NEDO-PR - 9609 003

¥/£ 9 3 M

rw-ttMctrr \S UNLIMITED ne THIS

#2: $.;[/#-

MBiix 3a 9 ^ 3 vM 1 8 4 m i :am, mm, miK | mm fjBr, c% Bm^mdefa/zAc, #$mm, ^fm^#^:mjt6z6(cA"3T, DISCLAIMER

Portions of this document may be illegible electronic image products. Images are produced from the best available original document. 9 SrafW2£ffl±3lfiW&iS YSBtt 3$ II8 • -*ir*xie

y £ * 6 39 sir

6096- ti d - 0 a 3 N »s®agjae^$* 8 39 ± B

...... (m) mmae# (##&#)...... (iv) ...... W) Abstract ...... (ix)

1$ msmmomg ...... i 1. 1 3 i. 2 ...... 4

2$ ...... 7 2. i # ...... 9 2. 2 momma##### ...... 19 2. 2. i ...... 19 c i) m#mm#&#Ykows - #?## ...... 19 (2) 27 2. 2. 2 ...... 33 (1) KPACAP-38C «k...... 33 (2) MAJOR BRAIN REGULATOR -  ; >I:j;6mP9vA7X^O|l§@i## - 41 (3) ...... -...... 48

...... 53 2. 2. 3 ...... 59 2. 3 ...... 68 2. 3. 1 ...... 68

...... 68

...... 75 2 . 3 . 2 A##— a —7 )]/%• y b • A y ~f ...... 81

...... 81

(i) (2) 7^vh 93 2. 3. 3 ##%#!:3^:] 99 (1) 99

...... 101 2. 4 #2#ftmm3>(:^-f 106 2. 4. 1 106 2. 4. 2 110

[ Mt i cam) ...... ns

#3# m%##J ...... 127

S4» tStlf ...... 131

Apendix ...... 135 A. 1 &### -...... 137 A. 2 151 A. 3 m ...... 167 (2) 177

(ii) m®#i#6##®A

^m®3@®m#^^^^ /m, Ammmm, Ammiem®4#^m %L^ m'&fau^a^'^&d^Ao m^mc/iatctiicmAT, AmcgT&mA

&o C^g)®m(D#^lz^^0-zL-c, A-3 C® liLT®S^Kol'T(i^lt^5^ B©$jsoVT(it A+#1:'te A Vc C C CBm

%#^MLTiEm(D^b&%A6Amm#^ m#®mtLT®mAg^^-c(i, mm®

^A-y% qrm%6^m#]&^#Af e c t a j; (), a awm L#^f - a#A m, Am%E#. Amm##. Am&%®##^e®j:3^@ammm#^t)'D^@#m[g# m®##6L-caa%3fi5o L^L, T&&o ^A*%®gmA tg ^ ^ % v \ o LTAZM^m&mO 9 6f 6^^.^ %% ##ma®#m ^ T ® ^ A & 5 o

(a{t###8r #miEM)

(iii) ft Inn 4 ft ft B S # K IN * UN -X ft B m ml m IN SB 4y # IN ® 0 B # f ftm # ft # B

m Iff Iff Iff ■iH # « W m m 11 $ HI A % mi sm BR 1 •§1 m B< BR & # e m IB % ft # IN %n aw M tM # iN igi # Ink ffen w ft s IN

— — m< m ft # ® m ti ft m\{ m ft ■a m E E E -U s ffi ft IB MB 5 Ml w % ft % 0 -rt e mi; m fe % # N ft ft B ft B B 7C mmg*#A### #«## eai %?###&### m#?## E- m?K#mA### #^### m^m#m#K#### m###^ SES m?K##A### m^mas an mu mm##mA##m m#7-mm^m+as #m Em m?K#mA### m^eas me### min m# mm##m&### me### mm^mmA### me### m# amm#;*# ## ### Mil ,s iE#^:# mm WH /j\# amme# mm### m###m i§ amme# m#### m#### #% m#### m###m Ujlllf #E ##/m<7### ###pmm il|± m±m xu7^7;i/i>%7A### m@### tr# #mmg§ ##mas m#### MB m#m# %#### #m mm ^t-7 tmm#### me### raj Tim Amtm m^##### me### mma# %#:m#K###t>^- #m cm #f#^#m##^#m^ ##^masm (m #m#mm##^^m# A@mgas# ih£ mx wm### /\\ff #y# m#K#^m###K###Mm# ##^^ SF^ ±0 #$ mmR%##Mmas ##^mm&ft@ A! rm m& ms mm (*) #ra#mm##^#m6 ##mmas cm ###^m##mmm^ A@^#as me###

( v)

a a c #, v % 7 Amigm ?xr ^D3>b°zL-^, K^J:7^6V7 h^^7cmai#&-e^

7/1/77 7^ 7{L^^a:BA%1##ma^, 7>7-7? ^^^mLT##$aai##(D

#TUao CO/:)6, #C, A6/ %2'##%a##§60fWI&-3%<"'?:/7 7:/7>7 7j:-X, ^^,(C, VX7A7>77^- % % 6'0#f % %#A& o#a o Z3 L/:#ftog*(=^/:#Ta#M^®yi/-^%;[/-^#a^mci6Li:, c®^#e %-e(±A®.%m#mcB8]i-a(9 ±(f, c^^-ctAM7;t/=AzXA0@#8g#vX7AT&%, ^yy^L-cm^aa, 6 a#^o L % y o h f 7 y#f dooomm^^A#m L#m[aim^ i 7 y y{b L u 7 ;i/7 7 A a i##mm[si%) &:&# L aimi^T, 1000 maam ^ L 7 % 7 A ® #fp^#E Ldcoifa^oT^ao C#km#03>t:^-f(DeWB0[(:&#fa62OOMIPS(;:#^fao ^y=, =k<#^fV7 h7j:7®#%t#^LT^ao -4Bo##%vz7A##(D#9s = m^@cj:',i:, aff#&@Cf/=6 om#mm7;i/y-XA^mmMcm#f a. #u#x.fuf, mo#m3#aw7;i/y-%A ^Etmi-aytve^o, m^^a#o77U7-7777f#3>t:ji-7&9(:a:E f aM^mm^ac#^L, m^-6 w A<#fLU7-7777t^#x.g)^ao a/=, EmeBgRfam#(d:, ^ ic#A u, samsma^Mam ^ fi a a^ % G, a^c-f a c 6 a

(vi) (fMRI) #bfrr#C6Tl'&o MC. mmommacaaaaw:.

y hagf^# rm^3>b°zL-^j co^7) j. am# rmoa#

Tl^o 3%, h7y7 - ai'-p^yxf-AoM

w< 7-^7?f-

ca^a-cua. HU PH 7 — 7'> 3 "j 7(International Workshop on Brainware)7:(d; i’mctffSMSCDjrm

*ys/a>j &o^r^x^-j 7tyy3>-eSi^^0#P9^(cj:a, c®^wo#&0K%

Mft&.ffiWjfait'ty-v-7sL-i? 3 >. MkvxT-Ay □'fkS®, -Yy - vSii • £i5W UT. #P!^^g(:#%g LM&ff ^ /:o

maj w®#p9^®mA^#T^mLAo

(v£) Abstract

In order to cope successfully with the highly integrated information society at the 21st century, it is said that excellent and various R&D, which themes range from hardware to software, such as high-definition television communication system, super high speed microcomputer, human friendly information processing, is required. In the middle of the ocean of the information from the multi-mediated huge intelligence fountainhead through, for example, internet , users have been requiring the environment in which they could obtain their truly crucial information and utilize it to their daily life or business. So it is necessary to develop a new information processing technique, especially R&D of man- machine interface, which connects the human and the information processing device such as personal computers, and more over, system interface, must be executed from a new standpoint. As one of the breakthrough to the new R&D which meet the requirements of the times, this Leading Research Committee (sendo kenkyu) focuses on the field of the function of the human brain, sense and perception, searches a new concept of information processing device which has a thoroughly different architecture from the one so far developed, aims R&D of basic techniques to build the new system. This is an endeavor to realize industrially the human-like information processing technology such as memorization, learning, association, perception, intuition and value judgment, all of which were primarily performed in the actual brain. To be concrete, a conceptual design concerning the basic element of information processing peculiar to brain function, and a principle of development of the new information processing system, are investigated.

Study of the basic element: The brain is a system which gains the algorithm automatically. Breaking down it into the elemental level, the basic element, which has unique traits of neuron, i.e., super parallel processing characteristics and knowledge accumulating characteristics, is attempted to be realized industrially. For now, a prototype element which imitates the neural networks is experimentally manufactured. A thousand neurons are completely connected and made the neural networks into 1 chip. The chip simulates a information processing network system of a million connections with real-time execution. It tries to test the performance of super high speed neural network systems which connects the 1000 neurons parallel with each other. This deserves 200MIPS of the performance quotient of conventional computers. Software which efficiently arranges it is also investigated.

Study of building a unique system of the brain: The brain is said to obtain information processing algorithm automatically for well-adapted behavior to its environment. In other words, the neural networks in the brain are memory buffers to store the algorithm, and brain is a sort of the memory architecture type computer. The essence of information processing in the brain is architecture which read out quickly the best output from huge amount of information stored through experience. A totally new mechanism is conceivably possible to perform it, that is, an architecture which have not only rough conceptualization of input information, skillful fusion of the information processes which test their hypothesis with top-down processing, but also excellent output

(ix) response which extracts required information quickly. Moreover, as for a mechanism of memory formation, it is important to make clear how it integrate the multi-modal sensory information, how the short term memory is converted to the long term memory. To investigate these, the newest technologies of optical imaging and / or functional magnetic resonance imaging (fMRI) are to be developed and applied. Further, encoding and decoding of various sensory and perceptual input and output are investigated to approach to the information representation in the brain, and a principle of interface designing fully corresponding to brain function is explored. This report of the Leading Research Committee has taken into consideration these technological movements, presenting the concept which takes in advance in “ Brain Computer ” , which is the final target of brain function oriented information processing, and making direction of R&D. That is, for formation of biological and industrial models from experimental brain studies, the strategies of learning and learning control for the brain to aquire the algorithm automatically (Brainware) are studied based on knowledge from measurements of brain activities which enable to make such models. Furthermore, computational neuroscience proposes neocortical neural network models which construct realistic neuron models to understand oscillation phenomena in the brain. Besides, evaluation of large scale neural net chip system so far developed confirms its ability to learn and process at high speed. Further enlargement of the scale of the Brain Computer as the first generation suggests construction of a new information processing system. To yield next generation of the Brain Computer, it is required to research and develop architectures founded on some new concepts, or elements and systems which realize them. Positioning the issues in information engineering field, and drawing their realizable images, are tried. At the International Workshop on Brainware, there were lectures on recent technological movements in this field by Japanese or foreign specialists in 7 sessions, namely, “ Information Processing in the Brain (1)”, “Information Processing in the Brain (2)”, “Measurement of Brain Activities”, “ Devises and Systems ”, “ Brain Computing ”, “ Panel Discussion ” and “ Poster Session ” , and active discussions were brought about. Foreign technological movements about sensor fusion, micro-technology of intelligence system, image processing and generating, was investigated by dispatching specialists to US. A new information system at which the objective of the R&D for brain function oriented information processing is desired basically very highly and anticipated enthusiastically, yet thought to be lacked a concrete developmental strategy. However, it must produce rapid progress to stack more of the outcomes reported here and to realize engineering application of them.

This research has been deliberated its line by planning and technological committees of Leading Research Committee - Strategies for Automatic Algorithm Acquisition in the Brain (Brainware) - consisted of researchers at national research institutes, universities or companies, and partly performed by the agency of cooperation with outside specialists.

(x) m G m E 6? & # E

W3SM3£a>«

1.1

#g. mm. mm. mm# #?. #@k#0& c 6 ^^ 6 % &«

£ £ S'i:tiST'Ste^S. KBW£ fc =k E' L T <, v>E> 0 Ifc^oT, 21#$2lc*%%(:@j a f 5 c 6 *

^T) ®m&^qr^-e&5o m##. ^#-ma?%#. m*##. mK#o#Am^m < c 6 T& a„ L^*^-c. (XL^^^T) oe^MmirE. m<#m6^^E%. #g. mm. (^#@^mo:t l:j:-3Ty*-XA^#%f6 6#(c. 8m##. #g%#. yxyAy-^f- ^ A ^ m u. mmmmT;i/x v x A mmm#-#-a c 6 &## 6 -#- & mm ^ > i: ^ - f c f^^^fi5^<#fLUvX-fAi)A^60i#&%^ »#%:A #AM@#OMm®x/f

— 3 — 1.2 mmmomx

A#±©m#6LT, M#x/^xioj:D(Nm7X7A©m%M^lj:, Atc@#&(Z)mu ©e^M^mu, mayxfAoMmAM#i:%ao ctib^, M^#v\ r##fkj

00A

A^m^b, i#m

&&A, ^©^K^fMPU, DSP, ytV#©ig##{W mm, g S, 44EWA^JE^-2>A©©®Ato&®P^^btiTVAU 0 :®S!fiM0 t> t A, A Pel, r, -c i ^6 ^ # A ^ mmmmmm# (71/07^7) (±cai:j6A6@f^M^T&6o ^T(±, i^#mm#@m9-tf7Hb(:j;a7-+7?ff##, ^^x- 7x7^^#, ^ 0 T A < jE < < A^ iCM'Z/f-XjL^x A'C^<@S Ax/ xib - ^/ X A AH"^-7cftm^fi6##!^@7x7A©^#fM^^ipM< ^ Cj%(iAa.h©-r ^f-x^xd^mGMmx/^x&^Cfi/xAA© #%M#, an^, 4^##, tom-mmAaiAx /WxjoJ:0(yxAA#Mmg#©^#f(h^%(cm#LTwao #AA7-+A7f^©

x/wx^#+-

- 4 - #&%?%<, #WK

(YL/-r>^j:7) A^c6^T%m?%c%a:"3#fLi'7vi/f^f^7

%7A^0g#^^^^6C6l:j;-3-C, (7L/Y>^^7) (i. ?;!/ fv-r-f 7#o##mmm#§(f^!9T%<. %.i/^ #m- g#

(2)

ON#^#^6 tm < m^T# 60 #/r LTomr#

-3-C, m##0Z#%##aE#6LTO 0^(cj:-oT^0##mm0 mcD%#^77o-f &o

MO### 6/: /:6 3 o 7c)

- 5 - w ro #

u V rr0 M I

8

ft 35 $

SE2$

2.1 if h

E#i##ma©## (yyi/n'vxA) %7-ff7f+©#7 f >y >k°zL-^?&5o >k°a-^A<7;L/3*U X i f yX^LTm&z^mm^a-o-o&eo m#7 7t:a-^^x %^©x n-t'yf • <-% • 7-f f 7f©7 * >y f vy^n y t°ji-^ j:|n|S(-lfS’Mtt^t)

L < ^'6 7 o zcTkk m#3>yzL-fA<7;i/XuXA^gm##L. f©##y-a7;i/XvxA©g (X7f 77^7) &K©&©©##l:j:-3TWb;M::f act^ckD, mmvxf "C* 8 f"#-6o

(1) “SaOir ^2o077’D-5:

m^##mm©7;i/XvXA^@###f6y%f AT&60 c©/:a6x mz)##L/:7;i/ XVXA&##fM(;:#%L, e©7;l/XVXA^#%f6CtT 6U77X

#S,4lTUS7Xo-fT&&o C©7Xo-f©miT,(C=k% %#^^^©#f mm#gr^mf6o C©%#^lcj:^-C#iJ3tT,5m©m*(d:. #&© (Xotvf^-X 7-ff?ff(%Vf7>#) 37(:a-^f©^©T&6^ C©77b°^-^©V7h7 J: 7^f#6 6 L/:7;UX V XAd

cm:*L. M©#*^@©7;1'XUXA@###©##& (C0#«&©7;i/XvXA(±lAe #W##L^6©T^ C^X7-T 77^76^7) Z©X7f 77j:7&6 tl:7;l/XVXA@###©I^7%fA (C©I#7XfA^B#(7 7t:zi-^6^) 6U7^©#gRAt#^7Xo-f^6o C©# ^7Xo-f(i, JW^'HR^C6a<'e#&6©7##(c#BL^ ^^^C6©%###6 LT©^ff#©#0SM^6It^6C6^-e#j;7o Lx#^6mm^^g-eab5c6^#j(p, c©mm^%6c6-e, mf©t)©©#^%g,

- 9 - ^ifm#T#6O^0T;l/=rUXA(j:-#-<'^ N3%l'o L^L, m#|::Vt:zL-fd

VXA^tfrffu^ m##f57;[/XvXA®m#lcAU| m®

6@-e, m0##m@y%^A

(2) BiDvea-^ : Bilii^l'^^yt ’a-^^

w<&6^cAmmLy=7;i/^vxA)

^##fa##6LTm^3fL6o mac®##®/:*!)*;:, ^©ISMtcS-d^ AAttlE^FNEU tiiJj'fZZ.ii&i&Wt&tUXtiijj’t&o titfj'fZt^ 7VI/XVXA (#) O^M^iWcD#6o mb°j.-^-r 7 -'Jit, i!iy€lJ/<-X7-^f^ft©^7t>y^7> m2.1-m) . 6L^C6T&6o C^UC^L, 3 >k°jL-^(i7ot vU-<-%7-+7^f+ 0 7^- >y ^7>g(3 >l:a-^-eab6o # #(±, A^7>k°a-^(:yo/7A^#x.6Ct(:j;-3T. ##mmo7;l/XVXA^A XoX7A(c#^T&6c:6 Xny7A;:#l'TU%uc6MaA<%WT#m\ T "@Ui##%m^X7A" T^6o C03>b°zL-f(i?-a.-V:/X7y->T6O. ## C0/z6lcA#^C6(j:Xny7A6l'7B-e7-^7;l/^m f6Ch:^-^T^9C6 (i@6Tm L Wc

7oX7A&^.x.^<6%)#y#(W^@ATU'5o ^i3(C,

60T. -e&6o 8m:i>l:zi-f kL %-oT. maimcj;?

mo;

-10- 7;l/XVXA0#gE*^6 U3##-e, C07VI/X V ^ C 6##69% 6 0T&ac6T3b6o Cfl(:#L^

bf\ ^C0##^#"3 : (1) S* 6^JK\i~fu 'f 7 -Mi gSEM £ *15 © "£ 7 n X 7 A^fifciC

mmT#. &t,c (4) ^m#-e##^Ramc*LTti@@ikT%a.±(f T^) t°^- f 6^6?o

(3) ^KV5i7:7;H'JXA| r A mmmy>f7>S=]>t 0a.—^ V J

m®m#mm07;i/=r»; XA m#m# 8a:/tv ^-XT-^rrTf-V ya-trytf-^ —X7—%t77-v (7*>y^7>Sa>t"x —51) ©#% (7;(/3'VXA) 6LT©7i/^ ?a?9A >7^7^. 7P ^2 -fe-y+h LT# T—S y: EU(at66t)) ^n»7^-T-5®6SSf»

svj o&m b* m= aj * bsih **®S: *et#m##L A7A=f v XAT&60 3gm#M=#f®&S x Xt-:/X$e *e>*'L't*a®a @L'iws®a com*^ 7l/^r>7j:7(d:m6M# X*ffl®lzjsfLiii*r-SSA< ^P^xXCSL'rfe-SCfctA' !&$£8:h.tf@fiK*HS-C$5 SfSUcl' ^mm##L/:7yi/7vxA-e&^, t turn mm'&&tm.t>ix& U-ffsaffittAfSES^Tl-'-SAt ■e®fctoic-7-rL7;u$Sffi7'i. ^xa« me##(D7;L'=fvxA@m# C"EUS®B#iSr(bL®jM^g VXK0x7M#(=*f6*@i@ *?#* (yxh't ’xxy^-fvX) o >©) "Sr^X.-S{^(i-» ® T4 /\—H7i7: ja-®e*S6^fc(t / \—KOx7: ,r w#mm#44m (wmm) -e&at ^%l4k,Uf-®%X : Zfn-tL'y-V ® Si£fc(-#oTttB?"-5 MBA l^ 9 ]2#A 6©#^^##7& 6 9 3) o m2, i-m

—11— (D ### 6 %, #g{k0^fsiy:^#^01^

6^ &a#^0B#im##*fiao ^0#, c 0%am*^^A^$#^#T#mc4' &6, co^r 6c6^&6o-e, c@^r>/<;L/%Aia=' (#2.1-2EI) o #m(2l%(iA^M(:j;6(3g& 0#m@#&#@t: f()M#L, Lyz^-oT, wmmg&o^g-EmtkL A^m f »9 L /:

L<##Lf9tfat#, A

0^m#t t)9 —90^6 LT#^m# c#m) 0##^m#misi%(:^0$^Emfa Lfcd^T, 8B(#^(d:^^m#^@^0l3K04:(:mOi6A6#m'J^m i^T, ec-e^0m±&g^A^0f T 2'0 =k 9 CE# L mB L T u & &&#x. 5 6, ##^0 g <" (cMiSf &$###(:

&mii# #rfsittfzt#5o ws#fm 0#^: t #M%5#^0#±A#F^##

#^(:mgT&ac6^^ao ^f##g0ck9l:, & a$#CS (#i)±) ^A^7L/:&6^ %0$#UCS ^

12- A^ t L/: t # lc##r##W^#%f 61 u 3 60 j ^ ^OA^J^ Xj(t)^ i

A^,=c .^(/)^(r) ...... (4)

CCTc[j#@ig*-e&&o y=/:L. AA^I^WC^U 6 9 (9 , e f= A c ^« ASM6, A^J##Cck6##%####LTi/A7xaiS&&a5Z 6TNMDA^##0f-r c®A^70]K#^Ca-r^y^a6 LT^A^x L(fg) < c Ati. A^^#^W(:#mL#af ^gm^A^m m#H(t)6LT^gk0j:9(c%)66o

^(0=^(0 + 9^(f-l) ...... (6)

L (f 3 ^ ^LT U 5 6 6 C 61C J: -3 T. 5 #mU^ 6 # 6 o A^mmm^m ^THebb###j ^#iEf- % 6, A#w# 6 A

#2Ei(c^f A^0mmm<^i>^-y%(c^j$^

^0A^JTiWB^Af 6

A^,=c ^(0^(0 ...... (8)

C j; oLyc#B^ 6#(c^A L ^j:3(c#@-e#6o L/c^-oT, ^OAA^

c^^ma9%#m(cf ^A^m^comm^vA^xfcca^^>#mo«k?%BT

-13- a^x >#&#&&#z D#wt, *§^o#{bT% < cf$& tiTU&o C0C

AFF},=^.(^(/)-j,)^(/) ...... (9)

2(sl>s2>0)&#&„ s sl6#@%EM@3 2(sl> s2> 0)^06 #6LT, ^A#^OM^tLT#

c ®#mij (c j: *x z 3 c * ^^mm#Eim0##[m% A<##g ^ fia0^65 3^0 ic%-3%#*#%*&&0#fi:aL-cimmicE#: ##t0 * (c 0:Et@^<^# ^fi6C6(:%6o

^gR'i ^(0 = 9 -^(f-D + ^W VXA^@###f50/f5 3^o BkL g###f 6 Tvl/d" V XA(:*f 6##l: fi ((c ^ ^ -c ^ e ^

^0M#CScfUTA^ mi&omRma&mfo

C0/=A

67;l/d"VXA time

WTatfTt>tif=6#, A/c#f^RT^## 12. 1-2EI

—14 — (#2.1-3 El) o 1. mnmttKDm)

##&%©#©#%%< H y ^ 7" 7 y$\M Atitz fe&EIS. ^SffiWl

Cc^A:#A^T&&o -^WttlSl-h *»«3EA* (4) Vial^-ya^'y^TA

s. e

Cftb©77-r7£lME1"5Cti3 >h° - frlilttl*(Dfcto(Dffi®1W®

; a i/-y sc/Aijs-et, 2.1-30 < 5 3 o /:A\ A# ^ ^^ l/-ya>OA36. #ff#0OAT^@A<66o e©AA. m@y;^l/-^3>6^ffLT. 7^A7^0%0Cf6^AC#*L. #m7oty7--tU9BT%.; ^L/-ya Ltz (#2.1-40) o

A^^V7 h%Cf-5A^6l^9C6Ac%-3-cm^mmf CL/:o U6#. m%Voy7Alyl#©-!)-y;l/-f>T#^f6mg^l#©^cm# #%x_Cck3%&(Z)'e6&o ##©#%c%-3-c#m©#^j# %-o-c. m^ocpuc< cm<^%^^^7-e#6o #1^. C©yaty-f-A<±###©2?oy?T?#?f#-&m# #. 4 tfv HOWS4 0. 4 v ho##2 0. 32k:'y h©%#2 0. 16K' V h©ytV©m^ajL3 0. ##i&^10. 1W4 0. I:'y h77 h 2 0. ^7 Pl/XO ^7>^m#f2 0. &mm^3 0T&^o m#©cpu-e©m+^D'y7m^©m#^ 2 7 0 y 7?%mLT^6©-e&^o

15- /

**# SRAM

m*# |AA# /V7-/tVv<%

*4hfa —f tt*U5?X* as *?F9 —9/$—?%—X'/

c-

S2.1-4E ^o-feyt-0fi$^n -y ±ag*#(d:2#0**a§6a]%g, 7+^-Ai/-^, &o ^©fc&A concur vf—T(i#@© d:6©^^(imW±-(fa&6. %8a^#K©##, ##/ • / =e V t±7°a 7"7 A • 3 - K (-f >x F 7 2 7 3 >© V x >) 6#### (#4-©#6M L, />> T'ffl © x - 7Vb / t V &F# L, *7-y

1111| if|.| 11 \ 11

2.1-5^ 7oty7--Oo7#m@%(j:^35K7- f'T&eo 25WHzT#^f ^C©7ot-/-(i 200MHzO/<>f-/f7Aj:»)a^|:#mig|%o#tig^mfff6o 2007otv7-mm 2O-5O^#0mm^%taV7^^4'AKiKmm#f'5C6:)O(Z)#l#QFP (#3.1-50) T, Fi#SmSRAM^^#f6 C6"C\ !@7oty7-^#gRf6o 16#©7oty7 LT& ^ , 5OO^±0MA^j$-3$b 2 ^WLTl^o ^f#7ot'y7(iSRANW<^#e6%-3T^6^A ^yy^^mmLDRAMfbf^?%?&6o &&CA #mmi@-e(d:DRAMt)^mLT (7y7m% DRAMlc7oty7^|^@^-WrT6W9^#^59) , 32t:>®S0J/<'y7-^(2 jRfe £>tiJ: 9o CCDflM#/1V 6IM#© 7 V 7^20—30### L A 7 - F^''°V3>©X o y ct -5 77 u 7-^3 y - 7T& &„

(5) m v

7>/-^y (/-/) L^Lx f##i(iA (M) A (B) 60KI#(cjoUT, ##&%x.6Ct^mT&6o A W) (j:7o/7AOj:9(27 — x.7JlJPMm(iZ$t-oTnfr£tL6cr>-V( (: ^- / (d:A^ti:6©1*#©^ 6/:A(2(j:, A (M) 3OTiN'^§T\ Zb''zL-/-C&53o i:LT, m#3>l:°^-/(j:m%©7oty7^< -X7-4f^ft®3 >k°^-/®|#oAKlftP^ISB, • 5o ^oT. Lftf, M10-Ml00^pgg#^^^A fc6A y 7 7 B 6 C 6 ^ fftft J: -o ####© L & & 6 7 6 ^ 0- Af C t ft % 6 © ^ AS^ (i l: &b#

C6ft%%ft -e®A#(j:@6TAT&6o (#fK#mAW^^ %A 7C)

-17 1) Marr, D: Vision: A computational Investingation into the Human Representation and Processing of Visual Information. Freeman and Company, New York, 1982 % WK (§R) : , mm#, mm, 1987 2) niA#M = mm^#, mm, 1996 #%#ig, mm, 1996 4) Shigematsu Y, Ichikawa M, Matsumoto G: Reconstitution Studies on Brain Computing with the Neural Network Engineering. Ono T, McNaughton BL, Molotchnikoff S et al (eds), Perception, Memory and Emotion: Frontiers in Neuroscience, pp581-599, Elsevier Science, Ltd, Oxford, NY, Tokyo, 1996

1060-1065, 1996 6) Koerner E, Koerner U, Matsumoto G: Top-down Selforganization of Semantic Constraints for Knowledge Representation in Autonomous Systems : A Model on the Role of an Emotional System in Brains. Bulletin of the Electrotechnical Laboratory 60: 405-409, 1996

18- 2.2 moftn&w&momb (7^vi^i7)

2.2. 1

(1) • #?## a. liCfefC

6@Ax 3%-€-<&*&#

M-pTl^o ffl.2. 2(lHElO-%%:?mT %2#^ %-3-C##WBO#kTMT, &523=8#®#a<&&K:##'%l\, —^A^tf E«, (#)

% 1 #r#A^#yz/fl:#AL 9 & 0=K O3>ifa-^-^%7A

m#mm0=k 3 %##%##&& t?L5x^fC kh

(D-%kfr£.W(Dsp-C£'(D'k 9 mm A& 3 -o -cm#-#- ^ c t ^©^TN^JxJ^

3-2## A & 61 ^ c f: 7 X o

#2. 2. 1 (1)- f >/<>^-

-19- V'o (f tumT0J: 9 C ^ 6 9 (#2.2. l(l)-2m o

(i) (%m#m) <%###&&<%##

(2)

LTBrnwcmij-c#

9o

(3) — >

(5) l'

(6)

#2.2. 1 (l)-2g

— 20 — b. E#ES^£5Mb t$f£5rfi£ LTU a © C*f L. D A#T &c-c . m &# & C W: D ± t LT^2.2.1 (1)-2EI. (4) (5)(6)C t C d; D ^

cisj^d fc©t#Abtiao ym{b®m^o%^-e#Li##a (cc-e(iA^) -m.

