A super-informal discussion about science!
Just amazingly good.
Zeven thanks for this, this is really interesting stuff!
Cool idea, would love to see more of this.
Watched the whole thing. Super interesting. Thanks for uploading this for those of us that are not in your TZ!
At 50:58 you mentioned that autistics can have issues with cognitive flexibility. First of all, people need to understand that there’s a difference between autism and autism WITH additional diagnoses, what people commonly refer to is people who are diagnosed with autism, but also with an intellectual disability.
When it comes to the actual topic of cognitive flexibility, I, as someone who is autistic, don’t have a problem with rules shifting. If I’m in a store, and I know where they usually have a specific thing I want, it wouldn’t bother me at all if I have time to look around for it, but if I’m in a hurry, this is where it becomes an autistic issue, and this is why people think autistic people generally just have a problem with change.
Let’s take this further and flip it around.
The last few patches, I’ve started playing pretty well, especially compared to a friend of mine who usually had superunicum stats and winrate. He’s not adapting to what’s happening, if I see a Type 4 or 5 heavy, I don’t sit around and try to pen it in the front if I know that I can’t, I won’t go to the part of the map where I’m likely to find them if I know that I can’t fight them. I don’t sit around and try to brawl with heavy tanks in a light tank and whine about my pen being too low. This is the kind of things this friend of mine is doing, but he refuses to find another way, he keeps going where he’s always going. Most people I hear whining about the state of the game are people who are considered to be good, or people who are followers of that person. I sure don’t think most of these people are autistic. This is something people seem to have problems with generally, at least that’s my observation.
EDIT: I just noticed that you brought up that experts are less willing to change. 😉
Very good and informitive , would love to see more. As always Zeven………..your the man!!!!
Zeven is scary smart. WG should hire you as a consultant on how to improve the training missions.
After college I had a good contact with people of neuroscience from Duke, interesting times.
12:20 If 43000 games is to 5000 hours than there are a good bit of people who at 5000 hours I’m better than at 10000 battles.
EDIT: Just saw 15:00… Pubbies will improve very slowly 🙂
As a gamer and as a pharmacy graduate this is something!
hey man, I haven’t watched your streams in a bit because I’ve felt like I have outgrown your content. Not that I didn’t enjoy it but yknowwhaddaimean.. anyway, this video however is fantastic and explained really well. I look forward to seeing more videos like this. Thanks for sticking with WoT
Thanks Zeven, that was an excellent intro into neuroscience. When you mentioned flashcards and MRI, it reminded me of a keynote given by Mary Lou Jepsen at a recent tech conference I attended. She started a company that’s trying to build an MRI-in-a-cap: . During the keynote she showed some images that she claimed were reconstructions of what the subjects were seeing using just the MRI scan data. From what I can gather, it’s the same exercise you described, except the researchers used a machine learning neural net to reconstruct the image the subject is seeing. I don’t think they’ve published their results yet, so I don’t know how much of the presentation was real and how much was fudged. Anyway, just thought you might be interested if you haven’t heard about it already.
It’s interesting. I liked it, but I didn’t find anything in it that useful.
I think one of the most misunderstood differences between a player like yourself that produces guides versus you average audience member is familiarity.
(What I mean by familiarity is, for this patch (9.19.1) what tanks are worth playing, what tank types go where, what bush is likely to have a TD or scout in it, where are spots I can go and farm early damage, and is it worth taking this low percentage shot at this moment in this situation?)
Zeven, Thanks so much for sharing this with us, I really enjoyed it! I personally have studied and developed artificial neural networks during my Masters in Electrical Engineering and MBA and continue to find it a fascinating topic. ( I have personally developed several artificial neural networks including some that have been optimized by Genetic Algorithms)
On your comment on the brain being like a computer, many are of the opinion that the human brain computes in an entirely different way from the conventional digital computer. The brain is a highly complex, nonlinear, and parallel information processing system. It has the capability to organize its structural constituents, known as neurons, so as to perform certain computations (e.g., pattern recognition, perception, and motor control) many times faster than the fastest digital computer in existence today. Consider, for example, human vision, which is an information-processing task. It is the function of the visual system to provide a representation of the environment. To be specific, the brain routinely accomplished perception recognition tasks in 100-200 milliseconds, whereas the tasks of much lesser complexity may take days on a conventional computer.
In terms of the brain, I know that artificial neural networks are robust and can still work well even if synapses ( neuron connections) are broken. Hence they are resilient unlike a computer. Also neural networks have the ability to “Generalize”, that is to deal with and make decisions that have not been seen before. This is key to games in WoT as well as well as the human brain. A lot of this has to do with the ability of “Pattern Recognition” which is studied extensively for those fortunate enough to pursue a degree in Electrical Engineering.
With regard to WoT: A lot of games require that those involved are assumed to “behave rationally” and based on this you can adjust your tactics by predicting what “rational” players will do during a match. Rational behavior in this game is a fallacy, I believe this assumption of rational players works for Tic tac toe and chess but not in WoT. For example some players may not behave rationally based on (1) New to game like a player who is running a tier 8 premium for the first time (2) Competing objectives as not all players want to win as some are padding wn8 by farming damage or focusing primarily on a personal mission to target selective tanks (3) state of mind is not clear based on frustration with getting one shotted last match and are now playing too conservative by camping, etc etc the list goes on. It is analogous the to the stock market where we like to assume “rational behavior” but we see bubbles and crashes that cannot be explained unless we accept irrational behavior which is very hard to predict.
I found the shoe-tying example around 29:25 pretty funny…I remember my mom teaching me how to tie my shoes; I believe I was about 3 years old at the time. The only way I could learn, then, was with bunny ears…make two loops and tie them together…over-under just did not click in my brain. A few years ago, some 25+ years later at the age of 30 or so, I thought of that memory—of mom teaching me—while tying my shoes, and suddenly knew how to do the over-under method. The brain is a puzzling organ, sometimes 🙂
Oh, and also you are an excellent instructor, with a very pleasant speaking voice. Thanks for the interesting and well-constructed discussion!
From reddit in case you don’t see it there:
“Alright, every talk needs feedback. Especially since you plan to make an updated version of this at some point. Here are some things that I wrote down while listening.
– Intro needs information of being wot streamer. It’s obvious while on stream but on youtube I can show this to my friend who plays wot but has no idea who you are.
– You could cut out the part from 32:45 to 37:15. I didn’t feel like it added much to the point you already made with the much shorter segment on Sagan.
– You could show a clearer structure of the whole chain of learning a skill. As far as I understood it: Perception -> deposition as episodic memory -> transformation to procedural-skill -> memory retrieval, and retrieval triggers. Perhaps on one slide and then go into detail on each of the steps afterwards, with the slide that are already there.
– Handball example. You don’t state the goal until 2 min into the example. Throw accuracy.
– You said experts learn slower what the global parameters change, such as arty changes. I can see this might be true. You also said that they don’t react as well to changes in local parameters. So they have more problems with overstaying in their lane (55:20). I think that’s not the case. People staying in lane is information overload due to insufficient filtering as you mentioned earlier. This gets better with higher mastery.”
First liked video, and then im watching :3
I really enjoy watching these kind of videos on how to get better and such at a game, i found it to be very interesting 🙂
Your email address will not be published. Required fields are marked *