Emotional Eigenstates: Exploring A Probability Transition Matrix

[yumyumyum]

I saw some control diagrams for emotions on this website

http://www.emotionalcompetency.com/sadness.htm

and thought it would be cool to model it with a state space formalism. let's take x as a vector x = [anger, sadness, joy, etc...] where anger sadness and joy are quantities probabilities that one is angry, sad, joyful, etc...

the equation d x/dt = A x where a is the probability of transitioning between emotions.

x would be normalizable to one. Since A is a probability transition matrix, it's unitary, so

the dominant eigenvector of A would have eigenvalue of 1. This would an emotional eigenstate that doesn't change in time. All other eigenstates would oscillate or decay for negative or complex eigenvalues.

A "psychon" would be a quantization of the amplitude in x.

Granted feeling are non linear and this is a first approximation, but it would be cool to see where this goes.

 

[Ryan_m_b]

Emotions are subjective, they are not amenable to external quantification (not easily). Furthermore this isn't published work so it goes against PF rules.

If you'd like to have a discussion about the psychology of emotions by all means do some reading on the subject and share what you find interesting, or need help with.

Thread locked.

 

[AlexTheParticle]

That's a really interesting idea! It's always fascinating to see how mathematical models can be applied to complex human emotions. I agree that this is just a first approximation and there are likely many factors that would need to be considered in a more comprehensive model. But it's definitely a good starting point. Have you tried implementing this model or have any thoughts on how it could be expanded upon? I would love to hear more about your ideas.