Born Rule — Physica

Equivalent forms

P(a_n) = |\langle a_n|\psi\rangle|^2

\langle \hat{A} \rangle = \langle\psi|\hat{A}|\psi\rangle

One line converts deterministic wave evolution into the statistics of every quantum experiment ever performed.

Symbol	Name	SI unit	Dimension	Range
$P$	Probability densityoutput Probability per unit length of detecting the particle at x	1/m	L^-1	0 – 10
$psi$	Wavefunction amplitude Complex probability amplitude at position x (1D)	m^-1/2	L^-1/2	-3 – 3
$x$	Position Location where the detection probability is evaluated	m	L	0 – 1e-9

Unit systems

Symbol	SI	CGS	Imperial
$P$	1/m	1/cm	1/ft
$psi$	m^-1/2	cm^-1/2	ft^-1/2
$x$	m	cm	ft

Where it holds

Universal postulate of quantum mechanics: applies to any normalized state and any observable. Its interpretation is debated; its predictions have never been contradicted.

Dimensional analysis

P(x) dx ->

[L^-1]*[L] =

dimensionless

|psi|^2 dx ->

[L^-1]*[L] =

dimensionless

Both sides are pure probabilities.

\checkmark

Discovery

Max Born · 1926

In a 1926 paper on quantum collisions, Born proposed that the wavefunction gives probabilities. The crucial 'squared modulus' appeared as a footnote added in proof. The idea earned him the 1954 Nobel Prize and remains the operational core of quantum mechanics.

Try this

The wavefunction is everywhere — so where will the electron actually show up?

A wavefunction spreads smoothly through space, yet every detector click is a single sharp dot. The Born rule is the bridge: it turns a complex amplitude into the probability of each outcome, and it is the reason quantum mechanics predicts statistics rather than certainties.

Research status: stable

Real-world applications

Photon-detection statistics in quantum optics and lidar
Tunneling-current prediction in scanning tunneling microscopes
Readout statistics of superconducting and trapped-ion qubits
Electron-density maps in quantum chemistry (the rho of DFT is N|psi|^2 integrated)

Common misconceptions

psi itself is not a probability — it is a complex amplitude; only |psi|^2 is observable
The randomness is not due to ignorance of hidden variables — Bell tests rule out local hidden-variable accounts
Normalization is required: probabilities only sum to 1 if integral of |psi|^2 equals 1

Experimental verification

Single-particle double-slit experiments (electrons: Tonomura 1989) show individual random dots building up the |psi|^2 interference pattern one event at a time. Quantum-jump statistics in trapped ions and photon-counting in quantum optics match Born probabilities to extreme precision.

Derivation

The Born rule is a postulate, not a theorem of wave mechanics — but it is tightly constrained.

Gleason's theorem (1957) shows that in Hilbert spaces of dimension >

= 3

, the only consistent probability assignment over projection operators is

P = Tr(\rho * \Pi )

, which reduces to |<a|psi>|^2 for pure states.

Unitarity of Schrodinger evolution then guarantees total probability stays 1 for all time.

Limiting cases

psi is an eigenstate of the measured observable⟶ P(outcome

) = 1

Measurement of an eigenstate returns its eigenvalue with certainty — no randomness.

psi uniform over a region of width L⟶

P(x) = 1/L

Complete positional ignorance — every location equally likely.

N repeated measurements, N -> infinity⟶ histogram -> |psi|^2The law of large numbers makes the measured frequency converge to the Born probability.

What if…

What if probability were |psi| instead of |psi|^2?

Total probability would not be conserved under unitary Schrodinger evolution — the |.|^2 form is the unique norm that unitarity preserves.

What if you measure the same observable twice in a row?

The second measurement repeats the first result with probability 1 — the state has collapsed onto the measured eigenstate.

What if the state is not normalized?

Probabilities must be computed as |<a|psi>|^2 / <psi|psi> — otherwise they do not sum to one.

1

Two-level superposition probabilities

Given ·

c1:: 0.8
c2:: 0.6

Find · P(E1), P(E2)

Steps

$\psi = 0.8*\psi _1 + 0.6*\psi _2$ with | $0.8|^2 + |0.6|^2 = 1$
Born rule: $P(E_n) = |c_n|^2$
$P(E_{1}) = 0.64$ , $P(E_{2}) = 0.36$

Result ·

P(E_{1}) = |0.8|^2 = 0.64

,

P(E_{2}) = |0.6|^2 = 0.36

; they sum to 1.

2

Electron in a box: probability in the middle half

Given ·

n:: 1
region:: L/4 < x < 3L/4

Find · P(middle half)

Steps

$\psi _1(x) = \sqrt{2/L} * \sin (pi*x/L)$
$P =$ integral over [L/4, 3L/4] of (2/L)*sin^2(pi*x/L) dx
$= [x/L - \sin (2*pi*x/L)/(2*pi)]$ from L/4 to $3L/4 = 1/2 + 1/pi \sim 0.818$

Result ·

P = 1/2 + 1/pi \sim 0.818

— much more than the classical 0.5.

Time-Dependent Schrödinger Equation→Heisenberg Uncertainty Principle→Density Matrix→Particle in a 1D Box→