Tuesday, 27 January 2015

David Sackman - Virtual reality for positive change and behaviour.

quote [ My feet did not want to move. And I kept willing myself... I had been thinking about this for two months! I was certain that I would be able to do it. ]

If you happen to have an emotional attachment to the idea of Free Will, the next twenty years are probably not going to be very kind to you.

Teen girls report less sexual victimization after virtual reality assertiveness training

[SFW] [Virtual & Augmented Reality] [+2 Interesting]
[by steele@11:09pmGMT]

Comments

steele said @ 11:18pm GMT on 27th Jan [Score:1 Hot Pr0n]
lilmookieesquire said[1] @ 4:56am GMT on 28th Jan
(vr porn comment) The "is that MY dick?" line. Nice.
rhesusmonkey said @ 3:23pm GMT on 28th Jan
I particularly like the very last comment post review. I personally can't wait to get the HoloLens for work. To me the possibilities are endless. I expect MS to purchase Thalmic Labs soon (or someone, I don't think Intel).
lilmookieesquire said @ 11:52pm GMT on 27th Jan [Score:1 Insightful]
I find it upsetting that this
"teen-girls-report-less-sexual-victimization-after-virtual-reality-assertiveness-training/"
is a thing that is helpful in our society.

It's like "this will happen to you, so let's practice!". It's very practical, but also, it is very sad. It certainly doesn't speak very highly for our society.

(I assume it means "girls report that they receive less...")
bltrocker said @ 1:14am GMT on 28th Jan
I'm more an Oculus fan, but MS seems to be coming on strong with AR. Google Glass ended up being a big turd, but HoloLens seems like it could go places.

http://www.wired.com/2015/01/microsoft-hands-on/
damnit said[2] @ 5:33am GMT on 28th Jan [Score:1 Funny]
rhesusmonkey said @ 3:25pm GMT on 28th Jan
It still is, just a different use case. Particularly the voice control piece, I expect to see MS push XB as a home automation control point. HL is a different beast.
steele said @ 1:44am GMT on 28th Jan
I caught a vid of the demo they gave and I couldn't stop thinking of Project Milo and some of the bunk demos they gave for Kinect. I'm not saying HoloLens doesn't have potential, but I wouldn't hold my breath over it just yet.
bltrocker said @ 2:09am GMT on 28th Jan
I mean, a lot of journalists have been trying it out, and it seems to be impressing some of them, so it's at least past the stage of vaporizing before people try it out. It's not like the claims are that amazing yet--a lot of "mouse clicks" in space to do simple things to the added augmented layers. Fuck the stupid demo where the lady builds a quad copter then they 3D print it; I'm more talking about the Minecraft demo, Mars demo, and assisted DIY demo. Sure, I'm not holding my breath, but it has entered my radar as something to watch.
steele said @ 3:14am GMT on 28th Jan
Fingers crossed then ;)
steele said @ 1:48am GMT on 28th Jan
btw if you have cardboard Cmoar AR RPG is a cool little demo worth checking out.
bltrocker said @ 2:04am GMT on 28th Jan
Re: the main link video--The dude is describing like the oldest Oculus DK1 demo in the book for a third of the video. He seems really out of touch, so I'm not inclined to believe how he sees the future of VR; I feel I'm more up to date on both VR and scientific understanding than him. In fact, as far as I can tell, he's a scummy marketing company CEO and doesn't do any research or engineering himself. Human behavior is never as clean as researchers would like, so I'm skeptical that many of his claims have much merit. What are the long-term outlooks on people that took one virtual reality session on aging in the mirror or cutting down trees? How robust are the effect sized and how generalizable are the described studies? Long-term and true effectiveness of virtual exposure therapy with the PTSD patients? Is the VR as effective as if you use live elements of exposure therapy? I'm laughing at the weight loss one--that is just not going to work out very well. Don't get me wrong, I love VR, and I'm glad that there are people excited about the possibilities. I just want everyone to steel their excitement (or fear) with a tiny bit of pragmatism.

On upcoming tech, I'd also like to add this. He's not the greatest presenter, but this is some of the stuff that the Leap Motion guys think/know is going to come out in the next 5 years.
https://www.youtube.com/watch?v=Mg-zZXeWyJU

Also, what does free will have to do with an increase in immersive virtual reality tech? That human behavior is more predictable and easier to change than we'd like to think? I understand that I have a lizard brain behind my higher level thinking centers, but in the end, it's the higher order processes that allow me to choose to eat the piece of cake or not.
steele said @ 3:14am GMT on 28th Jan
On free will? The big one I'd say is when he literally is incapable of moving without external stimulus. It's kind of hard to claim free will when you're capable of being shutdown by things you are fully aware of being an illusion. Keep in mind the behaviours he describes were problems Valve also came across when they were developing their inhouse presence capable hmd. The right illusion can make all those higher order processes completely useless.