J; ^ < ©WH^'# ^£ ti a t © -5 C tA

g)#*###&&<. ag,(i#mU#@l3&C^ UT^m #®#Emi)&#mw<4HbLT

"####" #MT&a C t^-#%C^LC©(d. 1924^{C^^$ tlfcSpemannt Mangolds

2-oB0#w%a<#{bLT

-21 W<65 6, #]gmi::W5g##%^mm#A(mej*ft,

c. Spemann^F{£X %>^r — il'f'^ If—t J^ait$lll^3S;iiii?=l©filJILt$ftl'''T\ 'ft — ii~5~4 If—•£■#0 %#, (2) , -r -f - o##fpm ## & cj;ki6m^^%T6636#%^ti'e ©$#©m# a%#wwmm-cM#6%aB0BBL :o%mGL w.

^fi6B#BgKa?ofFmca^<6^^ o) , c©a?©^%i)m^to Spemann[il935^i-0 r##©jg0g%mc^(fa##fFm©#mj (:j:0/-<;!/*&##

######. BBBgR B icm 5 ^ -r*#^© ^^m-DTL^-oto 1987^, FGF7 y ; V -^TGF7 y 5

##. #^#©,

1993^Eic/j:^Tx #^(cj;bx "J / ii x-hMMMM(Dy]4Xl 7 ntSS^icfc VT(6, 7) . bFGF^^ -tf-maao####fFm <^fte L %, ^^k^Ernmm^ & m jgk (iMtt#EEM©feSaE5S^ES5Hb ZMr&zt^ £ ft t (8) o C©B#4:Em#©#m^##^ftac^(j:%<, XbFGF(cj:(9, j;!9#%©#E^E ^ ^ o #%©%E^Emmm^ & #;m%©&#ama##a ftac^ waft. FGFc*f6^#^Be/r^

^ t M1993^, Harland#(: j: -eft^ 7r*E###ftNfta 6 ^ ftT# t^f?©—o noggin^#Am69i##&-C(l&"3f:^ #E^6%^m^ftt^Em^ftfFm LT##m©# {k? —t — ^-##7 5 C 6^##^ftt (9) o $ b icSl994^ N Hemmati-BrivanlonUt:: J: Ofollistatin (TGF7z; V-ftJRfa activinC^ALTfOfFm^M^fa^f) ft 6 #mm#f#m©&act^#^^ftt do) . c3LTC<#aft%o,ftft^c####© #9Wft#Cf & g A#2 a C 6 ft % t o

-22- 1995#. (0.1-5. Ong/ml: S-BSOpliOCdtO^-^f-^lf W# L 3 6 C 3^lc L/:

(in o Kzo#m-e$*##m%0#Ko#Km#a%#m0bFGF#&TT##L A:#. RNA^-fititi L. 7 7 iiX-)ly^fr-o/=6c6x ©%^2m^#6#f4m©FGFl/ty^-^E%m#L'CW6 z 6A<%g,^(C%l)x #FGF^

(^tlfmL-l—4«k#^L/:) o ^/r©Am©#

5mxl©^?-e&6Ct^igg$ay: (14) o —^T. 7^Xfj:;l/#m©4]|g#mm#bFGF^W: L:a66f6FGF 7y; (15,16,17,) , —3". Wilson#(ibone morphogenesis protein 4 (Bmp-4)^ESlEflSOSSlBflS^-©^ ^^(Smf5C6(cj:K)x (18) o C©A(C HliSLTiEi^Sp^x J: D. Bmp-4 5^##:©##^-dominant negative StMS

#C=kDchordin 6MfT,5. mic#RLTx (20) o i2(±©%M^6 # - ^ ^ - C =k 5####©^f##6 L Tg(T©Ef 3W#&[f±d<-? T<6 (21) o Bmp-4 ©^mc=k()#^(:^Ykf%o L frLMy&fyMKJr — iikk IF—X)'>b5)‘^'$tl'5FGFx noggin, chord in, follistatin#©^}f ^

-23- d. 4i£/< > i;& 6 A A# Wi##W

lag * M z.) mnim 1) Spemann, H. and Mangold, H : Uber Induction von Embryonalanlagen durch Implanta­ tion artfremder Organisation. Arch, mikrosk. Anal:. EntwMech. 100: 599-638, 1924 2) Holtfreter, J : Der Einfluss thermischer, mechnischer und chemischer Eingriffe aufdie Induzierfahigkeit von Triton-Keimteilen. Wilhelm Roux Arch. EntwMech. Org. 132 : 225-306, 1934 3) Niukoop, PD : Activation and organization of the central nervous system in amp­ hibians. J. Exp. Biol. 120: 1-108, 1952 4) Slack, JMW, Darlington, BG, Heath, JK and Godsave, SF : Mesoderm induction by early Xenopus embryos by heparin-binding growth factors. Nature 326: 197-200, 1987 5) Kimelman, D and Kirschner, M : Synergistic induction of mesoderm by FGF and TGF-r and the identification of an mRNA coding for FGF in early Xenopus embryo. Cell 51: 869-877, 1987 6) Mitani, S. and Okamoto, H : Embryonic development of Xenopus studied in a cell culture system with tissue-specific monoclonal antibodies. Development 105: 53-

— 24 — 59, 1989 7) Mitani, S. and Okamoto, H : Inductive differentiation of two neural lineages reconstituted in a microculture system from Xenopus early gastrula cells. Deve­ lopment 112: 21-31, 1991 8) Kengaku, M and Okamoto, H : Basic fibroblast growth factor induces differentia­ tion of neural tube and neural crest lineages of cultured ectoderm cells from Xenopus gastrula. Development 119: 1067-1078, 1993 9) Lamb, T. M., Knecht, A. K., Smith, W. C., Stachel, S. E., Economides, A. N., Stahl, N., Yancopoulos, GD and Bar land, RM : Neural induction by the secreted polypeptide noggin. Science 262: 713-718, 1993 10) Hemmati-Brivanlou, A, Kelly, OG and Melton, DA : Fol 1istatin, an antagonist of activin, is expressed in the Spemann organizer and displays direct neuralizing activity. Cell 77: 283-295, 1994 11) Kengaku, M and Okamoto, H : bFGF as a possible morphogen for the anteroposterio- raxis of the central nervous system in Xenopus. Development 121: 3121-3130, 1995 12) Lamb, T. M. and Harland, R. M. : Fibroblast growth factor is a direct neural in­ ducer, which combined with noggin generates anterior-posterior neural pattern. Development 121: 3627-3636, 1995 13) Cox, Wm. G. and Hemmati-Brivanlou, A. : Caudalization of neural fate by tissue recombination and bFGF. Development 121: 4349-4358, 1995 14) Bongo, I, Muhamad, MA, Sugaya, M, Kengaku, M and Okamoto, H : Identification of the Xenopus FGFR subtype(s) involved in neural induction. Neurosci. Res., Suppl. 19, S116, 1994 15) Godsave, S. F. and Shiurba, R. A. (1992). Xenopus blastulae show regional diffe­ rences in competence for mesoderm induction; correlation with endogenous basic fibroblast growth factor levels. Dev Biol. 151, 506-515. 16) Isaacs, HV, Tannahill, D and Slack, JMW : Expression of a novel FGF in Xenopus embryo. A new candidate inducing factor for mesoderm formation and anteroposter ­ ior specification. Development 114: 711-720, 1992 17) Tannahill, D., Isaacs, H. V., Close, M. J. Peters, G. and Slack, J. M. W. (1992) Developmental expression of the Xenopus int-2 (FGF-3) gene: activation by meso­ dermal and neural induction. 18) Wilson, P. A., and Hemmati-Brivanlou, A. : Induction of epidermis and inhibition of neural fate by Bmp-4. Nature 376: 331-333, 1995

-25- 19) Suzuki, A., Thies, RS., Yamaji, N., Sang, JJ., Wozney, JM., Murakami, K. and Ueno, N. : A truncated lone morphogenetic protein receptor affects dorsulventral patterning in the early Xenopus embryo. Proc. Natl. Acad. Sci. USA 91 , 1994 20) Sasai, Y., Lu, B., Steinbeisser, H. and Robertis, BMD. : Regulation of neural induction by the Chd and Bmp-4 antagoinstic pattern! ng signals in Xenopus. Nature 376: 333-336, 1995 21) Doniach, T. : Basic FGF as an inducer of anteroposterior neural pattern. Cell 83 : 1067-1070, 1995

-26 (2)

BEAomc^ 6i^#-e fm±o v A 7%r ^AO#@in|%(j:6T-3^yi<#|#%t)OOj:3l:

^oA^mmm^o^omm^c^^^a 6 L' 3 #^6 &o co j: 3 EII:mcfUTr#±^^T<66eT^^l\ &L5,

% <, ##o&&fif:&m -5>fM$0^ *oi)''\z.ts. Y) ooli^o a. wnMM<»mm£.tmfm%L%L AAi#omma(d:^TmoogT2ba^ AnL. 4 — 6#-e6EAo###o95%icmf6o coA#omm#o#%(i##Wa@%oi#An I: =k SOTfj:^: <, f o^jgo#g^y U 7#B@oi#%lc j:^ c ma, coKr%(c(j:^^^Tm#j#mmmo^m$js/A^%oimw<#2c6c^,^6 %c/:o h uoHx#m4:m#T ^##6 8^3 9 a BT##mmi%^4,000-5, oooc 3^f&o 6C5^, 13BB^m^^6^1,000f:Mr,TL^3o CO^(j:fO#, #

(#2.2. K2HEIA) o CO

%W6;McjA<@#LTU5

-27- C0@#^V- (#2.2. l(2)-2g|# a mam%# b $>------0>------(g> ------<0)>------m) o

(ma< -3 ^ m < ^ ------<0------

#2.2. K2HEI -So tztx-li.^ A n0 AESEt:: fc (A)o AT(j:A#T^L 0T, ^T(j:&6^#mg^3|5]#Jio=k B-c^A^ cmm#0mmw^am&&bo<&5o c 0^0#Ji$ A 6o A^38 B^-eKm^LTL^3 6^3^0 cfL^o (#2.2. K2HEI) o

tmm:?# 1 #8, 8003*^, "3 3 8 f 7]$64>% < % ^ ^

6lc=k6#A6:. m^em#@m0m^(:j;&#A^&6 (#2.2. 1(2)-1EIB) „ b. ->d-7PX £ tz%* K L ft00!^* Lr> ^#0—^0##l:#f 515ilg bh^K%m##g (i?m 0vA-yx%%^mmM(cm^#% C=k66. 28igg0^%j:^aiA#^A^T(j:&^(9#%L^00l:. ^#2/r^A6 8^- eRm#j:(p%)(i6A(c^<^^o ^0#, i7g0#ma##-;s-c& 50(cvA7'x^m(j:AA^'CM^L-C0< "o

C0ck9(:^ ^#^%0B#%(cmL-c%)6Tvf-yx^m#L^ w< %&# mmg c m g, fAm^*0 0505 < A ^ 5« c. iifjMOV-T^^i#g@l ±^LAJ:9(:. #mmcio^5#mig%m0B5R[d:. %(:A0*A

5 Z 6 # 5 o Z 0##0g@(c # y A ^"x A Ai"## #B0@#^N4-f 5 c 6

-28 (#2.2. H2)-2igi&±) "o3)

2.2. l(2)-2E^±) o C(DV-f^>y^ vAy%#-^-D>(D^#(:##f6C6^7

w T L $ 9 ® r###jja(d: A^J^A -o T # T #%<%^TL,a3o C(Dj:3(cM^#^^< LTL^3 6v-y^ >yz)

SSn C0V —~r-f 's-flz'fj\/*7 X >#^##:

^A^A

& o

aukmm

7"hnKht'>>f$^

2.2. K2)-2ig m^gic^(f6m#jA^mmcDv-f-^>yo (Shatz" 1992j;%%^) d. i'l-zfxffist:si-7xmmMfe/mmmj± m#;/ A y % & v - ^ > y c ^ y y % c t 3

-29 MAI-j; 5 t#A 6#/:o oi^ zL - a I: I j: A Afg# I:## L/=MA^&!o. A^o/Mcu^yxktgmu (#2.2. 1(2)-2EI^T) o c©A^*:m#Ly:vAy%MA^(^ ^©#, ^m©

C©y^-yxMA©#%©%7v76 LTA^0^U#(i^#%$^a<

©J; o U'y-y ZZ tit 1970^ftA(c$# £> I- ^7v H#%g©tOA#Al:i5^TH/III@-a-o>©S/^^'%m&^ A^mm^5Hzx3o#-i^mwmf6 6. LtZo ttz, IVJi£ lHzj-7x 6ZthMtHUz6)o c © j; 9 % c/ ^©m^x^ 3 .% -o>±M^e©^b6^Ai:tm-©^X7-XAxg#xu^©^^3^ij:*/:?%x&5o f:A:, &m##©jgC 0 #%8& 61'(ii@&© ^^5#%(i

L/z^cT^ ag#A^#l:^C^v±-yx^ %i##i>A#maE:L >ay>-

C6^AL#2.2. 1 (2)-3m?^#/r©A^k##mm© h v (zui/v? A h v c©*;[/y^7A|:j:cTm&^fi5#%©[#Tt)6####8#+^--ifn^m#©*;i/yc7Aj#amcj:'3Tmav >m{w

6^6^677+ yt^f-tLxmuxi'atf &e&m

-30- 'mm 1 • *UL-->7A/ 'J ynx-r >++■—If • 7'nx-f >*y- —tzc #2.2. 1(2)-3EI 0 A^mA^AS y ^-yx##lc*yi/y ^ AA# Afa. +^vi/(vDcc)*\g,mA'#-a=T#mt*a. yyyx#%Ti@ iDL/c^Wi/^Ati^yVi/^A • *yl/tf a V if n^>g6S4-^ —tfC

-©SJCilWS,,

f:6AkL o-e&aB**## brain-derived neurotrophic factor;^ jo -€• £> < -y-j-yXmjSi^fEffl L^^HHesiSSIF OyAyxAmi##0m#Fl:M^f 6 6 6 Wca^ME&mAf 6 C 6&E# L/c" « y:/f L.

6^^111:^''^® £ SI £> jia o f. ±^LT#y:j:9(:x m^%&##mjSck[;m#jyAyx^ c0ct(±. -c#%uc6&a6t!-&o

@ y a -tr x ^ ^ y y x A^-^ y ± y x#mm0&#c a ^ c % oT^6o COJ:3%^&^£ac£l:j:-oT^

ai'o c £-e(±^^o t/c

±^u/:j:9(cc07Lm^m#mm, m#j^#6em%%^Aaor#eyyyx®# &C-E-, c 60T# am#T

-31- 2) LaMantia A-S, Rakic P: Axon overproduction and elimination in the corpus callo ­ sum of the developing rhesus monkey. J Neurosci 10: 2156-2175, 1990. 3) ShatzCJ:B0^mmK Bmif-^JL>%H:26-36, 1992 4) Tsumoto T, Hagihara K, Sato H, et al: NMDA receptors in the visual cortex of young kittens are more effective than those of adult cats. Nature 327: 513-514, 1987 5) Tsumoto T, Suda K: Cross-depression: An electrophysio logical manifestation of binocular competition in the developing visual cortex. Brain Res 168: 190-194, 1979 6) Tsumoto T: Long-term potentiation and long-term depression in the neocortex. Prog Neurobiol 39: 209-228, 1992 7) Kirkwood A, Lee H-K, Bear MF: Co-regulation of long-term potentiation and experience-dependent synaptic plasticity in visual cortex by age and experience. Nature 375:328-331, 1995 8) Kandel ER, O’Dell TJ: Are adult learning mechanisms also used for development? Science 258: 243-245, 1992 9) Akaneya Y, Tsumoto T, Hatanaka H: Long-term depression blocked by brain derived neurotrophic factor in rat visual cortex. J Neurophysiol 76: 4198-4201, 1996 2.2.2 £ ###%

(1 ) KPACAP-38lcd:5S.lvd-yxE3(DSm

%gNT#R^7f K0^-3-e&& Pituitary adenylate cyclase activating polypeptide (PACAP-38) ti, 7n5{t, (1), PACAP^#^T##:/f 12, is, is), PACAP-38®mmi:^ita^mf#mo#%cM;W0@#%AaA#g(D

Id:, %#(D#m&m(:*f5PACAP-380##^##f5A:6(:, 5yh%#X7^X&mWT PACAP-380v^yxemic^^m#^m^/:o PACAP-38<&##X7fVi/a->y$# £• LA£ < T, Schaffer collateral-CA 1 v-)-"/X0#AepspO##&CK#A###l&0#$@ L##^#Wc (#2. 2. 2 (l)-lg|a ±, mean percent of control : e. p. s. p., 87.6 ± 2.3%, mean±S. E. M., n =24, p < 0.005 ;#2. 2. 2 (l)-lEta T, population spike, 88. 6 + 3.6%, n =15, p < 0.01), $/:, #@##-@^@^±-7X1?^, 1^1:# m®PACAP-38&%6 6#^epsp0##&cmAS#m#0lB4e#l#:!R]La: (#2.2. 2(1HE1 b ±, e. p. s.p. , 120.1 ± 3.5%, n =10, p < 0.005 ; M2. 2. 2 (l)-10b T, population spike, 142.3 ±7.4%, n =8, p < 0.005 ) « 1O^^0B^#, PACAP-38*%umLf:m&, PACAP-381d:, (1//M) , CAllMd:

100nMT&-D/c (#2.2. 2(l)-mc)o Time (min)

200 (%)

line

base

of

percent

Mean

I 0 oo oo X X' 4, >J to to CO H- >4- Mi A 4- MT m

© 7 r> <4 ^ I cr CO oo E CO oo I P s A "i 5£ !$ 4 n 4- 4ir 9i m §

6 3 cr 8 9+ P to cr p M to A " hr

4 I# 5* 41

.

|

t CD O M p S U1 F" ^ y s •< * _ \

• column :

I I CO C0PACAP-38IC j; & a#^#6 LT, @ (LTP)6®ao#%^#^L/:o CA1© LTD#, Schaffer collateral (1 Hz, 156}) 5 ±##^ tl6 C G)#"C^a ( 7 , 9) o ^f, Schaffer col lateral (1Hz, 15#<y S6L T, 4t'y h) , (LTD)&aS%*thT:m'T&&PACAP-38 (1 //M) fg-^6 6, (#2.2. 2(l)-2Ela, 64.4 ±9 .7%, n=3)o (100Hz, M6LT, 6ty h) , (LTP)^%^4±Tj5 u-c^^PACAP-38 (ipM) /: (#2.2. 2(l)-2g|b, 119.7 ±4.9%, n=3) o f%6%PACAP-38#=ka^:M'yxM$iJ&j:#

###, mn&mtmz ltd^ u?tmmfmw^^Lfzo cAi©LTDfc£tmtt@© ltp ©###, (3. 7. 9, 16, 17) o PACAP-38l:j;a%m(::&#aNMDAgfVl/f ; K, D, L-(±)-amino-phosphonovalerate (APV) #7tT (50/zM) TPACAP-SS^-X^^f X1C-9- ^/:6ca, cAi, APv^^%vi#A6i^#^mm^mg,fiy: (#2.2.2 (l)-2|2Ic , CA1, 81.7 ±6.1%, n=5, P < 0. 05;#2. 2. 2 (l)-2|Hd , dentate gyrus, 117. 8 ± 1.8 %, n=6, P < O.OODo Ctl6©###, PACAP-38(:cka&%v:h7xM#J&j;##g&A<,

150 r 1Hz 1Hz 1Hz 1HzPACAP38 250 r PACAP38 CD cd cd cd ■ 200 1 * u y . 150 h 100 '••A *« *» 50 - 0) 0 60 120 180 240 300 0 15 30 45 60 75 OCL CO d co d d APV APV ill 200 r 150 r 150 PACAP38 125 ■ I PACAP38 ., 100 ---- 50 "^vc^jfsA av^V^A-1*-- 100 ------0 L 75 15 30 45 60 75 15 30 45 60

Time (min) #2.2. 2(l)-2g| PACAP-38 ©%m©LTP, LTDC^f 6j!)D##6APV©%# a : CA 1 T©## LTDIC*±5%%#. b : 6#(@T'©6&faLTPic21±3lJnSt£o c : CA 1 T© APV (50 zz M) #6T T © PACAP-38 © %]^0 d : #KBT©APV(50/zM)#&TT©PACAP-38©#&.

LTDioj:Cf' LTPtm^ay^z:XAT##^^az6^#fao

-35- &53^o PACAP^mj:, #%©6Z5, 4>%< (Type I6Type II) Type I^##:(±, tf C (PLCM^KAi<&O, Type II ^*(d:, (f©^(:#W<W6 (11, 12, 18) o WnFkL Type I^^rnRNAMtKimmmmoaco^mLTdoD, Type II SS&mRNAti, CA1 E^lBB-hS^IllEfilfflflSW^'iC^lgLTl^ (11, 12, 18) 0 L/:A<^T, %b<6mmmmo PACAP^##Tm©vy %, -e^cAi e^ 7 —mtMkCGE< cAMP^OTmc&Sf 6 6#x. A If (PKA)®#@^^#^a/:)6(:, PKA®#^Mim#^lT&6Rp-cyclic adenosine 3', 5’ -monophosphorothioate (Rp-cyclic adenosine 3’, 5’ -monophosphorothioate (Rp-cAMPS) (8) PACAP-38®%#(:^f^B#^##L/:o 100/zMCDRp-cAMPS-Cl n$Rm±^ h LTisi%^, isiemm©Rp-cAMPs^#//fcmm%-e% /:Ly=f J: >/<-(:#L, PACAP-38 (l^M) $§#(d:, CA1, #^0^^3®y^^%-et)Rp-cAMPS^%x.% (#2.2. 2(l)-3E|a , b , CA1, 82.9 ±2.1%, n = 10, P < 0.001; dentate gyrus, 122.1 ±1.4%, n = 6, P < 0.001) 0 PACAP-38tc j: i/±- CA1, PKAOamb^g^Utl^Ci^ggf'Bo #m%ft%#m^NB-oK-immT PACAP-38 et)PKAO@Wb^S

Type lSS#:©T'ME"eii, cAMPO#C, 4 J h-;C3 ') >E©MicE< #BiBl*lMvC7

ASS©±# t tt )V v 7 ASK±—-ifb ti-S) o tnVv'O A%### -f l/bC'f-T&aa/l/^x^yC, ;4rC/>B, XfC7cr%^V>©%#^m"B (10^M) id:, -7o7-r>^±—tfc (PKC)@###im#&j?&?), LTP©##^m#fa (s, 13,20) „ %^c7o%^v> (i //M) vC7A/*;I/tfziV>##%#8%V>#jb#$II(CamiI), PKA, PKC^(H'©##

36- a b

200 PACAP38 150 100 50 Rp-cAMPS o 0 15 30 45 60 75 90 4 d c r 150 PACAP38 100 • 50 - staurosporine

0 - 0 15 30 45 60 75 90

Time (min)

S2.2. 2(1)-3E Rp-cAMPS, % ^ C/ o % ^ U >#&TT0PACAP-38O% a, c:CAK b, d : ttt^Ulo

St£©6C 6, PACAP-38JCj;SC:tLt>©Sft^oy CA 1 Tr (d:TypeII^# PACAP-38C =k ^ TWmmcAMP#&(D±#(d:i@# 6PKACD&Mklc (d:^# L % U d; 3 %f6Fr##}0#a!|T#6o #&, m#PACAP-38(:(d:M#%l^<, 6o CAlTT(d:, (mGluRs) 6/3 7 Fl/f-V %Yk(c j: c T—m#® >/ -f 7x#7#|d<, C dr(d:#mp^cAMP#&® ±#(c (d:(&#-#- a d< H6HV >MWS (#icPKA)©r£tt!$£§, LfccAMP^'ftS£ti T7f^y yy ^>^#ccM-^-#-6c6 do). PACAP-38cj:5CA l o vy-yx*]$ijo#At cdr6^/cy * —XA^,<#>1 Wr%Uo #i&, y 3 O ^ 3 r>/

-37- Table 1.

CA1 dentate gyrus

control 87.6 jt 2.3 % (n == 24, p < 0.005) 120.1 ± 3.5 % (n = 10, p < 0.005) Rp-cAMPS (100 pM) 82.9 jt 2.1 % (n == 10, p< 0.001) 122.1 ± 1.4 % (n = 6, p <0.001) calphostin C (2 pM) 90.6 jt 2.2 % (n == 8, p<0.01) 117.2 ± 1.6 % (n = 8, p< 0.001) polymyxin B (10 pM) 82.6 i: 1.8 % (n = 14, p <0.001) 122.6 ± 1.9 % (n = 6, p< 0.001) staurosporine (1 pM) 82.2: t 3.3 % (n == 4, p < 0.05) 119.7 ± 1.8 % (n = 4, p < 0.005)

PACAP-38^'MWg[US§rtT;\ ####%(:#%-3 A:

& ©## (L -cmi^A A $ a &. © m Am i) 2) cA3###m (21)« ^LPACAP-38^jA(f^;i/t wi, #m*^©m#ncA 1cko$)CA3 "3^^, PACAP-38(cj;^-cmm#mmi%©A^/m^#m^l:f6 L^L, in vivo T©PACAP-38©#m^©6}##%^#@, eLTmmCA1^36CA3-#m*^^ ©A A#g^PACAP-38(c j: ^ T(iA19 (c^b LTU6 ©^ A^ 4#©#35(C8I#L A V \

#m#AR©m^^j;O#^fi/:t)©A<0Wo ^A. *##AM^&#@%Wb#6##©# AgF -3 A#?K#mA#^©%*A" ^#%#A^#LAUo (m?K#m^w^T # m # #

##£$£ 1) Arimura, A. (1992). Pituitary adenylate cyclase activating polypeptide (PACAP): discovery and current status of research. Regul. Pept. 37, 287-303. 2) Arimura, A., Uchida, D., Somogyvari-Vigh, A., Shu to, S., Shioda, S., & Banks, W. (1995). In vivo cytoprotective action of PACAP38 in neuronal cell death in the brain. 2nd Intenational symposium on VIP, PACAP, & related peptides, Abstracts 70-71. 3) Burgard, E. C., Decker, G., & Sarvey, J. M„ (1989). NMDA receptor antagonists block norepinephrine-induced long-lasting potentiation and long-term

-38- potentiation in rat dentate gyrus. Brain Res. 482, 351-355. 4) Calabresi, P., Pisani, A., Mercuri, N. B., & Bernard!, G. (1994). Post-receptor mechanisms underlying striatal long-term depression. J. Neurosci. 14, 4871-4881. 5) Cheng, G., Rong, X. W., & Feng, T. P. (1994). Block of induction and maintenance of calcium-induced LTP by inhibition of protein kinase C in postsynaptic neuron in hippocampal CA1 region. Brain Res. 646, 230-234. 6) Del porte, C., Van-Praet, A., Herchuelz, A., Winand, J., k Christophe, J. (1993). Contrasting effects of PACAP and carbachol on [Ca2+]i and inositol phosphates in human neuroblastoma NB-OK-1 cells. Peptides 14, 1111-1118. 7) Dudek, S. M., & Bear, M. F. (1992). Homosynaptic long-term depression in area CA1 of hippocampus and effects of N-methy 1-D-aspartate receptor blockade. Proc. Natl. Acad. Sci. USA, 89, 4363-4367. 8) Frey, U., Huang, Y. Y., & Kandel, E. R. (1993). Effects of cAMP simulate a late stage of LTP in hippocampal CA1 neurons. Science, 260, 1661-1664. 9) Fuji i, S., Saito, K., Miyakawa, H., I to, K., k Kato, H. (1991). Reversal of long -term potentiation (depotentiation) induced by tetanus stimulation of the input to CA1 neurons of guinea pig hippocampal slices. Brain Res. 555, 112-122. 10) Gereau IV, R., W., & Conn, P., Jeffrey (1994). Potentiation of cAMP responses bymetabotropic glutamate receptors depresses excitatory synaptic transmision by a kinase-independent mechanism. Neuron 12, 1121-1129. 11) Hashimoto, H., Ishihara, T., Shigemoto, R., Mori, K., & Nagata, S. (1993). Molecular cloning and tissue distribution of a receptor for pituitary adenylate cyclase-activating polypeptide. Neuron 11, 333-344. 12) Ishihara, T., Shigemoto, R., Mori, K., Takahashi, K., & Nagata, S. (1992). Func­ tional expression anf tissue distribution of a novel receptor for vasoactivein- testinal polypeptide. Neuron 8, 811-819. 13) Lopez-Molina, L., Boddeke, H., & Muller, D. (1993). Blockade of long-term potentiation and of NMDA receptors by the protein kinase C antagonist calphostinC. Naunyn-Schmiedeberg ’s Arch. Pharmacol. 348, 1-6. 14) Masuo, Y., Mat sumo to, Y., Tokito, F., Tsuda, M., & Fujino, M. (1993). Effects ofvasoactive intestinal polypeptide (VIP) and pituitary adenylate cyclase activating polypeptide (PACAP) on the spontaneous release of acetylcholine from the rat hippocampus by brain microdialysis. Brain Res. 611, 207-211. 15) Masuo, Y., Ohtaki, T., Masuda, Y., Tsuda, M., & Fujino, M. (1992). Binding

-39- sitesfor pituitary adenylate cyclase activating polypeptide (PACAP): comparison with vasoactive intestinal polypeptide (VIP) binding site localization in rat brain sections. Brain Res. 575, 113-122. 16) Mulkey, R. M., & Malenka, R. C. (1992). Mechanisms underlying induction of homosynaptic long-term depression in area CA1 of the hippocampus. Neuron 9, 967-977. 17) Sarvey, J. M., Burgard, E. C., & Decker, G. (1989). Long-term potentiation: studies in the hippocampal slice. J. Neurosci. Methods 28, 109-122. 18) Spengler, D., Waeber, C., Pantaloni, C., Holsboer, F., Bockaert, J., Seeburg, P. H., & Journot, L. (1993). Differential signal transduction by five splice variants of the PACAP receptor. Nature 365, 170-175. 19) Tatsuno, I., Yada, T., Vigh, S., Hidaka, H., & Arimura, A. (1992). Pituitary adenylate cyclase activating polypeptide and vasoactive intestinal peptide increase cytosolic free calcium concentration in cultured rat hippocampal neurons. Endocrinol. 131, 73-81. 20) Wang, J. H., & Feng, D. P. (1992). Postsynaptic protein kinase C essential to induction and maintenance of long-term potentiation in the hippocampal CA1 region. Proc. Natl. Acad. Sci. USA 89, 2576-2580. 21) Witter, M. P., Groenewegen, H. J., & Silva, L. d. (1989). Functional organization of the extrinsic and intrinsic circuitry of the parahippocampal region. Prog. Neurobiol. 33, 161-253. 22) Yanagihara, N., Tachikawa, E., Izumi, F., Yasugawa, S., Yamamoto, H., & Miyamoto, E. (1991). Staurosporine: an effective inhibitor for Ca2+/calmodulin-dependent protein kinase II. J. Neurochem. 56, 294-298. 23) Zhong, Y. (1995). Mediation of PACAP-1ike neuropeptide transmission by coactivation of Ras/Raf and cAMP signal transduction pafthways in Drosophila. Nature 375, 588-592. 24) Zhong, Y., & Pena, L. A. (1995). A novel synaptic transmission mediated by a PACAP-like Neuropeptide in Drosophila. Neuron, 14, 527-536.

