Hello community,
we’ve discussed Lisa Feldman Barrett a couple of times already. In my brain, Numenta’s ideas go very well with LFB’s ones. I liked the video format very much. The topic is an important one, as one can often see somewhat misguided machine learning / “AI” approaches to humans like sentiment analysis, trying to measure something based on assumptions (that emotion can be “detected” via statistical means).
enjoy: How Emotions are Made (cinematic lecture)
Cheers
Dmitry
Prof. Barrett relies on linguistic expressions of emotion.
Sylvan Tomkins and the “lie to me school” rely on the face as the home of the emotions.
mark solms and jaak panksepp search the brain.
Three approaches, each with something to add.
this is a misunderstanding of LFB’s work. Anywhere you look, you’ll see a central theme: emotions cannot be detected via any means, even linguistic, because these are constructed, based on the basic building blocks of prediction.
What you probably got from the more popular literature is the proposal to increase the emotional granularity by learning an enhanced vocabulary of emotions. This is not the basis for any theory, but rather a practical advice on how to change your predictions (categories).
Lisa Feldman-Barret’s research does not fall into the category of “detecting emotion”, thus the conclusion may be erroneous.
see:
Although, I must say, I don’t agree with any theory that comes from a specific author. I think, polishing and refining ideas should be a continuous process, and thus, professors and any other humans can err too.