The majority of techniques that VR can use to modify behaviour are really nothing more than manipulation of pavlovian associations. Matter of fact, when designing VR worlds it helps to realize that you're basically just implementing a giant mnemonic delivery system. Take those mnemonics and combine them with basic reward structures and as long as you can maintain congruity with your narrative you can manipulate them to your heart's content. And that's not even taking into account all the cool tricks you can do to fuck with how people identify with avatars, which is what a lot of the empathy and generosity studies he mentions take advantage of.

There are plenty of possibilities using techniques we're already aware of works thanks to video games and storytelling, it's just a matter of time until the tech comes along that's really going to turn the world on its head. Because keep in mind, it's not just the presence capable delivery vehicle that's coming, combine that with data mining algorithms, iris tracking, consumer grade EEGs like the emotiv for emotional feedback, etc... and you can basically build a new you.

So, yeah. Hard to take free will seriously.
bltrocker said @ 5:13am GMT on 28th Jan
"The right illusion can make all those higher order processes completely useless."

He's overstating the power of the illusion. I've done that demo, and it's fun, but I had to actively trick myself into believing that there was a drop all around me. I took a step into "air" no problem, my heart rate increased a bit, I felt ground, and couldn't keep up the fake world in the HMD. Sure, 10 years down the road, maybe we have a truly convincing audiovisual world inside a HMD. But that's all it is. As long as you don't have a complete matrix situation or whatever, you're still partly aware of the illusion and you have some control over your experience.

As far as experiences that are vivid enough to change behavior, I completely welcome that. I see a lot more good coming from shrinking the world down enough to fit within everyone's monkeysphere. The potential to increase empathy 1 million% far outweighs the chance that VR torture or marketing devices rise to prominence. In the end, though, it's up to the user to decide that the lesson they learn is so important that it's still worth rising above the path of least resistance to help out a stranger or pick up the trash on the sidewalk. Keep in mind that if you pair a footshock with a scent or a tone or some scenery, not every rat will be deathly afraid when put back into the situation. Pop psych doesn't have foolproof answers to match our varied physiology. I'm not sure if that translates to more or less free will, but some people/animals are very hard to manipulate.
steele said[2] @ 2:11pm GMT on 28th Jan
Presence capable is pretty close to the matrix situation. After your brain has enough time to associate your movements with what's happening in the vr world your grasp on it being illusion slips away. At this point we already know it's possible, it's inevitable. And even if we're not there yet, there are tricks to getting there. Remember the brain builds our reality through a weighting of our sensory input, the simplest way to maintain presence via our audio/visual illusions is to nullify the senses we aren't using. See: body parts you don't think about unless they hurt, meditation or dreams with stillness of the body, or internal hallucinations in a sensory deprivation tank. Matter of fact, the sensory deprivation tank is a very good example of how presence works, by nullifying the senses you aren't using the hallucinations take on the illusion of reality. That's why they're capable of becoming such emotional experiences, it's basically giving your imagination presence capabilities.

Hard to manipulate doesn't mean impossible. ;) Remember in my projections I'm taking into account the fact that one day we'll be able to get EEG feedback (imagine if fMRI ever became that convenient. drool) on the fly. Combine that with data mining techniques and in a dynamic illusion, if a "footshock" doesn't work the illusion can reconfigure itself till it finds something that will.

---------------------------
If you're interested I made a comment on reddit a while back on how something simple like eye tracking could be used. It's very simplified, but it gets the point across.
Reveal

As an example, one of the first upgrades to the standard VR HMD is going to be Iris Tracking.

Let's say Facebook creates their Metaverse, what this means is that they now have an incredibly real simulation that your brain will respond to as a reality. In this world, with an Iris Tracking capable VR they will know everything that you look at, how long you look at it, and what you did next. Keep in mind your conscious mind isn't always aware of everything your eyes do, so they'll now be recording more information about how you respond to this faux reality than you ever realize is being given away.

