Syllabus Lesson 18 of 141 · Fixation on yourself
Fixation on yourself

The fixation study

Why it matters The moment you see an answer, it becomes the shape every later idea bends towards. If the first answer you see comes from an AI, your own ideas never get a turn.

In a study of designers generating ideas, one group worked with an AI image generator and the others without it (Wadinambiarachchi et al., CHI 2024). The group with AI help fixated more on the examples they were shown, and produced fewer ideas, with less variety and less originality, than the group without it.

Design fixation: sticking to the features of an example you have seen, even when you were asked for something different and even when the example has flaws.

The study was about visual design, not code, so treat it as a warning rather than a law for engineers. But the mechanism is familiar. Paste a problem into a chat, get a tidy answer, and the question in your head quietly changes from "how could this work?" to "is this answer fine?". The second question is much easier, and much narrower.

The habit this module trains

Before you ask, write down three approaches of your own. They do not need to be good. They need to be different from each other, so you have explored the space before anything fills it for you. For the next two lessons the tutor stays locked until your code assigns them:

my_approaches = [
    "precompute the answer nightly and store it",
    "keep a running count that updates on every write",
    "sample one order in ten and scale the result",
]

Recap Seeing an example early narrows what you come up with. Three different ideas of your own first, then ask.

Spotted a problem in this lesson? Report it · Ask about this lesson

This exercise opens soon

Finish the lesson before it and this one unlocks. The whole course is free, no account needed, and you can read this lesson right now.

Start at lesson 1 Back to syllabus