I know this article is about google but I only had a good Apple anecdote so there you are.
83 karma · joined September 27, 2014
My wife HATES all her apple products but still won't switch. Similarly, while I love Macbooks, iPhones make me sick and developing for them is like Apple personally giving you (the dev) the middle finger every time there's an update.
I don't know the solution. Wish I did. Wish the internet was as free and wild as it used to be. But you could have a foghorn go off on an iphone every 3 seconds and people still wouldn't switch.
What I would honestly recommend doing is a bootcamp for full stack, or even just backend or frontend, or native, React Native, etc. Start there. You must learn the basics of code first. Not only that, but you will need to know how a non-academic team works. Github for starters. Also if anyone ever tells you that "elegant" code in one line is better than "readable" code in 10 lines, punch them for me. That is not how reality works in the workplace. Clear variable names. Clear function names. An understandable architecture for what is trying to be accomplished. These are the most important things, not reducing a perfectly understandable algorithm to something akin to calculus.
I actually made some of the first BCIs for the PC a long time ago, but already was a certified architect in C#, Microsoft, etc. These days you want to go for Node/JS or Go or Java, etc. Once you learn one language it is simple to learn another, particularly C-based languages (JS/TypeScript/C#/Java/ObjC/Swift and so on - notably they use brackets instead of indents or words to denote functions and objects.
I used to do AI work, but mainly medtech these days like digital therapeutics, and paradoxically I also do bank security (startup I was working with got bought by a bank, thus the disparity between the two. hey, you make friends along the way, help them out, and occasionally a hustle becomes your job)
I highly recommended becoming intimately aquatinted with databases, both relational and otherwise. And learn how to optimize them. You WILL have to work with a relational database at some point and thus you will need to learn SQL, indexing, etc.
Along the journey of my life, I've become quite the physicist and like to think I have a deeper understanding of math than most people. The only benefit I see to a degree is you might learn some of that. Physics, signal processing (from an EEG for example), and so on, are unavoidable. HOWEVER, there are a TON of libraries that will do all of that for you, and along the way you will learn it. I highly doubt anyone writes their own Fast Fourier Transform (FFT) these days, but to use it you have to understand how it works, what it's giving you, and what it's not, in order to avoid egregious mistakes. I've had to correct many people on what a Fourier transform actually does, many of them who have related doctorates. There are some truly stupid highly credentialed people in the world. IMO even the wikipedia article on it has some serious mistakes.
Honestly? Everything you need to know you can learn yourself online very cheaply through online courses. Get yourself a mentor or two. People LOVE teaching others. A bootcamp, however, will land you your first job.
As far as getting a job, when it comes to programming or engineering, "the proof is in the pudding". I've interviewed hundreds of programmers. At first I read all about their achievements at MIT and all of what they are experts in. But, in the end, EVERY INTERVIEW will consist of a live coding session. 90% of the people I interviewed could not write a solution to total softball questions like, "Alphabetize this text" or "Parse this JSON". I would even let them google and they still couldn't do it. I've had people right out of bootcamp that were better coders than the most highly credentialed CS graduates.
Which brings me to my final notes: 1) Live code as much as possible. Learn to love it. I failed the crap out of my first live coding interview, but faced my fear and now I'm the only one in the bank department I'm in that live codes. As I'm sure you can imagine, product and design LOVE IT. 2) FORGET ABOUT LEETCODE or any other such crap. If I was ever given a leetcode question in an interview I would tell them point blank that people just memorize the answers to these, they are not applicable to real-life application, and then I would walk out because that company is clearly not serious. 3) Whiteboard interviews are also going to happen. Super easy. Flow chart with how the frontend connects to the backend, branching enough into each until they know you know what you're talking about. 4) Don't forget the old school! If someone asks you what "PUT" means in a request and you don't know the answer, that's pretty embarrassing. Learn about basic networking, from old to new. 5) LEARN DEFENSIVE CODING: This will give you a leg up because I've never met a programmer who knew it who hadn't also been a hacker at some point in their life. Basically what this means is how to reduce vulnerabilities, but also how to straight up trick would-be hackers. For example, error messages. If you have an error message that says "License invalid", it will be ludicrously easy to find that string and thus where the license code is likely to be. Instead, screw with them. Say your program detects that a user has hacked into your app, if you simply cause a crash or exit that just gives them an easy way to find and remove the problem. Instead, perhaps there is a commonly used number in your app. Change it so they get weird data. Or maybe division of numbers is needed at some point. Change the divisor to 0 to cause a random crash. Some defensive coders will simply use a timer, set it to a random value, and then cause a crash in some way. Whatever you decide to do, use multiple defensive techniques, and always always screw with the hacker so they have no easy way to determine where to look to bypass your protections.
