Wave function collapse.

In QM, every quantum object has an associated wave function that is used to keep track of the possible states it could have due to the fact that its properties are not set until measured. Once measured, wave function collapse takes the object to a certain state. In fact, every system (even the whole universe) has a wave function. The many worlds interpretation is based on the idea that the universal wave function describes everything that could possibly happen.

Wave function collapse is one of the most controversial features on the Copenhagen interpretation, because it needs an observer to collapse the wave function and this leads to all kind of discussions about what an observer is, when the collapse actually happens or how it finally takes place (measurement problem).

But there's only one possible state for things moving at c, because they feel everything in an instant. Our idea of a superposition of states evolving in a wave of possibilities until we cause a sudden collapse in all space at the same time is absurd for the carriers, because they only see the factual realization of the instantaneous exchange. Every exchange is an agreement on the conditions that allowed the emission-detection connection to happen (much like the Relational Quantum Mechanics interpretation states). Wave functions just describe our own ignorance about an event or interaction we know “started”, but has not become real yet.

All classical analogies for explaining the quantum world consist of a “preparation” stage (hidden variables) and a “revealing” stage, where the prepared system evolves in time. That's why we can't simulate the quantum world classically, because, at that level, the preparation stage IS the revealing stage.

Take the Monty Hall problem as an analogy for wave function collapse. The probabilities for the outcomes evolve by how information is revealed during the show, but nothing really changes behind the scenes. That's how entangled carriers can beat Bell's inequality: they switch doors while we choose randomly. We are clueless players in a show where quantum entanglement sets the backstage each time.

Our spacetime descriptions always exclude part of the information the carriers have by living an instantaneous reality. We can only access information as it unfolds in time, so we model reality the same way. Collapse is needed because it reintroduces the part of evolution we left out by extrapolating behaviour only from emission point.

Carriers know nothing about things happening “before” or “later”, so thinking wave function collapse is a real process makes no sense. It only proves the limitations of our predictive models, as we're trying to describe final outcomes that have not happened yet while they're unfolding in spacetime.