Changing Visual Perception of Motion with Neural Implants

 

Hey, 

Question: what am I doing right now?

Answer: I'm writing a blog post so I can explain to myself what I learned from reading this dry-ass paper: https://www.nature.com/articles/346174a0  

Let me tell ya, it was a slog to get through! And if I'm being honest, as I type this...I'm not completely through it yet. My strategy for reading dry literature like this is to copy and paste the text into a Word document and then have Word read it aloud to me. That helps push me through the paper. Otherwise, I spend 30 minutes reading the same paragraph over and over, absorbing none of it. Another thing that really helps is uploading the PDF to ChatGPT and asking it questions about what the paper is telling me as I have it read aloud to me. Should I be ashamed to admit how much assistance from technology I require just to read a few pages of scientific literature? Possibly.

Well, enough about my basic reading comprehension skills! On to monkey abuse in the name of science. 

So, in this paper from 1990, a group of researchers from Stanford used a brain implant to stimulate neurons in the middle temporal visual area of rhesus monkeys. 

Rhesus monkeys are actually the worst and I don't feel bad for them. I had to deal with them when I traveled to India. They will do all kinds of asshole things like steal your backpack and then you have to fight them for it and they have no qualms with biting.

The middle temporal visual area is indicated with the arrow in the image above. The V5 area is responsible for processing motion. V5 helps the brain detect and interpret the motion, speed and direction of objects in the visual field. This is where researchers put an electrode.

So, they stuck electrodes into the brains of these monkeys, right in the V5 region. Why did they chose the V5 region? Probably because neurons are organized in a columnar fashion in this region and neuronal columns with a similar functions are grouped together. This makes it easier to target specific neurons and reduces the chances of accidentally stimulating unintended neurons. 

The implanted electrodes had two purposes: recording electrical impulses and delivering electrical impulses.

After implantation, the monkeys were made to look at dots on a screen that were moving in different directions. The electrodes recorded which direction made the column of neurons fire. This direction was then dubbed the "preferred direction"  They trained monkeys with a "liquid reward" to look at one of two LEDs that corresponded to the direction of motion of dots on a screen. This is how they determined which direction the monkeys perceived the dots moving. 

Ok, so now that we know which direction of motion causes this particular column of neurons to fire, researchers sent little pulses of electricity in to stimulate that column of neurons while the monkeys looked a screen with dots. They found that there was a statistically significant increase in the number of times the monkeys perceived dots as moving in the preferred direction even when they were moving in the opposite direction, moving randomly, or not moving at all. 

SO, with a small amount of stimulation in a specific region of the brain, we can change the way monkeys interpret what they are seeing. At least, we can make them perceive dots on a screen as moving in a direction that they aren't moving. 

Let's brainstorm here, what are some useful applications for this information? Why would you want to perceive something as moving when it's not? 

Maybe it would have some cool VR applications. I could see it being useful as a training device for pilots learning how to maneuver space craft or something like that. Or maybe if we could figure out how to completely control perception of reality with neural stimulation, then we could just check out of the real world all together and live in a VR paradise.

Like, maybe someday our skulls could be full of slots to plug in different programs, with brains fully wired to stimulate any neuron. I suppose neural structure and response to each stimuli is unique to each brain so there would have to be a calibration session before you could actually enter your chosen VR universe. We could figure out which neurons fire when you experience the color green, or the smell of chocolate chip cookies, or the sensation of walking barefoot on the beach. But that calibration session is so time intensive, maybe the first few decades of your life is the calibration session for your neural implant and the whole goal of life is to experience as many stimuli as possible in order to map that specific perception onto a brain structure so that it could be artificially experienced later by stimulation. Maybe we are in the calibration session right now! Or maybe we are already in the VR session, calibrated from a previous life.

Maybe the useful application here is artificially regaining senses after they've been lost. However, in all these cases we'd need some kind of arbiter to determine the "actual reality" that's causing these neurons to fire. Like, if you were going to use stimulation to bypass vision, you would need a camera and maybe some kind of chip that could interpret the incoming signal. You would need something that can make decisions like "that's a yellow ball moving quickly to the right" and then convert that decision to electrical stimulation of specific neurons.

Ok that's enough nerding out for today!