-------
Ok so I've overwhelmed you and you may be rethinking your decision. Don't. Jump right in! Learn as you go, keeping the above advice in mind, and you will succeed. Take a few courses on the basics of the actual reality of working in the private sector, so you know the basics of github, but make no mistake: show me a coder that says he doesn't google at least once a week, I'll show you a damn liar.
As far as aerospace, I've no clue. I have friends at NASA who are bored out of their minds, but the are all credentialed. Even so, the exactly University you go to barely matters, as far as I can see. Again, coding and the understanding of fundamental concepts is something that can and will be tested during an interview, and maybe even before the interview with a take-home project or a general phone call.
Above all, have confidence in yourself, and walk out of an interview that is stupid or has a bad culture. EVERYONE has imposter syndrome. Make THEM court you. I've straight up never gotten a job I wrote a cover letter for. Usually out of some company that is truly interested calls YOU. They can see your resume on LinkedIn like everyone else. Which reminds me, up your LinkedIn game big time. Yeah yeah everyone hates it but you start writing things, people invite you on podcasts, the snowball starts rolling.
Well that's all I've got. I hope it helps!
Good luck!
"He doesn't know what he is talking about. Means do not have to be a standard deviation away to make the result significant."
https://www.brain.fm/pdfs/EEGFocusAnalysis.pdf
It's pending publication, but it'll get there. And now we have enough info to start a double blind. Sleep one is also pending publication.
The HRV (stress/anxiety) trial is already published, and was completely independent. We didn't even know about it until it was published. There's more in the science section too.
I remember when that Lumosity BS came out. We'd been around for 4 years already at that point. Now that it's finally us taking off, we're pouring everything into the science. We're going to do this right.
(oh, sorry, I just realized I'd already replied to you above. I'll leave this here though, for anyone who's interested)
Yeah, I have to say it has not been a great experience here at HN. But, maybe I'll win them over yet. After 13 years, we're finally taking off, and the first thing I did was pour everything in the science, so we'll be there soon.
Well, enjoy Brain.fm! And feel free to drop me a line with an update any time.
- Adam
http://chetansurpur.com/blog/2013/08/emergent-intelligence.h...
https://en.wikipedia.org/wiki/Emergence#Computer_AI
We're not using that term anywhere in the marketing either.
This is just me talking here man. I'm not a marketer, or even really a good businessman. I did this for 13 years and we're just now taking off. I do music and programming, that's it. :)
But I do think Junaid did a great job on the site. Yeah there's more work to do, always is.
Tough crowd here at HN.
- Adam
Definitely don't want you to be wary of the skip button. That's an important feature. You should love the music you're playing.
So here's a pro-tip: as an anonymous user, play the session for just a little bit without moving the slider more than a minute or two. Click skip within that time period and you're golden.
The skip does have more leeway if you just register. It's free! Plus, as an anonymous user you're not going to get near as much variety. Every time you come back in incognito you'll get the same stuff. We have so many people doing this, but I don't know how they stand it! Truly though, I do not care if people go incognito.
But we just kinda need to know a bit about you to make sure you get the right sessions, instead of generic one-size-fits-all sessions. Yeah, I know how that sounds, as if I'm just dying to get your email or something, right? Use a fake one, who cares. I only need the survey information for one reason: Everyone's truly is - to quote Tyler Durden - a beautiful and unique snowflake ;)
It gets better and better afterwards and the more you use it.
Enjoy :)
- Adam from Brain.fm
This is a goldmine for me.
Thanks a lot for posting, and for trying it out. Would love to hear your feedback, or anyone else's from this thread. Email me any time. Just my first name and brain.fm.
Next week I go back to working on more music themes. We hired a new guy a month ago and he's ready to take over the app, so, this is perfect timing.