-40- ( 2 ) MAJOR BRAIN REGULATOR • 4^7 =

a. eic ufix&i'm • &###&(%)## E63>k°zL-^-0[n|^b,^y:*U(j:2'Zlc&6®/:53^o S^^t>7D j; 3-c^6o c ^U#Lkk ^

E0jEU#@Ki#%&(:^LT^6o CA%[g c s&o LTW60/f59^o jE#m^#mic(iA#:7;>«kck(^6#im^#^tiTV'5o C^6(± S^W###"Mm##S8K ^t»a%m-e&6C(!:A<%6fl-C^6o e»Jx^A#gm^T

9^0to hzi>^A#%%#J^@-3T^6^7o h-XD##^#x.am%T2bao !5#^#^0^#K(i3 7#0-m^A7x0ck9%$)0*

6%. m&Hf-3/=o i-0$#m, A#:7;>(m*Am^#m0^A7xm0^m^ffl\ E�#m#%6LT, l^f^MAJOR BRAIN REGULATORS LT0M&mf:LTl'&C6d<, #150#4 0—*0#f%(: J: -3 Ti# &&(;:% -3 /:« b. -^mmm^mt lt0^75 >: uses^BE0wess 7 7x 0 77 7 s x a m#T—ir0#^mSS^(c^LT^#:0^#(:%6^0^^ ^/:(J:^0

-9hV0#m05HTmm#m±i6Bm#;#jcj#;u mkmi@Ts°-^s%o,

-41 m^a'o a 5im^m^^-wra 6, ^0m#M:d:^T6#^T7o%y^'yxm^@Tf a'o sHm^o^Ayx^A^oo.s^ ^T-e&aoT. M^L%y;h7x0(m6/La'#^am%#m0^yx(b#;i&fi5o 65, —#a9;:#aofa5HT^m^o#-ey6y%omBm^{SLTuy:0T&ao kho

66o L&L, v tyx^mB^^^a" wrm##m" ^cc-r^asHTWc^, c6^T%36(c$6T

6K#L^MT5HT(j:vA7xic^A%B#^^'CUao^, CO^(c#X.a/:6^^#6 *^0^9 h V0 5HT^im^M6LT^/:'o f a6vA7x@fm(j:#AT60%^L/:o ^ e OM^OgAl ' # 5 6 C 5 (3 ^ A # < , 5 HTW#A<% l ^ C 5 T(±>/A7%%AK^{l:^%^^Ao ^#^#^#2^0^# (?) z:9h';;:im#Dm#& ^o$§m, ±tcmmA^^(j-a#A-e(±^6-y%%mm8%W{WaA\

5, ayzyAyxM&^i'-yhvti^UT&ao L^L@uAT(iyA7%%m(±#TL, 5 HT#mojK6^% < , #4 0 5 < M g)6Ao 5HT(ij5%#:^5 vA7xo##lc^g3 LU« yA7x#^#36a

LT^aoi?(d:^U^6#ATUao gP5, 6%#MT(d:^6'y%OfF()#x.^mmfac6(c j;(), 5HnmiCRT#%^5y:3LT^aoir(j:%lV:59*\ #m#uAT(mm(c#U5HT 5HTlcj:a^A7xofP^#A$^#TL, «!##%)<%< ^aoT559o yA7%^#a9^#ja?&ac6K#Ra^#/:6a^$)L6^:Uo L^L—7 h V kAT(i hyperstriatumOvA7x%*A<77%i#AOL, 24# fRH&fcta&tOAlc^TLa 9 \ #/:ytv 7

5HT2A#%#:(j: 15@m 5 a 5 HT^##:7 77^709 5 6 0& 6 a # &

— 42 — 5HT2A^g#^b ^<U&oT##d<6&o 5HT2A%#:(DM^(:j;!9^f'y%0f#!9#^$^@TL, ^ -y w L /: o^ y=swatm&f# a>/ y & % < % 6 c

m*l:=k6 5HT2A^#®5 v j:6 ^m6-#kL/:o L^L%*^-3T#/:#%m(d:5HT2A^#:(Di#%m#60Mm^#<^ #L^=o 5HT2A^*(±A^M0###B®(ai//6'^T^#cT^(), #*# 5HT2A^#(d: 6 C Aghajanianii 5 HT2A^g#

d. '>i-yz&mmftwtmT&6y;l/7 Kl/^-V > (NA) . K-/^^ > (DA) , 7t

5HT6NA#mo^#(±7y h t-7 h VT±# <#%6o -7 h &(:^^6%)

Hcf ^%l'o Am^m-e0tV7^ vy h#f^%-eDA##(i6f^^#

— 43 — A# 1 - 2 mofm 5 HT6NA&-& ^ /:[d:|g|Be(C^ LT 6 f. 7'yhm##g-e30-50%(D^fyx%m(D@TA

0#^M%LT^a@T(d:/E^/c5 3^6#X.TU^o e. SrU'Bf^S : fi#Ht9BKJ:6&l^SKl*©JBi£ £.fc7 i y^vi-^xeffMEftic^S^^ftVTV-B c t, z.tit'efcftfr -^/c#LU^(cj;&#%A<%(c/E5o m#(cJ:t) 5HT6Ach^#%lc. 3/c(m#6&% <'#-C6T##0y-f;fX0#^k&-&&C,!:d<-e#&o 5v He 5 HT6AchO^*#g^im #lcCA3T^^y%0MW^#<, CA10#-e& -3/:o ffiWW @P&Tyy(c^fb(j:/E^ -3 /Co v f--/%OMW(i 5 HT6Ach& 1^ ^#/c CA3-e(i#^T50%&$ 0 vf-^Xgg m^@TL/Co 5HT6Ach

a#?y h^xn/x'^eAe^v h[/x^-e5HTam;t#%##&^y-&o &a??&50 /f^<&, X hL/X(cj;bl 5Hm/:(/T(i/E< v^7%#m#)^bf6or##^#x.>,/l6o handling

-44- ^#©5HT2A^##%&(d:Xh h©l/3(:@Tf6Ci^%6^C^i%y:"o Xhl/X*-

9 U- /"c x v hTi^M©xA7xS$[{d;M;£©A:5 9 Ao handlingL/:#^ A# 3 i§ i 6 @(z x v h^mocAitcASTOi/^-yx^m^^k^m^o sa-fid##© StiETr+$£/N°—& > hhandling$7'vA7XiS$;^riEAo Ao 6 i!©CA 1 Xr(ihand 1 ing$| h 6i>77x%a frMfi'-otzo ^/cfr#ic x 7 X £• x y h ©7 —iAxAtlS is handling^ ^(j-yzxx H±j;(9^l:7^x©mv^^lV=Ds @<@A/:Os ##(cx7x&m#L/:o ayz7*xm©m&iLThL ^m#-C(d:iaiAi'#

 >^vA7xB6Emj$^mf6f;m^-9if % ^ix(fs (PKOT&a^s >©mm#m^z^©i/2^6i/3(c%a'=o ^#vA7x*

*^m©PKUm#id:-#^XNc####(c^"3TL^ 9 qT#e^#x^^^o m^PKUtxvl/

7 y hTktb h©PKUiMUJ;9l:ty 7 ; >Bm%Sm^@TLs e©9xAB^m©^A7

#&PKu = ^-7> hx7x-rc©{K#%&m^A t/7; >###:%i^mf6CiK=kD##m#^Ba(f^s #fL ^/:PKUM©%mi$##&©@T^/:(d:±

#7& Ci d% & ^Tio (p s 3©K#T L 5 9 o g. £ttb

^^(CMAJOR BRAIN REGULAT0RiLT8B(%#^$J#L TU^,©T&5 9o c©#x^(±E(-##(z#EL^:(±i3(^^:^^l^

#m^mi##©m%ic7 i/-r 7 x;h-^&/: qr#*(±A#^ ©m@rEPIGENETIC%g#i 6or#W<%^#a^s ^#:7 ; >(i^©##^#< (%%A# R9 M # M)

—45— (B»)

£

£SeEiS*f (H]{§)

t/7S> •fen h~> *T"3-;i/75> y;u7Hi/±u> K-/\°S> 7KI/d"U> 7fef^nu>

52. 2. 2 (2)-

— 46— 1) Okado N et al: Dev Brain Res 50:217-222, 1988, 2) Okado N et al: J Neurobiol 24:687-698, 1993, 3) Huttenlocher PR et al: Neurosci Lett 33:247-252, 1982, 4) Huttenlocher PR: Brain Res 163:195-205, 1979, 5) Chen L et al: J Neur Transplan Plast 6:41-48, 1997, 6) Doubell TP, Stewart MG: J Neurosci 13:2230-2238, 1993, 7) Popov VI, Bocharova LS: Neuroscience 48:52-62, 1992, 8) Niitsu Y et al: Neurosci Lett 195:159-162, 1995, 9) Blin J et al: Brain 116:497-510, 1993, 10) Chen L et al: Neurosci Res 19:111-115, 1994 11) Bannerman DM et al: Nature 378:182-186, 1995, 12) Saucier D, Cain DP: Nature 378:186-189, 1995, 13) Stott DH: Develop Med Child Neurol 15:770-787, 1973, 14) Peters DAY: Pharmacol Biochem Behav 31:839-843, 1988, 15) Smythe JW et al: Dev Brain Res 80:183-189, 1994, 16) McKean CM: Brain Res 47:469-476, 1972, 17) Nigam MP, Labar DR: Brain Res 179:195-198, 1979

-47 (3) wmfffflmmtwmn a. mmoawm

KOJ:l#gg^6cT(j:eAT,

Al'o b. EIS®J®

7:6^0 cmwm) #. Tu^o camm cf < 2 -3 %, t icmmum### BS^Jicn > h n—;l/^tiT^-So

^m#^(±y=g)#^L-cu6o —i/A@aR(D#@#(cw^60(±j:(pT&o*m. #x m

6#mom#r&6o ##m#0-a-o

O %"l ' -So 2 c0m$&:iJA<&6o # 1 EBR(i 1 & V XA m20mw, # y x%^ o6=k -o xmae O!K6:m6^-^0mam(:JR$6j:3#a^ai:^6(kW3 &0T&6o COj:3l:, #|$]#(:XaX7A^ at. mis] #6=k3(:%-3-CU6o C02-o(D&Mij(±/=}(i

— 48 — C®,#mWlT&3E/lTU6

##(ZXI:A-W#E#m##

@E#R0%#J&##fN;L %0J:9(c%6o L/:@E

0LT(,'&O #E^#E#LTU6t)tfT&^o

L/z^-D-c, mm 6##o #*T&5o 6%^fll:U6o E6C^(m#T^

&6 (#±, 1997) o coj: 9 (:%#%mEo$mm#A

-49- fkf6Z6^mc%^/:^^T&63o LTA^ L/:0-e& ^o d. ##^(il974#(:Bfm3y h0B#&b r@gmm## (SPS)j 1983#61990# (:0 V ^>6 GSSG SPS®W%^aLTim^Ly: (Inou6, 1993)o OVO>t esse y:u^/L*m&(re&()##Mr&&o ?f#i/<;wz)@BR ^®=k3(:mm^^T^6 (^2.2.2(3MEI)o ov^xme'j###^®#^??;/## (GABA) yxi/<;l/T(mmf6j;3(:(d:y:g)aW0yt-y^-^^^6) OV^>6mi5L/:{b# #sw\ 3 v

J]/V9Hfflir 61 #*. b tiT ^ £ o GssG(m###m#0y;i/^y- h^me^.^-o>0XEma#^v±y%i//< ;(/TM$ijf6=kO(:(d:y=6a GssG #9#cmBL, #m6LT6^(:@BK^emf6o C^^,0 SPS^fi^G,!:. ±0ckO% #?##WhLT. m%om##0##(:##LTU66#^6ii6o OV^Xi- cssc^mTc# yvl/^^6A@T, C.Z1 -n XD#fg#l: j: ^ 6 (Inoue, 1995c; Inoue et al, 1997) 0 0 V >;>®limmWm:B@LTkL >l/AmBR(D#^(:^#-C65o L^L, CCDgB&^mm^fiT^T&O V L/A@E(m#R&T$m^fiTW6C(!:(:%5o ov^xmo@aB#m^m# #483## GSSG j: ^. - o XD^##^## fa#Bd<&&0

50 GABAergic neuron Glutamatergic neuron

GABA ...... wit. -> G n

GABA GABA Glu A u::: $ # GABA BZD-receptor Uptake Glu-receptor Glu agonists I 4 antagonists Uridine GSSG (Inhibition) X y POT!?! B3HWJ teWWM 1_1 —v------TI ...... 1 ^ I 'LrJ i Barbiturate-#BZD-R GABA-R 1 Postsynaptic Glu receptors GAEA recepor-ionophore compJex Cl" channel NMDA / non-NMDA

NREM sleep and REM slee Wakefulness i ------1

Complimentary, reciprocal sleep-modulatory actions of uridine and oxidized glutathione (GSSG) at the synaptic neurotransmission sites in the brain. Uridine facilitates j -aminobutyric acid (GABA)-mediated neurotransmission by binding the postsynaptic GABA receptor (GABA-R) as well as the uridine receptor in the GABAergic neuronal system. Similar actions can be induced by several benzodiazepine (BDZ)- receptor agonists at the same receptor complex. Astrocytes may contribute to the turnover of GABA. In contrast, GSSG inhibits the binding of glutamate (Glu) at the postsynaptic Glu receptor which belongs to the receptor complex of excitatory amino acids like W-methyl-D-aspartate (NMDA). Similar actions can be induced by several Glu-receptor antagonists at the same receptor complex. Astrocytes may contribute to the supply of GSSG and GSH as well as the turnover of Glu to glutamine (Gin) by glutamine synthase (GS). Glu is biosynthesized by glutaminase (GT) from Gin. The NMDA and/or non-NMDA receptor-mediated enhancement of neuronal activity causes an elevation of vigilance level, i.e. wakefulness. Either the activation of the GABAergic neuronal system or the inhibition of the Glu-ergic neuronal system eventually induces natural sleep, both non-rapid-eye-movement (NREM) sleep and REM sleep. Modified after Inoue (1993). #2. 2. 2 (3)-m

— 51— 1) Inoue S. Biology of Sleep Substances. CRC Press, Boca Raton, Florida, 1989. 2) Inoue S. Sleep-promoting substance (SPS) and physiological sleep regulation. Zool Sci 10: 557-576, 1993. 3) #±m%i%. ###;&&#& 39 = 57-68 , 1995 . 4) Inoue S. Pharmacology of the CNS Peptides. In: Handbook of Pharmacology : Pharmacology of Sleep, ed Kales A, Springer-Verlag, New York, 243-277, 1995a. 5) Inoue S. Challenges to the ancient concept of humoral sleep regulation. In: Sleep - Ancient and Modern, ed Liu SY, Inoue S, Shanghai Scientific and Technological Literature Publishing House, Shanghai / Asian Sleep Research Society, Tokyo, 55-88, 1995b. 6) Inoue S, Honda K, Komoda Y. Sleep as neuronal detoxification and restitution. Behav Brain Res 69: 91-96, 1995c.

22-5L 1997. 8) Inoue S, Honda K, Kimura M, Okano Y, Sun J, Ikeda M, Sagara M, Azuma S, Kodama T, Saha U, Musha T: A Function of Sleep: Neuronal Detoxification in the Brain? In: Sleep and Sleep Disorders : From Molecule to Behavior, eds Hayaishi, Inoue S, Academic Press, New York, in press, 1997.

-52- (4) - y

>om#0m%*xir^o@#^"DTU5o L TML. # /: 6#%^ - >oa#^-M]A-Cgij0/<^ - ><&]##&## LTf ^uc%^#x5o c0j:3%mm/^->0to^#x0mmcAmgm# mm, mmw Bn^LTi^6#x^^6o am# (^#6 mm). 6o -eafa(c##&9%m#K#Av>wr(i. h>mm-rkL mmm -eo/z&A L^L.

6 (igij C. (c M6 5 B ^ L ^ o j^mm^F#) (cm*)mLT^6o —/j#T0# cmmg, m#me#) tcm&fiao L^u. *am# /ztxti. m#m

^% ;:##& fi-rm#-#-6 KFfcsm^Am L. A-^ - n ^ - o

m@PO#@A^(cL/:^-3i:m#^9C(kl:M#LTW66#x3#6o ^6(z, 4^%(Dfij »?, (ff#) /<^->o#^'tuoexKtmm #^M#LTu^a#xG,trru6o &E#^##L/zm#^mE#K#om#-e(icfuc^ #Ly:mWM6a6o a. SEWS 1Cfc1}5BSfiMI$S

@^(c(±^ ^ /: < t 3 c -E-fUc-o&frnEWco ^0(c^#6C6 ^ < mcm&c/: Dj; 3 L/z"« ##^^-#89(c^ LTf o

53- cm#) ^/zC6^^LT^5o Wmom^#^##Ly=4;3T(j:. 65i>%ATff#< 8ij0^-ota$ckui)(9%^#mmLi:uT6. ^fuctooe^act CfW, j;-3T#y:iX h V &#&##<ffil:#

->&&;%./= «x ^cfGUT^ommoi//';- h

(±. m#0^#c#A<@TL, o^xn > W > • t>- K* 7—r>r y*f • -TX h-ew\ MMM$kKUftWSiK.%k^'TX>b

IW D/^^-yOjilljMS^ig^tLT^cDjlif/^yBLr < StcoiiT. jlS/tiS rg@^J^^LT<6o C0m@Ct)gm#^g##L-CU66#x.3fie'°' o

(:#l^ C(D#jm6mE#o#gR#^T&6m^#®JZa-o >a#60N#^-#^)6y:6 C5T^,6o (E#IRO o b. Sia^0Aiii*HftfcJ:tf=a-n>lal»

hfF#%- #@%A^^A^.

— 54 — mm## M$mm

±& HR##

Alexander and Crutcher (1990)1) JcOt&Sg ^2. 2. 2 (4)-m

ams, amme, mmamg. ##Mamg#®amMa#g^j;[;#umaAe, zoj;

- mmm#gpct#m<6A3fi&ws6 LT. #%02o0#%^$6"o ###®AA-zi-a>#GABAf;##T&&d<, AA-a.-D>

(GABAfp##) .

tSck^-#-^ SE#®### (fLTWz-MaU K/<; x±.

— 55 — —A. T(j:%Uo mm'J:%A5—tf%6

() A< ^.>^7 7 U h V AV-A6^^^07 h V y ?X Am^mAA^. % h v? t' u y 0&f>-C6&o ?h w^0AA^&o. coAAm^ff# (mm)

a/c, xhUA'/-A^3mm#M^®^m<*6c6(j:, %cfL\TK/<; P/<; XiA ^sE#cj; ^mm##cm#### ^ u-cu-c. a - o >mm#A? K/<; a j; 3 c £^i3) x HS^AT: RV-'N >^'&{fj§tiASlS©-zi-n >TtiRjEKti^m©^^&f6m%tm %cfGCAmm/<^->oto!)#A'm^(c^^^TU6or#&e^#^fi5o 62-o0m%®f#m^6iMT&6c6(d:. cfi&o

^cMmLTU6qr#eA<#Ag,^^o ca&2o0m#ogfm##ai:BmM@m/0##%##r^<#AiT/6o c. ^3 a - □ >©?g»

#/r (i4; 3 -rm^am^mmem^m-o -c^ a^-om aj: 3 A c t -r

^(C^Tl'6C6T&6"o $3kl\ ^VWCit^T##'E#^A^6(DT. ## (C#mf 5C(j:AL' h L, i:0^®^a-o >@#0##^#^?^

LA« sm® 6 C 5 A# 3 #0LED^#&# L A eomms&a^LAim-emm&ff-DTi'&o mm^^^oLED^A^r 3OO;V##C^0 LED &imm:A#3-&6Ak -A(±iE#0 (iELw) #A(i^E#0 (m^®) #M-eab6o JEL^#m(c%m^lRl^6 6#mi^4:A'2,#-C^®B$fTA#T'#-6o cmm&iFiya,!:

-56- LED o-5@) m

#LT6&$Df c® - amwj jigL^wao comm-?#, (mmowm &

(/^-» omomm&mtk #m<#b%%<%c/::|mTkL Zmv D- >oto *) #A - # mc#fGfa&o«k#x.&fL&o Hz -z: > y o%aa c Ati# L a ^ -r

6A<, h ec-o^TmmLTafa^o[ 2im< 4-3(j:CO#m^g-e#6C(!:A(iCO#jg(c ^'Ock3(:N#LTL'ao^#tlTLT^6 6C5T&6^ ^LTG&a^aKofLTa# OA#$^^k$#6-^-o>^j6o/:o c$-t±T&m#d

ATU60 (%&A# A g & #)

1) Alexander, G. E. and Crutcher, M. D. : Functional architecture of basal ganglia circuit: neuronal substrates of parallel processing. TINS 13, 266-271 (1990). 2) Aosaki, T., Graybiel, A. M. and Kimura, M. : Effect of the nigrostriatal dopamine system on acquired neuronal responses in the striatum of behaving monkeys. Science 265, 412-415 (1996). 3) More, J., Meyer-Lohmann, J. and Brooks, V. B. : Basal ganglia cooling disables learned arm movements of monkey in the absence of visual guidance. Science 195, 584 -586 (1977). 4) Gerfen, C. R. : The neostrial mosaic: multiple levels of compartmental organiza ­ tion. TINS 15, 133 - 139 (1992). 5) Jimenez-Castellanos, J. and Graybiel A. M.: Compartmental origins of striatale- fferent projections in the cat. Neurosci. 32, 297 - 321 (1989). 6) $4# 55, 680 - 689 (1985). 7) Kimura, M. and Graybiel, A. M. : Role of basal ganglia in sensory motor associa­ tion learning. In M. Kimura and A. M. Graybiel (eds.) Functions of the Cortico-

-57- Basal Ganglia Loop, Springer, Tokyo, pp 2 - 17 (1995). 8) Kitama, T., Ohno, T., Tanaka, M., Tsubokawa, H. and Yoshida, K. : Stimulation of the caudate nucleus induces contraversive saccadic eye movements as well as headturning in the cat. Neurosci. Res., 12 287 -292 (1991). 9) Lees, A. J. and Smith, E. : Cognitive deficits in the early stages of Parkinson’sdisease. Brain, 106 252 -270 (1983). 10) Mishikin, M. and Appenzeller, T. : The anatomy of memory. Scientific American 256, 62 - 71 (1987). 11) Mushiake, H., Inase, M. and Tanji, J. : Neuronal activity in the primate premotor, supplementary and precentral motor cortex during visually guided and internally determined sequential movements. J. Neurophysiol. 66, 705 - 718 (1991). 12) Rapoport, J. L. : The biology of obsessions and compulsions. Scientific American 260, 82 - 89 (1989). 13) Schultz, W., Apicella, P. and Ljungberg, T. : Responses of monkey dopamine neurons to reward and conditioned stimuli during successive steps of learning a delayed response task. J. Neurosci. 13, 900 - 913 (1993). 14) Strick, P. L., Hoover, J. E. and Mushiake, H. : Evidence for "output channels” in the basal ganglia and cerebellum. In N. Mano, I. Hamada and M. R. DeLong (eds. ) Roll of the Cerebellum and Basal Ganglia in Voluntary Movement. , Excerpta Medica, Amsterdam, pp 171 - 180 (1993).

