Understanding User Interest through Emotions

elvis
DAIR.AI
Published in
3 min readJan 9, 2019
Abstract Illustration by Elvis — Intent

Does understanding the relation between motivation and emotions help to discover user interests and intent? Lucky for me, through investigating this question I was able to obtain a master’s degree in computer science. So here I will share a bit on what I learned about his topic.

As you can quickly tell, there is nothing that really connects “motivation”, “emotions”, and “intent” on the surface until you actually understand the psychology of interest.

Paul J. Silvia wrote a very famous book about the topic called “Exploring the Psychology of Interest” where he claims that in order for you to be interested in anything, you must first be motivated in the first place. For instance, in order for you to be interested in basketball or any sport for that matter, you must actually feel the need to want to see it in the first place. This initial condition is called motivation. So the other question is, what motivates a human being?

source

Although scientists, psychologists, and philosophers have argued about the existence, definition, and the very nature of emotions, they hold the answer to the question above. In his book, Silvia discusses the role of emotions and how they help to spark motivation.

Imagine a scenario where you are feeling excited about some event, say a “basketball game” in this case. Let’s not think of how you became excited for now — this could have been through a friend or just something you are anticipating. That first jolt of emotion you just experienced has motivated you to want to see the game. Without it, it’s hard to tell whether you would feel motivated to want to see the game.

In essence, emotion, in particular, positive ones, such as joy and excitement, do seem to play a role in motivating us to want to do something, in turn exposing our inner intents and interests. We know emotions continuously change over time, our interests convey the same pattern — very sporadic. One day you want this and the other day you don’t. And you respond, “because I don’t feel like it”. Who is to blame for this behavior? Well, now you know that it is somehow tied with your emotions.

This hypothesis still needs further exploration given that we still don’t fully understand emotions. I respect Silva’s viewpoint on the subject. I think it provides a nice intuitive explanation of how our intentions change every day and why they change.

In building computational systems that aim to understand user interests and intent, I could see how these ideas can be further explored given the right datasets and problem formulation. You can also observe the importance of reasoning, causality, and external information. More of this in a future post.

--

--