Let's see what me and the AI-guy can come up with. :)
- Adam from Brain.fm
Don't stop exercising though ;)
- Adam
Honest to god we just used a standard one we got over a decade ago, and customized it with our name. We might have even gotten it off of legal zoom or something.
- The HRV/(stress/anxiety) study is peer-reviewed (and independent, we had no part in it).
- All our other papers are pending publication.
- Giovanni's work on acoustic stimulation is peer reviewed, and from Northwestern. Again, we had no part in that.
- The meta-analysis one of our previous neuroscientists made is also peer reviewed.
- Dr. Olmstead's paper is peer reviewed, so's the Fatigue one we have up there, and all the other ones in our bibliography: https://www.brain.fm/pdfs/ResearchLibrary.pdf
- Hell, we're even having our statistical user analysis peer reviewed, the one we have up (before we got popular) and the new one we're making now.
Tryin' our best here :)
- Adam
Junaid my co-founder shares a lot of your taste, and so does my sister, so I have a few willing guinea pigs :)
- Adam from Brian.fm
I'm also not sure how technical you want or how much you know already, so apologies if I'm too technical, and apologies if I'm not technical enough. I'll just try to start from the beginning and lay everything out from there, as briefly as I can. Tragically, however, I was born without brevity.
1) Brain.fm: The Brain's Perspective
When a sound enters your ear, it's converted from a pressure wave to an electrical impulse, which you can measure in the brain as a kind of spike on an EEG, or a cortical response. If the sound plays again, another spike occurs (although typically smaller). If a sound keeps being repeated in a fast and precise enough way, the cortical responses start to resemble rhythms already in the brain.
For example, if you tap your finger rapidly on the table, you might be tapping at 10hz (10 taps a second). 10hz is also a very dominant in the brain. If you could tap your finger precisely enough, and for a long enough time, what you would see in the brain (if you know what you're doing) is new activity at 10hz. This is called the Frequency Following Response, and it's well established. Check it out on pubmed, etc, if you're interested.
Now if you keep tapping for even longer - and again, if you're super precise about it - your existing brainwave patterns near that same frequency will start to align themselves towards the phase and frequency of your tapping.
BTW sorry about the delay in responding to you. Believe it or not, in between some meetings I was searching for a good article this whole time to show you. Here it is: http://www.pnas.org/content/111/41/14935.full
This effect is called lots of things, including, confusingly, the frequency-following response by some in the private sector. It's most often known by the public as brainwave entrainment, but that's rarely in the literature. In the literature you see all kinds of words for it: neural entrainment, auditory driving, auditory entrainment, steady state responses of various types... The level of confusion over what to call this thing is only eclipsed by the number of names made up for methods to induce it - all referring to the same thing: rhythmic auditory stimuli. No, better not call it that, in this paper its gonna be sinusoidal amplitude modulation, click trains, acoustic iterance, isochronous auditory stimuli, temporal sound... So, it gets a bit crazy when you try to search for this stuff (so we'll be adding even more to our bibliography/research library soon. Let me know if you haven't seen that yet).
For the sake of this post let's just call it Entrainment.
Something that's also being figured out is how to define when exactly Frequency Following ends and Entrainment begins. Some say it doesn't matter, others say it's when the phase lines up, or if the new rhythm persists even when the external stimuli is removed, as in the article I linked you to. There's also ways to approximate the root source of a neural rhythm (external or internal), if you also feed the stimuli into the analysis at the same exact time.
Anyway, to sum it up, what we're doing is creating a rhythmic stimuli that resembles frequencies already in your brain. We play that stimuli long and consistently enough to actually register a pretty solid response on an EEG.
Of course, it gets more complicated than that. This is the brain we're talking about. The waveform of the audio modulations matters (brainwaves aren't perfect sine waves, after all), the way the frequency is changed over time matters. When do you take a break from the stimulation? When do you tone it down, or turn it up? And how do you adapt to people on an individual level? How do you measure it? Audio processing can change dramatically in the brain depending on the content of the audio!
But at any rate, changing the brainwaves of a person like this can also change their mental state. I'm not saying it takes over the brain. It just.. nudges it a bit in the right direction.
So that's the basic outline from that perspective. Hope it makes sense, of course if you have any questions don't hesitate.
