Below are the five Twine stories I created for the work in this article.

1. The prediction one: Here we successfully predicted people's final choice based on the event indices of their earlier choices.

(Section IV.2 in the article)

2. Influence attempt #1: We railroaded people through the beginning of the story, using specific paths designed to make them choose a particular ending according to what we had learned in the prediction study. But they didn't!

(Section IV.3.A, "no choices")

3. Influence attempt #2: Here we added some arbitrary choices throughout the story, in case the lack of interactivity was a confounding factor. But it still didn't work.

(Section IV.3.B, "unrelated choices")

4. Influence attempt #3: We re-engineered the story so that people were given choices about the key events that still wouldn't break our experimental design. I was really confident this was going to work, but once again, it did not!

(Section IV.3.C, "low agency choices")

5. Successful influence: Having finally really learned something about why our original prediction study worked, we made a slight adjustment to our experimental design that allowed us to give more "meaningful" choices for the key events. This time we successfully influenced people to choose our targeted ending by ensuring that it would remind them of their previous meaningful choices. Huzzah!

(Section IV.3.D, "higher agency choices")