Catwatch

A short observation exercise — follow one designated human cat across repeated interactions closely enough to tell what you actually saw from what you assumed

Due: Wed, 17 Mar 2027, 08:30 · Weight: 15%

Your fieldwork#

Watch your interaction with one human cat closely enough to tell the difference between what happened and what you assumed happened.

Catwatch asks you to do the thing Week 1 argues is hardest: observe before you respond. The task is not to solve the cat. You are not building a model yet, and you are not fixing anything. You are training yourself to notice how much of what feels like “reading” your human cat is actually invention — an inference dressed up as a fact.

What you hand in#

Your Catwatch follows one designated human cat across repeated interactions — the same cat you will carry into The Care Model and Know Your Cat. If you are not already supporting a human cat of your own, the course assigns you a case cat for the semester. This is fieldwork, not an essay: a log someone who was not there could read and tell what you actually saw from what you guessed. Every entry is tagged Observed, Inferred, or Unknown, and every Inferred entry says what it was inferred from.

Revisit the log at least once. Somewhere in it, a claim you first tagged Inferred should visibly change once later evidence gave you reason to doubt your first reading — downgraded to Unknown, or read differently. A log with no second thoughts anywhere in it is a log nobody has looked at twice.

Where this goes next#

Catwatch is not marked as an isolated skill, and it does not end your acquaintance with this human cat. What you notice here — which signals you kept misreading, which “obvious” facts turned out to be assumptions — is exactly the raw material The Care Model asks you to structure next.