It may be a stupid question, but why particularly for probability density expression $k~|\Psi|^2 = k~\Psi^{*}\Psi$, it's assumed that $k=1$? As it is now, then in a complex plane probability density is just a rectangular area for a complex vector. But why it has to be rectangular specifically? Why can't be $k=\pi$, so that re-defined probability density $\pi |\Psi|^2$ would mean a bounding circle area of complex vector:

enter image description here

Or any other complex plane area scaling value $k$? What would be implications of that to quantum mechanics?

  • 1
    $\begingroup$ I mean, you want your probability density function to be normalized so that the $\int_{\textrm{all space}}p(x)dx=1$. So, provided that $\Psi$ is normalized in such a way that the integral of its squared-modulus is 1, this must be the choice. It's really about our definitions of probability functions, not about quantum mechanics. $\endgroup$
    – march
    Feb 10, 2022 at 18:43

2 Answers 2


It's a normalization convention for $\Psi$ - indeed, the only sensible one. If the probability density is $k|\Psi|^2$, just absorb a $\sqrt{k}$ factor into $\Psi$. This density should not be interpreted as an area. Indeed, the real reason we square has nothing to do with $2$-dimensional geometry.


As it is now, then in a complex plane probability density is just a rectangular area for a complex vector.

I don't think it's useful to visualize $|\psi|^2$ as the area of the rectangle whose side lengths are $\mathcal{Re}[\psi]$ and $\mathcal{Im}[\psi]$.

In the standard formulation of quantum mechanics, the states$^\dagger$ of a system are represented as elements of a Hilbert space $\mathscr H$, and observable quantities are represented as self-adjoint linear operators on $\mathscr H$. The expected value of an observable $\hat A$ in the state $\psi$ is given by $$\mathbb E_\psi[\hat A]:= \frac{\langle \psi,\hat A\psi\rangle}{\Vert \psi \Vert^2}= \frac{\langle\psi,\hat A\psi\rangle}{\langle \psi,\psi\rangle}$$

To simplify calculations, it is convenient (but not necessary) to choose $\psi$ to be normalized, i.e. $\Vert \psi\Vert^2 = 1$. If we make this choice, the expected value of the position operator $\big(\hat X\psi\big)(x) = x \psi(x)$ is given by

$$\mathbb E_\psi[\hat X] = \langle \psi, \hat X \psi\rangle = \int \mathrm dx \ \psi^*(x) \cdot \big(x \psi(x)\big) = \int \mathrm dx \ x |\psi(x)|^2$$

Comparing with the expected value of a random variable from standard probability theory, we recognize $|\psi(x)|^2$ as the probability density corresponding to the position variable.

Finally, note that if we had not normalized $\psi$, so $\Vert \psi \Vert^2 = C^2 \neq 1$, then we would find the probability density to be given by $|\psi(x)|^2/C^2$. As a result, the fact that the probability density is given by $|\psi(x)|^2$ with no additional numerical factors is merely a result of our convenient choice of normalization.

$^\dagger$In actuality, this is true only for so-called pure states. There is a more general notion of state in which they are allowed to be mixed, but that is beyond the scope of this explanation.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.