This is so wild. Honestly thank you for all the effort you put into this because wow 🤯 I am so freaked out by this😂 I remember talking about the whole mind reading for advertisements and tbh I was hoping you’d come back with a big nope. But 😅
Also, I wanted to add, I worked with several non verbal kids with autism in 2017 - 2018 and man do I wish they’d had telepathy😅 Would have made all of our lives much easier. What actually happened was we spent countless hours working on functional means of communication. And it is *hard* work, for them, for their families, for teachers and aides. There are a lot of tantrums and tears in reality. I had no idea this was a myth about autism, it’s so funny to me the way people perceive us.
But also as an autistic person, I wish I had telepathy… decoding normies is tough 😂
Yeah this was a wild one to work on, but I'm happy with how it came out! Admittedly a little less happy with how creepy this tech is and some of its speculative uses 😶
But yes, I completely understand what you mean about autism. As someone who worked as a behavioral tech before I started researching the topic and grad school, it's been very very weird to see this myth especially gain popularity over the last week or two. But thank you for sharing your perspective too, I appreciate hearing it!
Thanks again for reading and for giving me this idea! Always such great conversation with you!
I’ve saved all of these installments because they are so packed with interesting stuff I need a few reads to digest it all. I’m already looking forward to the next one!
Thank you for reading! This is an absurdly complicated topic, so read as many times as you need! I am also available for questions in the comments as needed!
I think it was some simple machine learning prediction training? Applied machine learning. Picking out features best suited for prediction of machine failure? Idk, maybe Bayes got around and I used something slightly different he tagged
Hi there! So sorry for not responding to this sooner. Not only was I busy, but I also took some extra time to think about this! With the caveat that this within itself could be its own essay, let's get into it!
I'd say you're broadly correct, but as one of my professors once said, you could pass a biology course with a C if you answered every question on every exam with 'it depends.' That always gave me a laugh, but it is genuinely true in a lot of cases!
I think part of what we need to discuss here is the forest vs the trees. Keep in mind, single neurons are fairly useless. It doesn't really matter if you have a single functional neuron, because one neuron has no feasible function without other neurons for it to interact with. We need that 'forest' of brain cells for anything to really function effectively.
Not only this, but while you're mostly correct about functions being unique *sometimes*, that isn't always the case. For example, we know what certain region of the brain do. The go to that most people know about is the amygdala (major part of the lymbic system), but we actually know even more precisely what different parts of the amygdala do. I for example specifically study the basal lateral amygdala (a very common area of study currently as it is closely involved with anxiety response). The BLA is very small in the grand scheme of things, but if we were looking at someone's brain for example and saw that the BLA was overactive, we can be fairly certain that this person is anxious. And for example, if we saw that the connection between the BLA and the praventricular nucleus of the hypothalamus (one of my favorite parts of the brain personally), was especially overactive, we'd know that this anxiety was social-oriented. Connections are extremely important.
single regions can't possibly doesn't address what is causing that area to be active. Because the brain itself isn't just isolated regions activating here and there. There are regions connected to each other, and those connections and interactions are what make up our conscious and unconscious experiences.
I feel like its important to remember that while yes the closer we zoom in, the less we fully understand, once you zoom in too far, we actually understand a lot and what we understand doesn't mean anything in a broader context. This is because we broadly know what regions do, we know how memories function, and we know where these things occur. I do think to 'cold read' from an individuals memories, things will start getting way more complicated than I think we're capable of at this moment technologically, but I do think it is possible! I'm not entirely confident they'll be perfect image reconstructions or interpretations (not in our lifetimes at least), but I do think there's a chance that it could eventually become that precise! Not just from several iterations from one brain, but just from getting even further in-depth understanding of memory coding and retrieval and where memories are coded, how the brain behaves when its retrieved, and interpreting what part of the brain is responsible for specific human experiences!
Now when we get into neurodevelopmental disorders (specifically anything that impacts neuron development), we could see things like decrease in neurons present, missing regions of the brain, decrease/increase in the size of those regions, neurons that are immature, or something else I probably can't think of. In these cases, the regions of the brain that are pretty well defined may not have as clear parameters as it otherwise would in a neurotypical brain. BUT those regions are still present (unless they're not) and they are still performing the same task (though maybe in a different way or not exactly as they typically would). In that case, I think it may be harder to make interpretations because the connection between regions, the activity of neurons, and the delivering of neurotransmitters can all be compromised or too far deviated from the average to be generalized by some sort of machine learning system.
All this to say... well, it really does depend.
This is without getting into the caveat that there are in fact different types of neurons in the brain (in terms of what they transport: dopaminergic, GABAergic, glutamtergic or on the end of cellular subtypes: pyramidal, unipolar, bipolar, psueodounipolar) and we can tell a lot about what a region does based on the type of neurons that are present in a region.
All this to say, you aren't wrong per say, but there's a lot of caveats and 'it depends' to it, that could overcomplicate things.
You've got the basics down, yes! There is of course more 'it depends' to it (because there always is), but that's it! Now admittedly, I don't know how colors specifically work (I've not read up on it), but I'm willing to bet there's some papers out there that specifically focus on this.
Also, it could be differences in region size, if the region developed, or even potentially it could be that those specific regions are not "functionally connected" as we'd expect them to be which something that can be explored using fMRIs is functional connectivity between brain regions that are 'communicating' with each other
This is so wild. Honestly thank you for all the effort you put into this because wow 🤯 I am so freaked out by this😂 I remember talking about the whole mind reading for advertisements and tbh I was hoping you’d come back with a big nope. But 😅
Also, I wanted to add, I worked with several non verbal kids with autism in 2017 - 2018 and man do I wish they’d had telepathy😅 Would have made all of our lives much easier. What actually happened was we spent countless hours working on functional means of communication. And it is *hard* work, for them, for their families, for teachers and aides. There are a lot of tantrums and tears in reality. I had no idea this was a myth about autism, it’s so funny to me the way people perceive us.