This may seem a little spooky, but not nearly as much as it is. Because now we introduce the "STANDARD A/B TESTING" that the defenders (of facebooks experiments) in this thread are touting as harmless. What the A/B testing does, is over time you will experience a large number of ads that have been specially designed to be at extreme ends of a scale of features that they have devised so that they can record what features your eyes lock onto and how long they do so. As they record more and more of your responses the individual ads will shift in design and format until they've created a custom set of personal advertisements that attract your eyes almost as powerfully as an addictive substance. All the while you never even realizing that anything out of the ordinary has occurred.

Now consider the future roadmap of VR. Iris tracking is just one of the first features. Then there's heart rate tracking, which they will use exactly the same as the Iris Tracking, but now the Metaverse knows what features of advertising you're looking at and whether or not it gets your heart rate going. And, of course, the advertising will adjust accordingly.

Next on the VR Roadmap, EEG aka brainscanning much like the Emotiv Headset At this point, not only will they be able to test a wide variety of ad formats, record your responses, and build the ad that your eyes can't resist now they'll be able to tell how these advertisements make you feel. And then they'll adjust the ads accordingly until now the have advertisements that your eyes can't resist, makes your heart race, and makes you feel whatever they want you to feel.

And you will never even realize any of this, because all of this is happening in a world that they can control completely. The very point of VR is that much of what makes VR real is happening behind the scenes in your head. Those subconscious responses, like your eye movements, your heart rate, your feelings, all those things that you aren't consciously aware of become an open book to the fully immersive metaverse of the future. And at this rate facebook is going to be in charge of it. And they are showing themselves to be the last organization we should allow to be anywhere near it.

Keep in mind, I used "ads" for simplification in this example. The reality of the situation is that in the future the "ads" will probably take on the form of dynamic avatars that will adjust over time till they embody a form you would find irresistible. Imagine if you were in a metaverse environment like WOW or something, walking through a crowd and the software is tracking every avatar you glance at, which features your eyes linger on, and how they effect your mood. Suddenly an avatar comes up to you that's a total knockout, drop dead gorgeous, your mouth has gone dry, your palms are sweaty... And she wants to know if you're happy with your long distance provider :P

And sure it might not work the first time, but the more data get they more they can refine their approach. They might notice you make purchases or "decisions" immediately after you switch from excited to bored. So they might create scenarios that would change your mood in that way to make you more approachable. Targeted advertising instead becomes targeting the version of you that they're looking for to serve their purpose.
---------------------------

The idea of free will really falls apart when you consider what the words 'decision' or 'choice' imply. You have to remember those words are meant to be metaphors, representations of the behaviour of a single entity, our body. But you and I both know that that our brain isn't under the control of one single entity, neural networks are weighted systems. They're basically giant multi-teamed games of tug of war using a net enacted on a very small scale over large periods of time. The word choice implies that there's a tiny homunculi sitting in our brain with his hand on a toggle or dial switch determining which things we're going to do, but then you have to figure out how the homunculi makes his decisions! It becomes a homunculi all the way down problem. The reality is we're much more like a disused suit filled with wee free men (or smurfs if Pratchett isn't your thing) fighting over what should be done and our neighboring disused suits are so busy with their internal battles that they don't notice how disjointed our behaviour is :)

I do agree with you though, I'm very much in the pro-change-via-VR camp and have been looking forward to what's coming for quite some time. I've literally been researching all this stuff with these purposes in mind for about half my life. Just been waiting for the tech to catch up. I've even been talking with some doctor and physical therapist friends on some techniques that may be usable in "powering up" physical therapy techniques for cerebral palsy patients (and I've got one that might work for MD, but it's a total long shot.) But that's also why I think we need to put a reign on facebook and the data they'll be allowed to collect, because I can come up with some real nightmare scenarios :(

Oh, and in case you're wondering why I keep bringing up data mining algorithms. With data mining we don't necessarily have to know HOW we would train the user to do what we want them do, we just have to know WHAT we want to train the user to do and provide enough iterations and feedback data for the algorithms to adjust the approach until the job gets done. It's basically using a neural network to train another neural network. That's why I'm so confident in my assertions, whether my techniques work with pinpoint accuracy is irrelevant because as long as the brain has specific processes it uses to learn and behave, proper data mining techniques will be capable of taking care of the rest. Data mining's versatility is part of what makes it so dangerous.
bltrocker said @ 4:18pm GMT on 28th Jan
We are not approaching matrix anytime soon. Sense of touch, smell, taste, and even hearing are way off (3d sound in VR is in very early stages). Taking away senses for those that have perfectly functional sets does not enhance presence for those that are left--that's crazy. If I see a pizza on a table, but it doesn't smell, taste, or feel like pizza, you aren't going to trick me into thinking it's a pizza.

