Ok you normalise the integral of psi^2 between the infinities (+ve and -ve)to be equal to one, i.e. you normalize so that the probability of the particle being somewhere between the origin and the infinities of all spatial dimensions is equal to one. Sensible no? Ok, if the wave function does not go to zero at the infinities then the probability of the particle being somwhere in space is equal to infinity. The reason for that is that if you integrate the probabilty density function over all space from infinity to infinty you get infinity, if the function does not go to zero at the infinities..., hehe, and the maximum possible value for a probabilty is 1... so the wavefuntion has to disappear at infinity.