Fall Research Expo 2020

How Reality Could be Otherwise: Counterfactual Thinking and Preference

View Poster

Humans often think about counterfactual possibilities of an experienced reality and imagine how it could be otherwise. Counterfactual thinking plays a key role in both individual and business decisions. However, there are infinite possibilities out there and we rarely think about all of them.

Here, we synthesize knowledge in memory and decision making research to specify the formation of counterfactual thoughts. We show that counterfactual thinking is highly influenced by semantic similarity and subjective preference. Not only are we more likely to imagine a possibility when it is more semantically similar to the reality, we are also more likely to imagine a possibility when it is more preferred.

PRESENTED BY
Feiyi Wang
MindCore
College of Arts & Sciences 2022
View Profile
Advised By
Sudeep Bhatia
Dr.
https://psychology.sas.upenn.edu/people/sudeep-bhatia
Ada Aka
View Poster(PDF)
PRESENTED BY
Feiyi Wang
MindCore
College of Arts & Sciences 2022
View Profile
Advised By
Sudeep Bhatia
Dr.
https://psychology.sas.upenn.edu/people/sudeep-bhatia
Ada Aka

Comments

Thank you for this intriguing study - I think you helped to explain some of my own tendencies with your determination that preferred scenarios are more easily imagined.

Good luck with your continued work,

Dr. Christine Muller (CURF)

Hi Feiyi! I like your project! I was wondering, do you think the results of the study would be different if the order of session 1 and session 2 was reversed? Maybe having session 1 first got people think about how the countries were similar as they rated where they'd want to work.

Thanks!

Abby

Yes, I think reverse the order will definitely change the results. In Session 1, we are trying to get at people's baseline preferences, namely, how much do people like to work in these countries on any random day. We intentionally separate Session 1 and 2 by a week because we don't want subjects to remember what they have filled out for the baseline and we only want their counterfactual ratings.

You're right that ethnicities would matter because people in general would know and like the places they live in or close to more than others. We haven't looked at difference across subjects because we have assumed that a large enough sample size would swarm the difference if the effects we found are significant enough. Hope that answers your question!

Thank you Dr. Muller! Some other studies also found that scenarios with higher perceived likelihood are more likely to be imagined. For example, it's hard for us to imagine running into a panda when driving in the countryside compared to a deer because we know that it's very unlikely that we'll see a panda roaming in North America.