Could you specify what you mean by "semantic validation"? Let’s say you have two different blocks, including a gain and loss block with positive and negative frames. In the gain block you give participants a text message in each trial like “you receive/lose $100”. Do you want to check if such a message is perfectly understandable by any participant?

Hi Rafal,

Apologize for delays in responding to you. You are almost right, however it goes beyond making sure that recipient of the framed message understands the message. I am wondering if there are guidelines/work that can tell if the choice of words in framing loss/gain will trigger loss aversion.

Regards,

Jay

DISCLAIMER: I'm neither a semanticist nor have I enough knowledge to provide an appropriate answer. That's just a try.

Hello, Jay!

If I were finding a way to validate if something is framed as gain or as loss, first thing I would recommend is to think about the "gain" and the "loss" frames. Maybe you can find something better at Frame Net, but for the little I know about frame semantics, an example of the gain frame would be:

Gain - An [operation] in which an [actor] gets a [positive outcome] in terms of [gained thing]

Operation - What the actor have done to gain something (i.e. a bet)

Actor - The entity or person doing the operation

Gained Thing - What the actor obtained a positive outcomes of (i.e. money)

Positive Outcome - An outcome in which the actor will have more of the Gained thing after the operation than before

For the "Loss" frame, I would define exactly the same, but replacing the 'positive' outcome with a 'negative' one (the actor will have less of the lost thing after the operation than before).

Thus, if I were checking if the frame of a message is a gain or a loss, I would check if the message has the elements in the frame. In Tversky & Kahneman (1979), for instance, the gain frames were questions in which the operation was a lottery with certain amount of chances of each outcomes. The actor was the person participating in the experiment (would YOU choose A or B?). The Gained Thing was Money (except in the travel problem - not sure if it was in Tversky & Kahneman (1979), though). And the positive outcomes were in the fact that the chances, in the prospects, were of the kind "p% chance of GAINING X".

On the other hand, the negative outcomes in the loss frame were in the fact that the chances, in the prospects, were of the kind "p% chance of LOSING X".

Anyway, if your messages are in English, take a minute to find for Gain/Loss frames at Frame Net. If you find a frame there, it's likely to be much more accurate than my examples. And of course, take my examples as just that: examples, of questionable accuracy (I'm just getting started in frame semantics).