigenvalue. They're the dashed lines above.
In quantum mechanics, every measurable quantity is a matrix-like operator, and the values you can actually measure are its eigenvalues. An atom's energy levels are eigenvalues. This lesson is why quantum energy comes in steps.
Problems
Apply \(\begin{pmatrix} 2 & 0 \\ 0 & 3 \end{pmatrix}\) to \((1, 1)\). What's the determinant?
Answer
\((2, 3)\). The determinant is 6, so areas grow 6 times.
Apply \(\begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}\) to \((3, 1)\). What did it do?
Answer
\((-1, 3)\): a quarter turn anticlockwise.
Why does \(\begin{pmatrix} 1 & 2 \\ 2 & 4 \end{pmatrix}\) have no inverse?
Answer
\(\det = 4 - 4 = 0\). Its columns point the same way, so everything lands on one line, and you can't get back.
Show that \((1, 1)\) and \((1, -1)\) are eigenvectors of \(\begin{pmatrix} 2 & 1 \\ 1 & 2 \end{pmatrix}\), and find their eigenvalues.
Answer
\(A(1, 1) = (3, 3)\), so \(\lambda = 3\). \(A(1, -1) = (1, -1)\), so \(\lambda = 1\).