人腦工作的復(fù)雜程度遠(yuǎn)比我們以前認(rèn)識(shí)的高。神經(jīng)回路的信息處理過(guò)程中以前很少關(guān)注的是時(shí)間因素,。關(guān)于神經(jīng)細(xì)胞這一復(fù)雜網(wǎng)絡(luò)是怎樣工作的,，奧地利Graz技術(shù)大學(xué)的計(jì)算機(jī)科學(xué)家們此前提出了“液體計(jì)算（Liquid computing）”理論,，近期他們剛剛完成了該理論的第一次驗(yàn)證，并進(jìn)一步破解了神經(jīng)元編碼,。這一研究由奧地利科學(xué)基金資助,，有關(guān)成果發(fā)表在12月23日PLoS Biology上。
Graz技術(shù)大學(xué)理論計(jì)算科學(xué)中心的負(fù)責(zé)人Wolfgang Maass解釋稱,，腦部逐步處理信息的理論已經(jīng)過(guò)時(shí),，人的大腦不是按照流水線的方式工作。在加工信息時(shí),，時(shí)序可能比以前認(rèn)為的更靈活,。研究人員以水進(jìn)行了形象的比喻，腦部工作就像在水池中投下了石頭,。這些石頭導(dǎo)致的波紋沒(méi)有立即消失,，而是相互疊加并收集相關(guān)信息，例如有多少石頭被投進(jìn)來(lái),，他們有多大等等,。主要的不同是，腦中的波紋在神經(jīng)元網(wǎng)絡(luò)中擴(kuò)散的速度非?？於?。
隨后，液體計(jì)算理論——其奠基性理論首先由瑞士神經(jīng)學(xué)家Henry Markram和Graz技術(shù)大學(xué)的計(jì)算機(jī)科學(xué)家Maass共同提出——首次被實(shí)驗(yàn)驗(yàn)證,。然而,，對(duì)驗(yàn)證實(shí)驗(yàn)結(jié)果的評(píng)價(jià)和解讀，則構(gòu)成一個(gè)新的挑戰(zhàn),。研究人員需要破解大量神經(jīng)元以分散方式編碼信息的編碼機(jī)制,。最后,，借助自動(dòng)模式識(shí)別方法，研究人員破解了這一編碼機(jī)制,。
研究人員表示,，從計(jì)算科學(xué)理論中衍生的人腦計(jì)算組織結(jié)構(gòu)的假說(shuō)，通過(guò)神經(jīng)生物學(xué)實(shí)驗(yàn)驗(yàn)證并最終得以證實(shí),，這項(xiàng)研究是計(jì)算科學(xué)和腦科學(xué)交叉成功案例之一,。（生物谷Bioon.com）
生物谷推薦原始出處：
PLoS Biol 7(12): e1000260. doi:10.1371/journal.pbio.1000260
Distributed Fading Memory for Stimulus Properties in the Primary Visual Cortex
Danko Nikoli?1,2#*, Stefan H?usler3#, Wolf Singer1,2, Wolfgang Maass2,3
1 Department of Neurophysiology, Max-Planck-Institute for Brain Research, Frankfurt, Germany, 2 Frankfurt Institute for Advanced Studies (FIAS), Johann Wolfgang Goethe University, Frankfurt, Germany, 3 Institute for Theoretical Computer Science, Graz University of Technology, Graz, Austria
It is currently not known how distributed neuronal responses in early visual areas carry stimulus-related information. We made multielectrode recordings from cat primary visual cortex and applied methods from machine learning in order to analyze the temporal evolution of stimulus-related information in the spiking activity of large ensembles of around 100 neurons. We used sequences of up to three different visual stimuli (letters of the alphabet) presented for 100 ms and with intervals of 100 ms or larger. Most of the information about visual stimuli extractable by sophisticated methods of machine learning, i.e., support vector machines with nonlinear kernel functions, was also extractable by simple linear classification such as can be achieved by individual neurons. New stimuli did not erase information about previous stimuli. The responses to the most recent stimulus contained about equal amounts of information about both this and the preceding stimulus. This information was encoded both in the discharge rates (response amplitudes) of the ensemble of neurons and, when using short time constants for integration (e.g., 20 ms), in the precise timing of individual spikes (≤~20 ms), and persisted for several 100 ms beyond the offset of stimuli. The results indicate that the network from which we recorded is endowed with fading memory and is capable of performing online computations utilizing information about temporally sequential stimuli. This result challenges models assuming frame-by-frame analyses of sequential inputs.