-58 2.2.3 B% ~ A —

mamm (MEG) #o#&c=i;(),

i(ct>HS^'ab§ tSt>tL5o

(1) M&}:sXrJx£ LTOE a/:6iz, m*#M>/XTAT%< T(^ i:Mx.§o #j/UfBscandar ^ c j; a, 6 & a „ -^0^ fia co@ig#d^ u ^ ^ 9 ^ -w- a

®#±&c^a6#x.6tiTuao ^oT#mm-^-o>^3#e, fL/:, (m$2g|B) , 2SB{rB-ti-btia@S (SMBS) , (3 >r-?x h) ©S^(zA #M^-o/c6R^mm^fl/Co Vaadia3(cj;(9G0/N0-G0#m^m^faf-;l/

M^-3TU/:o ##a ^_k

5> ti'€h) a o

(2) mmmm ^%®9^. #(:9# %7 Hz) (MHz) famwmc-ouTkk #i)%m^ot%(c*%^K^t»-3Ti\a t faa^ < #M3fiTu-cv\ao ^ ##, #1^3 W, o'Keefe ^iCckD. vxA^f^ &amm#mm®%/w?of-r; >^'6x c 0 9 vxAo<&mtoMW\ cw^fo#^#aL^-3&aoa^;y&%oc,&a®&

-59- Nakamura#CJ:-o"C, 1h;t/©#^@^#©%%$ii#fc:#:# ©-o.-n>©%^C^ #lti#L/W:$#C#LT©<&, tZo

Gray, SingerjoJ:^, Eckhorn #©%^ec^^T$ijmi&#e©m#^#*W!l^ fL/=o !i#gc#-m©m^cj:^. mmm, 6. %A^#. a/=, mm/#Nmw©&#&, m^MmA©^p*gm A/fL. ^<©m&d

j;6%A6 ut 3 Murthyg,(c:j;cT, ^yi/©#mm#^mc^u-c. %m6mm©mA^gf Il^t^ia —D >©|a^'^O^o/-Co MEG©mm^ ^0^#9ySm/<^-y©mmi^Rr#(cLy:o Llinas#©^#Cj:6 6. 7^#^^%^* C—CM g, 6##r%\ C 6 C R E M@KB# C9#-?Amfr#

(3) $r&ffY£ft®£j£&tff3ffl#ra

m^mmAc@#^%#j^mA L%^6 6#x. c © 7 (i. 4-^C%mc%-3-CL^L^ Amitai#©^#C=k^(:\ ^^(±#^7##^^ 5lMOo ttz, Whittington bCckfttf, M©tiSJtt-^-n y|l±©^ y bV~? -e7^ ^ ^ ©mu&^#T*#e-^.-o>A

immmm©## c Mf ^ t f ;h tm#- < ^ ^= -e ^ © % ^ c c%-o © A # % * —3(j:, V ; y Mh-r^vhAi/l/-^^©#af$MCj;et)©T, von der

Malsburg^ Singei\ Sompolinsky b/^'% D fib OkudaUCck oTI hh Li AAL, c©##^7Aoy#©m#^mmatc^#6f6o ^©A&cAm. mA^m©#mmm^#6L, #A©#

^T&%.&m-'bm##%%]M6#AL%l'o W^)©*A=fV-W\ hV^-T^A^F tltz'f y$— ><)b9 4 v —C Kuramoto^C J: V' < o^©/< V Jz~

AL. #/r©zL^-o>©

-60- B#6%-3TU6o &i#l^#@a6"C##-e\ J^±©S6(c j; D> @A"Ca#L/:-a-D jgfpLXy MH|± omaf#mic j; D u 6#x6(a3^^6m6 Llinas^(cj;mf, #*om%mm

sc^Cck^Tymmo ^ - n > m# ic^# L %#-%###,<& # & -o # ciGBf & #s g i # am & . MT$-fZ> o

(4) fft&KYmommim vi^iStiz^ i&’Mizi r^j n^j r^j rvvi/j r / <' '.y j rz=.f% j £n—K^-5—^— n y^' 6-3/:6f6o C0mm^x^A(:#U7;l/6#0/

—o(i. von der Wsburg(z#a&##A%AtT;l/-?\ COtfVWzkL A^gAtT^^/zm^A^ecDm^mm^lz# ft&y=*x omam&Atr&oATd^y h7-70$§A^j±c@AiA$^i:i\^^m^^ 3%Uo ^or##^Ayx#0K&{d:#^6frCwa^ #m®ff<*#m##6fiTU%Uo 4—C0^^zfu-(d:, Crick ^@#^^-6 r^Ao, mRCckS-f-f-?^ h«J (z% Niebur. Kochg,(z=k6tf;[/@f%^66o #&g±C^Ly:%m09f,, om##w\ a^TMizML%^^^%:^m^A6-wr&or#(z^aa^A#Uo L*^L, at. /<^-

-61- (5)

U-'>3>i:«kot4t5iLt, tufcfSSS^-a t&6

1) m*

-h&ca&o

<

2) &!@0%#iCck%hv#-*w=m#m#7mm®7 4-K/fc, 7SB© l J=l

@m<&6 6#x.3^6o m#m)

(6) Sift $MtrtT\ • {s#M-EET{I^ri£na-^-n xiWIMS-eab

frti6o ##^##^#"3/c36C(i#;t®AW^#T^ D, BtltoifcW^tL-BlBEl^icjoU

Bf cL t, <- a 3 ^#m-m L a fr#® y: A c Af & ##

& cx.%maL#jiuam#

(^HSUiW^B

-62- iff ^ * “ A

o,

® ^flJgC-Sl \stz-j--a> o - KA'y 7tCj:t; -gJJtlC U tzifr> TJEtiT^f $0

V, #mcU# 2 # 3 ##t <7)-|S$g^$ ti5«

TC'fO'A'OfV >^'^9)1

iSBJtx/\°-fT-^U. 7±o7ffi£8(±"a$n4v

#2. 2. 3 - 1 ^

-63- IK X ^------^

mmm<7)x/\a'f^iz,t-SfvzLl/ —va >lilij(7) h U — tvi Lx-va X: J: 5 Tt£M<7)JIIf£tf

7/H>7lN

X,

t ^ >*k

t JiU x

7SJ$IC<£:3 V $. V h-y-'TfJl'M&T (D%#^oAL j:

^2. 2. 3

— 64 — mnxrn Emad N. Eskandar, Barry J. Richmond, Lance M. Opt lean (1992) Role of Inferior Temporal Neurons in Visual Memory I. Temporal Encoding of Information About Visual Images, Recalled Images, and Behavioral Context. Journal of Neurophysiology 68-4: 1277-1295 E. Vaadia, I. Haalman, M. Abeles, H. Bergman, Y. Prut, H. Slovin, A. Aertsen (1995) Dynamics of neuronal interactions in monkey cortex in relation to behavioural events. NATURE 373-9: 515-518 O’Keefe, J. & Nadel, L. (1978) The hippocampus as a cognitive map. New York: Oxford Katsuki Nakamura, Akichika Mikami, Kisou Kubota (1991) Unique oscillatory activity related to visual processing in the temporal pole of monkeys. Neuroscience research 12:293-299 C. M. Gray, W. Singer (1987) Stimulus-dependent neuronal oscillations in the cat visual cortex area 17. 2nd IBRO-Congress, Neurosci. Suppl. 1301 R. Eckhorn, R. Bauer, W. Jordan, M. Brosch, W. Kruse, M. Munk, H. J. Reitboeck (1988) Coherent oscillations: A mechanism of feature linking in the visual cortex? Multiple electrode and correlation analysis in the cat. Biological Cytbernetics 60: 121-130 Andreas K. Engel, Andreas K. Kreiter, Peter Koenig, Wolf Singer (1991) Synchronization of oscillatory neuronal responses between striate and extrastriate visual cortical areas of the cat. Proc. Natl. Acad. Sci. USA 88: 6048-6052 A. K. Andreas, W. Singer (1992) Oscillatory Neuronal Responses in the Visual Cortex of the Awake Macaque Monkey. Europian Journal of Neuroscience 4:369-375 Andreas K. Engel, Peter Koenig, Wolf Singer (1991) Direct physiological evidence for scene segmentation by temporal coding. Proc. Natl. Acad. Sci. USA 88: 9136-9140 Andreas K. Engel, Peter Koenig, Andreas K. Kreiter, Thomas B. Schillen, Wolf Singer (1992) Temporal coding in the visual cortex: new vistas on integration in the nervous system. TINS 15-6: 218-226 Malcom P. Young, Keiji Tanaka, Shigeru Yamane (1992) On Oscillating Neuronal Responses in the Visual Cortex of the Monkey. Journal of Neurophysiology 67-6: 1464-1474 Venkatesh N. Mur thy, Eberhard E. Fetz (1992) Coherent 25- to 30-Hz oscillations in

— 65 — the sensorimotor cortex of awake behaving monkeys. Proc. Natl. Acad. Sci. USA 89: 5670-5674 Rodolfo Llinas, Urs Ribary (1993) Coherent 40-Hz oscillation characterizes dream state in humans. Proc. Natl. Acad. Sci. USA 90: 2078-2081 Y. Amitai (1994) Membrane potential oscillation underlying firing patterns in neocortical neurons. Neuroscience 63: 151-161 Miles A. Whittington, Roger D. Traub, John G. R. Jefferys (1995) Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation. NATURE 373-16: 612-615 G. Avanzini, M. de Curtis, C. Marescaux, F. Panzica, R. Spreafico, M. Vegnes (1992) Role of the thalamic reticular nuleus in the generation of rhythmic thalamo-cortical activities subserving spike and waves. J Neural Transm [Suppl] 35: 85-95 Martha J. Johnson, Kelvin D. Alloway (1994) Sensory modulation of synchronous thalamocortical interactions in the somatosensory system of the cat. Exp Brain Res 102: 181-197 Daniel S. Barth, Kurt D. MacDonald (1996) Thalamic modulation of high-frequency oscillating potentials in auditory cortex. NATURE 383-5: 78-81 Ch. von der Malsburg, W. Schneider (1986) A Neural Cocktail-Party Processor. Biological Cytbernetics 54: 29-40 Andreas K. Engel, Peter Koenig, Andreas K. Kreiter, Thomas B. Schillen, Wolf Singer (1992) Temporal coding in the visual cortex: new vistas on integration in the nervous system. TINS 15-6: 218-226 H. Sompolinsky, D. Golomb, D. Kleinfeld (1990) Global processing of visual stimuli in a neural network of coupled oscillators. Proc. Natl. Acad. Sci. USA 87: 7200-7204 H. G. Schuster and P. Wagner (1989) Mutual Entrainment of two Limit Cycle Oscillators with Time Delayed Coupling. Progress of Theoretical Physics 81-5: 939-945 Koji Okuda, Yoshiki Kuramoto (1991) Mutual Entrainment between Populations of Coupled Oscillators. Progress of Theoretical Physics 86-6: 1159-1176 Yoshiki Kuramoto (1991) Collective synchronization of pulse-coupled oscillators and excitable units. Physica D 50: 15-30 David Somers, Nancy Kopell (1993) Rapid Synchronization through fast threshold

— 66 — modulation. Biological Cytbernetics 68: 393-407 R. Eckhorn, H. J. Reitboeck, M. Arndt, P. Dicke (1990) Feature Linking via Synchronization among Distributed Assemblies: Simulations of Results from Cat Visual Cortex. Neural Computation 2:293-307 H. G. Schuster, P. Wagner (1990) A model for neuronal oscillations in the visual cortex. Biological Cytbernetics 64: 77-85 Peter Koenig, Thomas B. Schillen (1991) Stimulus-Dependent Assembly Formation of Oscillatory Responses. Neural Computation 3: 155-178 Olaf Sporns, Giulio Tononi, Gerald M. Edelman (1991) Modeling perceptual grouping and figure-ground segregation by means of active reentrant connections. Proc. Natl. Acad. Sci. USA 88: 129-133 Christoph von der Malsburg, Joachim Buhmann (1992) Sensory segmentation with coupled neural oscillators. Biological Cytbernetics 67: 233-242 Hartmut Neven, Ad Aertsen (1992) Rate coherence and event coherence in the visual ccortex: a neuronal model of object recognition. Biological Cytbernetics 67: 309-322 Tsuyoshi Chawanya, Toshio Aoyagi, Ikuko Nishikawa, Koji Okuda, Yoshiki Kuramoto (1993) A model for feature linking via collective oscillations in the primary visual cortex. Biological Cytbernetics 68: 483-490 Geoffrey M. Chose, Ralph D. Freeman (1992) Oscillatory Discharge inb the Visusal System: Does it Have a Functional Role? Journal of Neurophysiology 68: 1558-1574 Francis Crick (1984) Function of the thalamic reticular comlpex: The searchlight hypothesis. Proc. Natl. Acad. Sci. USA 81: 4586-4590 Ernst Niebur, Christoph Koch, Christopher Rosin (1993) An Oscillation-Based Model for the Neuronal Basis of Attention. Vision Res. 33-18: 2789-2802 W. Softky (1994) Sub-Millesecond Coinsisense Detection in Active Dendritic Trees. Neuroscience 58-1: 13-41 Miguel A. L. Nicolelis, Luiz A. Baccala, Rick C. S. Lin, John K. Cchapin (1995) Sensorimotor Encoding by Synchcronous Neural Ensemble Activity at Multiple Levels of the Somatosensory System. SCIENCE 268: 1353-1358 Ad Aertsen, Michael Erb, Guenther Palm (1994) Dynamics of functional coupling in the cerebral cortex: an attempt at a model-based interpretation. Physica D 75: 103-128

-67- 2.3 ES3 > b° *- :J-

2.3.1

(1)

a. tiD&IC DC. fP-DT^&j C6l:=k-DT#%^mAj;36fa#gR%7ya-f^^6o C077a-f^

-3T#/:o ^ tlT # LA:tf

^ao 6(^ ^ 3 ^ ^ T # /:# m$ # 6 ##^3r'j#-#- (& L 5 3 ^ C t V ^ C i- mm - ) b. ### (ilwri • b>“;to) (^^j(ci%r>tz) L

»», =£&#*'&&s

Wa##A#t##0IKA®#-C6!^ #A<#c;-cu^o

< =

(DSSttMMto £;*£?" 5

##yAyxA^(mm&®m^@^^y=^L, #$imAA^me@^6/:6fo xA^xA i #^ cf mi&mkjK-w-con#* fi. i:gam^mx. & a

(Das* fg£ icis x ± 7x a* &mv$> z x-f-XxA^A<$)A:Waa<#A&#A'#'5®(d\

- 68 — c. S&^m6AMC& o j; 3 6« (D#^S$,&%#{& &%&-#-$ 4:fE#m#-ejG##W<^-3 b#L < ##2mri'&MakL 3o tiAm - /:LT^5i#A6flTjot)x 100^q^3#;V

60 WukL

svoboda #(1997)^7 v ALTjol'T. by-\®$#K*LT##^^A F V

3.1 (D-lE)o

puke repeal 10 s

FIG. 1 In vivo two-photon excitation laser scanning nAavM^Xjuv microscopy (TPLSM) combined with sharp microelec ­ trode recording in the primary vibrissa cortex of anaes­ thetized rats, a, Schematic representation of the _____ |20 mV experimental set-up, and £>, representation of the record ­ ing and imaging geometry (neither to scale), c, Three- dimensional reconstruction of a layer 2/3 pyramidal neuron (projections onto the x-z (top) and x-y (bottom) planes; stack consists of 110 sections, 512 x 512 pixels each, acquired with 2-pm spacing). In the x-z projection the surface of the cortex is towards ililJJL the top. d-f. Cellular responses to whisker stimulation for three representative cells. Left, recordings including one entire stimulus train; right, expanded views of sections indicated by grey bars. Action potentials (lower traces) were associated with dendritic (Ca2*) transients (upper traces), with two action potentials (black arrows) produ ­ yvyA/JVvN cing larger [Ca2+] transients. A spontaneous buret con ­ taining four action potentials (e, grey arrow), produced a still larger [Ca2*] transient. Whisker stimulus trains are shown at the bottom. Fluorescence was measured 150 pm above the soma and 290 pm below the pia (d) , 60pm above the soma and 170 pm below the pia nnnnnnnnnn j—| |—i |—LJ— (e) , and 82 pm above the soma and 240 pm below the -1* zoo "» pia(f).

S2. 3.1 (1)-10 Svoboda K. et al., ,Nature385(1997) 161-165

-69- T^^). (##%: Yuste & Tank, 1996% A a#^jgx/<^r ^ a mm & & -? y=a% svobodam®#

%Ew% 3 %Tu%w\ m% ? ^ %#$&muy:#^^ 6Ca^##0@#*@y,#^ O^tlT^a (Lev-Ram ii, 1992 ; Llinas ft Suginori, 1980) 0

a#*<^-3"CUa (Miyakawa^x 1992) * ^ tiy=$#-c #mA L#u c 6 -? T u a o

at%at. me (m#^6&#^a0^^9W<^@6%ao m^maaeML-c^ sakmnnc#%xvi/-7W C0Pn®|:^aff^SnlT^a (Stuart & Sackmann, 1994; SprustoniS, 1995) 0 JT$

^#K%ed:()x ^A^xA^e^fa^. m&^mfav^Lyzo-r&a m2.3.1 (i)-2m 0 e#A Lt a t f a## ^ c #Ramo^#(d: $ ye-ou %u%u (Turner#^ 1993) 0

Somatic current pulse soma soma dendrite dendrite

|25 mV

10 msec Dendritic current pulse D r-n soma

dendrite

25 mV j soma 200 400 6< 10ms Distance from the soma [/tm]

&K&foM%&coMmfot A,IWK*-yv-i:/uE#LKrSSjlEfc A : Ml&m'hiotMLTzmW)W$L. b : fc«MIItl:i^ c: ^i-vxxtnzx-ox^Ltzssbme. a, b, c, t>, mMftx-nfeMm&tffflttmmizycfr ix^&. d : 0. IS m/sec. :Stuart and Sakmann: Nature, 367, 69 (1994) i 0 ^2. 3.1 (D-2EI

-70 — ©##s5eoammm#&ic#o Tmmmoca^ ^ yiis»<±#T5 mmw^#tca2+#ma^^#(c v#m^(cmmmoca^#mA

mmoCa"#m^±#fao^^moctT&a)0< (Lev-Ram#, 1992), NaWWB### C#r,T^Ca^#m^±#f6 (Jaffe#, 1992) o ###momm#^%B^&^#Afa 6, #^im(cNa###ygmm&^me#LT, -m%oca^#&±#^im#a#ic%ao ic c#tt^r tfy^t- *W5*s u r t' a (c i j:mmmoa%^@ c =k ^ a ^ ^ %##o y 9"

v w%yA-yxA^icj:^T m##Mg#at, mf mm&ome y- % h-#-& ao caf -F * immmoca' w yam±#^ &/= 6

? +■ * JV & LX l' a K-rd- y ^#^%(c#mt a -ryf ^ jvwci±, w 3^om@-emmi4ca2+^y#&A< ±#-a-6tMo3-awd<&&o mm^ca2w^-y#m±#o/:6 (ccof+4;yw&#ibL, mm&om^@tm@RoM^ig#ao cot#, ##%y^ "yxA^omm#:^o6#(d:@##&o^o##T(j:m% a ?& a„ mm#, Mieaic»iii‘>tyx^iTt'§ mmo#4^@ga(cy^-yxLTua (Buhim, 1994 ) „ mietmA^^f

KcDfiTicj^oTfrbnao ®w,f4yA'/xAA©^Abi-F ti[^§MWu£m#lTa%^#^-%

(fao-e&a cmMM) 0 m^g^romwM^^Tow^A^^^Bic^-act

(Tsubokawa#, 1996). 2/c, NlMA^(i##%A^(cj:aCa^##a#maom^&& ffiS^a (Callaway#, 1995) 0 ^ ty±y%wm% y^-yx^^mmmicie^^aiamm^^aiTuao ^g^#=-e&ae, uu, §#uyAyxA^^mm#(cAoA/:a%^@tma'A6#a t y±-yx 6#d

-71- c 5 (S2. 3.1 (1)-3E3)(Magee & Johnston, 1997: Markram M, 1997) 0 r#EtoS ! jjVuNAAis sl_ +TTX -TTX JUJUUL -20 0 20 40 Time (min) ®##:H6^(^T^6(Hebb, 1949). F

—

6 LV'o In =11) •20 -10 0 10 20 30 d. tf;i/tern^a6## Time (min)

±ic5E^ Jb •& t' (^/ N — h' 7 burst. (B) Plot of EPSP amplitude for the same neuron showing that paired stimuli without back-propagating action potentials do not modify EPSP amplitude, whereas subsequent paired stimuli with back-propagating action potentials result in a long-term, large in­ crease in EPSP amplitude. (Inset) Average EPSPs for the last 2 min &/<< of each period (control, +TTX, -TTX). (E) Grouped data showing normalized EPSP amplitude after paired stimulation with and with­ h/ > btf/U out TTX application. (F): Summary of mean LTP amplitude under various experimental conditions. The amount of EPSP potentiation, plotted as percent of control, is shown for all cells under each $cg-d < V 7 b 7 ^7-ffiMW^lT: condition. We calculated potentiation by dividing the average EPSP amplitude at 10 to 15 min after stimulation by the average control &5o. EPSP amplitude. V'7^7t S2. 3.1 (1)~3|21 Magee, J. C., & Johnston, L -cm# L A< W# T & 6 o. D., Science 275(1997) 209-213 i) tti * ©Y*^Sti:it!6S$a^Sffif cd

—■'flSpjT

-72- up tabled LT^YL^EI'SC 6 (isr#T&6o 4) 6J: 9 Kf 60

6) mijA^

3J:9lcf-6o

«t 3 C^6o e. ISiE 67:6(c###m(d:, 3^. #^7#c =kk» sm^m a #i -o ^ a c w: ^ 6 cm Ft##a<#g-T&-3f:o A%#mm?:ef;w:j:a##%mm5kk «k 3 6 t)0"e&6o co#;Wc#a-e&M;%#c%:3:frT #/=

g m # #)

Buhl, E. H. et al., Nature 368 (1994) 823-828 Callaway, J. C., et al., J. Neuroscience 15 (1995) 2777-2787 Hebb, D. 0., The Organization of Behavior, Wiley, New York. (1949) Jaffe, D. B., et al., Nature 357 (1992) 244-246 Lev-Ram V., et al., J. Neurophysiology 68,(1992) 1167-1177 Llinas, R. R., Science 242 (1988) 1654-1664 Hinas R. R., & Sugimori M., J. Physiology 305 (1980) 197-213 Magee, J. C., & Johnston, D., Science 275 (1997) 209-213 Markram, H., et al., Science 275 (1997) 213-215 Miyakawa H., et al., J. Neurophysiol. 68, (1992) 1178-1189 Spuruston, N., et al., Science 268 (1995) 297-300 Stuart, G. J. & Sakmann, B., Nature, 367 (1994) 69-72 Svoboda K. et al., , Nature 385 (1997) 161-165

-73- Tsubokawa, H & Ross, W. , J. Neurophysiology 76 (1996) 2896-2906 Turner, R. W., et al., Neuroscience 53 (1993) 949-959 Yuste, R., & Tank, D. W., Neuron 16 (1996) 701-716

-74- (2) u

o memmMiWa ^/:

"3/:

a.

;lz/f6 61#^&<%w t (Dt%6o cc-eti. #wm i/^a T0##^ em#^ HcE^LT#^

w0^ AA/<;[/%oR% (AND,oR@%) oor##wfrcua[i]o

&6S^M#a-e@@LTU6o L^L.

-75- &&&&<. fAT0##&##f 6 (C 60 > i/ f vl/0±#(: ck D 6 &##&&

5#4t)66o L&U

^ ^ ^ a # > / v A > wso mamm#t /<

;l/%ajy70b%hf7A0N#^M<^TW6[2]flT^6o f0#l:t)^ o ^0##M#a-e 6 A^j ^ ^iTi^ 6 ^rc L^&o Lf=a<^T. 69i#Abtl6o

LTt^ %0t)0*<& 6o (DL I Ft^;l/ : A^A#^t 6t)KmgBm&6 LT#

®ii *i 7—nOtrA : A^StilHI DSSoWSfM^sTN ft £ WS£E A 6 t'<

vl/X^m^L, Mm^<±#LT#^6 6^(cA:Km6#f#^^6o ®SAMtf;L/:#/r0m^6ef;L/T, A^B(i^C#m0W^#^T, #g|3m@d<

^2. 3. 1 (2)-m ;l/0#mm%#(: =k 6 V ^ v 7EI A : L I FttESd^T^K B : S AM#E#f C : 'bW.bnts -txVl/A*< B6C

^7-r7-J:a0tT;l/(^ ^#^0#LW##f^#"Dm# -#m0tf;l/6W(;m6[3:]o C0tf;[/(j:, U

©AA(i%<, #^e^(@^/<;l/X#6Aa^fr^ms&C^f6 6j;lA COtfVwa B9@ » ' ^AM4>^#&©/< 7 ^ ^ 6 C a (: =k 0 v #E#m©l ^ 6 W5^^®^#(cai##%^#/:-Wr^c6%)m*6o fJAlf, ##(:A#%A#a<&6 6L #3 <#eAL##6Ctf;i/^^L, ##[sI%(:^m%e^#A LTU6 [5]„ V ^->7 y 6 6 S AMty" ;i/^@0A©m^^6a6X7Axef;i/6Rej:9i:%6o Aft/:^;i/V7>7^->tf;i/t)ab6o ^^©#EWBlc(±##^^^^»)&5©-e,# fvi/c w#©##g&m % Afif: a c i c «p < 6&#A6o

b. ^Sgij mmmg&kk A^m#^msLT^©Mm^m^f66im^(c, A^m^Cck^^muy:#

##A6 6#, A A ©EH^EIi 6 L T @g§ f*i ic A #fl £ ft 6 S i:: & 6 o aV^Afitf, #@ CckD##mm^Dtx^gammA9(:##f 6 6U3#-e&6o #a©#E##%#%sK#^g)C©#M(cM#f 6 ^©6 LT#Emm@5©mAf$^©m o, ^© >/A y x-e© oi#%©#m<% o co& s [6], *-XA^©/

Hebb©#@;iJ (j:$-#Emm©A^J 6 6 9 ^ A 6 A^E^f 6 t ©T, mm%tmi::j:6#mW-C&6o j^g,©A^^Xj(t), i©^#^Yi(t), c<^# @a&6-#-66, Hebb©##mj[f:^©^:© j: 9 ft6 o

A ifjz=c«%jfuym.).... (i)

-77- Ai#,&T©AA#A©l5|B#@#'fl:c j; o 6#@©tt@6-6 % ^ T© So ^-ce.

6 $)6%#Af 66. %© J: o C# ^ ^ 6 o

A wji=c.%j(uym;-d.im .... (g)

Ifi^A ##C =k -o LTB^^©A < #6ti% < 6##<&j$y:#3fL6o eo#O%a6LTA^'A^0#gl/<;l/n,m(n. AAA^-e^JWA ©6£M$ti. iilT©6^MS$-tirbti6 6 !^A© j; 9 C#$#.

A Pj^c.^ijftAWfyift)-,!;) .... (3)

C6a<##6o eux.^. m=01Tn^Wji6'#-66AA<^ h^Cia^U T?f < 4" n^mmm^#^w©^#Kj±^jLy:#[6-#-6 6^©mm© 6$Am - $^(6©B#^#j#f6C6t)^g-e&69^. #WFA-C(i. -e©%mt)%g,^T%©C 6^. #wmmtAtia A A c (j: t A;[/ @ ck ^ c-#- 6 ©-e. c Afi6::6#Bm-c&6o ±c^ $ /: AM LTu 6Ad. %M#©mmA

^ A -y% ©mm^Etm^ A: cm 0 Afi 6 c 6ii. v A y % -?© -r A > ©ma

C6 6'66Z6T&6 6L/:#igM^^6o fCTht. AA©##^^@CAA^r#^ MC##LMmf-6g#^m^. AA#@mH(t)6 LT^©j;9 C^A6o

Hji(t) =Xj(t) +qttji(t-l) • « • ♦ (4) tztz L. q (0

-78- A###K(!:#

60

A fji-Cji . (gjm)-si)yi(U .... (5)

m # 6 C t ^ (p ^ f:„

6 c t 6o $)0##in]%-eA#:

V7(t)

#2. 3.1 (2)-2EI VI/ A :##a?

c. 33b U(C

3A^6^-n:^/:o *g#«tLT,

-79- L A 9 C 6 #. C zT^A L t J; 9 j: SfiJA^#+#0

Xf A&f#0 ±#St36(C#. SC«h^##%0±-?^^^ tMt>ti&o L-rex Tff

f-^^#E'#-6y;^l/-^6LT^Wmtt6o #^0^i##m@-?#. w#: ^#%^0^@i3$m0%%m#0%!@(e%]AT. i#m%2:Vg@i3-?^f 6mmi*#tck6#^

0$m ^mg-?. ####^^0j: 9 #A^-?0!#g^#Em0f@#mm0d:ma-*#%«p Afi A@c0a#%#%0mm-?&&6[9]o Afwej:(pmm##!^#mLt#fLuf*

(m?g##A#^^ # 46 0E ^)

1) Fujii, H., I to, H., Aihara, K., Ichinose, N. and Tsukada, M. (1996) Neural Networks, 9, 1303-1350. 2) Sakuranaga, M., Ando, Y. and Naka, K. (1987) J. Gen. Physiol, 90, 229-259. 3) Nagumo, J. and Sato, S. (1972) Kybernetik, 10, 155-164. 4) Shigematsu, Y., Matsumoto, G. and Akiyama, S. (1992) Proc. Int.Sym.bo, N. I..S. Iizuka, 130-133. 5) Aihara, K., Takabe, T., and Toyoda, M. (1990) Physical Letter, A. 144, 333-340. 6) mmgsj (1996) m-'b- 3>b°zL-f. BA#@#AW. A#30-58. 7) Artora, K. and Singer, W. (1993) Trends in Neurosci ., 16, 480-487. 8) Shigematsu, Y. and Matsumoto, G. (1994) Proc. Iconit>, Seoul, 1113-1116. 9) Matsumoto, G., Ichikawa, M. and Shigematsu, Y (1994) Proc. IC Soft Computing Iizuka’ 96 , 15-24.

