Computational Analysis Methods and Issues in Human Cognitive Neuroscience


Google Tech Talk
January 14, 2010


Presented by Bradley Voytek.

There is a massive, relatively uncoordinated effort underway to map out the relationship between brain and behavior. Human neuro-imaging experiments abound with approximately 30,000 neuro-imaging studies performed in 2008 alone. Most of the data from these experiments are analyzed on an individual desktop or small, local cluster. Neuro-imaging data contains information about neural activity in both time and space and can easily exceed 1GB per subject. In order to analyze the functional properties of neuronal networks these data can be decomposed in a variety of ways (behavioral condition, principal and independent components, phase and frequency components, graphs and digraphs, etc.). This exponentially increases analysis time and database sizes creating bottlenecks in the analysis work flow. I will discuss a variety of neuro-imaging methods in terms of the sources of the signals measured, what these signals actually inform us about how the brain gives rise to cognition and behavior, and how this information can inform medical diagnosis and treatment. Furthermore I will highlight how advances in computational processing have improved data analysis and discuss the computational roadblocks that impede research progress.
Video Rating: 4/ 5

Please follow and like us:

Like it? Share with your friends!



Your email address will not be published. Required fields are marked *

  1. Super interesting and thanks for making publicly available. From a CS perspective, the biology itself/effects are kind of irrelevant aside from the underlying architecture (neuron, dendrite, action potential). We still lack an understanding of the fundamental mechanism for the formulation of synaptic connections between sensory neurons, cortical neurons, and motor neurons etc. I'm convinced if we can abstract all the unnecessary information (necessary for bio but not AI), a more general "algorithm" will appear. The part on speech comprehension hints at this goal a little bit (do comprehension neuron activations have some time delay function, allowing for easier reactivation when the same sensory inputs?) does some bottom up memory layer prompt neurons for activation with some degree of error in during prediction tasks?

  2. hi,

    I have a question about the mind and whether it produces a signal? I learnt that the mind produces alpha, beta, theta, delta and gamma waves which affect neuron activities and also that the neurons emit electrical pulses. I have this research which i am doing which aims to know if there is a signal it can produce. Any info you can provide would be useful.

    Thank you

  3. If you are talking to a bunch of techies, its probably worth mentioning source localisation with eeg, and event related fMRI. also, its BOLD, not 'using blood', i dont think anything in the world 'uses up blood'…. (@ 22.20) even if you are offering a low level of explanation, you should not be saying things like more blood comes up to replace the other blood etc…. you are a scientist, come on!

  4. @bradleyvoytek
    thanks …i have a feeling that it may have already been done.

    do u do work on the pineal gland
    can u explain this ritual carried out by black children
    mental projection perhaps
    utube: kite + hit + steel + plane + must

  5. @bradleyvoytek
    as there is a lot of talk on collective consciousness..
    have u tested signals from two brains while stimulating one brain ?
    i had an experience with a lottery ticket …
    the host of the TV program asked the audience to picture the numbers and the following week 3 of those numbers turned up …
    the program was pulled off the air after that

  6. Have we made "any" advancements? This guy keeps reiterating the idea that we know nothing about the brain and all the science is just a pure crapshoot.

  7. he talks fast.
    distributed computing is not applicable to brain sim.
    samplingrate isn't the problem, (cell)-resolution is.
    CERN/LHC is collecting from many sensors at a time.

  8. The cpu cycles needed for simulating the human brain are crazy huge. The most powerful supercomputers in the world can barely simulate the brain of a cat. I guess something like Folding@Home would help, but yeah, cpu time is a major issue to overcome.

  9. How Binary is the Brain, today's computers are 100% Binary, but what about Neurons, are they really either on or off, or are there states in between that make a true correlation between computers and AI programs and brain and human behavior not entirely possible with pure computational analysis as done with today's machine code binary code that is computers?

  10. @Leobons
    Well, probably there's also people who didn't do med school 😉
    I'm exited about it.. guess my sleep has to wait for another hour 🙂