We have all seen the baffling two slit experiments with electrons in several forms:
A source of electrons (beta radiation, old school television, tesla coil, whatever) is facing a barrier that is impenetrable, with the exception of two narrow slits. Behind the barrier is a film that tells us where the electrons landed on it.
Result: Like previously seen in light, which is known to be a wave, after a period of exposure, we find that a diffraction pattern has appeared. QED, electrons diffract with one another as if they were a single wave.
Using the same slits and detector, we ensure that only one electron is emitted into the apparatus at a time, expecting that a single particle cannot diffract itself.
Result: While a single electron hits the detector forming a single dot, the position of the dots over many trials reproduce the very same interference pattern, leading us to believe that not only do electrons interfere as they travel together, but that a single electron is capable of self interference (a wave property) but is observed at an apparently random location whose probability is the very same interference pattern.
Even spookier, the final experiment is the same but adds a detector to each slit. This is to determine which slit the electron goes through in each trial. After all, an interference pattern can only happen if the electron wave goes through both.
Result: not only is the electron detected at only one slit, but no interference pattern is obsverved; as soon as it's measured, it acts as classical as a cannon ball.
This final point is a problem. The experiment's measurement affected the outcome, which essentially breaks science. In theory, all experiments have meaning only because we assume the system we observe evolves according to its intrinsic properties; an observation infers a model of the system's causality by casting its factors as necessary and sufficient to reproduce the observation. But, if the observation itself is necessary and sufficient to determine the outcome of the system, the veracity of all models is undermined because the independent variable we manipulated to probe the system is precisely observed. If our experiment measures the impact of a cannonball, we may have forced the system to behave like our ballistic model just by measuring the cannonball itself. The model can be verified regardless of the system's nature. We have ruled out the scientific axiom, there is no hope of understanding any system in a model if we can't observe an independent variable and assume the system will produce an outcome that reflects its nature.
Thus, my first question is why I am wrong. We continue to do experiments, construct models, and pretend that systems are thereby understood. We use models to predict the outcome of a system in its natural state when all we know about it is how it responds to arbitrarily selected conditions, that we may have forced our systems to involve cannonballs just by weighing them when they contain no natural regard for cannonballs. What is the source of this confidence after we know the results of these experiments?
In particular, my field of study was once biochemistry and neuroscience. We routinely added and subtracted cannonballs (I am talking about molecules and genetic codes and stuff, this is a metaphor) and measure their impact. We routinely concluded that the system actually involves cannonballs if adding them produced an impact (sufficiency) and subtracting them produced a state void of impacts (necessity). Why should this continue to work so well if the the system might behave totally differently as soon as we stop flinging well-measured cannonballs at it? Rest assured, we will never run out of cannonballs to bombard our system for the purpose of validating our ballistic models.
One experiment that would help ground our scientific axiom in reality again is the second component of my question. In the two slit experiments, we only concern ourselves with the electrons that reach the detector on the far side of the slits. In reality there must be some electrons that impact the barrier as well - it's not a barrier if no electrons are impeded by it. In that case, we may be measuring more about the barrier and the slits than the electron. In my experiment, the barrier is or is coated with the same electron-responsive material as the detecting plate. Like experiment 3, we have 2 results, but instead of a binary observation at the slits, we obtain 2 images, near and far. If this experiment in any of its forms has anything to do with the nature of electrons, the near and far images should be consistent with regard to the particle's wave or stochastic nature. I label the results as follows:
a. Near image is totally blank with no electron deposition.
b. Near image can be explained as the results of classical particles peppering the barrier at a uniform density according to the constraints of the electron source.
c. Near image cannot be explained classically but does appear to satisfy its term in the wave function.
d. Near image is not blank and appears to record single events concurrently with the far image i.e. one electron is recorded on both images.
e. Near image is not described by any of a through c.
f. Far image has no intereference pattern, resembling the result of experiment 3 above.
g. Far image has an interference pattern, resembling the results of experiments 1 and 2 above.
h. Far image records more events than can be explained by the random exclusion from the near image.
i. Far image is not described by f, g, or k.
j. Far image records fewer events than the number of emissions less events recorded on the near image, but it is not blank.
k. Far image is blank.
The combination of these results, denoted x & y where x and y meaning that both x and y are true of the result, ~ x meaning that x is false, and x , y meaning x or y or both, can be sorted into conclusions:
A: (c & f), (b & g), (k & ~a). It can be concluded that the nature of the event has been systematically altered by the application of observation itself, as the results are not consistent with either previous experimental results when all emitted electrons are accounted for. The system under study can't be reduced to an electron behaving on its own, but must include the observer as a causal factor. Because the experiments apply to electrons, which are observed everywhere, and are believed to be generally true of particles we observe and incorporate into models, all models reflect our interaction with a system and not necessarily the system itself. Science fails to deduce anything other than the effect of observing a system. A measurement cannot prove or refute a model's reality.
B: (a & ~k), (b & f). There is agreement about the particle nature of the electron in both images, or in some way the two slits can be assigned causality and the observer is ruled out. We have been studying an artifact of slits all along. Closely spaced voids form a wavy interfering path that free particles may follow, and electrons are free particles. In the case of a we should probably explain why they like such paths so much, but other than that, we can fix the standard model with math, probably.
C: c & g & ~(d, h, j) The universe can be waves all the way down, but at least they exist independent of observer. The cause of experiment 3 producing classical particles is not necessarily known, but we have ruled out observation by observing all electrons with an accurate model of the simple role of the barrier. The method used in experiment 3 should be examined carefully to catch the causal factor that differentiates it from experiment 2, as it appears to make the difference between quantized states and continuous fields.
D: d, h, j The barrier itself is not well understood if it allows an electron to interact with both images, so the previous two slit experiments can't be interpreted without knowing the effect of passing through the barrier, or propagating through the barrier in some way. Neither a piercing projectile nor an interfering wave including the barrier contradicts any of the experiments.
E: e, i Either the experiment has erred or we don't understand the electron at all. New models are required to explain the event that we currently designate as the presence or absence of an electron.
Has this experiment or something similar been planned or carried out? What are the results? What are the flaws in my analysis?