On 26 October 2025, four days into the Game, a post told the viewer they had been detected re-entering the feed less than five minutes after leaving it, and set a task: type the real reason you opened Instagram right now. The caption ended: you did not open the feed, the feed opened you.

3,419 people answered. Sorted by how often a reason appears, boredom comes first, then curiosity about what was new, then wanting to stay awake or put off sleep. Sorted by how many likes each reason drew, the order flips: delaying sleep first, the low moments second, and boredom near the bottom with fifteen likes across 579 comments.

This paper reads both orders. One is what people say. The other is what the room recognises itself in.

Asked for the real reason, what do people say they open the feed for, and which reasons does the crowd endorse? The post could not check anyone's honesty. It could only count what was said and what was liked.

• Instruction: Your task: type the real reason you opened Instagram right now.
• Where: one Instagram reel, 26 October 2025. 3,436 top-level comments in the archive, 3,429 with text, from 3,419 people.
• Categories: each comment was placed in the first matching category from a fixed list, in this order: a low moment (numb, escape, distract, cope, stress, sad, lonely); sleep (sleep, bed, awake, tired, night); boredom; a notification or someone sending something; work or school being avoided; habit or no reason; curiosity or the feed itself; the toilet; waiting or in transit; eating. The order matters: a comment that says bored and sad counts as a low moment.
• Other: the third of comments that matched none of these. Reading a sample, they are mostly memes, single words, and reasons phrased in ways the list did not catch.
• Likes: summed per category, as archived.

Boredom is the honest majority answer and it is nearly invisible in the likes: 579 comments, 15 likes between them. Nobody sees themselves in bored. Delaying sleep has fewer than half as many comments and nineteen times the likes, because a reader at 1am recognises it and taps. The like column is a mirror the comment column is not.

The low moments are 8 percent of the answers and a quarter of the likes. A comment section reached through the recommendation feed, at night, is where people say the quiet thing, and the room answers by agreeing rather than replying. That is a real function of the medium and worth naming without dressing it up.

For the Game, the post was proof of a premise: the feed does open you. Four days in, 3,419 people told a stranger why, and the reasons were sleep, boredom, and wanting to feel something else for a minute. Every later round was run on that audience at that hour.

• A third of answers are uncategorised. The list catches the common phrasings. Memes, single words and indirect answers fell into other, and the true share of every category is higher than shown.
• Categories are one-per-comment by priority. A comment that names sleep and boredom counts as sleep. A different order would move a few hundred comments between the first four rows without changing the shape.
• Likes follow position. Early comments are shown first and collect likes. The sleep answers were early; part of their like share is the platform, not the room.
• Self-report at the moment of the impulse. People wrote what they were willing to say right then. The honest reason and the written reason are not guaranteed to match.

The same question at a fixed local hour, using the check-in data to post when most of the territory is in its morning. If boredom stays first and sleep drops out of the likes, the night is the variable. If the shape holds, the reasons are the audience and not the hour.

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

The post has no Chronicle entry of its own. It sits in the first week, between THE INTRUSION, ⟡012, and THE COMPLETED, ⟡013.

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.