2) Brain.fm: The Music Perspective
From an audio's point of view, we're doing a number of things to subtly create that necessary rhythmic stimulus.
a) We're modulating every piece of audio. It's these modulations that create a rapid rhythmic stimulus. If you listen carefully to Brain.fm, you may hear a fluttering sound. It's more obvious in some parts over others. We're using filtering techniques to target some frequency bands over others (and by frequency here, I'm now talking about the frequencies of pitch, not brainwaves, where a cello would be low and a violin would be high in frequency). Doing this, we can single out instruments, modulate parts of a nature soundscape but not others, or disguise the modulations as vibrator, tremolo, or the natural vibrations of instruments, strings or the LFOs of electronic music..
b) We align everything else in the session to the brain-frequency we're trying to hit, so there is no interference from a stray drum beat or a snapping twig. Not only is there no interference, but the entire audio "scene" is being actively used to guide your brain toward the goal state.
The 3D audio also factors into this a lot, because the mind pays attention to motion, and that helps elliminate habituation. But that's another story.
Hope this helped explain it. Completely understand if you need clarification, or an entirely new answer hah!
:)
- Adam from Brain.fm
But I actually like it like that. I want a double blind study with identical-sounding placebo and more subjects. After that we can study how it compares to regular music commonly used as study-aids.
Regarding the statistical analysis, result interpretation, etc, I feel I should let Giovanni himself defend his paper. He's dealing with family / health issues right now, but I hope to get him on here as soon as possible. I promise he will respond eventually and clear this up.
I apologize for the delay, he's dealing with some family and health problems right now. But I promise we'll clear this up.
Not to add to your expenses, but definitely invest in some "sleep phones." I like the Took's brand, and it's the cheapest. You can lay completely on your side and the wafer-thin speakers don't bother you at all. Pretty decent sound quality too.
Sleep-phones are unfortunately essential to get the most out of the session, especially as an insomniac. (to be clear, I have no affiliation with any headphone brand)
I'm a lifelong, severe insomniac. Have restless leg too, though minor in my case. It was a major frustration for 11 years that I couldn't truly help my own insomnia with the audio stuff I was doing. There were protocols others had used/recommended, but they were pretty limited in their effectiveness. I kept experimenting with things, testing mostly on myself. The real breakthrough came when we perfected our use of HRTF ("3D" audio).
My problem was mostly sleep-onset insomnia, and this enabled me to create an interesting, constantly-moving 3D scene that was able to dissociate me from my daily worries and chattery mind. If you notice, the sounds rotate around your head. Our neuroscientist Giovanni Santostasi hypothesizes that one of the reasons the thing works so well is because it mimics the effect of rocking in a cradle or a hammock, since the ears do somewhat have to do with balance (he mentioned a study in which they analyzed the effects of rocking and it enhanced the quality of sleep from an EEG perspective - can ask him for it if you want).
After that it was a matter of finding relaxing ways to implement the stimulation. Giovanni is an expert on what's called slow-wave sleep. It's the deepest stage of sleep, and is where memory is largely consolidated. His work at Northwestern is almost exclusively developing audio stimulation to enhance slow-wave sleep. They recently got a ridiculous large grant from DARPA for it. Look him up if you're interested.
So, after some talk, I created a protocol to stimulate slow waves that also ended up involving the 3D audio.
It took some tweaking, lots and lots of testing, but the end results were... staggering. Frankly unbelievable.
The first time Giovanni did a sleep study on me with this new protocol, he thought I had faked the EEG data somehow, and I don't blame him. It was frankly more likely that I had somehow generated fake EEG-like microvolts for 8 hours than get the result that we did. Then he tried it on himself and got nearly the same result, which was a surprise because he can sleep standing up if he's tired enough.
What he saw on the EEG was that the sleep protocol enhances slow-wave oscillations by 20-30%. To put this in perspective, if you decide to pull and all-nighter tonight, tomorrow when you sleep, you'll see a slow-wave gain of only about 10% to compensate.
It's an insane result. Absolutely ridiculous.
But the thing is, the result is clear as day if you do a sleep study. We confirmed and confirmed and confirmed it. Take a look at the sleep study analysis on the website. I know it's hard to parse - we need to convert it into non-giovanni-speak. But look at those correlations, and how consistent the result of the stimulation is between subjects...
