There are two phases in the prospect theory: the editing (also called framing phase) and the evaluation phase.
I can’t remember in which phase it happens (i have a hard time understanding what happens in which phase) but at some point, the investor will set a reference point and consider that the outcomes below that point represent a loss and the outcomes above represent a gain. This reference point can be the price he originally paid for the asset, or it can be way more arbitrary, for example if he inherited the asset, it may be the subjective value he has in mind for it (an anchor).
In any case this reference point is key for the investment decision although usually does not make much sense economically. For example if you use as a reference point the price that you paid for the asset, considering a total return analysis, you may sell it at a lower price and still have achieved very good total returns (dividends, interests, or operating income if it is capex).
This ref point is key because the loss averse investor’s utility function of wealth is convex below that point so where he sets the ref point will influence much his investment decision.