-80- 2. 3. 2 — a — Jl/%- "j h • 51 "j ~f

(i) ssm^yt°3.-?mynt'yy-•

a. J? %#" (Do W'%#

mw\ m(±. &(cwf L eomm-e, t0#mcNf5% S^/^tifcb^tiTE'E^ LTEI^tix tWS?S#^EIt(c j:-DrME^tir^< o #

O?66o

#-CkL #&, #/r0W^T^^m/LTU6m^3>k°^-^0yah^^y (#m-ya tv if-) (COUT,

b. ttgllSrotrA

#®mA ^ L < mw-& c tzL -^®#A

0^^T^6o ^2.3.2(l)-mA(:^:L/c®(±, ##0CAl##:mm0m^®B ^H/-%L/c^0T&5o RC, ^

2.3.2 (U-mi:^Lfc,k3%#&%&

— 81 — tfm_Lc>WT#%A'o

TU%&r,%6l'3A#%aaA<&&o 6C5;K f>FVT^

(2) #<#&. yrntm^L/co f>KVT^ y^ - >Kv4')') c mm*, e z ^ z mm ^ fi 6 < v J: - =, > c mz, f:

-o/ctl'-Sob, eLT^

Z0%^-3^TW:, %#T

A B C

2. 3. 2 (1)-

^=po^^yx*<*D, mmofmww (CAimaao#^ cAs^^e) ^g,o^'^id:^ti:!]A#<, iooo^@mo^<$& £> U'o ^>t<0, l>io-e#&)^Wiir5 J: 9 UAt)t 1000*7^ AA^Stf &tfftii'SSbLtivJ; 9 &8E^'S£LTU5o HBtittlllfflf§jx^7f^ 7 ^ KffflL'^t>©-e* 6„ ->i-/XlcA*5tl/c

8 S#:©*ti:^EA-E>o Cil;6a A- K • -i' >^A'X£tti*t'-S>o E3C ti^Sc: (f) (Yj) m@Ro#m6o&Aojaw±@^ @?@m#e#&#B%#6LTu&A6eim@m±d

^2.3.2

##6LTkL - >#mThk c® j: 7 L/cv%%-%#%#&

ca^BLUo g(Tx m2.3.2 (i)-mA@cAi##mmoB#&ma6.

CAl##:K0@|3{6fcj:*P#%6A^ ^^TU6o v^-7 7-@m6e(itia#ig$Tm3aT< ^>K7-r (LTP) o)

#g*6#^7T. l$0A^-el)CAl###B^^A^-Wr&C

A^^fl^(f^(fCAl%K#Al:#K54'l'kL A#%

T&fX ^^%A^7T6%#®A#% A^J^b^ T>K7< COck9%##^#%^#ta#A/:Wil@(±^ #H%%m 2.3.2

##«t-f^^mT#aj:9%6®-eab6Ag^*57o ^0/zAKfi, d-T^fh®

6 >7'T&&o

®#®MS^'#T^Xf-A&#G%f&C6#T#%Uo T(±, — f#&Z##(c%#-e#5®-C&5 7&?:ErftkL a C 6^6^60m 2. 3.2 (D-2EI(d:#/!r0ef7l/T^#^fl 6&$#%4-70ai3,5-e&6o jf0j:7%#^Wa@'e&ABRM ^ c a.

7 Ka<##%aaw\ mo-f-f ma a^A^^AT&^o ioo##^±&^, i#o#g@^iooo^±o##

-83- -Y y/WxAA fiEiaea

* dendritic weight anode

m2. 3. 2 (l)-2g|

SRStidend (#K%ii) node (8E/ - K) cell (/hSi-IIHSS) body (WS#) 0 4*?* 6. MAIf, dendti-O/^ A/xA*T36ttilti)]'C%6= c ft ^&&&&-&*)#6 6#0

#A9%yDtXT#^^^Tl^6(±T. ##Mma(i#ATab6o V7h0v;ji

L/-va:/?#0m, mgRT^^Uo #m^-KT&#LU0-eaba^ %%d:LT^V7

c. Sffl^n-fevthME

7d -fe y 'V' (Neural Net Processor) I$M[hIS#;S-'> ^ ^ U— Y~t

tz’fu-teyVr — ’Z&Zo #^7"Dt vlNC SS^lEllTx A##%7; a. k-X a >

5 c t*

5ef@$m&%kL 7o t y -f-mcjm LTmmmmt a c a« y7$^l#t4#oytV -fy? SRAM) 0 5#0%aT#BE^i% (m2.3.2(l)-4BI), f0##-?#ialOOO#0WStlOO

#%/W h®ytVTlOOOJ2t±0Ay]^^()-#(±act^-C#a}O\ #@M0mT(i^

-84- 3*&9L WJJM

|tt«M* $**@8 la** |ab» mm,o^xns. ~ ;(77-

**T=3-y 70-bX##S# w&vz* c *Vh7 —<7-/*-3i-X ® @ W»B8 a.fe*£» BS** Bsiaew-reifiro 7n"tzyy —09 *yH? —7/

S2. 3. 2 (l)-30 An-tr -y y 7 0 7^a-Al/-7. a. #Ea#wm##s#&a6amt%, -Y>7vy>7%6'o0 eo/:A, CO^otyf — 0^##®##, #@/<5/-7%^7o77 Ati'blSS-tSo /'f y • **') iff oy5A.3-^(^>xh7^^a>oVxh) L. /<7 7 - " ^ t v 6o 2 &#(:: LtaAkk /tV -7ft;%B$im&mmf&/:a6C)ZAT&a. ^ >f V 7 - + + V V' ^ y e V fi 1 SL^AEKti 3 #Orn^m%^ ##/>j:*r> > ^TSEiESflTl'-Bo 7 y h 9-^§iJ®PE]£§-!i/>ymo7-7vl/ytV&mmL, MWtflti/<$-y

Kd<&&&0(zx

cmmmf

- K ' /

T, @B0@#^#9 0^x &6M;L ^m0M^30M#%#7 0^T#^^fL'5o

CCDt&^gMI# (#^0%#) (:^LT0m%6#^M#"e0#@^9C6}6

W#a^AL/cC6^#g0hV^6L-C^ fC#(;:A;f706&@c#& T&6?o C0#^, A^omsm#,

L/:SAM;iJ (4) 0J:9(C, #@f/<#A#ty SJ:D#jl6#A hk coj:9K#m7'otyif0#^0^ NNP VI PROCESSOR

MMDTO-16 BFDTO16

TC558128AJ-201 TC558128AJ-20 TC558128AI-20I TC558128AJ-20

MMADO-16 BFADO-16 AO-16

MMCS BFCS MMOE BFOE MMWRL BFWRL MMWRH BFWRH CPLD XC7336 IORD- 1FRD 10WR- IFWR NETDTO-7 NETWORK select IF ADO-3 NBSY BUS NRST HI-ADRS LS245 NCLK Direct connecting IFDTO-7 PBSY to other chips PRST MCLRCS1 CSO PCLK XTALOSC CPU BUS 25MHz CHIP ID IN/OUT RESET Cascade collecting to next chip

S2.3.2 (l)-4@ 0S/WH&

-f □ -tr -y IF-iiBSicy * v ft X t y * V ictiS ® CMOS — SRAM -$• fll 11S o h ^ yo 6. CPU (*X h • ti ISA/

f v X I CSI6CS0# ts x y — h 5 o

e

LTl'&o #m^oty^0yciy7AmK^#Ly:NNPDL0K#f##^##(:LT^/=/f#

86- /=Uo ) mEoyoy^A (j:. f-fwmoem&;aau &am&& ^#f5C6^BM(C3-K^^6o %#####% ct-oT&^m^/^yt^act^fm^^ao ^myotvyoyay^A^ mmLTuao^. &<#-em#om%tfom@#&om&'?&fK ##fo^o-r(i%Uo #m (i. #gi:j;-3Tm#*tia^#w-c&6<, y oyvA^+^T^^tK^ ^2%. #myoty yoyo/vAtd#^(cj;cT 0#&##y&%&0#m&E#yactT6&0 -eL-c. ya /?toy 3^ gat ty a##^m#T# aoAx tu3@AT±&o#^/f^^m#yaoT&ao d. Sfflyn-feyy-OtSrS CO@"C(j:, m##%T^L/:#myatyy-OK#M^^vt>X^#0&'#'ao CO y a t V y- (d:%^0 C P U t (d:M% -3 /: 3 >t y Hc^ # #&# $ fLl:U a ^-w

^^#^(j:y>y;l/Tabao 16tvh (1V-K) o^^-tl —647-FOg[#([^ ^O^gB(c(±gE^O%A<#E^^^"Cuao 16tv MO^O%±<6 4 tv #

-87- loadreg^fr □— H • 1/S^X##^ (4.2 £#i@) (t#^D—bflfi©!/ yXX©e9Jn — bteS^SA,, 1 6fi4t§aiWtD-Ft«^ hit-ISo 2see (iH'^F^^a) 0000 [-eg 0001 count reg: L > X X ©IF ij count: &«. '• - ¥© U y X X^ E&l±kyxfCD-Ff5#&count2#%#*&*f5. m (16BSSB) 0012 4050 3020 : 1/ yXX 0 £ 1 it 4050 £ 3020 £n-Ft5»

initgrp, initrand, initconn if A—y°W^b^ MWSlMih loadreg ©#&^tl/^X^©D-b@B^^*t). o 0 0000 reg init count init: 1001 initgrp. 0101 nitrand, 0011 initconn initgrpid:, ?;b-Xim&U;&w&%©am-&^Xb7^b©mmf cm<. C©#^"C, initrand (D7°ci-kZ(DmffllZ(DftMfftZc V X ^ — b 7b-7©5feji(t&fiOTl£©!:t&©(4. 5 #BB)o initconn ^fiSLEtoicSE^SW-rs^X^a UTWyjlT-x VXX-b • ^-Xto^sIttH < = 7yX7

initgrp ^(16 BEE) 0991 0008 initrand #il (16 BSE) 0C52 0054 1000 : 51$ £ 54 T">- b LT%mtb. * y b 7-X £ 4096 IE8S

initconn #| (16 BSE) 0A32 00C0 0040 : #@##©&$tg(&ty b It. J&K£«J8Hb

setwta#^- b?X— b • 7—74V • 7 b VX (wta) Z ©#<&#. 'StiU7)!i\ M,V© 4 If y b 20 If y b©#-f >^#©T&1 6 If v b-C&%, ^4 -f > - V©%S7 bl/XC

# (16 BEE) 1201 0000 :w t a $ 20000 itES

tabled 7^-feX • 7— C©#^(±. 7XtX • #&©T@8 tf y b tt7-:/;b©7§ £ £ 7- b$@-CEBL, g%(d:7-7;b©@ (dt^x^m^xtyb#) ©VXb"C&%. 0010 0000 count-Rbit count-gbit: ©,*;&,% (%A$) $1 (16 BSE) 2004 0000 0005 000A 000F : 4 offl 5 o^tb©4U • *7 -k y b £ 7X-tr X • 7 -XMtSSBtr am = #m, E$©am©g#i±o-c&t). 5#Ri±2©#$EEt!iEm©EB^-r#%.

tablerd^ 7XtX • 7-7>Ef y7Xt V©##6#tE©^(tMUSo - cell/dend**hfr

t&c:aA^#5. etf^b0##B#mm@&TKK:^fo ______bit 15-12 11 10 9 8 7 6 5 4 3 2 1 0 sec base rsv connection mode input number access mode cell/dend on 0100 0 pla inc wt til l chO tbl rand this ext odd even base dh^©&$3 — b 0100 rsv ?# h y h 0 connection mode @#©Z— b 2> tf "J b pla: nJM&.<&> D inc: t> ^— b • Z — 7)1/% ii Ubi'y'^ltol- wt: Z x.— b • Z—tOb £:££#) input number Atl'F-l > $ tbl:Z7tA - Z—Z/l/©3b7-lz v b&^< >^l:Al#LTZ74zA rand:^-r>f ©#T@[1 Hv initconn^tab^mij cell-dend AAx ^^>bZd"b (a*c@) C©Hvbl± n-jvr— ■ /z V©##z b 1/Z©g±iu (£y b 1 6) CMtBf &o W^X, ^zbz^b (Hvb&icfs)

this:titi*#gs, C©gS©S«7: This hV’id ext: f>b?-Y odd: #8 % / even: (mjK#A©@Bc©AA;m^. a##AAr#ma&6M#(X fZbZ^ b%

z©3?s©4? f y 7, ^nf-^^z.iro-sr, a uW^'-o "< t *,%#?*<:,. ^%©A;,ff > V &b,r^f?(j:. c -Xiz® 2. .'I;:;. ^ 4 ©A D4i! A©.b * :- * £Eii'd < M (16 BEE) 4700 1000 0000 0000 UWtjbZA*, 'fWlMHJA) Z\ CCD assii iooo #mcm±L, 5t.MmfiEfo&u (oooo p Ig^ftT^saA #-Ct) At4,0000 SScDffljgaA^A m (16 BEE) 4100 1000 0000 0000 qmtiii&Vx Zx-b ffl (16 BEE) 4000 1000 0000 0000 C©:@£\ £x-bliU y7)> 0, 1 (@££1-b Vi?7,$) zx-bz-z;bi±#m^a%^. m (16 BEE) 4780 1000 0000 0000 0200 Ctlfi, 2 ^©AAdU >7££f7>l?!k 4f»©;t7^- 2o%©^Z>^o m (16 BEE) 4720 1000 0000 0000 Z-Ztb • Z7ZX • Z-b&fg^

i)=T ■ T—-?fis ■ 7i>±7~ 7>yj, • 1: zztz -1-bK, +-;>>= {0.2,4.6} «R •*}#»<&»-<£ 5

©z bt/A&m-#f b?&&a P P £B©^-TJ;5fcA |B|D4##c©#b#k:mw6fl5 P + 1 p+i A*Wy7y-±©EZ bl/Xli^- bt'/tyx P+2 f . >Pt^. P+3 — y 3 >Asf#?)fi§o ;£ ;£n Z — Zlb ■ 7 7 -fe X P+4 P+4 Z>ZA - Z^tAl±#m#W@6-C&S. f ©#^, P+5 P+5 P+6 A-XMt 1 t”'7 b£->z b^n§©T% Z-7> P+6 P+7 P+7 * Ptt«K+>^roigSi-57" Ku*

-89- am 2: mm*- k(±> mm7 V /< > b (inc If 7 b) setwtoTifiii5 h' v b 6s 0 v^^tDBS^x— b LteT Kvx" £***#*>«;> %o wt Hv b# 1 CDBtli, H >7"E U \Z.mULtzt> ?-- b • 7- 9 =— b 9 —7/1" 2**>roMStf«3 ^X- b - 7-7A©#7 bbAld: setwta # 7 x— b9 —7/b cell inc If 7 btf 1 Tdt, 6 3»A©wia^te5 0©=k-5 C, ^©HtTffiStT bb7&@#^b7V^bbf& b 9—7/1" W@£©8irm, -fiWC, inc tf7 b&lCf-So cell #* i n c Ify hOSb*i7)EM0 logic >j>m.Ati;££, KF cell/dond: £.RAS dend & £ 1

reset 804X #lfO output set 8F4X 25IZ1 output buf 8A8X IESIe input output inv 858X Sis input output nand 87CX NAND in1 in2 output and 88CX AND in1 in2 output or 8ECX OR in1 in2 output nor 81 CX NOR in1 in2 output xor 89CX XOR in1 in2 output rsff 84DX ?P7?£LRSFF set reset output srff 82DX 7P77&LSRFF reset set output dff 8AEX 9P77##DFF d~in ck output rscff 84FX 7P77##RSFF reset set ck output and3 88CX AND3 ini in2 in3 output ©Xt± cell/dend T-^-fhf 3 = If 7 b 3 ip F>^l£© cell/dend tlSK&^-rSo m 1 (16BSS3) 8A84 0000 0000 cell (00000 Stt) * b dend (10000#%) m2 (16 BSE) 89C0 0000 0001 0002 XOR 0#%k 1 SittfA*, 2 SifetftiJrt &m:dendm#©@a&, 10, 11 (SL, SH) Tb"© b v 7 9C dend l®7^fi' k'f'fSD-FtUo C©#^&#oif A D 7< A < 7^%^ 7 b 9 — 7# #Brf g&. 1100 reg 0160 10# (€r# hex) rag:Vi>X*#* m (1651*13) C3480020 7 b '> a Jl b • b^l/§f> b 7 "f b © 20 #% (10020 #%) ®((t t &. &#: ,I©i&^'T@m-eg&bS;7fW:0,l,3,8,9,A,B(by7f*©#Ui8##©bZ;7f) TtiT8 tf7 bCB“ b^n-So

-90- errcell/cntcell -tiM* MISSUS©^—££ n — H ##©M@H:Mda, err##k cntf#^^, »®S»*^D-FtSo 1101 0000 0100 0000 (DO# hex) errcell 1101 0001 0100 0000 (PI# hex) cut fell m (lonaiB) do# iooo ES: ^-*6 5'n-h'£toa©ti:

clrwrk/break/stop dn&^ttrD-fe^SJffllofeftCfllv^ftao clrwrk b*9 7 break lira-bX©*^ stop fcm7Siesta0 Piffle 3.5 ^#®-&«fc0 11H 0001 0100 0000 (FM0 hex)

9- p?8F&$4!re&a#, <-©&&# ioAg-c&a. clrwrk ©tg-6\ FI# 0000 ©7 o C V 5.-ffl 0000 £ 1 7- H’6{fSn'ta =

e.

aooo#— 10000#) C0|@gB#(Z)f:*k LSI (Am##

(c##LAM^Xf-A^#(fpLAo My%AA0mm#A(im^03 TU3

-eomox m*o##-emmi?#set*<^-DAo

LTU6o

f. ££&>

A#.

Rl#-e$,S e to &%#&<&set L/co

-91 m m m #)

1) %$7C- (1996) 2) Iijima, T., Ichikawa M. and Matsumoto G. : Imaging of dendritic spike propagation in CA1 neurons of hippocampus. fS## 3) WAS&B : 38, 34-45 (1994) 4) NC94-131, 131-138 (1995)

s) (1972 )

(1) 1996. ll^S#m###^445. 6 Modeling of Artificial Neurons with Complex Dendrite Structures. Michinori Ichikawa, Hitoshi Yamada, Toshio Iijima and Gen Matsumoto (2) 1996.11 7 9 #1060-1065

(3) 1997.2 (^-A%) 1997^3^^ mia#

(1) ^8^11^14Bm^#rWE#

(2) ^8%12^28B^B#mE# (3) % 8 ^12^2081)-^ 7% (Scias :###B) :EM# ^l#l 0 0#^®#

92- (2) 7 td’y h 10077 *> i- L A1000 - a. - □ > £/ X ^ A a. CtD&A b7-^^VLSIf e®AAA^#%-a-7/l/4'y b7-^®#

1. #fm - V >y

2.

3. $i#mm gn^/i/f v>yw\ /<^->mm®AA®###m®%#j^@9o t>®

wwakL &

=k9l:%6o A®B®##tbb#f6A)6A. AKR®## %±) 68%#®####-?#

'Sretinad ganglion cells®## (S43®) LTV-Bo I407-^Xf-'>3>

-e#mRfK-^-7/i/4y b7-^®##^, fgM®#@T^LT^6o ±(:^LA^mt*^^-Wrfi(^ B#^»^^7//l/^ V >y^@7-^.-7/l/4v b7-^

®##(£n convolution neural networks T'tf L ASSJbie®-E>0 3 91®® #"44X'AoVT^n- mm#ai^7A6®^^-7/i/4v c fi6®—a-7/1/4;y b7-X®#^(i, 4AAxi%A<— 6T&60 —a-o>^^EM%#"e&6o SiJ|^jM@^-jfi"5 — 3. — 7/VA y h 7 — 4®##(ix feedback neural networksT'TF L A—S

^c#J®#A(c® 60 S3E® #k®2#lclt^J-#-6®T, 4A

c®#±A^LA[]®7-?#, ^Tcm^fapDM^/f^f/i/

— a. — 7A/A y h /”' y X Axis

3@®#A^^A6^T^LTW5o >A7xm#^#l)-zL-n>Bk®2#lclte'Jf

'E.o AAL, A®i«tiN convolution neural networks ®##tN feedback neural networks®##^#

& L A2A®#®^ A"5o AAA»L>> modular neural networks

®'C6;H(dA E(e^LAA##^—^-7/bAy #ft®7-^XA-43>(i, EKe^^#A##®z.zL-7/l/4v f^-f&A###^')

93- Size of typical neural networks.

Human brain

Feedback neural networks Modular (TSP) 1000-city neural networks

Feedforward neural network: (3-layer, pattern recognition)

% 10

100-city Retinal ganglion cells

Chinese charae] 128x128 (pixel size )

64x64 Convolution neural networks A Motion detection 32x32 ■ Line detection 10-city • Edge detection

10* 10 Number of neurons

#2. 3. 2 (2)-m JL - 7 VI/* V I' 7 - ? om#

©#H

1. looom&UiovLsif ^ 7^## L 2. 3.

##

94- -7^4;^ h7-7©y-fA; -7Vl/$ y h 7 — 7 n #fg

b. lOOO-a-nyi/XT"JxfDMftJMM 7 tzy 1000#©-a-a>?

— — y J\/% "j h 7 —7 yXy-A^-^^L/Co ^SL/"czE-T?;t/ti^ 7 ^ — K/ <^©##B—^#d>A@it(:% -o T#{^f 5 o

TV -y* (t) + 52 wu7j (*) + Ii(i) (1)

y>k*to] a if a > 0 0 otherwise,

::t\ /tiii#S0-a-D>0Htg[, j,’ (t) #Mt|:i#B©—a - > l i (t)|iBeSiJi(cjb'^§iSg©-jL-n ^©TLSATrC&So ^ V?Ea] te 7 A n 7" #—^-o >0##Bajd]M#76^o 4T0-a-n>!i, ±^7^^ &o £T©—^-n 7{i3AiL©7 n y ?fc£iQ ##^#T^5©T, S—ai-n >^b©ttidj/''7L/XtiS|BlSgiC^J5Kd:tl5o

(Dvd-yxBK it (2) £ (3) T'^Jg$tlT©^> X -5 fcN — =l-u >©f*|£P1tf£^IE©J§£\ ^©##A #©-zi-o>©Ad]

d>

®#ttfc =k rXtoS'Jtt 7 77x $ tiA/n°;vx (i, ^iiz:icoR@g§T'inSS ii 5 = it

-95- hyperbolic tangentmtOJE®##

®(C% LT, COCiK-ed^Bao#^^ LT7 7 C yL &^-Wr& C6lc%6o

^ (1) T#m&Hy=###@zUl,

N dr -v!(t) + 53 wijyj(T) + + y*( 0), (4) fj>i 3=1 y,'( 0)#%%8 m&&y-o m^(112(fy h®7^7. yC7>*Cz>^-emmLTLi6o ## #7777(3K^^®/<;i/7^7')' 7A^^, %M#7777^g,0/<;t/7(l77 7#J^A^i f5o ma##z)4:(Z)mmg(;k m@l3m&om##(:lt#L7/<;i/7^^7-h7;(/7777 -WCck^T^gRL, l^@;m@^lEO#A(iy^7@J^, m®#A(17y7m^A^f5C6lr m#L-CU6o ^y:, com#m@%(c6l:'y ('®7-('7;l/7777t^#AL, 647<;l/ 0#^W3&^T#&ck9(:%-?TU&o

c. 1000-zL-O 7777AO#R% #2.3.2 (2)-2g(clOOO—a-n7777AO#jWtEI^L76 777A^#(1, 20 x 567y 70E^jT#6K^^Tli6o AT®7777^#%, *0B#:[sl%O7y7'7777777

VLS 17 y 711250, 0007- h (DO. 7 n mCM0S7- h7 7 7 L 70 —o®7v 7 mi=18#(Dmm#:[m%6^Wm#:|El%i:'3#51#(D7777(5)1% (A@Br918#® 7777(3

TU&o ^mm#:@ll%0 7777l%&%gmi-6y:6i:, #2.3.2 (2)-2g(K^L7j:3(:, #Tff®7 V 7(lW8a#[m%6 777%(3%^mv\ ^0#07 y 7(17777(3%®^^mv^T%#:# AS!0777A^#gRLTli6o ^y:, -c®^-n:20m®7y7^##L, 56#®#-K ^AL"C777A^#gRLTVi&o 777A®{t#^#(:^L7o #-a-o7#(r(ll,008#T^»), #-zL-o7^70 1, O2O@077772#-3TU5o -jL-o 7^®^(C#m^H61, 008#®7777^(4 012#®7777(1, WA^m07777tLT#mW#6%-D7H6o

-96- lOMHzO|)0(d:2

1000 Neuron System 12 external inputs

Host Computer /A High-Speed

56 chips ■ Neural chips usig cell body circuits □ Neural chips not using cell body circuits

%S2.3.2(2)-2EI 1000-^-o>vXf-A<7)#B%B

^2.3.2 (2H^

m

—a—a># 1, 008 — ZL-O 1, 020 ^rv:f-yx#& 1, 028, 160 12 #A&^#12(: y h IE {fill tf y b 7 l:''y h 20MHz 10MHz X(±5MHz MS#m: 6 Lfv h% V ^ 1,120 3.3V

97- d. l£ £.6b 7lf v KD#&^#clOO^#0^f7x-e^A#ALY=lOOO-zL-o>vX7A^^ PDM l"7-f OOf ^-^^X^^—o-e&^Winaer-take-alUg^&mi^T^&OV-^XT-y a >

6 ^ L y:##, 10, 000#i@mT& & C 6 ^#E"e # T U 6 o mc^L/cckoc. 5o mm0#t#7vi/d'uxAt)^/e^f:^%-e&6o

AM#c^y=-3T, 8^70 cmmA# ;p # w s)

[1] Hirai, Y. : VLSI Neural Natwork Systems. Gordon and Breach Science Publishers: Reading, Birkshire 1992 [2] Hirai, Y. and Yasunaga M. : A PDM digital neural network system with 1, 000 neurons fully interconnected via 1, 000, 000 6-bit synapses. Proceedings of I CON IP’ 96 Hong Kong, pp. 1251-1256, 1996

98- 2.3.3 v t° a - ^ y tzco'&m

d) ^ > e a-^ 5&

#2-^^LTAA^7f6Z6l:j:"3T, av:m ^ cm ^ a j c 6 c & -o /:«

#4A^I:

i-fiSHK L tz z. t ttw -5 £ Tr b tzwc T^y;i/^m?K3>EzL-y0fJA(d:^#(cAA'r6&ao r3yEjL-y^#H0%W iCfcWTj ti-&t^ HEK(£tiy:x.^ c 6(d:. wm## Rr#^#a^#6"3TAT^x.±(fTjo

£tz^ TAoSfqlKtoicS-ft; LTfc , ^tif-^y-S^fjESk

^ -Bo m

^WW:7^mc#H##T0'&WM-6o cmm#^T0##T& 60 r%]#j c#f&2 -30#g%7yo-f 03^®^'3^S3 C C^L^T03 >b° mu/:Wo m^:#mg%M&3c&(f5o

AFa^SMo^flSti. ^l5©E^fc

-99 ^B#^^0*#t@^"oW%(c^Aao y-A0^7^7x-%6LT^ 5 ^;c%#(bom^A6-w-^%#^#ao

f%6^o^y hT&a,, C0^Wff##^^TV6 #A®M0(j:77o-70^(0^65 7 o mcW%#^77o-70#iE4rff-D-CV5^ C#^=kD^< mUmA, LT#>LTU < 0&mav„ Dd(y h, ^y ho^y

Vo

(c) V7 h7x7x — 'J x. y V / ^ y h 7“^i_'7 x 7 h -f 7^-Ay 77 h7x7ii^y h7-7±^##Bl6B(cmA%VJ: -7i > b t lTft&i~Z ct-5 icttofco UU77h7x7Bx ZtltTcD? yt°^L-

L/VL^/^b—^"etis -fe + i U 7^r Sic(i77 h

Ox7%.-^x>h^om#aa#(D^#%%^<#ALT^t). Mb^0#T r#^#^6 07^ - K^y ^t-o/:77 hJi-^x >

!n >h°J--^{c^-rstffgI^b©77a-7i LT. WWii:OT©—o£3§-/lTV50 (j:, mo##&9m#(i:e®tf7i/0#m-czb6o c^^-ecv <-7^0^771/^0^ M#@iC%LTfFb#T#A^ ^yzLT-eflb^m^^-Wr&CtiCck^Tx H^0A#:^t7 ;HbT#60^61^7 6 A# &o -e6i-6A±-C0#8ijt 771/0##^t

m0m

W<@#"e&5 7o V7h0i7 y-r->'*;u J-—V % y h y —y X-->'x> h ■{>3-71-7. (mm##A# # mm#) I-VX>h (□db h) #2.3. (1)-

-100- (2)

a.