Not only this, but we're showing an increase in what are called Spindles, which are highly associated with memory consolidation. It's very likely that we are also increasing memory consolidation with this, in the same way that his department at NW is showing that using a phase-lock-loop auditory stimulus enhances memory consolidation.
I know you didn't ask any of this, it's probably boring. I'm just very passionate about sleep. Have a lot more to say about it in general but will stop here :)
To sum it up: What I'm getting at is that even if you weren't an insomniac, you can still very much benefit from using the sleep sessions!
AND I will bet you that if you get some special headphones that allow more freedom, it will help you get to sleep as well.
Please keep me updated. Message me or find me on brain.fm. Would love to be updated on your progress. Tell you what, if you commit to getting some sleep-phones, I'll give you the price of the headphone's worth of free Brain.fm time to test this out. I'm just curious now, and love helping a fellow insomniac. :)
- Adam from Brain.fm
heh, "emergent." Sounds like I'm asking you look up the latest Hunger Games knockoff.
There is discussion about extending the trial, and I agree completely.
And we're almost done with both the iPhone app and the Android app! Stay tuned :)
Good point about Spotify. I could have sworn I pay $10 for it. Around $8 for Netflix, unless I'm mistaken. At any rate, I try to stay out of the business processes that go into making those decisions, for my own sanity, with the exception of occasionally giving discounts to my 10 twitter followers ;) haha. I've relayed your feedback to Junaid my co-founder though. Thanks again.
It's definitely not meant to replace musicians/composers. Actually I still work with musicians/composers to create tiny bits of sound (individual notes, etc) and stems for the AI to work with. More importantly, I'll work them to help me figure out how I can improve the AI for certain genres. Indie music / folk, for example, are particularly difficult. Rock, near impossible, so far. Could be the limitation of the stereotypical 3:05 long song, but then I think, Mike Oldfield.. haha. Fun challenges!
I would LOVE to make a tool for musicians though. The 3D-audio stuff alone could be pretty useful I think. If I can do some cool stuff with it, I'd love to see what other, more creative people can do. I've never seen it done this way. There's some VST plugins, but they're not that great. Of course there's a ton of binaural microphone recordings, like the barbershop. But it's a whole different level to be able to move thousands of generated sounds around in 3D space.
Maybe sometime in the future when things calm down :)
Here's the Terms Of Service for Yale:
I AM RESPONSIBLE FOR REVIEWING ANY WRITTEN CONFIRMATION STATEMENTS PROVIDED TO ME (ON PAPER OR IN ELECTRONIC FORM) REGARDING ANY INSTRUCTIONS, CHOICES, OR REQUESTS THAT I MAKE THROUGH THIS WEB SITE. I AM RESPONSIBLE FOR REVIEWING FOR INACCURACIES ALL OF THE INFORMATION TRANSMITTED TO ME. IF THERE ARE INACCURACIES, I AM RESPONSIBLE FOR CORRECTING THEM USING THE TOOLS AVAILABLE TO ME ON THE WEB SITE OR BY CONTACTING THE BENEFITS CENTER TO POINT THEM OUT.
Here's Amazon:
AMAZON ATTEMPTS TO BE AS ACCURATE AS POSSIBLE. HOWEVER, AMAZON DOES NOT WARRANT THAT PRODUCT DESCRIPTIONS OR OTHER CONTENT OF ANY AMAZON SERVICE IS ACCURATE, COMPLETE, RELIABLE, CURRENT, OR ERROR-FREE. IF A PRODUCT OFFERED BY AMAZON ITSELF IS NOT AS DESCRIBED, YOUR SOLE REMEDY IS TO RETURN IT IN UNUSED CONDITION.
And here's Pfizer:
PFIZER MAKES NO WARRANTIES OR REPRESENTATIONS OF ANY KIND AS TO THE ACCURACY, CURRENCY, OR COMPLETENESS OF ANY INFORMATION CONTAINED IN SUCH WEB SITES AND SHALL HAVE NO LIABILITY FOR ANY DAMAGES OR INJURIES OF ANY KIND ARISING FROM SUCH CONTENT OR INFORMATION.
