Here's a video of a guy learning to ride a "backwards bike", if you turn the handles left, then the wheel goes right. It took the guy forever to learn to ride a backward bike. He kept remarking that "any disturbance would make is brain switch back to the old 'normal' bike control system".

Later in the video, he did something amazing. He tried to ride a normal bike and apparently no longer could. Then after about 20 minutes worth of trying and failing utterly, suddenly the understanding of a normal bike snapped back into his mind, and in the span of about 30 seconds he could ride the bike easily. How did that sudden change happen? What state was his mind in that he couldn't reach back to the old system? And then what suddenly happened so that he was able to?

More generally, is there any research in this area? I think it would be awesome to hook up an EEG to someone and watch as they make this mental leap from one control system to another. Has this ever been done?

Note here (because there appears to be some confusion) that I'm not so much interested in brain plasticity or negative transfer. Those are interesting in their own right, but for this question I'm most interested in the "switch back" phenomenon. Why was the individual somehow blocked from reaching the normal bike control system, and how did this barrier so suddenly and so drastically disappear? Apparently there are now two systems in his brain, and they are somehow quarantined from one another. What mechanism makes this possible? And why would such a mechanism be beneficial?

I realize now that I have experienced the same phenomenon with language. I studied Spanish in high school and was pretty good at it. But after learning Japanese it's difficult to "jump" back into Spanish speaking mode. And if I'm in Spanish speaking mode it's hard to jump back to Spanish.

  • $\begingroup$ I wish I had better tags for this question. $\endgroup$ Commented Apr 30, 2015 at 13:33
  • $\begingroup$ Highly relevant: en.wikipedia.org/wiki/Prism_adaptation. I'll try to expand into an answer later. $\endgroup$
    – Josh
    Commented Apr 30, 2015 at 13:58
  • $\begingroup$ Possible duplicate? Different task, same answer. cogsci.stackexchange.com/a/238/2868 (edit: guess not) $\endgroup$ Commented Apr 30, 2015 at 13:58
  • $\begingroup$ How do I find out if someone does the EEG test?! That would be amazing. $\endgroup$ Commented Apr 30, 2015 at 17:15
  • $\begingroup$ Updated question. To clarify the idea I'm most interested in. $\endgroup$ Commented May 2, 2015 at 12:56

2 Answers 2


While both prism adaptation and negative transfer are pointers to the right direction, I'd see this as a question concerning brain plasticity (you may want to tag the question with that, I don't have the required points to create a new tag).

The guy could indeed learn to ride the reverse bike, but he would have to work hard on it, and would have a hard time reversing to normal bikes later on, as you report.

Take a look at wikipedia's Perceptual adaptation page:

Experimental support[edit] In the 1890s, psychologist George M. Stratton conducted experiments in which he tested the theory of perceptual adaptation. In one experiment, he wore a reversing glasses for 21½ hours over three days, with no change in his vision. After removing the glasses, "normal vision was restored instantaneously and without any disturbance in the natural appearance or position of objects."

Modern version of inverting mirrors with harness. On a later experiment, Stratton wore the glasses for eight whole days. By day four, the images seen through the instrument were still upside down. However, on day five, images appeared upright until he concentrated on them; then they became inverted again. By having to concentrate on his vision to turn it upside down again, especially when he knew images were hitting his retinas in the opposite orientation as normal, Stratton deduced his brain had reprocessed his vision and adapted to the changes in vision.

So people look at the world literally upside-down, but then adjust abruptly.

A good read about this is On Intelligence, and in my research with chess experts I have argued for a cognitive model of "experience recognition" in here and here.

An amazing video of the beginning of the learning process is here: Amy hears sound for the first time. At about 1m45s you can clearly see that she is confused about about her sensations---she doesn't have the experience of hearing, and her brain is just now trying to make sense of what this new influx of surprising information means and how it should be processed.

Or look at this small girl, who has had her entire left hemisphere taken out of her brain, with few behavioral problems.

As a thought experiment, imagine you could suddenly "feel" the radio waves flowing all around us. At first you'd be extremely confused. Yet, after some time, you'd be able to discern people turning off their phones on a plane, turning on their gps, or the "silence" in rural areas. If you had the experience for a long enough time, you would be able to ground it to meaningful events surrounding you.

  • 1
    $\begingroup$ Nice answer, and interesting to read about your experience recognition model. I've been steeped in Chase/Simon/Gobet/Ericsson from my very first academic steps and consider the questions they ask to be central to psychology, so it's always interesting to see new perspectives and critiques. We've covered their approach quite a lot here thanks to Malcolm Gladwell; maybe you could set up an apt question to introduce and contrast experience recognition with their approach? $\endgroup$ Commented May 2, 2015 at 9:16
  • 1
    $\begingroup$ I'll try to think of a good question, Christian. It actually may help better distinguishing between the models. $\endgroup$
    – linhares
    Commented May 2, 2015 at 10:34
  • $\begingroup$ Thanks for your interesting answer! But I'm actually most interested in that bizarre transition from one control system to the other and why they should somehow still exist, but be walled off from each other. I've updated my question. $\endgroup$ Commented May 2, 2015 at 13:02

Pending Josh's better prism-adaptation-based answer, I believe the general cognitive term for this is negative transfer.

Most of the time, when people learn something new or improve their skills in one domain, we observe related improvement in related skills and domains. Sometimes, however, it's the other way around―typically motor activities. A relevant previous question here on CogSci was based on negative transfer in writing, "By learning to read and write upside down, what did I do to myself?", where Jeromy's answer should be of interest.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.