On the second day of the Game, 23 October 2025, a post captioned SLEEP CYCLE TEST told the feed that the system had detected increased early-riser activity. Comment 8 if you woke before 8am, it said; we will verify with psychic powers. Late risers, report 404 and prepare for recalibration.

40,149 people answered. 37,768 wrote 8. 221 wrote 404. That is 171 early risers for every late one, from an audience that is, by every other measure we have, mostly young and mostly on their phones late at night.

This paper treats that number as what it is: not a measurement of when people wake, but a measurement of what people say when one answer is a badge and the other is a confession, and nobody can check.

The caption asked about sleep. The question this paper can answer is different: when a crowd is offered two unverifiable self-reports, one flattering and one penalised, what does the split look like? The post was not designed as an experiment, and the paper says so. It was the second transmission of a Game that did not yet know what it was. The count is still the count.

• Instruction: Comment 8 if you woke before 8am. Late risers, report 404.
• Where: one Instagram reel, 23 October 2025, the second post of the Game. It reached far beyond the account's following; 39,889 distinct people commented.
• How we counted: top-level comments from the comment archive. A comment counted as 8 if its text began with the digit 8 in any form (8, 8 with an emoji, 8 followed by anything). It counted as 404 if it began with 404. Everything else, including the 22 comments with no text, counted as other.
• A second signal: 284 comments mention being on their second, third, or nth video from the account. We counted those separately because they are a measurement of something else: how many times the feed had already shown these people the Game.

Survey researchers have a name for the tendency to give the answer that looks best: social desirability bias. The usual estimates of it come from comparing what people say with a record of what they did, and the gap is typically a matter of percentage points. This post has no record to compare against, which is exactly why the gap is not points but orders of magnitude. Nothing anchored the answer to the truth, so the answer floated to the better story.

The design of the options did most of the work. The early-riser answer came with a joke (psychic verification) and an identity (the system has detected you). The late-riser answer came with a word for failure and a threat of recalibration. Given that pair, 404 was not a report of a sleep schedule. It was volunteering for a penalty, and 221 people volunteered.

The second signal is the one we did not expect to find. 284 people wrote that this was their second or third video from the account. On day two, the feed was already serving the Game repeatedly to the same people, and the people noticed and said so. That is the earliest evidence in the record of what the Game would later build on: that the feed delivers a series, and a series makes a crowd.

• Nothing about sleep. No claim in this paper is about when anyone woke up. The instrument cannot see it.
• Not a designed experiment. The options were written to be funny, not to be balanced. This paper reads the result, it did not set it up.
• The 8 count is generous. Any comment beginning with 8 counts as the badge, including comments that go on to say the person did not wake early. A sample of the thread suggests those are rare; we have not counted them.
• The nth-video count is a keyword match. 284 is the floor. People said it in ways we did not search for.

The same question with the options swapped in tone: a badge for late risers and a penalty for early ones. If the split flips, the finding holds and the number is the pull of the badge. If the split stays near 94 percent, something about early rising itself is doing the work, and this paper is wrong about why.

The transmission this paper reads is one post. Its numbers on the platform, and what the comment archive holds of it, are below.

The Game opened the day before this post, filed as THE FLASH WARNING, ⟡011. The Sleep Cycle Test has no Chronicle entry of its own; it sits between that opening and THE INTRUSION, ⟡012, two days later.

The instruction as it went out: the reel frame beside the caption, captured from the operation's own signed-in account. What the paper counts is the answer to exactly this.

Native captures of the comments this paper leans on, taken from the operation's own signed-in account on 31 Aug 2026. Handles and like counts are as they stood that day; the numbers in the paper are the round-time count.