I mean, I can talk to my lawyer about changing it if you really think it's that bad. :)
haha, I know I shouldn't be surprised by now, now it's still funny to me that I am sending a heartfelt thanks to a one "slagfart" :)
1) Research:
I responded to your other comment with some research links. Here they are again just in case: https://www.brain.fm/pdfs/ResearchLibrary.pdf https://www.brain.fm/pdfs/ElioConteHRVandBWE.pdf https://www.brain.fm/pdfs/EEGFocusAnalysis.pdf https://www.brain.fm/pdfs/EEGSleepAnalysis.pdf More on our site as well, and we are working on the bibliography still. Some of it won't relate directly to your questions, such as the slow-wave stuff, because it's meant to be a general reference for what we're doing as a whole.
2) Pretty Charts:
The pretty charts you see are direct copies of the EEG analysis by Giovanni. We did try to make them pretty, but if you're interested in the originals check out the analyses Giovanni did. He'll be on at some point to explain them further if you want to direct some questions to him directly.
3) Your study procedure recommendation
Regarding your study procedure recommendation, that's a great idea. We did something very similar to that in the Focus analysis above. 3 conditions: No Music - Music Without Stimulation (placebo) - Music With Stimulation. To the untrained ear, the music was indistinguishable. Most people can't tell which is which, and the feedback we got from the subjects confirmed it. It was actually really interesting: some of them asked to be able to keep the placebo-music (not calling it that, obviously), saying they worked better on it, when their results often told a different story.
Next study will be double blind and have more subjects. Giovanni feels we have enough information now to do it right.
So, someone above asked about the AI thing and I'm happy to explain it. I'll paste a lot of that below. Although I have to say I'm not sure what textbooks are calling anything these days. When I talk to programmers right out of college sometimes there's a bit of confusion as they talk about patterns I know, but use different names. I grew up programming on an 8088, and I've been working for myself for 13 years, mostly by myself, so I'm sure I'm behind on a lot of things.
Having said that, I'd probably call what I'm doing "Emergent" because there are many competing little pieces. Let's see if I can explain it quickly here, though please don't be too hard on me I'm trying my best to answer a lot of questions, but I'm really happy to go more into later after all this dies down :) Conceptually, you could think of it as first creating what I'll call "song-bot", which acts a kind of overlord, and has some instructions from me (like maybe there's some specific chords I want it to use, keys to avoid, tempo, brainwave protocols, genre, etc, etc). This song-bot guy then spawns a bunch of little other bots that compete with each other for the right to play/fade in/or generally be a part of the final result. These little guys have different characteristics, like a "drum-bot" might has different places it "wants" to be placed, and so generally competes with other drums, but not always. Sometimes I'll even have little "bots" for individual notes of a certain instrument. They obey certain rules of course, to form a background, drum line, or a simple melody, and then they pass that information along to subsequent incarnations (there's some learning involved in that process, though I hesitate to give it any textbook term - man, you have me terrified here of defining something wrong! :) ). Through that learning, the pattern the original little guys made has more weight and will tend to repeat. But again, not always.
After a while, out of all these smaller, simple processes form a very complex song. Complexity "emerges." So I call it a kind of "emergent intelligence."
The resulting song can be quite complex and varied. Of course there's more to it, because songs have sections and clever variance, but all in all the genres I'm using are very structured. Techno especially, is very easy for the AI. Sometimes I'm tasked with creating a theme in a particularly difficult genre, and that's when it gets really tricky/fun (such as some of the Indian ones, which were labors of love). In these cases, I may have to re-rerun many "generations" of the AI through the same song from beginning to end, with different parameters/instruments, but with the same patterns/learning, because what can happen in these cases is that it starts out simple, and increases in complexity as it goes, so the start can sometimes be a bit boring.
Also keep in mind this is just a conceptual explanation. The code is much less amusing. Thinking of naming processes "bots" just for fun though.
I hope I explained it OK. Listen, I'm not saying it's Watson or anything, or even that I'm following the definition of "emergent intelligence." I didn't go to a textbook to make this, it's really just a necessary step I had to take in order to make the computer do what I wanted it to do. I tried out some different ideas, and this one worked the best, in my opinion at least.
- Adam from Brain.fm
Here's a bibliography we're still working on. More to add, but a good start so far: https://www.brain.fm/pdfs/ResearchLibrary.pdf
Here's an independent peer reviewed study on HRV using our tech: https://www.brain.fm/pdfs/ElioConteHRVandBWE.pdf
We're planning on following that one up with a more robust study, because our users do get great results with HRV and it's an interesting topic.
