Math ★★☆ Moderate

📍 Brouwer Fixed Point Theorem

Every continuous map from a disk to itself has at least one fixed point — a point that the map sends back to itself. Drag the target point or pick a preset to explore the vector field.

Map: Contraction
Fixed pts: 1
FP at: (0.00, 0.00)
Max |Δ|: -

Brouwer Fixed Point Theorem (1911)

Proved by L.E.J. Brouwer in 1911: every continuous function from a closed disk D² to itself has at least one fixed point — a point x where f(x) = x.

The visualization shows the displacement vector field v(x) = f(x) − x. A fixed point is exactly where v = 0 (no arrow). The theorem guarantees such a zero always exists inside the disk. The glowing star ✦ marks discovered fixed points.

Contraction: the entire disk contracts toward the draggable target — fixed point is exactly the target. Rotation: every point rotates by an angle that shrinks to zero at the centre — only the centre is fixed. Squeeze: x-axis contracted, y-axis stretched, yet the origin remains fixed. Twist: a vortex-like swirling that also pins the centre. Inversion: maps x → lerp(x, −x, t) — as t→1 the only fixed point is the origin.

Drag anywhere inside the disk while in Contraction mode to move the fixed point.

About Brouwer Fixed-Point Theorem

The Brouwer Fixed-Point Theorem, proved by Dutch mathematician L.E.J. Brouwer in 1910, states that any continuous function mapping a compact convex set to itself must have at least one fixed point — a point x such that f(x) = x. The intuitive version: if you stir a cup of coffee and then let it settle, at least one coffee particle must end up exactly where it started. More precisely, any continuous map from the closed unit disc (or any homeomorphic shape) to itself cannot move every point; there is always at least one that stays put.

The simulator visualises this by colouring each point in the disc according to whether it is mapped closer to or further from a fixed point, using a vector field approach that makes the fixed-point location visible as a "sink" or convergence point. You can apply different continuous mappings — rotations, compressions, shears — and see that none can avoid having a fixed point, while a mapping that is merely "nearly continuous" can avoid one if it tears or folds the domain.

Frequently Asked Questions

What exactly is a fixed point?

A fixed point of a function f is a point x in the domain such that f(x) = x — the function maps x to itself. For a map on the plane, it is a point that does not move under the transformation. Fixed points arise throughout mathematics and its applications: equilibria of differential equations, solutions to iterative algorithms (Newton's method, value iteration in dynamic programming), and steady states in economic models are all fixed points of appropriate maps.

Why must the domain be compact and convex?

Both conditions are essential. If the domain is not compact (e.g., the open disc, which excludes its boundary), a map can push every point toward the boundary without ever reaching a fixed point. If the domain is not convex — for example, an annulus (ring) — a rotation can move every point without fixed points. The standard counterexample: the map f(x) = x + 1 on the real line has no fixed points because the line is not compact (it is unbounded).

Does the theorem tell you where the fixed point is?

No — Brouwer's theorem is a pure existence result. It guarantees that at least one fixed point must exist somewhere in the domain, but gives no constructive algorithm for finding it. For numerical computation, Scarf's algorithm (1967) was the first practical method to approximate Brouwer fixed points by computing a combinatorial "triangulation" of the domain. This work later led directly to the development of homotopy continuation methods for solving systems of nonlinear equations.

What is the coffee-stirring analogy?

The classic analogy is stirring a cup of coffee: if you stir continuously (a continuous map) in any pattern and the coffee stays within the cup (maps the compact convex region to itself), then at least one molecule of coffee is in the same location after stirring as it was before — it is a fixed point. The analogy is vivid but slightly imprecise: real coffee molecules undergo thermal diffusion and the cup is three-dimensional, but the topological argument holds in any dimension.

Does the theorem hold in higher dimensions?

Yes — Brouwer's theorem holds for any continuous map from a compact convex subset of ℝⁿ to itself, for any dimension n. In 3D it applies to solid balls; in n dimensions to the n-ball. There is also a generalisation called the Kakutani Fixed-Point Theorem (1941) for set-valued (multi-valued) functions, which was used by John Nash in his proof of the existence of Nash equilibria in game theory, earning him the Nobel Memorial Prize in Economics in 1994.

What is the connection to game theory?

John Nash's 1950 proof that every finite game has a mixed-strategy Nash equilibrium relies on Kakutani's fixed-point theorem, a multi-valued extension of Brouwer's theorem. The "best-response correspondence" — which maps each strategy profile to the set of best responses — satisfies the Kakutani conditions, so it must have a fixed point, and that fixed point is a Nash equilibrium. This mathematical underpinning transformed economics and has been applied to auction design, evolutionary biology, and international trade theory.

Can the theorem be proved without algebraic topology?

The original proof used algebraic topology (specifically the non-existence of a continuous retraction from the disc to its boundary circle — equivalent to the fact that the fundamental group of the circle is non-trivial). In 1978, Hirsch gave an elementary calculus-based proof for smooth functions using the intermediate value theorem in higher dimensions. Combinatorial proofs via Sperner's lemma (a coloured triangulation argument) are also fully elementary and are often taught as a first introduction to topological fixed-point theory.

What is Sperner's lemma?

Sperner's lemma states that any "proper" 3-colouring of a triangulated triangle (where boundary vertices are coloured by specific rules) must contain at least one "rainbow" small triangle with all three colours. This purely combinatorial fact implies the Brouwer theorem in 2D: by taking finer and finer triangulations and applying Sperner's lemma, one constructs a sequence of nearly-fixed points that converges to an actual fixed point. The proof requires only compactness and the intermediate value theorem, making it accessible to undergraduates.

What happens if the function is not continuous?

If f is not continuous, the theorem can fail. A simple example: the function that maps every point of the disc except the centre to itself, and maps the centre to some other point, is discontinuous only at the centre but has no fixed point in some constructions. More dramatically, the disc with a hole in it is not compact (or not simply connected), and a rotation of the annulus has no fixed point. Continuity is crucial because it prevents the map from "jumping over" the fixed point.

What is the Banach fixed-point theorem, and how is it different?

The Banach fixed-point theorem (1922) is a stronger, constructive result that applies to contractive maps on complete metric spaces: if f is a contraction (i.e., |f(x)−f(y)| ≤ k|x−y| for some k < 1), then f has a unique fixed point, and iterating f from any starting point converges to it geometrically. Unlike Brouwer's theorem, Banach's theorem provides the fixed point explicitly via iteration and guarantees uniqueness. It is the mathematical basis for Newton's method, fractal image compression, and Picard's existence theorem for ODEs.