h+ Magazine

Fall 2008

Issue link: http://cp.revolio.com/i/393

Contents of this Issue

Navigation

Page 33 of 41

34 #1 Fall 2008 Neuro Overclocking the Human CPU A primer for the future of human intelligence James Kent Although the human imagination is capable of many things, it is very difficult to imagine being smarter than we are now. We may be able to envision a life where the average human can hold hundreds of facts in working memory and manipulate them all with perfect accuracy and efficiency, but it is hard to imagine what that would feel like. How much more would we "know" due to the heightened capacity of our super-genius intellect? Would the feeling be cold and computer-like; would it be eerily prescient and clairvoyant? Would it be god-like? ese are more than just rhetorical questions. While we 21st century humans are currently locked within the framework of our genetic neural architecture, our spe- cies has gotten to the point where we can routinely tweak and build on the physical traits we're born into with some training, chemical or surgical tinkering, and/or tar- geted genetic alteration. Messing with the fabric of human intelligence may be an eth- ical black area in today's climate, but super- intelligence research is well under way in many forms right now. We're heading into a future we can hardly begin to imagine with our primitive brains. Human intelligence is already progress- ing in ways we cannot accurately measure. e sheer force of evolution, culture, and centuries of written language has imprinted our neural DNA with the networks needed to process abstract symbols and draw com- plex hypothetical conclusions based on available data sets. is is the core of hu- man intelligence: the ability to compare, contrast, and juxtapose sets of data against each other in order to draw accurate con- clusions and predict likely outcomes. Unfortunately, our mental toolkit is comically weak, allowing us to only hold five to seven variables for comparing and contrasting at any one time, and constantly needing to "dump" whatever is in work- ing memory when distracted by new tasks. Lame! Not only is our working bandwidth low, our long-term memory is lossy, leav- ing us to rely on external storage methods (ideas encoded in symbols or bits) to com- municate rational output to other people and keep track of all the new "information" we create over time. For creatures that have short unpredictable lives, this limited setup might be okay, but for modern humans it leaves us wanting more, better, faster. Since we have external memory stor- age down (thanks, Internet!), this leaves personal working-memory bandwidth the most lacking of human traits in our time. In biophysical terms the bandwidth of our intelligence is limited to a tiny conduit of neural cables running from our working memory in the brain's frontal lobes, back to the abstract symbol processing networks in the parietal lobes, and back to the work- ing memory again. is intelligence circuit is where all the heavy-duty puzzle solving goes down when you're reading a map or working a Sudoku grid. Human problem- solving requires that data moving along this circuit be fast for focus and precision (good conductivity) and robust for complexity of thought (dense wiring). Increased speed and connectivity along this circuit is where the future of human intelligence lies, and there are only a few ways to get it moving in the right direction. At one point in time it seemed that drugs were the answer to this question: Dex- edrine and piracetam, cognitive enhancers, ginko, ephedra, nootropics, and the like. While these supplements are indeed nifty for achieving short-term focus and mental clarity, they seem to only milk the limited capacity of our current wetware without providing the instantaneous multi-point IQ boost we would expect from our "smart drugs." Drugs can increase human intelli- gence temporarily by increasing the speed and conductivity along the intelligence circuit. However, most of the evidence to date suggests that the brain will eventually begin to power-down or tip into psychotic states if this method is used or abused for too long. To build long-term conductivity you need to train your mental reflexes just as you would train your hand-eye reflexes, and like any training this takes long peri- ods of discipline to see even limited results. Books, video games, and websites that fo- cus on multistage puzzle solving in strict time limits (yes, I'm talking about Tetris) are probably the best way to get the logic circuit wires crackling and ready for more complex problem-solving, but what about improving the robust capacity we crave? Data capacity, bandwidth, or robustness along the intelligence circuit is the main shortcoming of human intelligence, and what divides the geniuses from the morons. In real terms, this metric defines how many abstract symbols we can hold in working memory at any one time while still per- forming rational analysis on those objects. For instance, how many words from the last paragraph could you recall if you closed your eyes right now? Could you remember enough words to complete a simple seven- teen-syllable haiku in thirty seconds or less without any errors? No? Why not? If you can do it you're probably a genius, because that means you have the capacity to hold at least ten or more random words in your working memory while perform- ing rule-based contextual algorithms to rearrange logical syntactical output under strict time limits. A computer could do it in a snap, but the limitations of our working memory make this all but impossible. is capacity is a trait we cannot easily improve in a lifetime, not without radical mental training, dodgy neural steroid hormones, or even dodgier drug-induced neural plastic- ity. What we do know is that this capacity for robust intelligence is genetically inher- ited, which naturally gives some people the upper hand. According to Richard Haier of the University of California, Irvine, one of the initial founders of the Parieto-Frontal Since we have external memory storage down (thanks, Internet!), this leaves personal working-memory bandwidth the most lacking of human traits in our time.

Articles in this issue

Archives of this issue

view archives of h+ Magazine - Fall 2008