Ultimately a pressure curve is a mathematical function that takes input logical pressure (p) and returns an output logical pressure (p’)
that is it maps logical pressure to logical pressure. The logical pressure comes in as a value between zero and one, and the output is a logical pressure between zero and one. The specifics of the mapping of the input to the output are completely arbitrary and we can make pressure curves do whatever we want.
In reality we wouldn't just have one pressure curve function with a single input logical pressure parameter like that. More typically we'd have a pressure curve function that accepts multiple parameters. By tweaking these additional parameters we can control what the pressure curve is actually doing in a dynamic way.