But also as an autistic person, I wish I had telepathy… decoding normies is tough 😂
Yeah this was a wild one to work on, but I'm happy with how it came out! Admittedly a little less happy with how creepy this tech is and some of its speculative uses 😶
But yes, I completely understand what you mean about autism. As someone who worked as a behavioral tech before I started researching the topic and grad school, it's been very very weird to see this myth especially gain popularity over the last week or two. But thank you for sharing your perspective too, I appreciate hearing it!
Thanks again for reading and for giving me this idea! Always such great conversation with you!
I’ve saved all of these installments because they are so packed with interesting stuff I need a few reads to digest it all. I’m already looking forward to the next one!
Looks like I'll need to start thinking about what the next one will be... 👀
This is fascinating. I think I have to read it again to fully absorb all if it. Way to dig deep.
Thank you for reading! This is an absurdly complicated topic, so read as many times as you need! I am also available for questions in the comments as needed!
This is so fascinating! Can’t wait to read your story!!
Wild stuff, right? This was such a trip to read up on!
Can't wait to share this Friday, it's already coming together...
what a fascinating article!! and man those pictures are haunting…
Right? The pictures developed are so creepy! And also very inspiring to me...
Nah it’s your proofreading brain saying that, I like your intro more
Me hating that I knew what Bayesian estimation was, used it once, and now totally have forgot🥲
Oh really? I'm curious, what did you use it for???
I think it was some simple machine learning prediction training? Applied machine learning. Picking out features best suited for prediction of machine failure? Idk, maybe Bayes got around and I used something slightly different he tagged
Hi there! So sorry for not responding to this sooner. Not only was I busy, but I also took some extra time to think about this! With the caveat that this within itself could be its own essay, let's get into it!
I'd say you're broadly correct, but as one of my professors once said, you could pass a biology course with a C if you answered every question on every exam with 'it depends.' That always gave me a laugh, but it is genuinely true in a lot of cases!
I think part of what we need to discuss here is the forest vs the trees. Keep in mind, single neurons are fairly useless. It doesn't really matter if you have a single functional neuron, because one neuron has no feasible function without other neurons for it to interact with. We need that 'forest' of brain cells for anything to really function effectively.
Not only this, but while you're mostly correct about functions being unique *sometimes*, that isn't always the case. For example, we know what certain region of the brain do. The go to that most people know about is the amygdala (major part of the lymbic system), but we actually know even more precisely what different parts of the amygdala do. I for example specifically study the basal lateral amygdala (a very common area of study currently as it is closely involved with anxiety response). The BLA is very small in the grand scheme of things, but if we were looking at someone's brain for example and saw that the BLA was overactive, we can be fairly certain that this person is anxious. And for example, if we saw that the connection between the BLA and the praventricular nucleus of the hypothalamus (one of my favorite parts of the brain personally), was especially overactive, we'd know that this anxiety was social-oriented. Connections are extremely important.
single regions can't possibly doesn't address what is causing that area to be active. Because the brain itself isn't just isolated regions activating here and there. There are regions connected to each other, and those connections and interactions are what make up our conscious and unconscious experiences.
I feel like its important to remember that while yes the closer we zoom in, the less we fully understand, once you zoom in too far, we actually understand a lot and what we understand doesn't mean anything in a broader context. This is because we broadly know what regions do, we know how memories function, and we know where these things occur. I do think to 'cold read' from an individuals memories, things will start getting way more complicated than I think we're capable of at this moment technologically, but I do think it is possible! I'm not entirely confident they'll be perfect image reconstructions or interpretations (not in our lifetimes at least), but I do think there's a chance that it could eventually become that precise! Not just from several iterations from one brain, but just from getting even further in-depth understanding of memory coding and retrieval and where memories are coded, how the brain behaves when its retrieved, and interpreting what part of the brain is responsible for specific human experiences!
Now when we get into neurodevelopmental disorders (specifically anything that impacts neuron development), we could see things like decrease in neurons present, missing regions of the brain, decrease/increase in the size of those regions, neurons that are immature, or something else I probably can't think of. In these cases, the regions of the brain that are pretty well defined may not have as clear parameters as it otherwise would in a neurotypical brain. BUT those regions are still present (unless they're not) and they are still performing the same task (though maybe in a different way or not exactly as they typically would). In that case, I think it may be harder to make interpretations because the connection between regions, the activity of neurons, and the delivering of neurotransmitters can all be compromised or too far deviated from the average to be generalized by some sort of machine learning system.
All this to say... well, it really does depend.
This is without getting into the caveat that there are in fact different types of neurons in the brain (in terms of what they transport: dopaminergic, GABAergic, glutamtergic or on the end of cellular subtypes: pyramidal, unipolar, bipolar, psueodounipolar) and we can tell a lot about what a region does based on the type of neurons that are present in a region.
All this to say, you aren't wrong per say, but there's a lot of caveats and 'it depends' to it, that could overcomplicate things.
I hope this made sense!
You've got the basics down, yes! There is of course more 'it depends' to it (because there always is), but that's it! Now admittedly, I don't know how colors specifically work (I've not read up on it), but I'm willing to bet there's some papers out there that specifically focus on this.
Also, it could be differences in region size, if the region developed, or even potentially it could be that those specific regions are not "functionally connected" as we'd expect them to be which something that can be explored using fMRIs is functional connectivity between brain regions that are 'communicating' with each other
But yes, you've got the idea down!