4 ideas
In the mid-1980s in western Maine, Bernd Heinrich came upon ravens gathering noisily at a moose carcass. Their loud 'yells' seemed to invite competitors to a scarce winter food source. Over four winters he hauled carcasses of cows, goats, and deer through deep snow to bait sites, watched from blinds, wing-tagged birds, and kept captive ravens in aviaries to work out why a bird would advertise food it could keep for itself.
Raven 'sharing' is not altruism. Mated adult pairs hold territories and quietly defend carcasses against lone intruders, so hungry non-breeding juveniles cannot feed. Juveniles yell to recruit a crowd large enough to overwhelm the residents' defense, so each recruiter gets access it could not get alone. Cooperation emerges because a crowd breaks a monopoly, not because the birds want to share.
Juvenile ravens spend the night in shared roosts. Birds that found food the day before return and signal it, and at dawn naive birds follow them out in a group to the carcass. The roost pools scattered, unpredictable food discoveries across many individuals, so each bird's odds of eating on a given day go up.
Heinrich listed every plausible explanation for the yelling, including kin selection, reciprocity, predator dilution, showing off to mates, and overwhelming defenders. He then designed field and aviary tests that each predicted a different result, so outcomes could rule hypotheses out. Many experiments failed, gave ambiguous results, or were wrecked by weather and bird behavior, and the answer came from slowly narrowing options rather than from a single decisive test.