Al^fltll (Artificial Intelligence) 3 > t: ^ - f ^ A^C^flt ^ ^1: A^0ft6 m#&a63'LTl'ao M (1960^ft@) ALTl^o e0#g(C3b60(j:##(Reasoning) T&*). Ky^>'-T>x

;!/-;!/) AXW^x-fAliMM^oy^A (Lisp. Prolog. 7oyy

^3>x%7A#am#m^^) (Cck%m&f6C

g, 6o#%K Agu^aj: ?«

mm, Ob® #%^) @lW%0y:)6lcB^tf';HI:LT#-3Ti\6o 3 1/1° a. — $ (Computer) 3 > h° a — 9 ®Ki:o t'>T(d:IIP1W(1){c^#S^. J 4 x > (Von Neumann) #(c a#^fi6=k9^. m#f-&3>b"zL-;Mi7o?'7A##giT&&d<#;uc. (abem#? (j:) ^Ty^7>gl3>b°^-^T&ao x/’f-f X (Device)

X^#. ##0% (10 &*$?-?&&hv> wos(CMos). wos^^iaicaAwao ^ -eii. c#64"30m%^^^L^0}6 t° ^ — 7 . Computer) . D (x/^ ^X. Device) 6Wa$flTU6o 1##X^(j:3>b''^-^C&[#^(hf6X#T$,(). V-T

7>#3>l:zL-^±CD7;l/d'VXA&aiXj:^Dy3A0^i)#^#jC'6LT^Co A X%#A(i. ^m(:^#L/:7;l/=fVXA6W9#^mx..

m^07vl/=fVXA-e&60i:*L. AX^#T id:. ^@g(D6nH^r^ti^V^mM%^#^®/:A(D7;[/d'VXA^#^LT#T^6o A 7

-101- Artificial Intelligence

LTU6. #2.3.3 (2)-lEl0±## (A 6 0 v hkL -zL-o h9-?6LTtf;Hk*%T&6J2m#$y ^7-?efVi/6LT0m#&cmA&#L TU60 1980^ft T# AI%]

5;L/$'y &-c\

(T/<^X) 6^90A

2.3.3(2MElOT^ (C6D) #2.3.3

(A6B) 6T^ (C6D) ^^CX-o^60(±, m#%@# AI%# (A) 6B (B) (Cognitive Science)T^ao (D) 6f/WX (D) (j:. #(:#@HcMaLTUao (A) 6 (B) #2.3.3(2)-m0A, B,

o > h° jl — ^Olftt ^ i > *} x.7 (Brainware)(4)i l O M #o T # # L ^6#. yi/^>C7jc76l'3#±(i, #2.3.3 (2)-m-#-/It't,A. B. C. D0#A6LT

102- b. nmziMitmM

+% h ?^5o VT/l/fYA^m&^^facefcab'Bo mmtL-ni, %*# rmmj rm#j

mm, #;s, f-+xh (^zc;eo#A)

rm^J # (#f-:fV) (cgcf

r@#j o#A. A^#^#^fat)6K%6#^A i#T&(oi#-e&^ # a6fa0d<#^T&6&g,. rm^j -co#^A T&53o t;5Av cceid:, ##(: ra#j (b

^oy5Am#|gre0#*0^#0 106 LTSfEt^SSSro (Operational Semantics)#^ -E>#A CLTiiStt^rV-y XDW}^ (±#^6 LT®^(9#l^<-#-TC#mLT^ao

£?ti%>£)''£>'V'$>%>o

3>k°zL-f

^oy7;>/^#@T^3 6C6(:&6o m^03>l:^- Tvi/^vxAo#^ yoy?Af#^ k°zi-^T(^ r#^Kj;a#gJ Tfft)#j:9f560T6ao %-3-C. c. MMo>t°i.-?(Dtbg'Ty6^0^0^##^-(i-j:9^L-CUao

103- Difference Level 2 > on Structure

Difference Level 1 on Fundamental Patterns time i time i+1 2.3.3(2)-2gi

7;i/^v

§o 4") Ramose

#W^0Aa04:(:^@LAAA^f#&f 5AG,, f AT, Rm^^LAMmotAvl/ Am^$ti5T&5 9o eA;l/Am#^^iA&6(±, %Oo) o#^(c%5o r^c^a^gj iryo/ 7^>y(:#^A5C6^^-WrA96LTA5o #AB, Ctl^T0zL^-7;l/4;vh®#% T#f#/&9c9#6f57-X&&^AA, —@PT&5C6A^ ^5(%^fiA^mc#LT(j:W#T&56U3A3(c#A5^#T&53o

6AT#5« o) ^mo^#r(W#T&5A, 7;i/=fvXA0#^<@m^#^ amny^-f m^03>l:^-^0^##^6C6^ #3C6A%#^^5o WA<(±$6B2c(D#^K^B3tl5o 2f, 7;l/^VXA-eo^

^iJA(i\ BA vX7A©P4IIAAH)t$ titzt L J; o 0 LA L, 7;l/d-vXA(i^M^B"3< d" V XAAg,A#:^K^L=k9 6f 5 6, ^#LA7i[/^U XAA^m#(c

-104- -e&^o yoy^AofpBE^am^w?#*^. mamctmfff&7o?'7AT&^TW

g%m%W(=kL m^oy^^vxAc

«t'E,^m#^®Rg^0#&f6Z

U7;l/^-rA(:

TA^tM#m±0v7;i/f-rAmm&ff6c0=k3%i>%7A %LT%6-c. h (^m7HA^ai 0mm(c&/cam^3>t:zL-^^6R-#- &T65-5o (^SA# MO: &)

1) mgA;mA<^3i>-?T3>t::i-f&#iJr)/=;K *^Uj%(1995) 2) NC&iVLSIzLzi-on^yzL-f. ^^(1991) 3) m^ec cm) = H,si, m?##@##6d996) 4) M7^m^##^#^#NEDO-PR-95009(1996)

-105- 0 1 u V rr p N3 -fcx ro £m E na tnif T$ -± 8 0 Jv rt 8 DO 4^ h-L d d 9v d ) Si m rv ft d zy- d Si S^ cd ror 4 Ok tk # y nn- 1$ an> C C VN B1 EE - 1 • H rt 8 8 V rt. 4, c~ d X VI- d 04 A ow Sit- 0 (X y 4 IS % $ y 4 4 4 # 4 if ­ • ft 8 8 ft d* A d 9t Sf 9t fV (X li fM y if 4 if il if if 4 y # X X K 4 32 P SB 4 %\ # <- 0 0 8 jffl r 04 A R 4 VI- Ok r^t- *> C" St- X # Mn # )JD- s ? m EE y y IS 4 zx u„ 8 V d d d V V d X. rv 9+ «- 4 c~ d rv rv (X can m d Mi "77, s 0 # 3d y # # 13 rm- n% 4 * m 1 *. • 1

u — 8 V DO 4^ h- 1 d d rv 9t iy- rt ft rv 04 c~ d 3 ffilt c Oik y 4 n$ # % 4 * S - zx 4 m % # 1 M- d V d rX X a w 0 #

- yt 4 y ft 8 .c\ d 04 ' x^r rv ft 9- rv K c~ Ok 3i* 4 nw m # 4 4 y 4^ # rt > )# % ss _kj 1st 4

^ $- — rt 8 d 8 rt 8 rt V % 1 00 d 04 rv ru i X K 4 St- sl- N4 4 3 4 iW 0 y 1$ X m 32 4 # SI 4 o> ?$ n 0 0 8 rv X d d' 04 d 8 d c~ d Ml A zy- ft IE (V X <~ y m vl- “ S ■V/y r< rv ;y- » rv 4 ix 04 # y X % m 4 s3 8 8 8 X x c~ 04 4 % 39 # X s # 4 m 32 <

• H r 8 8 d 4 A St- 04 ow m y # @1 s g: # nl % 8 d. V 9? 04 4 2 ft 04 d vl- SB > 4 # St- 4 3t IS

<

3 » r n % d 04 04 d St- 04 Or 3 St- 0 4 * a $ 4 IS *L ,JvVt //ins ws

“ © © r< © % rv C d X X vl- * 5Dr S $ y % 41 1 3H 9+ s # ss » Mil rv W 8 6 0 ft 9-v" X d w- ll- &r C y yx X IS m 4 % a 32 4 MB ri 9+ ifr a- rv -M * rv w I * 4- * I 1 * — 0 d rv d d A St- 04 §a 4 X td & - nti[i null I DO d Ml 4^ Ml A _ A c: (4 rv 04 % * 4" 1 mt 32 H #

*■ 1 • H — V rt n I CO co 05 4 d rv g: 4 0 / a * 4 mb • © X ft d r 4 S % 4ir if X if y # EE # a igf 4 s 8 d 0 0'S S* 0

#2.4. i - m

' i

— t

05 o i—1 /X d4- 1 CD A 0,. 4 hi 5® d x 9 © 4 Or 4 4 d s 4 at® 4 4 0 a > Or 4 Or CD <— <~ rt in -d 0 A a X 35 d St @a 4 S Sit ° d a g 4 ow 94 tX IXD y y 4 A X d d 0 4 4 w Or &v Mi fin £N © # $ (X 94 4 94 4 8 f t xj< Or Em ant / -C— Or n> n> Bti CP" o V ft m (X i® 4 ft 51 c~ at® <5- j 0 SF # VI- X 8 # # 5® out 4 4 d4- D m m- 4 z • % St a m Or PP d a y % d ik 4m % 4 S # Sit a$ N> iff® a S Q 0« Z ov % % m > V # E d4- d 9 i# /X 7$ T # y Si C~ d U. 8 Ml 4 ft f# 8 3 # > 4 4 o< © A 4 to g X' 4 d # s y & at® m Or s / A y # m d a 9- CX O W l 4 cx T3 4 4-. 4 EE- X 4 Sv % 8 ft % X Mm a c~ 9- 4 # 9 i—1 \y< d Or a z m- % V ft 4 LI r< %% X 4 ?x % A 3s 8 4 > Or d tX 94 m «— do H y y c~ z > A cH i Or A © 4 A © ° 4 # S# Sit # # 8 4 ° ii # o< > hr Or ss Or d X Vi 4 A V H # $t 8 # 41 to HO rt > 4 4 X H I®as 4 © s® 35 A © c* s ft no- o< ow u 8 © $t Or 5$ Sft at® Or & C 8 g: S > A CX # a W wa S® d Or 0* rn % M d V ffl Or a Or a i—k t® 4 rt S® 94 9: X # © M> s © s& n> 5tT Wt C* 0 mg % S® Sh n: X # C X 0 8 to % 0 % ot rv "3 no d X m St d S® s A MT 0 C ant- # M 9* ant S if gt r 94 d © X pp X at® PP X ra Q V *t © a % m MI 4 4 4 d y 1 A St © s Mi 4 4 □h 4 9fi Or a # 9 sH ©f u. d 4 St * “N 1 m # # 4 Mi V Etl m # > u„ 8 ° fX £1 4 4m % 8 © ant* d 8* m mp CX r -r V 4m d g: -X X at® Or S) 4 4 tX 8 PP 1# ft 4 HO m 4 CO 4 8* X (X d)> ° y 94- 4 X o X 4 V H a © gr Sit 0- m a ft A # ° ° y o 0 d rt S4 9© St FtT V 9 A Mi 4 ii c~ 8 8 MT ot 4 W d 4m -4 d4- r< % y I G 2 HO CO IN) 4-y 94 4 ru 4 y » d jSS 0 1 C 8 8 ow X8 > d irfr 4 eX 3 s ft #11 8 nn- 4Jt 1 9-t flute- a! Or d » $ # EE A 8 iv to Si to y Si 9 iltj 4 o« Cx *g S' |Dr A 8 A Or S y 3 C~ %P SI d 0 3 iite4 ant CT# 94 PP u # y y n> S 4 rx f# 3 C©###l:^f6i&e?#6Ll:©DNA^#^fl^6f6 6x

#A3fU:9o C©C6(j:x m^##t)7 vi/^vXAoi@@#^#^WL^ ir©#k#%#)& (d:^c/:

7A#d

(3) ^mm^^AtLxom

mi±x i#&f#m<35#^^eT#/:7vi/7';xA0@mm#om#& (7L/^>^^7) ^ j:-3Tmw^iK+^"cm#mfG^0##%myx7A(c^f6o -e©##& (7t/-r>7j: 7) tLTx #^%(ix #mmmic#A3#T^Dx ##$11% ttteia©7®#jg (AP©IOTt£*£lf) tLT#A^,^TU6C6(j:x 2.1 T##%LAo C A(ix M ' 'f» <«, • v --/-- --^ “ i.*-*'- - ~ ’ ?*?'■- *5ll-|n]A’'-DTl,'i)i:SB3ttAtJ:A<-iio *)

A%©#im' ^#© 4f:4. fd. 4* I emm C©/=36x 2-3. g#^#K • x^;^-£$?*) Ati • tti7

-108- a (#tm 6#x.6o

(4) &TV

<##LA-3Tl'6(Z)T?&&o &A A) m%xm 1) ?. y^yxni/?' i. 7vd-^># (#*7C'trmms:R) r#^ - g, f#m. 1977#)

2) 1. 1. 1987#) 3) eA?c# (#%^x#%$4#7^T'7V-42, 1996#)

4) rRNA±0f*#^3b°-f^RNA®#jm-^®igmom#^#%®mjgj /

—109 — 2.4.2 • 7-*5 L^91^©S2-ti:ftBE=i

(1) JH®3>tTa-^©1ftftf-Ot'T

MO##tf;L/6LT0^.zL-5;L/^y h9-^^<-X(:LT'3< 6^6-zL-o3>l: 6##f C 6"'o % ot\ y t U<-% « T-+-r^f- + ®^ littftfi—F^xTft

Ltcb cot LTaeSIS iiJ: -5 o rnMni/tczL-f vXTA®#eN

/eu<-% » B&tT/l/if&7-+7?f

T&53o

6 LT(D##(c-3l\-C^6o

(2) mkomm.7-*7-9T+ a. Mic&v&mm^TJi'

T. ±Tm#^c<

#2.4.2-1 EI(j:mEfrt^l/^|@@(:#6-±(f6C6T, f;i/^#6RLTU6o coctkk 1^3 C6&##LTW^o b. mmmmcmcKmmtfvt/ — xL-7;l/^yh(Z)f^;f%x 7^-K7t7-K#^ 7/f-K/<')/^E0=k9(z^#f6

"'o y %T 4" 7 V V KE(c#f5 7 V v

ARTCAdaptive Resonance Theory)C&-j' < $ v h 7 - ?, ART4; v h 7- ^T6 6o (ART * y h 7 - ? kk #PM#0 ^lu^bGrossberg Network^&Pfl-£tl&0 )

-110 HVC

#2.4.2-1

S L T V £ £ V ? SHE T'ab >0 „ t bl^T-^-r kUni^tU ^-K^x #2.4.2-2 E0=k9 (c^

/'- e,fiT^6o c. yiH^-BPgJg^TOl/^&So tfJ&j&ti—*.-n 'y^K-Xt LfcPDPCParallel Distributed Processing) 3$ ^ V a—XA0at#(C^O, sWs^M^tiWLt LTU€>0

ATTENTIONAL ORIENTING SUBSYSTEM SUBSYSTEM GAIN

INPUT PATTERN

#2.4.2-2 g| ART^yh?-?

-Ill- #2.4.2-3

cm@) eomm^±oi/

6, 6W7^#&T&5o

^-^e —-SlEti^-^e —-5^)6l31^ 5>f<£ 575k -th7"v > ^A/EEtcS-dUT, ^ PDPO2 00 L/<;l/^E^ f&„ h7-^LNet(PDP^y h7-X) 6^^ l^ffll'5i|l/kl/©^y h 7-^HNet© 2 o© VKjUD^ >y h 7 - ^ led; *9 M@^~5 (S 2. 4.2-3 El) o 2o®L/<;l/0ma^m6LT/\-tzL-^&56^%fC6^T#j:7o

r^u 0#%0#mc65o

(3) i2tft®l3Vh'a-^

t)65o %-oT,

'7-+T^^^;5k ^2#ftim!=]>tf:L-f

—C0##6LT^ X' 7-+7^^^aL'nd, m2. 4.2-5 EI®=k7%tf';l/^^

-112- Cerebral Cortex

(slow but logically elaborate)

Activating Thalamus factors .

Amygdala multimodal (fast but crude) sensory inputs S2. 4. 2-4 E

LT, (LTM). a. > U £ £’

amEtm, (#ACJ:-3T(i. Ocko^u

#2. 4. 2-5 g|

-113- %8 :am#±:ammg cpum©i/v%^6U9*^-3^<3^=k 3o b. mm?yEI.-?om^mm12?

Am©#@l3#d<

6%&o t;5/„ e©&6, cxxw^m#L/:6#xxw^fte) ^imm. ^##/j#©

3 S*###m©&-3#6<&WLTl'

—A> y ^atmi^SU^Ki^ gAjFfiE^©k©//$Bf[;£tiTU£;fc&Ht§tttiil < ^ JS L^L, m©to#&#m<%fK 9Bm©#sd<&&e%u#6w<&&o ytv^-% -7-^f#co#@(;:&ago#

#AA© fc&gfty- 0 fc © K t£ 6 9 o d) # 2 mftmig; 3 > b° ^ ^ ©m#~i% ^ -c ytV<-X'7- + T^ff©#2#ft^3>tfa-^©##(i (1) (AiB^fJ© cko^) ##©#A#:-e&^x #^©mmfpmi:j;o. mi#ftd:^a;m^©##}6

(1) filteff (Behavior);#'1 Z>&tzWJ(k'C'&> Y) ^ (2) (ifltfiSc (Architecture)#'1 Z>&tzW$k -e&6#A m2#ftm^3>t:zL-f©m#t-e(i. (i) t (2) ,<7

±&#c±&© am##&&-32l/<;P©|#m#Wl'3©##(!:LT8&&T&59o C©#A, @#^©(±,

#© r#m##j ±#m# rm#j ^##^#%7-

L^L, ytV^-%-7-+7?7f©m2#fW#:]>b°^-f^|D|#T©m#k|\ B# 3 >k°jL-^©#%#<#L V' X 7 v 3" I: A 6 C «h& A#L 0 , 4"#©^#^#$^o (jsmA# M a: &)

114 1) t v, voi. 32. No. 12. pp. 1301-1309 (1991 ) 2) ATm:%#m###^2r-A%(1994) 3) voi.77, No. 12, pp. 1256-1262(1994) 4) G.Matsumoto et al.: "Brain-computing” , Extended Abstract of Brainware Workshop, FED-148, pp. 31-36, Tsukuba(1996) 5) T. Ae et al. : "Two-Level Processing for Real-Time Image Understannding” , Real-Time Imaging II, Electrical Imaging ’97, San Jose(1997)

115- mu M'Xmmmmmmtmm am) (f MR I ) UT-

Peter Bandettini (Medical College of Wisconsin) and Hideaki Koizumi (Hitachi Central Research Laboratory)

(i) f mr i mm

Koizumi : & A © 5^ < ©iro(i n #(CfMR I (functional magnetic resonance imaging)

Bandettini : mic^cTf MR I L/:o

t Ltzo Koizumi :#&#,/'-/<-KA#TM<#&fiaL/:aA 6^6©#^?-#"^? Bandettini : M¥©n—If >1f±©i^ 6tr-To vly'y%k&m&Z (MG H) ©0 f

Koizumi : C 6Tf L^ ttz^ SBfLV^ ^Wtoo

Bandettini : ^ztlt^^tzOli^Mtz. 0 Koizumi : f MR I ©Iw^Cfi^ £tz-h^MM^^z L Ateo %]im^ 'h ^ L/:^o Bandettini : y=- 3/:—ig^*WTL/:A

Institute for Health)©o/< — h • 9— j-—1#± (#%^ a > K> A#) tvx — /<- K©^r> » >y#±(3^©A^6%AL-C&Ki@6Tl^ LA« Koizumi : C©j;9 (C 1 9 9 2 tf © 6 ^ Cffliff lTHo©i^7A-^ b f MR I ©lsi5C

i'©Tt&? C7^X3>v>E$4A#, f LT/\- /<_ FA#©?"^ ztty Bandettini :#L//<7-^3>, Tl/T2#%. PET

-116- (positron emission tomography) A £\ t > h £ r> izT~fu — f^ilt Lfco LAL. €-©4l'C\ SO©#x.^'$ > h

L/:o Koizumi : f">B£. 'yKf-4 7-Tr&£ tz-fotfX'-tteo Bandettini: #rLl^g^a^6£#(:(d:.

U60 fftf'fk i- £'A£'A#Nm&#=U'©Tfo #. Wp(c.

(2) f MR I ##&% 1. h^'sZzrj y^JtiJf^- (#e#%) Koizumi : IMR!©#####^##?#/:©#. %©S#£NIH (^g) , eLT7 77A#±©'?y?xT'7:/?m%#r (MXV) ^£^ K>A#^mi9.

(:m-3T#m©#%^^m6T^afo #©s##gi#m##%%&i&-3Ti'a fMRiom^^ 2©j:-p(:^LT^af^? Bandettini : g#^#LTl'5/'-/<-KA#£'^X3>y>E#A#K:'3(''TTfd<. #

x.Ti'e6muafo ^-LT. #mM%#m##©^a-mL/ch%m^mL-c^a-#-o c©*ga

-117- 2^0 Koizumi : < h 5 l/Xf-'j '>7° 0 A V jr -i — > (&$}#%) h ^ o Z. t & =k < If 9

8m4#&o##K i&K m:#, #

y 7LTU&fc#0##Tf ;K r,T^ & 6(c±#c#@K:#gf-s# 6#^.T

LT^<0-C(i%^TL j= 9^0

Bandettini : < h y >XxV is~7l) A U tm — >^ i; $ icx ^ ©

Koizumi : A < -€-© t 9 0 tr'^feo Bandettini :^<©^KmA'^R$#6m-^(!:LTf MR I (ij:V'#A

Bandettini : Koizumi : * #i#f"<5, ^ < (DM'ftSf^tzfzilfM^ -o fzfzifU'ht^k.S 4-aT0^<0%m#w^LTuafo #(zm##m#©ck9 A^lA< K@L/: V -y-a<tzh'’tM^'>tto

(3) mrnm^rnm^? Koizumi :AWH©fMRI-C^ 1.5T(1AX7 = 10,000^X7%) ©Ut»^6ig##6^ 3T- 4T© f MR I ©^@K#-e&5* ; -71/ -0-y/i:Vi/#± (;4;V^A^) (d:x 2dvU:7TlcMf:U

-118- Bandettini 6, ^fWR##©S/Nlt (#^*##lt) BOLD (blood oxygenation level dependent)n y b y x b kliAL^y©t?Ac?&flM

A%©{%#%#^ (Tl) C#bH -o©&Aa9g#^bx

*-#l:';l/#±©

TkL 3 Tomamm (BP!: echo-planar imaging) T, ®{C 0.6m©^FB1M L^L. ^^^##c-ouT^am#^b@am#^©#ff(j:mm am#i:w <©&*&<&% afo oc,'m#%#m) ^#jmf-ac6i:=k^Tx am#T&imm##(;:&#&^<©

Koizumi : 607y o -^ A##

Bandettini : •€" y 0 £ bi^> aMI#©$A:#(^My 6SAR (surface absorption rate) ©

am# #m(izR.m©Bgmmt L-cme&fiTi'a©-?. -j8:K:K#m^m^a<, m# gm©m:#ai's/Na<#&^&6i'3C6-e-#-o @am#^m-cm#&9%s/N^##^m&f 5©id:, mm ^#3#%##A©^(:^r)T#^-4-o 4 TT/<- Ky-^gOQDn^;!/^ ^LT^#^#6©(igm©#Tfo 3 T-T&#LT##-CK6%a-&/Lo Koizumi : #m- %#m©#A^y^- t'^Lcmam#mm©e^+^gi#ai'#-c6id:@ m/:a#3 Z6-M-&0 Bandettini : ^#6, ##L:^%L'3'36& f MR I ©^WT. C6d<&03fo |s|# C. ^#©M (artifacts)&#bfi#U©-C\ @i#©lW#{(:(d:+^&m^g ctTfo fMR i©+^%g#%m%LK#bay:imie^mmf ac& till"'i i>X(Ci oT ■fcilGil^tt^r ti b A^ yo

-119- Koizumi

5^0#JC % % #a om#d<6 66^3 C 6Tfkio C6T[d:6^^-l±/Lo

(4) Efc&'^X h/x^yV A£;£AT Koizumi

7 fV A0^lj^^ L%m(f ^ ?

Baudottini : $\z.\i''—KA^T"7>tw — • A—-i$±h>A#0Xfw-7>' tz-y-V^^ttPETA^^y-fA Ly:o f MR I (j:PETj;l9 EPR (evoked potential response) d; D 6 ^ 9 C 0 IB^®/<7y^A(d:PET^bem^^/:^0^<, PET0##0#@ ^-ei^&^xHcRB&frruaLf:,, L^L. ^O^-p^/^^y^ACMb^i^ j:9%, 20-30%>^yxy^^V\ #02O-3O#imiH&L%U6l'9mk;iEL#

frtit'

&6C5^-%]6C6A<-e#5(D-efo Koizumi : ?£ •€> ISiz'o Bandettini : ^3C#g^:0(±, m%##T#U5U5^%-?%$#&^&C7>yAC^;t 6#M^g6Ll:^6C

f^U±#^##T100 msd--y0#

Koizumi :#%%/<7y^A®aiJdiCkl\ Bandettini : ■£<£> biStyHiTo

120- Koizumi 7l/yv

Bandettini : X 7T^fctotl^-ti/vo

(5) *3©*$fciSttf'(7)

#^7^-X6 LTii$ &%< l#6<

LT. A^#B%f 6 & A 0 6 L7 j: - X A 6 El' a f ^<0

Bandettini : ^6 -9 A £ S V ^ Xo Koizumi fMRI0#^%ii£'O2bA()A&60XLj:7^? ^JA(AE ##^4:^6 LA#Em#g^#6C:£(d:# (d:ALT#^%£#A&0"eL * 7&0 Bandettini : ##(:oUTX=f *<, EOSESMAltSJMA^US £;: £>T'1"o tt\ S

LA^ A0XA^'# TLjzOo ftU:. -3^0. igAf MR I Aj:^T#m&#£#§r fMR i WA##-?v k° >XA(f? A < . c® X 3 AE 6 A^A# A A 4T#m^ #ATua j:9(c#e$fo Koizumi :#^A, EAj5(±6#m@#£mM^#^O^#^O#%(j:#mt#J0m#A

Bandettini : &3l:%A%a<#0XAyX£LTa#LA#:Nf%&%l'0# £ u •? SS©SS^ if A b A VT' L 9 fro 05) A^MgAOd, ##A#^##£. -eLTfMRI^i:^-C^a#a##(D^03;ziJ:X-7 3 7-efo #JA#\ -u- 7A^f^X (v-^^###^^r) O7V%h7'3^*1#±0j;3A#^f06 0&m%LTU&AkL 2'7i)-3Tm^#$[|LA^

121- %-D-C#^fo m&&m%LTl'&#W:A4 AT7;fn- a/f/f^WMRI&#^Tl'ath4,o ^©cko^A^tOn;^-

>r — 's 3 y^AWf t -1U'£ir0 £ £ lz<$■$&£ o tzlifr <0 t SA& Tf* L <* 7 0 Koizumi : h 7 >X7> y 7 V A ') — ©119 Hi c A-?^ Lfcfao

Bandettini : tiV0 %©##=f—C0I®Wb13 ; a.—A — y a >©n ^ ^—>r~7

Koizumi A^©m*'e©7a>T^7Tf^o

(6) ffij*'\©sn Koizumi : fMRKi. CtUi. m#g' mm-mm#

< 0BG#ma<#6frr < 6 Bandettini : iiV0 ## - ###(: WmM%77c-f^#6m6fl^fo mtfSSfg 7^Xl/^y7 (dyslexia, Tld:, #5###:^ (MT, mm^s|6#g) C@mi:A6 fi&ckoc,

£tc^ © < fu3l^©Rf££M^©Ct fcliL©tii£tlT©£1“o £

Koizumi Bandettini : ^ < ©A/^'ltg^I? i#x.Tt't t«= SiAESEIf_®JSS7' lv >7y

(7) $6[c^*/\|p|(^T Koizumi : Hifi&g £tlT©£7 >7 7 fatitzltt 5 tlfci^^T©

Bandettini : 7>77^#%^rT©mm(i. A/rCctf^ ©A^^#l\T©a6l'9 C6lrUy:o c«k^<-xc^<©a^Tir^6o @*^%5^6T7TLTmA6C6^AtnT(±iZ©^6#L:$ L/:o ##]©*§#,

Koizumi : C7#%$#lT© 6 <##z^> (complex systems) t © 7 i)0lcol'Ti"0 J;

7 (C#C^ L/=^? %*^6#%$$l"C©6 77X(caotic system) t

122- Bandettini : LT jo D,

>/AT'Cft£ fc@autSlgte&Vo WISi:jboT, ^oif

i9iij% (emergence)^ # X X idJrb^T fc-o £ (robust) fx Koizumi : £jl&Wd\ ##&aMiJ::/l'nt:-£. #!g&ai4i%.$vl/^-£U'O%o0

^oja. ^0^T, m##^£B#m##t#W<0rij:L,

@xa £#m^i(c#joj%a(>0T^ y%f-Ad<§ ^^ma-a@@£#XTi^'#-o Koizumi fMRI0%lc^60(j:^0j:3%^m# ffimi'af-d'? Bandettini : o —Ajq ^oTF'-^fa* • * ^4% Slfc© ^&Iro @ lEa-i] jo ii: A: £ E V ^ 'fx f fMRi/NiRS (&#f(.^)^y- fMRi/MEG (m#M-) £#-oj=ma-Aj3e-e-#-jao fMRit^© m^jo-wr-etL m%ejk*F0/\t/D « f MR I £ M E Gjotir-e(i> f MR I ©M#©^6}##^#a £ £ j)< T#6—MEG0^-ei@W^^M#^#^££*

It$cdM if b -? £ =fc (. 5ft)'' -3 %/ -S "Zr L <* 9o Koizumi :NIRS^y-^>/, #^b^/7X^-l:cUTtd:^omjojT,^'#-^?

