Protecting Our Brains From Datamining

Jason Koebler writes: 'Brainwave-tracking is becoming increasingly common in the consumer market, with the gaming industry at the forefront of the trend. "Neurogames" use brain-computer interfaces and electroencephalographic (EEG) gadgets like the Emotiv headset to read brain signals and map them to in-game actions. EEG data is "high-dimensional," meaning a single signal can reveal a lot of information about you: if you have a mental illness, are prone to addiction, your emotions, mood, and taste. If that data from gaming was collected and mined, it could theoretically be matched with other datasets culled from online data mining to create a complete profile of an individual that goes far beyond what they divulge through social media posts and emails alone. That's led some to develop privacy systems that protect your thoughts from hackers.'

Comments

Q: Why did Apple buy Beats headphones?

[QuietLagoon]

A: To get larger sensors closer to the brain.

Re:Q: Why did Apple buy Beats headphones?

[gl4ss]

Beats buyers have brains now?

Yeah well,

[dale.furno]

Just wait until data from people like me with ADHD and PTSD starts corrupting their Hey look a Squirrel!

Increasingly common?

[Anonymous Coward]

Is it really? Or is it a click-bait headline that really means here's a couple of companies who have a product which does it but nobody else does?

Re:Increasingly common?

[tnk1]

I agree, to an extent. These devices are hardly going to read minds in the sense of providing all of that detail.

However, whatever they lose in quality (of resolution), they may make up for in quantity. A poor quality device may still be able to provide some useful data points when applied to a larger group of people. Put some branding or situations inside a game, monitor for coarse grained interest or emotion, and you might have something useful to marketers or game designers. Or not.

When things like this start approaching mass markets, people start thinking of other uses for the data. Working in a field where people are spending good money trying to vacuum up all the data on the Internet, even shitty Facebook posts, I see first hand how people get excited over any new data point. Most of it is crap, but there's some gold in there, for sure.

Click-bait, but still interesting to consider.

Ridiculous

[sexconker]

Here we propose an integration of a personal neuroinformatics system, Smartphone Brain Scanner, with a general privacy framework openPDS. We show how raw high-dimensionality data can be collected on a mobile device, uploaded to a server, and subsequently operated on and accessed by applications or researchers, without disclosing the raw signal. Those extracted features of the raw signal, called answers, are of significantly lower-dimensionality, and provide the full utility of the data in given context, without the risk of disclosing sensitive raw signal. Such architecture significantly mitigates a very serious privacy risk related to raw EEG recordings floating around and being used and reused for various purposes.

So MIT pisses away cash on research that comes up with "Just anonymize the data, sorta, before shipping it off to advertisers and you'll be protected, sorta."? And of course it's peppered with meaningless shit like "personal neuroinformatics system", "smarthphone", and "privacy framework".
Hey MIT, give me a research grant and I'll come up with an actual solution. Hint: Don't let people put EEG sensors on or around your head for a game, a video, etc. in the first place and you won't have the problem of them selling it to nefarious parties who would use it against you. Much more effective than the proposed equivalent of "Do Not Track" for brainwaves.

Re:Ridiculous

[s.petry]

The paper, if you read it, also discusses the "Why" you need to do this (as most scientific papers tend to do). It is not just about building a method of giving the data anonymity, but telling people why it should be done. While a bit deep for some, I highly recommend reading the paper.

Companies are already trying to figure out how to cash in on your EEG data. What's the big deal you ask? Well...

Using more direct attacks to reveal EEG information, Martinovic et al. investigated in [28] how the
109 brain's response to a particular stimulus (so-called P300 paradigm) can be used to narrow down the space
110 of possible values of sensitive information such as PIN numbers, date of birth, or known people. The
111 tasks required the subject to follow the experimental procedure without explicitly revealing the goal of
112 the experiment: for example thinking about birth date while watching ashing numbers. Although the
113 presented attacks on the data may not be directly applicable to preexisting EEG data, as they require
114 fairly specic malicious tasks, we can expect | as the subjects participate in multiple experiments |
115 correlations violating privacy could be obtained from raw EEG signal. For example, when a large corpus
116 of the user responses to a visual stimuli is collected, it could be used in P300-based Guilty-Knowledge
117 Test, where the familiar items evoke different responses than similar but unfamiliar items [29].

In other words, and without the formatting and line numbers, people could maliciously collect personal information simply by providing you visual stimuli. They can do so at such a fast rate that you would never know what happened, such as an a series of images or questions displayed for a whopping 13.3 milliseconds per frame.

Yes, I'm cynical. That said, I also pay a lot of attention to privacy and how things are used. Because of those two things, my answer is easier. If you are worried about privacy don't use this type of product. At which point, there is no need to harvest data even anonymously. If doctors need to share data for diagnosis, there are secure protocols and methods of exchanging data which do not include having your EEG data on an internet facing computer. The best doctors in the world can't evaluate the results of an EEG in a split second, so taking a bit longer to share data is not going to change how long it takes you to get a diagnosis.

If you want to share your EEG with the world and have all your Facebook friends try and diagnose you, go for it. Educate consumers to the dangers and I have no problem with it. Every EEG device sold to a consumer needs a big old warning label "WARNING! Use of this device may result in your personal life being completely destroyed, including the loss of all financial holdings and credit(E.G. Bank accounts, credit card, social security number, etc..) as well as pertinent personal information that would allow someone to assume your identity at will." This can be done without you having any knowledge of the theft, and can be done within a few minutes of using this device..

Some may claim "Regulation", but I'll point to the TV regulations which ignores current subliminal inserts in video and audio. (You can question the results of those things all you like, but they do happen and get caught from time to time.)

I'd love to have such datasets

[Mister Liberty]

... for all those that are placed above us to lead us,
and of all those that suck up to same.

Lets all take a step back to appreciate this:

[crioca]

This information is undoubtedly being caught up in the global surveillance dragnet, which means that government agents are literally spying on people's brainwaves. The most hackneyed conspiracy trope of all time is now a hilarious reality.

Re:Lets all take a step back to appreciate this:

[tnk1]

Edward Snowden will shortly be releasing transcripts of this too. Here's one example:

SUBJECT 1765467-2: K3yseRS0Z3 [[TRANSCRIPT BEGINS]]

STRAFE
STRAFE
STRAFE
STRAFE
SCOPE
FIRE
FORWARD (RUN)
TEABAG

[[TRANSCRIPT ENDS]]

I'm not worried. I'm already aware of what most 15 year old boys are thinking about, we don't need the NSA for that.

Phrenology much?

[Opportunist]

People. Please. Be reasonable.

I've been doing a bit of research in that matter (because, well, the idea of controlling a computer with your brain IS kinda cool), and we're FAR, FAR away from a mind reading device. If such a device is possible at all.

Every kind of "mind tracking" technology in existence not only needs a LOT of training (on both sides, the device AND the user), but most of all it needs cooperation to the extreme. Actually, it is pretty HARD to make that device actually "understand you", and that's if you WANT it to understand you.

Now going and trying to pick up subconscious thoughts is at best akin to phrenology, where you have some sort of brainwave patterns of known people and try to pretend that the ones you read on another person that resemble them have any kind of correlation. The whole shit smells like good ol' phrenology.