This question is connected with the so-called wave-particle duality. We know experiments which can be explained only if we assume that the quantum objects are waves. However, are there experiments which can be explained only if we assume that the quantum objects are particles?
The fact that on a photographic plate a quantum object leaves a single spot, is not a convincing argument that we have to do with a particle. People can say that the wave-function collapsed.
But I consider another argument, as follows: let a source of, say, yellow photons, be open for 0.1 picoseconds. (This is the coherence time of a down-conversion photon.) Assume that the wave-packet contains only a few photons, and that we can record the time at which each photon impresses the detector with precision of femto-seconds. I think that we will find that some photons impressed the detector in the beginning of the wave-packet, others toward the end of the wave-packets, and most of the photons, in between. This is not a wave picture, it is a particle picture, i.e. that a photons has at a certain time a position better localized than the entire wave-packet.
However, for relying on this argument I need a datum: how long takes to a photon to impress a detector, e.g. to kick out an electron from an atom?