"...one day we'll be able to get EEG feedback (imagine if fMRI ever became that convenient. drool) on the fly."

EEG feedback is shit. It's so limited in almost every area. Even with our current best setup, like a $10k set of amp, "high"-resolution electrode cap, and conductive paste all over the scalp, it's barely usable. The most up-to-date research on controlling typing/directing a robot or wheelchair with thought requires the participants to reiterate their mental image e.g. red ball moving upward about 10-30 times for a single command to be sussed out. fMRI isn't all that great, either, and it's certainly hobbled by the fact that the blood response is delayed seconds behind the actual neural activity.

I could definitely see your ad example as something that could happen. I'm fine with that, though. There's only so far you can go to try to make me change my mind and support you as a company. They'd probably find out that I only respond to real data, and I'd welcome companies trying to get my dollars through informative and truthful comparison tables. I can definitely see the positives of a more perfect world of targeted ads. If there are some retards that will buy one piss beer over another because the avatar has bigger tits, I'm fine with that.

As a neurobiologist, I really dislike the myriad of metaphors you're using to describe how we think and make decisions, but it's not so incredibly far off that I want to go into the subtle differences that exist between what is known and how you see it. As for data mining, you're limited to what you're able to measure with a HMD. Assuming people will actually tolerate VR infused with ads all over the place, you will have a few parameters to match with "did or did not buy a product". You aren't getting my greater physiology e.g. you don't know what my blood glucose looks like if you're showing me a Wendy's ad. You're not seeing which muscles are contracting. You don't know what my Wednesday schedule looks like or what was in the email I just read before I put the HMD on. I think the bottom line is that for the next 20 years, we will have to underline this tech as relatively *limited* compared to what we expect to ultimately do with it.
steele said[1] @ 6:51pm GMT on 28th Jan
Fine you're being literal with the matrix, I get it. You want neural implants, I don't blame you, but then you should be especially concerned with the path of VR because it's that path that will be building the foundation that leads there. Anything people become used to under VR will be a standard under the newer systems.

Taking away senses for those that have perfectly functional sets does not enhance presence for those that are left--that's crazy. If I see a pizza on a table, but it doesn't smell, taste, or feel like pizza, you aren't going to trick me into thinking it's a pizza.

I misspoke and now you're thinking about it backwards. I fixed the offending part of the comment. Presence isn't a switch for each individual sense, it's an overall congruence between the stimulii and the expected experience. True, I can't replace the missing sense, but as long as I don't invoke those senses I can still maintain presence. Incongruence is what leads to VR sickness. Matter of fact, that's an issue game developers are going to realize they're facing soon. Things like missing inertia in driving games may be enough to cause temporary VR sickness. BUT and it's big but, there are plenty of scenarios that I can invoke that you've never experienced before and thus your brain won't expect things like sense of smell or taste. For example, you've never been an amorphous blob, but by linking your movement to its appendages I could have you identify with the avatar and use it's experiences to evoke normal emotions.

If EEG is shit, we'll get something else, not worried about it. The arguments I'm making aren't dependent on the technology it's the data mining methods that we know we have and will work. The technology will catch up. I've been waiting twenty years, I'm expecting to wait longer.

There's only so far you can go to try to make me change my mind and support you as a company.
That kind of prideful statement is a marketer's wet dream right there.

As a neurobiologist, I really dislike the myriad of metaphors you're using to describe how we think and make decisions
I don't doubt that for a second. I've only just started learning how the brain functions on a biological level a few years ago. But I'm the kind of guy that teaches himself how to program in a weekend just because I had a site I wanted to make. Give me a few more years ;) Of course keep in mind, I've also got a background in electrical engineering, security, computer programming fucking everything, internet marketing, writing, game design, and psychology/mythology have been my obsession for many many years. My metaphors will probably always rub you the wrong way.

I'm already planning 50 years out. My ramblings to my doctor friends involve datamining neural interfaces, voice inflection, skin responses, facial expressions, blood content, etc. I don't give a shit if it's not possible yet, if it's a form of feedback produced by the body I'm including it in my projections to better figure out what we might be capable of.