Bandettini : )fc h d^/7 7 ^ jo £iS%y\:ey n t: ji^f ^ g,. Kib^#£ ji^o £ji(if MR amg%#^D^Xa££(c% a-rwoo vTii/f^A-em^m^b^mxa©^ ^T^M#M^%ac£A<-e#^fo cjtkk

y xt A x A a # a or##^ab a © T c 0^&^^o ^ r#e L

Koizumi :

-123- Bandettini : f MR I

Koizumi : |=I#&###T"C\ f6j%££ L(C|ftl^ Bandettini : t'C.'X ^iSffl^'afe-ST'L J; 9o Koizumi : t i?K Bandettini : -tfO—Wiiz-ptO tz^'il:"i't3io Koizumi (1997^2^5 8,

P. A. Bandettini and H. Koizumi, ”7r I

INNERVISION, 12(4), (1997). (g###tb, Tel: 03-3818-9821)

-124- *-9- K&1R

mUL $iJ@r, #2im.

A^-C^2-5mm03^T, ^

< wcj; ^-c^o. sm^Torna^o^ 7 A ^

##<&# mcao, 6„ ernm^mm

ic, A»iL-c.

& 5o Sf&I^MR I (functional magnetic resonance imaging: fMRI), v h n >CT (positron emission tomography: PET) &^f9\-it^ ^ — 'J y 7" (near infra-red spectroscopic imaging: NIRSI, or optical topography: OT) fr'&£0 ##(C PS IS (p#) (electro-encephalography: EEC) 4'b MS##ff(super-conducting quantum interference device: SQUID) fcpSlH (magneto-encephalography: MEG) i)<$>%>0 C tl bOA'/SIro^rE^ LT> SS> APa! mm -

(functional magnetic resonance imaging: fMRI)

%*®^^hocT (?ET) -c, m%0m^##(d:*<7 7 7T-(c=i;-]-cmuai £ tltzfr, 7"n 6 C d: (blood oxygenation level dependent: BOLD %^) &##LT, /:o #& *&g^Anmem&

-125- <,c,'a/<7fx A>

mW2#0@#&fXl'f&%&(:#, ^ a/:EIB(L-T^Rmc#^* W:EIB

SiniaB -iBo

7 xW&3 BS£l> 'y IS (#%?0^^^-^a##Ty-<;i/*)

^ j: 7 % t> 0 L, #B&M-?M0#J 1% %) 1 ^ @ A Bm L T ^ 5 o <##?& (complex systems) >©@^:^-Cl tl^b© 7 n >-r^ 7 ftM&Mz-tz g

126-

ss

a^M^mA-c^^rntr^g

m^g,^6j:3x io#@m (#^5$, M5^mm) L^L,

K z 6 & (f e & & „ a /=*y o ^ J: ^ h6&B^^Tio d

wmu-c& oa ^%j; o 61^oA^ 3#a^ L^o

-129-

S4S

4$ trttf

L^L. f##i(d:A(:^?#^X6C6A<6T&66f66, A (hONmcjoWT. ##&

A#yo/7A0 j; 31:7:3 jL7;t/i>Ml:%-3T#&&fi&0"e#%:l'0T\ #%07— zLTvi/ c^a/7A) %V^o 6f:)6l:(d:, AC% e^0 kk C0yi/-^%;l/-0WA^#^Mk:z2. —+ -^03 >t: a: —

llEO'/nt y If • <—% • 7 —4-r ^ 7-^07 * 's J -i v >In > t° j. —7 ti"/n 7"7 A (:%-36 0 3>k:zL-^A<#Ae/zAC^yo/vACj;^!:? ^jL7;l/^-fF^6C:tT&6o 73.a7vl/#3>l:a.-^[d:7Z:a7;l/l:(d:J&

^m07vl/^VXA (#) AA1##^#X^##-#-^/:360|%&6LT

0<&djAf5o AAfac6-e#^%m^L:, 7;i/X-XA (#) a<###b&o C0«t3 tv- ' <-% ' 7-+7^7f 0#/77>#3>t:^-

CZT0/1 v c0$#M%m#3 >k°^.-^0 3 Hd:. 7;i/xvxA^@m##f6mA^fX3h^^y(d:, &%) y V 3 LT^ #3L M#3 %^0yntyy » <-% » 7-77^7-^071 >777>

K % 6 % %^R0 3 > k° zi - f 6 ### C#f lA < T& 5 7 o A^##% (^&#%) (1, m#3>b°^-fa<7;t/:fVXA;&8#m#U 70##f6 7;l/3*vXA0g8t"M^@#^-e#6#lii& (7"l/7>^j:7) &B70&00#3S(:j: -3T4)%<3^cyac6(cj:0, Mvxf-At L-C0B#3>t:^-^^#^M^f-6/:6 0SI@^r> l' ( o^O^jS^ fco

133-

wmwm-tim e v swaa z v $sa gias^ i v xjpuadv

a. i 9tmm$L

mi mm

1. M S B : « 8 ^ 7 J3 1 0 g (7k) 15:30 ~ 17:30 2. M#*Br: cm) 3. ft & m: d)#m - #ma (mm#) - # m e# (mm#). mm (mm#) . #% (mm#). (mm#). ms cmm#). (%ms). sm (mzs). No (5y$mm). i## (s mib#). m± (#±m#). i^m. mil# (a$mm). /m (a&). (^/+-y) . #* (#@ftm. a^). *# (amftm. *@R#). (#?#) (2)m^m##

(3) NEDO

(4) $## (#K?m)

Subsbiis. mm. see E#*# Hsm^i-i W7^m#s# 8%%^#%#^ H8BR1-2 HsmKi-s g^9-^yay"/ M#^B(^) H8BK1-4 H8l^Kl-5 Sl^-fl/ay^ tvya>(^) H8B^l-6 5^9-^ vav^ m#m###(^) H8MK1-7

4.# $ (1)## :

-137 #¥£ LT9=# z -= tz0

%. hmm^gmL/c, 6« ' #%(ck 0 6 L T±(f ^ 0 , E# 6 (± M^E "30% 0 &o

(3) #mm^

(4) #%#### m# Hsmm-i 9-^yay-y^ $ w:o (5) f gR 8 #g^#e%K-o^T #m#m ($?m) C0^g0c.-X^K#

<5 o

*B^#RAT##f6o

gC @ v^E

(^2-3B@m o ^9-^ xa yVCol^T

### ck D ## H8SmS1™2 ~ H8$g8:l~6 ^ 9 — ^2 v a v 9®

C 6 6 L/:o

ma c^b^) ^ms^-a-a, -3s## = $5&#m -f? y v 3 >2A. ( 1 &) (i LeDoux, Goldman Rakic, McGoughi§-JS;©|III

138- (RRA) : ####

• -fe y ■> 3 >2. ( 1 4=r) Aertsen, Singer, Abeles#&®l#T3^#'#'

-1 v y a >2B. ogmm#### (2g) (##A^## = mm##) abt-2## = mawM### = mm##), ;nm i#*& wbA^##:####) 0m^r-e^#f5o K# • -tr y '> 3 >3. (3^) tiKoernerS, : &###) > m;n cmm# = ##). ;nx (ATR = mm##) a„ t y a >5. 6o ms####]###-',, • -f? y v 3 >5. ( 1 £) KochS, (^## : teASSS) >Leon Chuaft (fr^fllS Calif. Univ.: A###) Poggioft (3C## : %A## &) om(&T^#f6o mmmmm# (s^) ##) . 4^# (m&A : ##) . lii* (NEC : 153###) 6

p;i/i± 6A5C6C L. 7 ^T&a65 C 6 K L Ao £1 ±

139- W2 8 4# m 1 mm

\.mm b f ^ 8 ^ 7 ^ 3 0 B (1&) 13:30-15:30 2. mmmwi 3. titi s s

- #m& #m cm#) - # m #$ (mm#). (mm#), m# (^#). # (#iE^). NK os m*) . mm (s#m#). /j\# (sm

XI UtEEk # (3)NEDO ■M, icffll M (4)#B# (^?m) mas nib, mm,

H8BA1-1 H8BA1-2 H8mAl-3 BE9-^^av^ (#0 H88BIA1-4 SEV-^7 v a yX M'fX;?- • XX'J^-X H8BA1-5 @E9-^S/ayy tv^a>(^) H8HBA1-6 SE7-^c/3vy @####^ (^) H8BA1-7 25: ^ci H8MA1-8

4.

(1)## :

^7 moB 6^y:o

-140- tife O tZo

?#g%LTua - com#?) rB^.#j&j o^-e, ##

#V=l'o (3)Bg^### (ZK^) :

gm2fu=;&k h^momerx dcm##?#

L, 9 ^ s ^0?##*oBwa-mzmsM&a t& ±(fa^gd<&ao b. mzRmm(NEDO) =

#0%##^-^36TU#/zUo c. m&Rm =

M###74'f7d<@a:aC

g L /:o (5) ### (#iii E^M^A) Z«9. *# H8MA1-1 BtfF?-

##) $#CNL?g(8^7^10B(cM#$fiy:R#

CO^Oz.-X^K

-141- # O'# (C ^ 6660 -^##^®-#6LT. g^7-^V3'yygg#^?%L-C^0. #

£tZ' ###j;DR#k mm^-e®

mjMf a 6 ®msa<& 0 #m a #/:« (7) B%9-^ s/ a ^y "/IC-DUT H8BA1-2 - H8BA1-6 icgcf#. ^-^ygv^®^###^

(8) ^®#

(DsMfrfc-otz o a. *#g®wm%##. mm.

^T®fsm^® " ^ t V ®##^r±(f a (Functional Memories) £ 1' 9 60 • Saifi^^lM^fijffl LT. zl-*?4 >^7^-XCDB^SJ(37x-;L/-tr-7^

* L/cfleg^ft© V 5 -a. Is—is 3 >'SrMi^MIE'^"ao - @#%6U3#m^%^L'3c. gm%A-

- LT^ao-e. a®A^A$<^#Df^#^®@m - A^tty (L®^Ac6aK#6 LT. a - LT®B##C&B L. t>y6%^@%6®M^v%yAg* b. A#®9%^#0-

*^g®#^M#® #t%^4-#® % 0 a&® UT&T® mm#<&r, S/V 3 ><-X0#%&@AT@jg#3 il'3#gg&%&#&a& #®0f^N^^®^ (#&) & - y ,y F ^ Nm#®^ Ld

-142- m%o3uma

*fHzNmLT,

tl&o b tl^Bo

m ±

-143- riH8ltS1S$8&&©$fc£SF3£J % 2 ®&m =Mzc

1. m m B : « 8 ^ 1 1 ft 6 B (7k) 13:00-15:30 2. Mm&ar: (#) 6#^ 3. tti s s: d)#m • SMM - # m 9=@ (mmsn . (mmm . . Mo (s#m m . m± (B$mm . /j\m (B]%). m* B^). oamfta. *m^#). (#

(2) S®SS ik% xmi mm. /Mt. % (3) N E D 0

(4) ### (#?#)

f mss#* m i (^) Hsmm-2 H8^m-3 (^) H8^R2-4 Cfny^A H8MK2-5

4. is m

-#1#

—144 — '#3# -*§ cf

- y - v >y

c©m^c.

- &#@m#&Ax#m0m:m#&A6@m&©mm

• 7;l/3"V X'A©g|»ji • x^wjSffl©Mat

'a#©x#^m#0^(#?

vs'^D^-iiL'? h nu=A x£o&tf-Stew'd' < D ©i&SEtt • X-;[/©#%6 7aX%©XX^a-;Wk gA&XcA®

MteM^kPiir&ctteltzo (3)S^9-^ i>a

%io, (ftm #mcg*

(#3 0) % 8 ^12^ 178 (A) f%l#30^b^@f^C

£ Si r>tz o JW ±

-145- 1. m m B : « 8^12^170 OX) 13:30-16:30 2. = cm 6#^ 3. ft S S : d)#m cmm#) ' # 9=# cmm#) . (^#). me (mm#) , mi (mm#) . NX (nm^:), (%m%^). Mo (x#m#), ;%# (x#

(#±@#), ma# (0$mm , /^m (0^). (0^), 159 (1T#ft@. #T) (yt--^) . (&#fta, WR#) . - #m ($?%) (2)m# ## m ih% xi ME9> (3)NED 0

(4)### (m?m) ^EX Mg

@s#*# H8BK3-1 (m H8BK3-2 (B^c) (80

4. # $ (1)## :

(2) Su@SV:B©BE

(3) ?j5% 8 : #*#ma=k94^a©###kL

-146- r ^r-^CD h — >J

- K7'y7^%#L/:Uo

'#fL^7;[/^VXAom^ (7L/-?x;i,-) ^fGm(c(±#g - g<'0fGm^#L:5O(d:jlg^6 r&Sj - t g m 6 a

• Z>fr-C'tS.\^&l$-l3fi[\^7F.-tZ.tTfilr'

rm#3>t:^-^j cfuc^-oi: U6l'6%#m~077n-7A

i&S^'zb -E>> (h®##^

(4) ^0# mm^w=k^^T0^m<6-3/:o %8$m0#E?m-emm3>k°zL-^ - 7-^7^7^0@f%M^cg@LT, NEDoms0#&g%a?jEo

%gk K##M^(i%9^2^12B (*) 4:#imoej:^ (M) ####?#

Kl ±

—147 — w-m 8 ^ m 2 4 mmm

1. Bl AH 0 : ¥$ 9 %■ 2 ft 1 2 B (tK) 13:30-15:30 2. M### : (M) 3. as is m : d)#m - #m# e* = mm#) - # # ^# (mm#), (mm#). mm (mm#). (^#), # (m fGA), (mxA). mm - No (s#mm , /j\# (xm

s^) x lag (#m - = 4&T) . *# (#*fts, $m&#) [4$iii (a?m) ] (2)3

XI

(3)] ebsmik :b

E#*# H8mA2/#4-l %8^m # i (^)

H8BA2/K4-2 f B%8^m # 3 (^) H8mA2/K4-3 ^g%8^& m&#%#^##6% (m H8mA2/K4-4 %8#m Am#m6 H8IB(A2/K4-5

4.1$ m (1)^# :

ts -S cfc. 9 Ulit^cfc o tz0 ttz^ wiisix

ff(Z)##mmj 6LT.

148 (2)0 m mmAia 16# (^) &^#3[mm 16# m (^)

(3)f 5% 8 W: l^T : ct % # 3 ^ ^ 3ibS

£ ### ic—f£ $ tlfCo (4)BE7-^ v 3 "j ^MfgSt-^VT

(D: m a : ¥$, 8^12^10 0~ 1 1 H

)##'#xf-:#Bij#m if#, memm i5f#(m sm 29f# 197^ 279^ (P9 #SA 22^) 0#^(:cUT rnffmmA#^ r#;n^e#^w c^##% K##m) ,to. irn,

bi-ckoTM^^ti*^n-tr'yf-fco^T. $?ilt©i ^c. f v^(LSlXbjockc;##(D##(cc^T. #W&^y:o

f y-yoME^J ICOUT #mm J: 0 I: =k ^ (PJ) ? 7 o t y 1# f 'y B M £ L /: aT##^® ^ fi Ao mmicNf #e#^ a c £ £ u^o (7)^0#

T& -o /=„

- mcf< %- cm#f 6o L%>" eycf< Ip" Si4> -E>o

-149 - A

# uut y (zA#A<^omm3:mT# 6 w#mo ®#:#J h9-f-d<

' If'yyyo^jLy hT%<#@g!0yn^j:y ' %#Oct9(:kL g-A#0 7 ISM&c/SMti^b&Vo ®mm6A#0*f6 - A#6LTW##om##a

36 btl-E>B©5S^rffl#"f'5o -gim±#0Xn>;j:f hT##&=N\*<%*x z®^eo#^(±mmo m®A#% %AA^ 3 o% D #&&<##-? 6 & o " Z%&-CO###%Td'%D/-zL-d<&TUaC)-Cx C%&&#A##/r

j: t). % 9

Tl'^o - CO%36|:A -S Z. h.&5tW~$~-Bo . 4-^(m#3>k^-^^#:,i:4cm^m36T#/:^ commt&wmoAmAm

Kl ± A. 2 r«tEtfimi@Er7-1 is 3 v yj

International Workshop on Brainware m # : I WB’ 9 6

£ m : Bm&A

* *: % SE: iW#a#6, g$#Eiai%#A. B*#a#6 mmam % 8 ^ 1 2 ^ 1 0 B (A) - 1 1 B (*) 12^ 1 0 0 12:30 ~ 18:20 12^118 9:30 — 18:00 mmmm ■.

bSISjS : 7-kv^a > r#c##aa(2)j rma##aiu rr^'^f X/yXfAj r|gj^zi > b°o. — 9 j r/'^Alh'rV Xtl y '> a >J

o m m # ise cm ##%### 3 #) 2 9#

###A0# : 191 9 7^ (i) c?i'0^7)

5^8 512^108 (40

'* — -f—'s*f/~f\/’i- V — (mm : 12:30 - 13:00 iSiSSSS

#3'j (SE) »;H 13:00 - 13:30 6m M. Ito President, Science Council of Japan rK#j& "To Create the Brain - Perspective of Brain-like Computers ”

■fe y is a V 1 A. (mm = mm L 1. L. Cahill Univ. of California, USA 13:30 - 14:20 Brain Mechanisms for the Modulation of Memory by Emotion ”

L 2. ig'/S 14=20 -14=55 K. Tanaka Inst, of Phys. and Chem. Res. (RIKEN), Japan rwtmommamj Neural Mechanisms of Object Vision ”

L 3. ysig mz 14:55 - 15:30 K. Kawano Electrotechnical Laboratory, Japan

"Neural Mechanisms of Ocular Following Responses ”

3 — t — y'u — 9 15:30 - 15:45

■fe v a y 1 B. ®CDltffi*S(2) («* : SIS *- ;e?K«86»®S) L 4. A. Aertsen Albert-Ludwigs Univ., Germany 15:45 - 16:35 "Dynamic Organization of Cortical Activity ”

L 5. mm W; 16=35 -17=10 T. Iijima Electrotechnical Lab., Japan r (c J: 5

"Entorhinal-Hippocampal Interactions Revealed by Real-Time Imaging ”

-152- * y'> a > 2. : /J\# l 6. tpm j] WLt# 17:10 - 17:45 T. Nakada Niigata Univ., Japan

"MR Analysis of Brain Function: Present and Future”

l 7. mm 17:45 - 18:20 S. Kuriki Hokkaido Univ, Japan K a Yntxc a j Perception and Recognition of Letters - Magnetoencephalographic Study of Human Brain -

y )x 18:30 - 20:00 mm % (cm ##a#) S. Maeda

¥ES8^12^115 (tK) tr>3> 3. T/^X/vXtA cmm: 9=# M L 8. ## 9:30 - 10:05 M. Aizawa Tokyo Inst, of Tech., Japan =t C/— zl— D ^ y h 17 — y7 yiSEJ "Artificially Designed Formation of Biomolecular and Neuronal Networks on Solid Surface”

L 9. S. Ullman Weizmann Inst., Israel 10:05 - 10:55 A Model for Bottom-up and Top-down Information Processing in the Visual Cortex ”

L10. WH 10:55 - 11:30 Y. Hirai Univ. of Tsukuba, Japan F6 C' y h 1007] v AX (C j; a 1, 000.% ZL - o PAD TM 'J 9 zl — 7 )l/% "j h 7 — ^ yXfAj PDM Digital Neural Network System with 1,000 Neurons Fully Interconnected via 1,000,000 6-bit Synapses ”

lii. atm mm 11:30 - 12:05 K. Yamada NEC Corp. , Japan KM f ^ ^ a zi - 7 ^ h 7 - ^ j "Artifideal Neural Networks for Practical Pattern Proces ­ sing ”

# m 12:05 - 13:00

-153- t V v 3 > 4. — ~f \sM.y-?— v3 y 13:00 - 14:15

■fe "j y 3 y 5. (mm : MB #f& ;B*#M) L12. E. Koerner *BK¥f 14:15 - 14:50 Honda R&D Co. Ltd., Japan

yotyyy : ’’Parallel-Sequential Processing in Asynchronous Networks of Spiking Neurons: A Theory of Cortical Computing ”

ns. rfjin m# 14:50 - 15:25 M. Ichikawa Electrotechnical Lab., Japan ^n*7 pn4r-y+l©^^j ”A Newly Developed Micro-Processor for Brain Computing ”

L14. m# g# 15:25 - 16:00 K. Doya Japan Sci. and Tech. Corp. , Japan

’’Reinforcement Learning in Animals and Robots ”

Z? -- t -- ~7 1/ -- y 7 16:00 - 16:15 ■fe "j y 3 y 6. xti y Ss a y (mm: 16:15 - 17:45

"Brainware & Brain Computing ”

L. Cahill (Univ. of California, USA) A. Aertsen (Albert-Ludwigs Univ., Germany) S. Oilman (Weizman Inst., Israel) ¥# W= (mau#) (B iLMffpJr) mm (a$mm)

^o-yyf 7C 17:45 - 18:00 (2) mmom

(D t ^ v a > 1 A : Information Processing in Brain (StDlWIESZkS (1))

d © -tz >y v 3 >T'l

LCahiii#±(d:,

(#(C7Kl/AV» T=&%, 63—3(±

W i, l^i#±(m# LTU e o

mccuT©#^^#^Ly:o #mmK#m(Z)[email protected]; u(i^©iz -3

c©m&@#&?f-3-cust&c, 6y;h A > ^wm©w%/<^^ g, d©##(i, y f 3^T^6pT##^^'6C6^#L-CUeo R±, c©tyva>-e(j:,

6^Sia©fi(|@^b|^0^ LAtbllAchOo d© j; 9 1C, ^oTtt$&Ti6ll£JSfotlTUA:, #

^ ^ 3 -c # /:©A<

m # A)

(D fr «y y 3 >1 B : Information Processing in the Brain (9^©##%# (2))

d©dz y -> 3 >trfd:,Ad Aertsent#zb©”Dynamic Organization of Cortical Activity ” t iESt#dr©” Entorhinal-Hippocampal Interaction s Revealed by Real-Time Imaging ”© 2 o

-155- Aertsen1#±(i^ —o —c0z.^-o>O a#o^#C"3 L ###4; 3 o (c^-o-ca'jatx zL-ow^ i^c-zL-oK^L-cimemmim^m^T^^ct KMf&6 y-mm v \ im cornea#- am&a#^#-#- a c 6 c j; *), ^m#o&aR^#mfac6^r#ao eu-c. #ff@cm%aMfG(i. j;-3Tzmi3^a#4m4 6 ^fk^aam (ongoing act i V i ty) 6 L /: - ^ - o> OMfG 6 a C 61: J; *) ?#JT # a C 6 ^

£titz0 BftU^j&l^ni~6fSWi> BMM'Mte.M't6~zl — n >©SJtE{clt^Tfc y

(%7-rx) acticjiDx <

^@E^-a $)0T& D. !2#0#m^ LTl\ < ±T—30#@^O ^x.T < a^0?&ao

A; ? ?(cwcLTix c6d

Memo

(D tey S'3 > 2 : Measurement of Brain Activities *tyc/=,yw\ ^^LTAm^KCck-oTm^fiamm^mL, -#;:###0&^afiao ## o^&icA^'j^fiao ##a#c6t)^"D-c%fa###gK&avxim#^a^M- ##-a#&T&ao 4—#@%#(:6^^3^®mfT#m^k^#mijfaA*T^ao

(Superconducting Quantum Interference Device : SQUID)> ##I1##WMRI (Functional Magnetic Resonance Imaging : fMRI) tiao ^ ^ h n >CT

156 (Positron Emission Tomography) 5

L. fMRIi^-oVTbN $T^ SBOLD (Blood Oxygenation State Level Dependent) ^KoUT^btlfco dtUi^ ^"n J;< ^5 C LT&tX 1: a#®a##m.5T0yx^A (U6 m#yx^A-c#i5%i;:&mfa6L, @i%m#v%f-AASS.LA:o eft#.

btl/c. teffl&ftlfcMEGggtes 2 ^ >4;;l/0SQUIDT, st^mmtJEntex^t -#^#0 Ma?6#^(i$u%#250ms^mre(j:a#jm^T@; (i T)

C0j; -5 a c t & c mam# /j\ m ^ m

(4) "fe y y g y 3 : Devices and Systems (r/t'fXiy^jA) $tyygykL ^-eo^oO#^3(i^L#fiT, S^:f/W%gcVX4:<-^a'Oj;9 (:y%f-A04:T^jmLTU<^tL^, KO##&9%%_WV:o (miA) wsfsm&A %(:3> ho-vi/Ly:^,

^S^rfr o tz0 uiimani$± (cy^x^y###^) (mm#mmo##mm(:9 )UTyx^Am& 9 a#^

^ 9 mm 6. g,mAB^i=i^ 9 mmA<*^ y <;i/-e#sfpm& c t ^ j; -c,

cm%A) (±. l,OOO^-n>0f,fyf;h-^-7;WK'yh9-?:&#Jf;m CO @%ov^^a.

-157- Em#%m (NEcf#myf^7#%^r)

TSLAo fALAtyvaXi, #©tyya>£lt^ff:c>X©mm&&%5AT&&;K #K#

3= # @ m

(D t -y y 3 V 5 : Brain Computing (JK® =J > fcf a — £)

9 l-HUil'#'&MM t f tlicfUlf' 6 fc> ti/co ■by is 3 AESEfDAS®iEW-ir>^-6DKoerner1t±^ “Parallel- sequential processing in asynchronous networks of spiking neurons : A theory of cor ­ tical computing ” £H tz0 Koernerf#±(i> >/^;l/%© 6 “coincidennce detector ” iLtib£> Egic t;t £(DA^;fr>

£A6A#%#KA^m*LAo £©#$Micrt>oT,

“temporal coding ” (#£§f J;-Sfffgn — -r-f >^0 ©fili^-^^LT //>7^£e(^'540Hz@a©m%A3 > t: a —y b&ISH©—

^(:m?g#mA#^©mil#±. %A#±Ccka “A newly developed micro-processor for brain computing ” £® Lfc^S^frAtlfco F7 %.7

#M^oiryth- (Y-hTl/f) T&5o C©/:^ 7Ao^@%£lt^ >d. 100^3%^ y a >©Am##m[gmm©##^#^'rqf#T&6£^3o ^^^©-yotyf-^Ay h 7-^/

^/:#hRM7'ya-fl:j:&^©m#£U9lW#^b,tx Yl/f >3>

-158- j; 6 “Reinforcement learning in animals and robots ” #H#t#±liTemporal Difference (TD) ^#RiJ4r4^C'6: L

actor-critic system ” t LT!il''Tt'> a >Xti> temporal codingCDltllUL Z> ~J -< — K7 * U — Y-yj — K^'y ^MS©P4!l Vol-)UDT$^ MMzfu-\zy-y-—cr>/'—V'ry

-h(jf btl\ ■€"tl biZ'Oi''~C:^M.^^Msmi)^^S. cF tlfzo (8$%^ H I i i&)

159 (3) Workshop Schedule

December 10 (Tues), 1996 December 11 (Wed), 1996

(Science Hall)

(Science Hall) Sesion 3 9:30-12:05

L8-L11 P O S T Opening 12:30-13:00 Lunch 12:05-13:00 E R

Plenary 13:00-13:30 Session 4 13:00-14:15 E X Poster Session H PA-l - PA-13 PB-1 Session 1A 13:30-15:30 I P PC-1 ‘ PC-11 PD-l‘PD-4 B 0 I L1-L3 S T T Sesion 5 14:15-16:00 I E O R L12-L14 N Coffee Break 15:30-15:45 E X H Coffee Break 16:00-16:15 Session IB 15:45-17:10 1 L4.L5 B I Session 6 16:15-17:45 T Panel Discussion Session 2 17:10-18:20 I O N Closing 17:45-18:00 L6.L7

Banquet 18:30-20:00

(Restaurant "The Space")

-160- (4) CONTENTS / PROGRAM FOR EXTENDED ABSTRACTS

December 10, 1996 Lecture Session Plenary To Create the Brain -Perspective of Brain-like Computers - M. Ito (13:00-13:30) President, Science Council of Japan

Session 1A; Information Processing in Brain (1) L—1. Brain Mechanisms for the Modulation of Memory by Emotion ...... 3 L. Cahill (13:30-14:20) University of California

L—2. Neural Mechanisms of Object Vision 14 K. Tanaka (14:20-14:55) The Institute of Physical and Chemical Research (RIKEN)

L-3. Neural Mechanisms of Ocular Following Responses 17 K. Kawano (14:55-15:30) Electrotechnical Laboratory

Session IB; Information Processing in Brain (2) L—4. Dynamic Organization of Cortical Activity 23 A. Aertsen (15:45—16:35) Albert -Ludwigs University

L—5. Entorhinal —Hippocampal Interactions Revealed by Real-Time Imaging T. Iijima 27 Electrotechnical Laboratory (16:35-17:10)

Session 2; Measurement of Brain Activities L-6. MR Analysis of Brain Function: Present and Future 33 T. Nakada (17:10-17:45) Niigata University

L—7. Perception and Recognition of Letters-Magnetoencephalographic Study of Human Brain- 36 S. Kuriki (17:45-18:20) Hokkaido University

-161- December 11, 1996 Session 3; Devices and Systems L—8. Artificially Designed Formation of Biomolecular and Neuronal Networks on Solid Surface ...... 41 M. Aizawa ( 9:30-10:05) Tokyo Institute of Technology

L-9. A Model for Bottom-up and Top-down Information Processing in the Visual Cortex ...... 44 S. Ullman (10:05-10:55) The Weizmann Institute of Science

L-10. PDM Digital Neural Network System with 1,000 Neurons Fully Interconnected via 1,000,000 6-bit Synapses ...... 54 Y. Hirai (10:55-11:30) University of Tsukuba

L—11. Artificical Neural Networks for Practical Pattern Processing ...... 57 K. Yamada (11:30-12:05) NEC Corporation

Session 5; Brain Computing L-12. Parallel-Sequential Processing in Asynchronous Networks of Spiking Neurons:A Theory of Cortical Computing ...... 63 E. Koerner (14:15-14:50) Honda R&D Co., Ltd.

