Random Thoughts on Empathy and Compassion, and Neural Pathways

Earlier in my afternoon nap, I think I made the most interesting connection. It may well be wrong - my explanation is certifiably sloppy and unrigorous. It definitely needs a lot of refinement. Yes, this will be very bullshitty if you know anything about neuroscience or psychology. It also relies on some very unproved hypotheses of how brains work. I am mainly putting this brain dump down for myself anyway. But if you are a pop-sci author who wants to run with it, feel free to do so. I’d also be interested in investigating this in a more formal way.

There will be some things in this post that will be slightly controversial when compared to traditional norms. Thus I have provided a list of definitions on some specific terms I will be using in this post.


A passive emotion. Empathy means "to feel what another person is feeling".
An active emotion. Compassion means "to join in another person's pathos(suffering) and a compulsion to alleviate that suffering".
What and Where Neural Pathways
Parallel processing pathways in your brain. See Where, When, and How: Are they all sensorimotor? Towards a unified view of the dorsal pathway in vision and audition by Joseph Rasuschecker for a concise introduction.
The "what" pathway is basically the pathway that frames an object external to the body as the main reference - e.g. what something is with respect to the object itself.
The "where" pathway is basically the pathway that frames the body as the main reference - e.g. where something is with respect to the body.

The Idea

The connection I made is simply that empathy and compassion corresponds to the “where” and the “what” neural pathways are being activated respectively. The logical consequence is that empathy is fundamentally self-centred, while compassion is fundamentally other-centred.

Further Clarification

Now, I want to be clearer. When I say “empathy corresponds to the where neural pathways activating”, I do not mean that the “where” neural pathway is exclusively being activated. Most sensorimotor activations will activate both pathways to some degree *Barring some literal brain damage.... What I mean is that when you feel empathy for someone else’s emotions, the dominant pathway that is being activated is the “where” pathway.

Further clarification: the “what” and “where” pathways are well defined for sensorimotor inputs - e.g vision and audition. Emotions are traditionally not considered sensorimotor. However, there can be a good argument to be made that emotions are the results of simulations on the sensorimotor inputs - i.e. your brain is giving virtual inputs to these sensorimotor pathways.

Lastly, despite the fact that I described activations with an “order”, the order of activation isn’t really that well defined. It more or less happens at the same time. When I say “X activates Y pathway”, it doesn’t really flow causally like that. That sentence should be read to also mean “X relies on the activation of Y pathway”.

Thirst: Desire is a Sensorimotor System

Perhaps it would be clearer if I use emotions relating to desire. Desire is traditionally considered an emotion. However we can make a case that desire is a sensorimotor system (at least in the bigger picture than usually described for sensorimotor systems).

For example, let’s say I am thirsty We can reframe this as I desire water. Translated down to circuits, there is a sensor system somewhere in my body telling me I need to drink water. Now let’s say there’s a glass of water in front of me.

My vision sensory system sees a glass of water in front of me. The input from my eyes gets split into two pathways. The “what” pathway tells me that there is a cup of water. The “where” pathway tells me that there is something in front of me. Put together, my brain becomes conscious of the cup of water in front of me.

This activates the motor system which moves my hand towards the cup of water. As I move my hand towards the cup, the sensory systems on my skin, vision and other body parts keep activating the “what” and “where” pathways. This gives my brain a “real time” idea of how far my hand is from the cup.

This goes on until my desire for water is sated (e.g. I drink the water).

On Simulation, Movement and Motion

Movement doesn’t have to be physical. They can be virtual as well. Mirror neurones (which itself is part of the sensorimotor systems in our body) are the hot new things that are being studied. The science is not clear, but this is my understanding of how it might work:

If you see me touching a cup to lift it, your mirror neurones fire in place of the actual sensory and motor neurones in the sensorimotor system. If in a real situation your touch sensor system would fire if you touch a cup, then the mirror neurones will fire in place. As a result, your motor system may fire in response. It would seem that your brain is simulating touching the cup as well. Hence “mirror” neurones.

The simulation will also go through the “where” and “what” pathways. Some simulations would activate one pathway more dominantly than others. The “where” pathway can be thought of as framing the body as the main reference point. The “what” pathway can be thought of as framing of external objects as the main reference point.

The brain is also capable of simulating motor neurone activations. A very good example is athletes visualizing their workouts before actually doing it. This simulation activates the motor neurones, “warming” them up.

As such, it might be useful to define the very separate notions of “movement” and “motion”. Movement is the idea that your body (or parts thereof) has moved. It requires orienting everything else in relation to your body. Motion is also the idea that your body (or parts thereof) has moved. However, orienting for motion is in relation to everything else outside the body.

If you teleported suddenly from one place to another place, that’s a movement without motion. One way to frame it is everything else around you suddennly changed.

If you walk in place on a treadmill, that’s motion without movement. One way to frame it is that your feet are moving with respect to the treadmill belt.

Empathy is a Self-Centred Emotion

Empathy mostly activates/(relies on activation of) the “where” pathway. Empathy is fundamentally a self-centred emotion. This might sound a little strange given I have defined empathy above as “feeling what another person is feeling”.

Feeling is a qualia. You cannot really feel what other people are feeling. Your brain can simulate what other people are feeling.

My argument is that simulation of emotions require more activation of the “where” pathways. In order to simulate an emotion, you’d need a reference to your current state. Recall that the “where” pathway frames the body as the main point of reference. The body is the current state. So you would need more of that.

Compassion Requires Motion

Compassion by contrast is an active emotion. It requires active motion. Active motion - real or simulated - would rely on both “what” and “where” pathways. My argument is that the “what” pathway is dominant as it is the target. Information from the “where” pathway would be weighted less than information from the “what” pathway.

Thus one may think of compassion as a other-centred emotion.


The consequence of this random thought, is that television is responsible for increasing the levels of empathy in humanity. A corollary is that TV is also the cause of increased self-centredness in humanity.

Consider the medium of television. It is a very passive medium. You sit there and your vision and audition subsystem does most of the work. There is very little motor activation. When you move from scene to scene, your “where” pathways get activate to “reorient” your self. At the same time, the mirror neurones that fire in simulation of the emotions of others on screen. This further activaates the “where” pathways.

Per Hebbian learning, neurones that fire together wire together. So you’re learning to be more empathetic while watching television. The “what” pathways doesn’t really activate as much because no motion was achieved.

Contrast this with traditional compassion-training meditation. The instructions almost always introduce a notion of doing something. This allows the brain to simulate activation of the motor neurones.


This is a half-assed blog post about the framing of reference points with regards to emotions. This leads to some interesting ideas: that empathy is fundamentally self-centred, while compassion is fundamentally other-centred. These arise from the idea that empathy (relies on the activation of)/activates the “where” pathway while compassion activates/(relies on the activation of) the “what” pathways in our sensorimotor systems.

I make no further claims about these emotions (and probably should not given I’m not the best at expressing ideas - I don’t want to get cancelled for saying “empathy is not a great idea, folks”.

This is a very compelling pop-sci-ish post with a lot of plausible sounding things, but I make no claim on the veracity of it. It could be very wrong. However I think it might be interesting to entertain some of these ideas.

comments powered by Disqus