Have you read Ready Player One and Snowcrash? If you haven't you totally need to, if you have, keep in mind that version of the metaverse is what Palmer Luckey and the guys at Oculus are stroking it to late at night. The metaverse as a platform, something facebook specializes in. Right now they're trying to design the next step for the internet, a platform they control. They want to be the hub between you and all the other VR experiences you may interact in. They want to be the service that you give total control over your information that asks you if you want to share with other companies. You know, like facebook does now. The difference is facebook is built on the web, but they own the leading hardware for the next step, they are aligned perfectly to be or buy THE service provider for everything that comes next. Including your purchases, your classes, your social interactions, your sensory feedback data that may come available in the future, your everything. Personally, I would rather be concerned now and prevent worst case scenarios than just playing it as it comes. Besides, whether or not you decide to take part in their metaverse you've still got a live in the real world with all the people that will.

As for the ads, it will probably be a lot like how facebook handles their newsfeeds now. It won't seem like ads, instead the avatars you come into contact with will be filtered to relay information that will subtly shape your view of what the world is currently thinking. Random inexplicably smoking hot avatars may approach you to chat about nothing important while never mentioning the brand name or logo of the outfit their avatar is wearing.
cb361 said @ 5:05pm GMT on 29th Jan [Score:1 Funsightful]
Pretty sure Rush Limbaugh knows what it feels like to exist as an amorphous blob.
steele said @ 5:27pm GMT on 29th Jan
I got nothing then. :P
bltrocker said @ 8:01pm GMT on 28th Jan
I think I see where we're divided most. You seem to be thinking of all the possibilities at the end of this tunnel, 50, 100, whatever years out. I'm thinking somewhat more conservatively. If someone tells me something might be developed 20 years from now, I call bullshit until I see that the goal is imminent. Being trained as a scientist does that to most people, I think. I'm pretty sure that's why I'm so excited about current VR, that *some* of the crazy stuff like limited presence helpful AR are really happening right now. I guess I'm just not interested in the stuff that hasn't come within the "it'll be here in 5 years" i.e. "probably will exist at some point, but could take anywhere from 5-20 years to develop" window because it's not worth losing sleep over. We don't even know if Facebook/Oculus will be the industry standard 10 years from now. Steele--I want you to make millions or be the legislative hero calling the game this far in advance and being right, but that's just not me.
steele said @ 8:25pm GMT on 28th Jan
Then you must die! I mean, yeah I guess that's cool.

The way I figure it is we have to prepare now if we want to see this thing through. It's what we dream now that will shape what we do with the technology. You gotta admit though, I'm not claiming anything that outlandish, it's just a matter of waiting for the technology that supports the necessary feedback to become available.

Outlandish claims will come soon enough :D
steele said @ 10:53pm GMT on 19th Jul
0000 said @ 11:48pm GMT on 19th Jul [Score:-1 Boring]
filtered comment under your threshold
rylex said @ 3:41am GMT on 28th Jan
Am I the only one in the world who really dislikes TED talks?
papango said @ 6:24am GMT on 28th Jan [Score:2 Good]
I really dislike them. The link below goes to an article (which was actually a TED talk itself) about what's wrong with TED and why it's doing more harm than good.

http://www.theguardian.com/commentisfree/2013/dec/30/we-need-to-talk-about-ted
Jodan said @ 4:19am GMT on 28th Jan
You are not alone. I too roll my eyes anytime someone cites a ted talk.
lilmookieesquire said @ 4:52am GMT on 28th Jan
TedX talks are really... well... I even have a person I went to elementary school with in one. Regular TED talks? I don't know. I feel like they've gone from educating the masses to a starter company sales pitch about "how some great product/technology" will "change everything" (but never does). That said, that might be entirely unjustified- I haven't watched one in ages.
arrowhen said @ 4:58am GMT on 28th Jan
I dislike TED talks, but only as a subset of my seething hatred of internet video as a means of information delivery. If your subject is any more complex than "look at this kitten doing a cute thing", FUCKING WRITE IT DOWN you lazy dick!

(It probably doesn't help that I just picked up two new hobbies where apparently 90% of the available information about them is in the form of video tutorials.)
HoZay said @ 1:36am GMT on 20th Jul

the next twenty years are probably not going to be very kind to you.
no matter what you think about free will.
steele said @ 1:38am GMT on 20th Jul
True that.

Post a comment
[note: if you are replying to a specific comment, then click the reply link on that comment instead]

You must be logged in to comment on posts.



Posts of Import
Karma
SE v2 Closed BETA
First Post
Subscriptions and Things

Karma Rankings
ScoobySnacks
HoZay
Paracetamol
lilmookieesquire
Ankylosaur