Technabob » learning

Scientists Testing Automated Learning via fMRI: Shut Up and Take My Studies!

Lambert Varias — Thu, 15 Dec 2011 17:00:14 +0000

Remember how Neo ~~cheated~~ fast-tracked his Kung-Fu and weapon proficiency in The Matrix? That might soon be possible in the real world. In the false world. In our world. This one. Dammit. Whoa. A recent set of experiments have indicated that it’s possible to teach a person by copying the brain patterns from another person who is good at the task to be taught. The best part? The student doesn’t even have to be aware of the whole teaching process! Are you listening Lazy College Senior?

The experiments were conducted by researchers at Boston University (BU) and at the ATR Computational Neuroscience Laboratories in Kyoto, Japan. The task to be learned has to have a high visual component – fortunately, many if not most of our tasks have a high visual component. In simple terms, a person who’s good at the task is then put in an fMRI machine and asked to visualize the task. The person’s brain activity is recorded and then re-enacted in the student’s mind. After being taught the pattern enough times, the student’s brain becomes familiar with the pattern and the student becomes better at the activity. Watch the video below for more details (you can watch it here with captions).

So does this mean that in the future, our schools will have fMRI machines instead of desks? Will we even need schools? Will we be able to download LeBron James’ hoops skills from the comfort of our home and then be basketball geniuses after a power nap in the ol’ fMRI? Or will this be the key to making the robots-enslaving-man part of The Matrix real? What if this has already happened? I need to lie down. And passively learn the truth while lying down. Thanks, science.

[via National Science Foundation via i09 and DVICE via Geekologie]

Adorable Baby Tries to Use Magazine Like an iPad

Shane McGlaun — Sat, 15 Oct 2011 13:00:32 +0000

My kids have been using gadgets and electronics since they were old enough to sit up straight. I know I am not the only one that has kids like this. They both love the iPad and the iPhone and when it comes time to watch a show or play a game, they more often than not head for my iPad. My son also likes to read things on the iPad rather than print material.

I think that this video of a 1-year-old girl shows what is changing for kids today and the traditional publishing industry moving into the future. The traditional print publishing world has been living on borrowed time for a while now. Current and future generations of children are growing up with the expectation of interactivity.

Click to View Embedded Video Clip

To these kids, print publications are like broken iPads, just like this little girl. I imagine when these kids hit 5-years-old and start to read books at school, they may be forced to read printed books, but their preference will be for the interactive gadgets they grew up on. Many schools are already going to these interactive tablets and notebooks to save money on printing and improve interactivity. What do you think, is this video further indication of the impending death of print, or just that babies like to grab things?

[via Gizmodo]

What You Write is What You Get: Scribblenauts in Real Life

Lambert Varias — Fri, 01 Jan 2010 16:14:46 +0000

Made by Steven Zhou and Syed Olmer Gilani of the National University of Singapore’s Interactive Multimedia Lab, What You Write Is What You Get (WYWIWYG) is a visual way of learning about language. Just like 5th Cell’s critically-acclaimed DS game Scribblenauts, WYWIWYG can analyze written words and come up with a virtual representation of the corresponding object.

Even the set-up is a bit DS-like. Users write on a whiteboard, using ordinary markers; a webcam then takes an image of the written word or phrase. The real magic is done by the software, which “analyzes the image and interprets the handwritten words into different output modalities.” In other words, what you write is what you get.

Unfortunately, the demo video doesn’t go into detail about the capabilities and the limitations of the software, like the number or kinds of languages it can interpret, the types of words (adjectives, proper nouns, adverbs, verbs) it can interpret, the number of words it can recognize, or if the virtual objects can be made to interact like in Scribblenauts.

Nevertheless, this is an awesome project, and preschool and elementary teachers will no doubt fall on their knees and cry buckets of joyous tears if and when WYWIWYG comes out, because it’s a clever way of capturing the attention of kids and motivating them to learn. After all, you can’t make green poop appear if you don’t know how to spell “green” and “poop”. You can watch the demo video at the Interactive Multimedia Lab’s website; be sure to check out their other projects as well.

Robot Learns to Make Facial Expressions, is Rewarded With a Measly “Signal”

Lambert Varias — Fri, 10 Jul 2009 12:41:47 +0000

At least dogs get food when they get things right. This poor robot Einstein, made by computer scientists from the Machine Perception Laboratory at the University of California, San Diego, was given time to play with its 31 artificial facial muscles while “staring” at its reflection in the mirror. The robot was also equipped with a camera as well as facial recognition software, so that when it happened upon a human facial expression while it was experimenting, the software would send a “reward signal” to the robot. That must have been one hell of a signal.

The scientists’ next goal is to get the Einstein robot to socialize, and eventually robot tormentors tutors.

Click to View Embedded Video Clip

Marian Bartlett, a facial recognition expert who was a part of the experiment, said that the aim is to use robots to conduct one-on-one tutoring, which she says can improve learning “by as much as two standard deviations.” Sounds like they want our kids to go insane.

[via Wired]