Eigenspace Visualization
Run the Eigenspace Visualization Fullscreen
Edit the MicroSim with the p5.js editor
About This MicroSim
This 2D visualization demonstrates how eigenspaces are vector subspaces containing all eigenvectors for a given eigenvalue. Eigenvectors are shown as colored arrows, and their corresponding eigenspaces are displayed as semi-transparent lines extending through the origin.
Key Features:
- Multiple matrix examples: Browse through different 2×2 matrices with varying eigenspace structures
- Visual eigenspaces: Extended lines through the origin showing the full eigenspace
- Eigenvector arrows: Solid arrows showing unit eigenvector directions
- Transformed vectors: Optional dashed arrows showing how eigenvalues scale eigenvectors
- Matrix display: Current matrix A shown in the corner
- Color-coded legend: Each eigenvalue and eigenvector pair has a distinct color
How to Use
- Use Prev/Next buttons to browse through different matrix examples
- Toggle "Show Grid" to see or hide the coordinate grid
- Toggle "Show Transformed" to see how the matrix scales eigenvectors by their eigenvalues
- Examine the legend to see eigenvalue and eigenvector component values
Mathematical Background
The eigenspace for eigenvalue λ is defined as:
Key properties:
- Every eigenspace is a vector subspace passing through the origin
- For 2×2 matrices, eigenspaces are lines through the origin
- The eigenvalue λ determines how much vectors in that eigenspace are scaled
- If λ > 1, vectors are stretched; if 0 < λ < 1, vectors are compressed
- If λ < 0, vectors are reflected and scaled
Examples Included
-
Diagonal Matrix \(\begin{bmatrix} 2 & 0 \\ 0 & 3 \end{bmatrix}\): Eigenspaces along the x and y axes with λ=2 and λ=3
-
Symmetric Matrix \(\begin{bmatrix} 3 & 1 \\ 1 & 3 \end{bmatrix}\): Orthogonal eigenvectors along the diagonals with λ=4 and λ=2
-
Shear-like Matrix \(\begin{bmatrix} 2 & 1 \\ 0 & 2 \end{bmatrix}\): Repeated eigenvalue λ=2 with only one linearly independent eigenvector (defective matrix)
-
Reflection Matrix \(\begin{bmatrix} 0 & 1 \\ 1 & 0 \end{bmatrix}\): Reflection across y=x line with λ=1 and λ=-1
Understanding the Visualization
- Solid arrows: Unit eigenvectors showing the direction of each eigenspace
- Semi-transparent lines: The full eigenspace (all scalar multiples of the eigenvector)
- Dashed arrows (when enabled): Show Av = λv, demonstrating how the matrix scales eigenvectors
The dashed transformed vectors illustrate the fundamental eigenvector equation: when a matrix multiplies an eigenvector, it only scales the vector by the eigenvalue—it doesn't change its direction.
Embedding
1 2 | |
Lesson Plan
Learning Objectives
Students will be able to:
- Visualize eigenspaces as lines through the origin in 2D
- Understand how eigenvalues scale vectors within their eigenspaces
- Recognize the relationship between matrix structure and eigenvector directions
- Identify when matrices have orthogonal vs. non-orthogonal eigenvectors
Suggested Activities
- Predict eigenspaces: Given a diagonal matrix, predict what directions the eigenvectors will point
- Verify Av = λv: Use the "Show Transformed" option to verify that transformed eigenvectors are scalar multiples of the original
- Compare symmetric vs. non-symmetric: Note that symmetric matrices always have orthogonal eigenvectors
- Defective matrices: Observe how the shear-like matrix has only one eigenspace despite having algebraic multiplicity 2
Discussion Questions
- Why do diagonal matrices always have eigenvectors along the coordinate axes?
- What does it mean geometrically when an eigenvalue is negative?
- Why can't two different eigenspaces (for different eigenvalues) overlap except at the origin?
- What happens to a vector that is NOT an eigenvector when multiplied by the matrix?
Assessment Questions
- If a 2×2 matrix has eigenvalues λ₁=3 and λ₂=-1, describe what happens to eigenvectors in each eigenspace when multiplied by the matrix.
- Why does a symmetric matrix always have orthogonal eigenvectors?
- For the reflection matrix, explain why λ=1 corresponds to vectors on the line y=x and λ=-1 corresponds to vectors on the line y=-x.