Does the viewer describe a deterministic or a probabilistic future?_

The question

If the future is already fixed, describing a future target should be no harder than describing something that exists now. If the future is genuinely open until it happens, describing it should be harder, because there is less of a settled fact to read.

This study put the two side by side to see whether the anomalous effect depends on time at all.

The design

One hundred ARV trials, split evenly. The present trials asked about binary facts about the world that were already true but unknown to the viewer. The future trials had two outcomes that would only be settled later by the results of mixed martial arts fights. A blind judge scored each description as a hit or a miss.

The result

Both conditions beat chance. The present was read correctly 88 per cent of the time, the future 62 per cent. Both were significant on their own (present p below .001, future p = .027), and the difference between them was also significant.

Reading the undecided future was possible, but reliably harder than reading the present. The authors read this as support for a probabilistic future.

Why it matters to us

This is the empirical basis for a design choice we make throughout the platform: we treat the future as a probability to be scored, not a fixed answer to be hit. It is why single sessions are never presented as proof and why our scoring is built around calibration over time.

The honest caveat is the authors' own. The design was not balanced, the present condition always ran before the future one, so part of the gap could be order, learning or fatigue rather than time itself. They say so plainly. It points the way; it does not close the question.