Hello, I came across Tom's experiments and then found out he's having them performed at the university I attended, so that peaked my interest. So I looked over the experiments shown here: https://cusac.eu/experiments/
And it seems to me that if the assumptions are correct, there would be some big implications and real-world applications here. For example, if the assumptions in experiment 4, and experiment 2B are correct (that you can predict a random number before it's chosen, and that the result can determine the injection input data before it's injected), then you should be able to do things like predict the winning lottery numbers before they are chosen.
So here is a way I thought of to scale up experiments 2B and 4, where if those assumptions are correct, it seems to me the same logic should allow the following to take place, and predict the lottery results.
Based on https://cusac.eu/experiment-2b/
Scaled up experiment to generate yet-to-be-chosen winning lottery numbers:
1. After a particle is fired, the result on the screen (diffraction pattern or particle pattern) is observed only by person A. R1 is inactive at this point, and the signal is suspended between D1/D2 and R1, and if R1 were to activate, it would store the result.
2. This process is repeated among additional separate apparatuses, where the binary results of diffraction pattern or particle pattern (0 or 1) is concatenated and converted to decimal value (representing lottery numbers). Person A buys a lottery ticket with these numbers.
3. A period of time later (it shouldn't matter how long), person B chooses any number they wish (the winning lottery numbers in this case), converts them to binary values (0 or 1) and injects these values at the w-w injection point, after the signal passes D1/D2 but before reaching R1, and R1 then stores the result; this is performed among each apparatus in the same order the particles were fired.
The prediction is the number that person B injected will match the result person A previously observed on the screen (0 for diffraction pattern, 1 for particle pattern). Therefore person A can predict what the future lottery numbers will be, since those are the numbers person B will have to inject in order to result in the pattern observed by the particle. Otherwise, it would be possible to know the "w-w data" the particle traveled if the result was a diffraction pattern for that particle - which contradicts particle-wave complementarity. So you may be wondering, what if person B chose to not enter the winning lottery numbers in order to try and contradict the initial output pattern? Well that would mean the results wouldn't have been the winning lottery numbers to begin with, since they have to match what person B injected
. So there is no possibility of "tricking" the system. That's why it's necessary for the knowledge of the results to not influence person B's injection. That also answers this question by Tom in experiment 2B:
An interesting variation: if the necessary timing can be achieved (roughly, particle travel time from slits to result screen + algorithm computation time + light speed time from R3 to R1) < delayed detector signal travel time from slits to w-w injector located just before R1), the experiment would be able to turn the w-w device on or off to directly contradict the algorithms prediction immediately after the prediction has been made but before the particle reaches the w-w device.
That would produce wave (diffraction) patterns with recorded w-w data available and particle patterns without any available w-w data recorded — both in direct conflict with particle-wave complementarity.
In other words, the output (whether a person knowing the results, or a machine having recorded the results) can't do anything that would influence the injection (input). A different entity with no knowledge of the output results would need to make the injection. Otherwise, if the injector had knowledge of the output, they would have knowledge of the w-w data which would mean the result would have been a particle pattern to begin with, in that case. This is the same concept as in the delayed choice quantum eraser experiment.
Alternative method based on https://cusac.eu/experiment-4/
1. After the particle is fired, the result on the screen (diffraction pattern or particle pattern) is observed only by person A. R1 and R2 are USB flash drives, they are turned on but their data is not analyzed.
2. Person B chooses whether to destroy the R1 and R2 USB flash drives (corresponding to a 0 or 1).
This method can be scaled up in the same way the first method was scaled up (using multiple machines to convert binary "0 or 1" to decimal numbers).
The predicted result is that if person B chose to destroy R1 and R2, this would correspond to a diffraction pattern of the particle that person A previously observed. Therefore person A can predict what person B's decision will be, since they will always correspond with one another.
This logic seems sound to me if Tom's initial assumptions are correct, and seems to be supported by the delayed choice quantum eraser experiment. From those assumptions, it shouldn't matter whether a random number is injected, or a predetermined number is injected, as long as the injector doesn't know the output result, since that would influence the injector's decision.
Would it make a difference whether a conscious entity or a machine knows which numbers are being injected? I don't think so, as long as they aren't influenced by, or have knowledge of the result screen data. Whether a robot or a human injects the numbers, it shouldn't make a difference if the numbers are random or are any numbers that the injector chooses as long as the result screen data isn't known to the injector at this point. I believe both cases should work, if the input determines the output and the output determines the input (shown in delayed choice quantum eraser), since they cannot contradict each other in order to preserve particle-wave complementarity.
What do you guys think?