My discussion of quantum mechanics interpretations stressed the point that the interpretation of a mathematical formalism cannot be tested experimentally and must be judged on other grounds. In this section I would like to make a related point: while interpretations cannot be directly tested, it is possible for experimental results to favor one interpretation or another.
For example, suppose that a new physical phenomenon were discovered which, in its physical interactions, was qualitatively different in its effects when observed by a conscious and intelligent observer than when not so observed. Such an experimental result would not prove the Copenhagen interpretation, but it would tend to corroborate it because at the interpretational level the CI employs the analogous process of state vector collapse by observers. The absence of such a phenomenon does not discredit the Copenhagen interpretation, but its discovery would lend the CI considerable support. This then is what might be called a corroborative experimental result. In this section similar possible corroborative experiments for the transactional interpretation will be discussed.
What corroborative results might bear on the transactional interpretation? Experiments concerning absorber deficiency at cosmic distance scales, detailed studies of the character of quantum randomness, or searches for physical effects arising from unconfirmed TI transactions would all bear on the transactional interpretation. Further, a definitive characteristic of the transactional interpretation is that it describes causality as arising from precariously balanced cancellations that nullify the occurence of advanced effects in quantum events. We can speculate that for sufficiently small distance scales or sufficiently short time scales this balance might fail and violations of microcausality might appear. The observation of such effects would then provide corroborative support for the transactional interpretation.
It is therefore interesting to note that evidence for microcausality violations in high energy electron scattering has recently been reported by Bennett[10,11]. He has reanalyzed data from of electron-proton scattering and shown that the data exhibits a statistically significant deviation from dispersion relations based on microcausality. He proposes a semiclassical model that is ``precausal'' in that it contains acausal terms corresponding to advanced effects, and he shows that with such a model he is able to fit the experimental data.
In our opinion it is too early to base conclusions about quantum mechanical interpretation on Bennett's interesting results. Before it is concluded that microcausality has failed the data should be carefully evaluated and if possible remeasured, and other possible explanations for the observed effects should be eliminated. In particular, it should be clearly demonstrated that the reported effect does not arise from a breakdown of local commutativity having its origins in the quark structure of the proton. In any case, this area of physics should be closely watched, for its implications for the foundations of physics could be very profound.