One of our former neuroscientists published a well received meta analysis in a journal, but it's a bit outdated now re: what we're doing. Still, it's on our site if you're interested.
And then of course here's our own research that is pending publication:
https://www.brain.fm/pdfs/EEGFocusAnalysis.pdf https://www.brain.fm/pdfs/EEGSleepAnalysis.pdf
We also do regular analysis of user progress with a few statisticians I work with, and those will be published as well. Sorry, just takes time to get these things ready and through to a journal. Trying our best though. :)
In a podcast called College Info Geek and recently in I did a talk at Northwestern (The Garage) and I called what I'm doing "Emergent" because there are many competing little pieces. Let's see if I can explain it quickly here, though please don't be too hard on me I'm trying my best to answer a lot of questions, but I'm really happy to go more into later after all this dies down :) Conceptually, you could think of it as first creating what I'll call "song-bot", which acts a kind of overlord, and has some instructions from me (like maybe there's some specific chords I want it to use, keys to avoid, tempo, brainwave protocols, genre, etc, etc). This song-bot guy then spawns a bunch of little other bots that compete with each other for the right to play/fade in/or generally be a part of the final result. These little guys have different characteristics, like a "drum-bot" might has different places it "wants" to be placed, and so generally competes with other drums, but not always. Sometimes I'll even have little "bots" for individual notes of a certain instrument. They obey certain rules of course, to form a background, drum line, or a simple melody, and then they pass that information along to subsequent incarnations (there's some learning involved in that process, though I hesitate to give it any textbook term - man, you have me terrified here of defining something wrong! :) ). Through that learning, the pattern the original little guys made has more weight and will tend to repeat. But again, not always.
After a while, a song... emerges. A kind of "emergent intelligence."
The resulting song can be quite complex and varied. Of course there's more to it, because songs have sections, but all in all the genres I'm using are very structured. Techno especially, is very easy for the AI. Sometimes I'm tasked with creating a theme in a particularly difficult genre, and that's when it gets really tricky/fun (such as some of the Indian ones, which were labors of love). In these cases, I may have to re-rerun many "generations" of the AI through the same song from beginning to end, with different parameters/instruments, but with the same patterns/learning, because what can happen in these cases is that it starts out simple, and increases in complexity as it goes, so the start can sometimes be a bit boring.
Also keep in mind this is just a conceptual explanation. The code is much less amusing. Thinking of naming processes "bots" though just for fun.
I hope I explained it OK. Listen, I'm not saying it's Watson or anything, it's really just a necessary step I had to take in order to make the computer do what I wanted it to do. I tried out some different ideas, and this one worked the best. I like the result at least :)
Here's a bibliography we're still working on. More to add, but a good start so far: https://www.brain.fm/pdfs/ResearchLibrary.pdf
We've also done our own research that is pending publication:
https://www.brain.fm/pdfs/EEGFocusAnalysis.pdf https://www.brain.fm/pdfs/EEGSleepAnalysis.pdf
Here's an independent study on HRV using our tech: https://www.brain.fm/pdfs/ElioConteHRVandBWE.pdf
We're planning on following that one up with a more robust study, because our users do get great results with HRV and it's an interesting topic.
One of our former neuroscientists published a meta-analysis, but it's a bit outdated now re: what we're doing. Still, it's on our site or in the journal if you're interested. :)
Lots of science to this. Here's a good start: https://www.brain.fm/pdfs/ResearchLibrary.pdf
We're still organizing it. Apologies for any inconsistencies. There's still stuff we need to add as well.
And glad you like Brain.fm as well :)
To my great surprise, silence is actually the worst performer in our research and for other researchers I've talked to. check out our focus study, and also if you hit the "more" button you can see a kind of bibliography (that we need to organize better).
Silence can be a distraction in itself. Counter-intuitive, I know. I actually think I remember an article about this popping up in my news feed recently, trying to explain why. I was in a hurry so couldn't really read it, but I'll try to find it again after all this craziness dies down :)
But anyway, everyone's different, and different situations call for different tools. When the noise level is just too high for me (like, construction outside, vacuuming, etc) I use those crazy earmuffs meant for gun ranges. A shopvac could be right next to your ear and you wouldn't notice hah