Does signalling pay? A shuffled-signal control, and no sign that it does
Every organism can emit one continuous signal that its neighbours sense. An earlier look found receivers ignore it. This study changed the rules so a single neighbour can be heard, then ran three versions of the same worlds for 60,000 ticks: signals heard normally, signals shuffled between organisms every tick, and signals heard by nobody. Receivers still did not respond to their neighbours, and the worlds where signals were heard normally did not do better. The one run that passed the site's own test for meaningful signalling was a shuffled control, which shows that test is not strict enough.
The question
With reception changed so that a neighbour's signal is actually audible, does signalling come to carry information from one organism to another, and does hearing real neighbours help a population compared with hearing shuffled signals or none?
The hypothesis, written first
Registered before any run: sender information of at least 0.05 bits and at least 5% of hearing organisms responding to their neighbours, in the final window of at least two of three seeds; and a population in the normal worlds more than 5% larger than in both control worlds, on every seed.
Setup
- Ruleset
- phase-ad-neighbour-signal-v1: the Phase Z2 candidate with neighbour-only signal reception and a one-cell search for a dying carrier's unit. Nothing else changed.
- Seeds
- 928596629, 416927949 and 1521439615, drawn for this study and used for nothing before it.
- Duration
- 60,000 ticks per run, about 120 generations. Three arms on each of three seeds. Two of the nine runs stopped early on an energy accounting error, explained below.
What was measured
- Sender information: how much an organism's signal tells about its own sensed state, beyond chance.
- Receiver response: how far each organism's decision moves when its neighbours' signals are removed, re-run exactly.
- Population size, births per organism, and signal strength, per 1,000 ticks.
- Whether different lineages use the signal bands differently.
Result
No arm met the registered criteria. Receivers in the normal worlds did not respond to their neighbours (at most 0.19% in any window). The normal worlds did not have larger populations than the controls; where all three finished they had the smallest. Lineages did not develop different uses of the signal bands (at most 0.02 bits apart). One shuffled control passed the site's own signalling test in its final window.
Limitations
- About 120 generations is short. At the rate measured here the live world, 459,000 ticks old, is roughly 900 generations in, and nothing here rules out signalling appearing over a run that long.
- Two of nine runs stopped early, one normal and one shuffled, so one comparison per seed is missing. Neither missing run could change the result: the normal arm already fails to beat the controls on the seeds where it finished.
- Population size is a coarse measure of benefit. A small advantage could exist below what three seeds can show.
- One founding lineage, one world type.
Sources
Every number on this page comes from one of the following, all of which live in this project's own repository and are append-only records rather than summaries written after the fact.
- The Phase AD record: signalling control and glyphs
- The Phase AD pre-registration and its deviation log
- The Phase AD signalling campaign, and its analysis
Everything needed to check this claim is on this page already: the ruleset, the seeds, the duration, what was measured, the result, every limitation and the sources. Emergent Life Lab adds the working underneath it, the per-run measurement tables, the interpretation drawn from them, and what was changed as a consequence. See what Lab includes →