L—13. A Newly Developed Micro-Processor for Brain Computing 66 M. Ichikawa (14:50-15:25) Electrotechnical Laboratory

L—14. Reinforcement Learning in Animals and Robots 69 K. Doya (15:25-16:00) Japan Science and Technology Corporation

Session 6; Panel Discussion ”Brainware & Brain Computing ” (16:15-17:45) Chairperson: Panelists: G. Matsumoto L. Cahill Y. Hirai (Electrotechnical A. Aertsen H. Koizumi (Hitachi,Ltd.) Laboratory) S. Ullman K. Okajima (NEC Corp.)

-162- Poster Session (Session 4) PA—1. Hypothetical Reasoning in a Brain-like Modular Network ...... 75 H. Tsujino, T. Masutani and E. Koerner Honda R&D Co., Ltd.

PA—2. Synchronization of Processing in Asynchronous Networks of Spiking Neurons: A Novel Scheme for Gamma Oscillation in the Brain ...... 78 S. Nagai and E. Koerner Honda R&D Co. , Ltd.

PA-3. AnAutonomous Robot with Biological Brain Model Based on Basal Ganglia and Hippocampus ...... 81 J. Yamamoto and Y. Anzai Keio University

PA-4. Computational Physiology of Lateral Inhibition ...... 84 T. Kato, T. Sakamoto and Y. Kobayashi* Electrotechnical Laboratory *Toppan Printing Co.,Ltd.

PA-5. Dual Oscillator System in the Procerebral Lobe of Limax Brain...... 87 A. Iwaxna, A. Yamada, E. Kono, T. Kimura and T. Sekiguchi Sanyo Electric Co., Ltd.

PA-6. Neurotransmitter Levels in a Rat Brain Subject to Magnetic Resonance Imaging ...... 90 K Hyodo and K Homma Mechanical Engineering Laboratory

PA—7. Theory of Self—Organization of Spatio-Temporal Receptive Fields...... 93 M. Miyashita and S. Tanaka* NEC Carp. *The Institute of Physical and Chemical Research(RIKEN)

PA—8. Identification of Noradrenaline Receptor Subtypes that Mediate Synaptic Plasticity in the Cerebral Cortex ...... 96 N. Qkado, K. Nakadate and M. Matsukawa University of Tsukuba PA-9. Distribution Pattern of Serotonin —2A Receptors that Mediate Synaptic Plasticity: An Immunohistochemical Study ...... 98 S. Hamada, N. Okado and K. Senzaki University of Tsukuba

PA-10. Aggression and Serotonin : Contribution of Inhibitory Mechanisms ...... 101 S. Ueda, A. Nishimura* and K. Yoshimoto* Dokkyo University School of Medicine * Kyoto Prefectural University of Medicine

PA—11. Correlated Activities of Adjacent Neurons in the Hippocampus ...... 104 S. Suzuki, H. Kaneko, M. Akamatsu and J. Okada* National Inst, of Bioscience and Human-Technology ♦Kyushu University

PA-12. Dynamical Neural Responses for Vocalization in Guinea Pig Auditory Cortex ...... 107 R. Tokioka, T. Miyashita, N. Murai and K. Fukunishi Hitachi Ltd.

PA-13. A Role of Dynamics of Piriform/Entorhinal/Hippocampal Cortical System in Odor Recognition ...... 110 T. Oyamada, Y. Kashimori and T. Kambara, The University of Electro-Communication

PB—1. Dynamically Adaptable CMOS Winner-Take—All Neural Network ...... 113 K. Iizuka, M. Miyamoto and H. Matsui Sharp Corp.

PC-1. Image Enhancement Based on Lateral Inhibition and Light/ ...... 116 Dark Adaptation Y. Kobayashi, T. Kato* and T. Sakamoto* Toppan Printing Co.,Ltd. ♦Electrotechnical Laboratory

PC-2. The Gabor-Type Receptive Field as Derived by the Mutual- Inf omation Maximization ...... 119 K. Okajima NEC Corp.

164- PC—3. Illusory Line Motion by Two-stage Motion Energy Model for Area MT ...... 122 M Nomura NEC Corp.

PC-4. Dynamical Map Model Processing Temporal Correlations between Stimuli ...... 125 N. Kiriyama, 0. Hoshino, Y. Kashimori and T. Kambara The University of Electro —Communication

PC—5. A Model of Working Memory Playing a Leading Part in Awareness-- 128 S. Inoue, 0. Hoshino, Y. Kashimori and T. Kambara The University of Electro —Communication

PC-6. Active Perception Based on Coupling between Electrolocation and Jamming Avoidance Response 131 Y. Kashimori and T. Kambara The University of Electro —Communication

PC-7. Coherent Link between Dynamical Attractors as a Key Process for Odor Perception 134 O. Hoshino, Y. Kashimori and T. Kambara The University of Electro —Communication

PC-8. Recognition Mechanism of Quality and Intensity of Taste Stimuli in Gustatory System 137 H. Horie, A. Tsuboi, K. Waki, Y. Kashimori and T. Kambara The University of Electro —Communication

PC-9. Models of Electroreceptor and Electrosensory Lateral Line Lobe for Electrolocation of Electric Fish 140 H. Owada, M Kimiyou, Y. Kashimori and T. Kambara The University of Electro —Communication

PC-10. Possible Architecture for a Brain Like Computation Machine ...... 143 T. Qmori Tokyo University of Agriculture & Technology

PC—11. Inductive Learning on Special-Purpose Brainware Architecture ...... 146 H. Fukumoto, S. Hiwatashi and T. Ae Hiroshima University

-165- PD-1. Multiple —Dipole Source Localization Based on Information Criteria — 147 T. Kiyuna, K. Kami jo and T. YamazaM NBC Corp.

PD-2. Self-Repairing in Neuro —Computers under Defective Neuron Circuits ...... 150 M. Yasunaga and K. Mold* University of Tsukuba ♦Hitachi Microcomputer System Ltd.

PD-3. 3D Object Representation in the Inferotemporal Cortex of Macaque Monkeys ...... 153 K Tamura National Inst, of Bioscience and Human—Technology

PD-4. Development of Multi-channel Extracellular Recording System and Its Application...... 156 H. Oka, H. Sugihara, R. Ogawa and M. Taketani Matsushita Electric Industrial Co., Ltd.

-166- a. 3

(1) ir>if7^-va

a. lac&ic 1996^12^ 9 B^blSB^T, ^SKmiq]#, B(T®6^ 7^>b>D.C.TM#$a/:" MF1 '96"(Lm#^jTTU6t>-!)—KM -#-5m^##KmgL/:o @##^AT, /7V7^;1/^7T#^:#(C.I.T.),^V

oK#, -7^^ot>T-K#^2’&*)8K, aK#0#&,

b. mm#% " The 2nd MFI ’96 "KOUT 1EEE/SICE /RSJ (BTo^y h#6) ±#® "1996 International Conference on Multisensor Fusion and Integration for Intelligent

System" 61'7 &0T 6^T, #imi(±1994^K7%<^%-eM#^fiy:o fm:SI##l'T@#2imd<19964EK# *^KjEU$E@Km^TU5o /:

-167- ’’Including Sensor Bias in Shape from Motion Calibration and Sensor Fusion ” (5) ^((Navigation) ------18® ’’Adaptive Rule of Combination for Observations Over Time”, "Dead Reckoning Navigation of Movile Robot Using an Indirect Kalman Filter”

© 3§4r —(Varied Sensing) ------6® ’’Study of Odor Compass ”&df0 © 'J x.y Y vX-r A (Multi-Agent System) ------4® "Multiagent-Based Distributed Manipulator Control ”, "Guiding Distributed Manipulation with Mobile Sensor ” &df0 ® 3 > t0^. — 9 \i'J b >(Image Understanding & Computer Vision) ------4® ”Shape Reconstruction from Endoscope Image by Its Shadings ”, "Stereo Matching as a Multistage Grouping Process ”&£"0 © ffifi(Grasping) ------4® "Dynamic Object Localization via a Proximity Sensor Network ”, "Integration of Vision and Force Sensors for Grasping ” /idiV ® (Bio-Based) ------2® "Development of a Fish-eye VR System with Human Visual Functioning and Biological Signals ”, "Analog VLSI Circuits for Sensory Attentive Processing ” © (Stereo)------3® ’’Multi-Sensor Fusion Scheme for Calibration-Free StereoVision in a Manufacturing Workcell ”dfo © §###%6i##(Target Detection & Tracking) ------10® ”Human Motion Capture by Integrating Gyroscopes and Accelerometers ”, "Multiplicative Inhibitory Velocity Detector (MIVD) and Multi- Velocity Motion Detection Neural Network Model ” ftifo © B#lalSfS(Real World Computing) ------4® ”A Proposal of Novel Information Architecture—Open Cooperative Work Space—"fc dfo ® 6(Vision & Range Fusion) ------4® "Sensor-Fusion of Intensity- and Laserrnge- Images ”, ’’Recovering Shape and Reflectance Properties from a Sequence of Range and Color

168 Images ” U £'0 © toSCTactile Sensing) ------4 $S "Optimal Grasping Using Visual and Tactile Feedback”if0 © fbWj-fe > y >— > ^'(Active Sensing & Planning) ------3 Si "Multimedia Sensor Fusion for Intelligent Camera Control ”AEif0 © fSffi(Application) ------3 IS "Sensor Fusion and Planning with Perception-Action Network ” &if0 ® (Low-Level Fusion) ------4 Si "Adaptive Sensor Models","Optical Design of Distributed Signal Detection Systems ”ifo © VA — (Sonar)------4 "Fusing Bynaural Sonar Information for Object Recognition Systems ” %if„ @ — -l — h (Neural Net)------311 "On the Active Perception of Speech by Robots ” U if0 @ S lyKCD(High-Level Fusion) "Intelligence Fusion","Communications Management in Decentralised Data Fusion Systems ”%2'o © #ia(Archi tecture)------4 # "Real-Time Microvision System with Three-Dimensional Integration ”, "Video Compression and Enhancement Sensors Using Column-Parallel architecture" % if0

i.

tzo fcWS(Navigation)i^St"^rnSLlH^l^fr^tZo -t? y v a > b LT(i 1 8 iit'9 d ffe-ir >y -> a >K fflU>tF(r-ol' ‘Tfe#Sl-8E7)^Vo SlraftnrKv

— 169 — 6T^ Cfl^-#9/:6IC(!:^LT^#0t>y^#lzK7-7©#m"C^6o ^"7^%^^ lc#LTnd(v hAy#^© FcUH^ *) ±if z> nr © tz0 ##&od(y % mm#^:^6C5^#f6l:(j:CtiL^ bflf: (c%±(:|#9 T&TU& 6l'9©d<##T&6 9 ^t-ro%g§©%A «S# odfy H^t!Z;

o-^-A^©'7 + ^ an >/n°x UbAftl^b^S) ili^^xioi^ &Ei596]i-a#e& ii. §###%] 6 m#

M-fey h LT\ ^M'tKDHit!^^LT. 5 ^©6

iii. &#:©##©#& ftfitintfy h (cAFb I©J5Atil?#D AtU: 3 6 © 3 Mlql^£> ^ A^©##^#$m lcy-^6L-C4%#Lj;3 6f 6^A<$,6o m©#@G(ctDm&M-f^ + #fi(c j;-3Tt>-fA:%6©#ll3A:be&fr(,\ f-f©K©[p|±l:^)6Tl^o ^<©A©M;C^ i^A/f©^. ^fff6A^^#fawm6LT, ^%^#^#mLck9 6f^#%-e&^y=o QA©#^#(:j:'5 6©-e66o t>y6LT(i, */7, Aomg#. 5o 7-yy^f >ym&^<6(cio(,^T*%h7-©%W^g(c%5C6^&&©-ec©j;3 tm%h ET (iM©m©E^M^a'J^'#- ^ C6T#9ii©#j^lcm'G,aT,&o aa'9 0%©fiJ^(CB%aLTl,^5o A^^©#%K^LT 4#/W;w# hoz:^x6U3^g^mt;(:-3aT*#giE©mm^m

-170- iv. X — J — h X t° — X : T 3 > h° a. — 7 t" 7 a > J

6 < ^f,fTbfiT&^. 4-imox 77 A-? ## jWOX— J —• h X h°—ii — fth.\^X.^")a ckA^) © "NAVLAB" ©g(imi:#l*^L\t©T& -3/:o 1 0#WX#&6CMUn#'y #m©%K%^ A&@#$©M##^T&&o #A###m#ayXTA(cW4%T#%k i#g&©##<&^#f 6C6, !#6©gMiJ^ff9C6, L =ko(c U 7;l/f-T >^>f©Ra#6 LT##%##M%&©T66o ##©NAVLAB (^5##/:-oy:^?) k"y7/<-y&3BW#©-f7/-tfaX\ l%m%m%©9 9%^@#^ffL/: V C^^gR%L/:t©C-9-^ g,# tlTU 6%]^#* 6 m^r LT%6TW%%f*#a % 6 X:E©X7%^#o^v Hc^^T(±AX%#^#6©W^0^A#<. CW^G,I)t###A©^©A# %#%X-7t %«)3o c. eraummiss

(Dt> U 7 *X- 7Xf4A^(Caltech) W?$ YU-CHONG TAI##g©#^m^K^L, m#. #4^ff ^f:o TAI##g(i. * V 7 f ;l/- %X#^-7 ly-fRC^^Tx 74 ^ ngf i^lX©I^lfl<\ A^©xX 7 nfflIS©* ^X(f©y:6(:#eL/:o %^%/:4±©^A^T77 7UX/ #*(ff:ck9T^5o Chaltecb©7X7ox7>#^(4:^ 7V-7X-A^g%g#©1991^(c^L /:o m&(±, 2OA©#0S^#L, 10&0©A#%;fn^j:? h^m^TUao Xlc. ^MX ^#©STEELE Laboratory (C^mLTW&o @^'eT'cHrX^/\7FV>7'f6/:6©m#&T^ 7 n^y —>7"tlZ^'>6o 2%l'L4^>f©7:c/\d<%X3fl6oM&©JE;%kX 4,000%7^-h"e&6o 7 f h U 77'X7^ 6i@#^P©y:6©7 V - WL-A (7 7X100) A<800f^7^ - K /< ;i/77X7o%x^ #?7vX. ?x^#]&©y:6©^v-7;i/-A (77xioooo)A< ?oof A7^-h. ^m%x. RiE. /y©/:A©@m<5oof^7^-h. cAx 7Xh©/:A©@#m000^7^-|'. ##^<1000^7^-hT&6o 1995^

-171- (?7X100-1000)d

Hd:. WT®aOT&&„ (1) y V 3 ^ o@(f7% ho (2) *<)]/$'?'( {7 oJnXJCj;^ —~f<&Wfeo (3) (4) (5) (6) ( 7) tbLBrX >1h-0#fPo (8 ) (9) ^'>'J3 >Sf*STM / AFM 7o-7©*0 -

do) 7^^o#Km^#%#%m7o-f-v70mfFo

ARPA, NSF(grants and PYI Award), NIH, AFOSR, Packard and Lucile Foundation (PackerdFellowship), HP, Seagate, Rockwell International, Medtronics, Topomatrix, JPL (President and DDF Funds), Xerox, BEI Inc. and NSIC / ARPA.

®t] V 7 f ;l/JZ 7 ;*# o # 7 if 71/ % # (UCLA) ## UCLA-etd^ KRISTFER S. J. PISTER ###. CHANG-JIN KIM###. CHIH-MING HO##.

MING C. WU###&#mLf:o UCLArid:.

>###. UCLATli.

^ o##(:EEBLT#oi).

TKRISTFER S. J. PISTER###©9f^J

SU-5 ^(76-S^E5

LT^<«

(1) (2) ^M?4'^DAnXi:ckaV-7%7v/< ( 3) 7>f 7 a MM#

(4) ,DmE^OfR#^##j0/:)6(DMEMSy7t>#- (5)

(6) (7) ##ClWS(c&#a3D7^?n7t%# ( 8 ) ##CMOS K is (j-5W3 w JPiSStf*

rcHiH-MiNG Eommom^mj

#X7V7^-;[/-T^:#(USC) ^6UCLA^#^T#/:6^9o t6t(DZtX\ fe, —A©B*A© ’M#:A^E^S/^:EW£ LT, *

(k Ao #©#AAkk ” REVIEW: MEMS and Its Application for Flow Control ( Transaction of ASMS, Journal of Fluid Engineering, Vol. 118, SEPTEMBER 1996 / 437-447 (by Chih-Ming Ho & Yu-Chong Tai) (C||L < 5F^btlT©So C ©#)t©4b##©Yu-Chong Taift(iExE©Chaltech©S*|5:gT\ :®2A^\ 7^7n-i?>1f iAS-^fc o

(1) (2) 9 n-ti A/Br A (Shear-Stress)-fe >if" —ic J:

Ztib Fo-;bf 6C i^g^lcLTU 7^3>g)3#m©^ff#^C©A)&T3>

&s %Mi-AoA —oA(t ©-7-T 7 ny ■? y/n°£ ■o 1 ATS^®©3>ha-;lt§ o-v>y$qm<,

TMING C. WU##g©W^^J #hL ELECTRO-OPTICS, OPTOELECTRONICS ©#P9^T&6AA ?^7aK#^#mLT, - %©yV3>C7j:/\(D±(c^A%7Y:^^^fF(p±(fj:7(!:LTL'6o " Micro- Optical Bench on a Chip" 60f/L%rU5o Bench ALL y-T^-Klz —f-, 7^^o7l/^;[/(Fresnel) o-^V-tf-A xYl/y -f K* 7-A

-173 ft5®-e&6o m#180mrn, mAm8#500rnnk ±#(±yV3>C7^/^(D^#^b,2 5 4mm±Ao yU3>r>^/\®±(:-7^T^0±(766, #^(:#A^d;W6®C6Tab5o ^@#®7/W%(hL T, WW?&=k

rCHANG-JIN KIM *g©E^Sj ucLA0^u->;i/-Aom^ ft U 7 tJ],—7j&'<-9 ^ UCLA®-74' ^ 6y=AC##L/:6®C6T&6o f«i:v^^ngWS^btSiOCiT\ f 9%r abo/co b7^7 + -®#&#^3^ Mis7n-fc:xt£ff\ ii®il@^1"^rgES-tiT, 7^D77^'J3 £ -e&c/co -e®/:A(c^ 7 ^ >yyy%f > u&&%%(Z)A8%mmf &«boc6-c6&o

V L/-^(UC Berkeley) #m

®V-7^ A#®^C# BiJ®#S#^-3TU6tM

[RONALD S. FEARING JttStif ®9f^Sj £ n v-1°^ U-* tfc©a<&®—o®Jj£ #!?&5 (Robotics Reseaarch, The Seventh Inteirnstional Symposium, edited by George Giralt and Gerhard Hirzinger, Springer, 1996, 570-581) o CtUiN m^®±C-7-r^03^;I/7l/^ (D6c®3^;l/®m#2. 3 ; V)

9 & t,®^ 7^?n##®ma/&Ti\ /J\$Wgga,®#m%(!:l:fGmLj:9 6f^t)®T6&o %>oD£ c, #^^AfTC^6®(j:. 7^?cD—^v- y%7A7=jb6o #^®m#c(±, fpm^#LT7K^mi^m#:mmv%7A^mwi:wao

ATi^^g-eab!)^ 4-#^%#T#6ck9ir&-3/=o

174- LROBERT J. FULL

l&lis Department of Integrative Biology £ U *5 LT& D m£LTl'&0-C\ 4'B(^W^@^5^5%##0l:y^7v'y0

T & & & 3 £ L to y+ y v t b % *9 ® y y ^y^##LTjo!9o ##(:##(:%-3f:o 4#:o^#^Ll:W'5#gRyj, l\ @/r0#*&6^#^#TW/:o

C£&fr-3T&tK coyxf-Ald:, %#MfF(Dy=)6ic7-y-i>3>^my^ %U

«fc 3 &E^£ME L;fc©7"\ Mb LVfffgicfeMfoS £ t&Tgfco

H i$ $ %)

175 mn 0 am m m 8 #12^ #1 a 90 (^) ------>7'>>h> h > #3 0 11 (*) 77yhy^ —>o#y-^;i/X# o#y-tf;i/x #4 0 12 (^:) n-fy-tf;(/x yu-chong nmmmwm UCLA (* V 7 f7k=.7;A#o-f y KRISTOFER S. J. PISTER### CHANG-JIN KIM### CHIH-MING HO## MING C. WU### #5 0 13 (iifc) #y77yy%3/<-7k4'^ RONALD S. FEARING### ROBERT J. FULL## BRH #6 0 14 (±) #y7?yy%3# —>#*/6 #7 0 15 (0) ^m#*iBEE / sicECtmmmm#!##)/ RSJ (0a;#y F^e?) ^#©”1996 International Conference on Multisensor Fusion and Integration for Intelligent System ” (%Ifb 7 X t" A © Ac di> ©H# t y # - A c (2)

a . Electronic Imaging ’ 97HS#S 1997^ 2 ft 9 0 cfc *) 140 11:7 / ') il San Jose Convention CenterlcrM $ tltz Electronic Imaging ’ 97IC 2 ^100 -lSBOieSSSnL/'Co ^©Convention Centers jolt -g>@ SPIE(The International Society for Optical Engineering) ©ZllfB'i* 5Photonics Westic^ IS&T(The Society for Image Science and Technology)^#© Electronic Irnaging*

#6o IA©#SnLtzImage Processing Methods© ^©Real-Time Imaging IId:Applications of Artificial Neural Networks in Image Processing II© 2 L IIA

])7Jl7-i A-T y — it's? (Real-Time Imaging II) fay y a >(£ Image Understanding, Tracking, Video, Real-Time Imaging System© 4 o

-i y — '7 MM (Image Understanding) yfa'y fafa^T:^ h y 7° X7 r@^j ^©IHhiXilfiUo U 7 fa 9 4 h,7 A (Real-Time Imaging System) ->XfX o tz0

d’UXACLTt)^ ###ai7;l/d’UXA©%^d:ir©^ >7')/ >d:l'7EK%

So

-177- ^/-^mac8#6A:[-a.-7;l/:t'y b9-^©fGM (Applications of Artificial Neural Networks in Image Processing II) t y * a >©#&##© 6 jo 0 c /:o Neural Networks Architecture for Image Coding Neural Image Segmentation, Compression, and Texture Classification Neural Networks for Object Recognition and Classification Optimization and Learning

#3>k°zL-^0#2#ft^#^6®(:##6^5ART (Adaptive Resonance Theory. #16*

ART* y KGrossberg Net £ k9 )i£fSA3. (2MSI©* 9 oT<^So STM (M (±A^7# (T) >/) -Po LTM (d:±T^©ma^#^ff9o C©±T©^-C. ©m^*^ ?*£#m©So < ma-^# ^tidtlTVSo ART* y K g&tilOlWJ^AE^btobtlTUS *>©-£&•£** ART©/\- KC7J:7fl:. 7;

ATTENTION AL, ORIENTING SUBSYSTEM SUBSYSTEM GAIN CONTROL DIPOLE FIELD STM Fa

STM RESET |WAVE

STM

GAIN CONTROL

INPUT PATTERN

#A3. (2)- 1EI ART*y h

178- 96x128) & 19x25 (Tzfd'L. 3x5) ART$v f- ART-emALJ:9tL/:t)0-e^'5o

®-o|±E^,i:-a-7;^y 6^9Al^T&a^ fe-5—ottdoUX®

#6^|6i-e&ao y:/d'U, ##0j:9(:. ^#Ly:mm^M^mA^#ay=A(:(j:. -3.-7 ;l/^v C@^##ART@#A, iKA®6C5K Fuzzy Logic&#AL Fuzzy ART6#LT^a (#^0^aB(iCCC&a) „ m#^(l(j:Aircraft(F18) 0# A-effjog 6t0T&-3/:o c ®

->&!WcLTi'a#m®m:aaTi#&ac6d@#j ^a. /<^->^B%LTi'a#m^#%

/-x°^- yMj<UaiiiH^ b(Dt B#fyW<£ b ®L A:

Host Computer

Structure. Level FBN L

Level

FBN feedback Network

FFN feed-forward Network 55A3. (2)-2E

179- b. 6F^f$ll5tt|olSllE —Stanford University©Professor John Koze — 2 ^ 14 84^#^ Stanford University© Professor John Koza^KPdJ Lfz0 (KozaXVb —X

©Forrest H Bennett III &I9SS tlfco ) Professor John Koza(i> »feWT/F-i'iJ X A (Genetic Algorism, GA6 #0#) ©#J%#Tab 3John Holland (University of Michigan) ©txhT PhD4r V — K$tlTV£ (B^^A

4^Jhtfb tu General Chair) o

1) GP(cj;&m#%#a©8mmiS 2) 3) #UBGPy-;t/+y h©fF6& ::m GA6GP©al4C^UTm%LXJo#A:Uo GA (Genetic Algor ithrn)&###mU= $< 6 Procedure GA: begin initialize Population P(0); evaluate P(0); t=l; repeat select P(t) from P(t-l); recombine P(t); evaluate P(t); until (terminate condition); end. l60 X-X/^XACgcf#, m^?X-;l/[J:%%#ftP(0)^b, m^(select),

(recombine) „ fFffi (evaluate) £ It ^©#ftP(l)^^l P(2), P(3), P(4)...... (?) TJl/XUXA^%0, ##

ab o EHW(Evolavable Hardware) K J; o Kfe ~o t, iSoi ^'iK tlntzo GP (Gene tic Programming) CoUT iis GA^MfWft^Xn X 7 Aid|iJ^3 Ltz fc©N <" b l'|c,So TXXc^X 4"Blk Professor John Koza^MPol GAchGPOPaKcii|2|pff2. 2. 3 bW> A3. (2)-3EI© =k 3 Atmu©&6 C

-180 5SA3. (2)-3EI GAI:ɻE

#A3. (2)-4 El

GA0#fF (recombine) £ Ltli> 3512 (crossover)^ %^^S(mutation)btl^o 35K^Ml:£6£x 3582(381 Pft2. 2. 3© =k 9 b£>$ tl5o —GP(3##A (Expression Tree)^-^^£"#"50T^ 35 E(d:mA3.(2HEI®j:9(c@B^*0-o^^x.#f#6%6o #@±(3. (30=k9l:Mx.a^ xhV>y?# a£-#-6C£f, IcmMLTl^o #i&0#^(C3L'T(3. ^0 2 £ jo () -5o

#i#^35& (Grammar)5o SitiA(Derivation Tree) ^#x.jl(3\ 35E(38l|(#2.2.30j:3%A0±-e035E£^&o John Koza0^9GP(:jo(35'/oy7A(3 ^#*(3I#T&9. ir®35Et)m^#m*±-e&g,6

(m#A?^x.(fm(ieaf)0^j) -e&o, (instance) £07 yf->/0^U##A0#g%A

-181 (instance)©#A^g,. (^2 9, #m#@©

—GP

h? hy-eff&o A- h-? p >6 (6^©jgm)

ck©#^ A©m#l'6#mA(imm©M#

& 6. A# -f > -/vi/^ij ^ -e © © ^cp^ff ^ -n ^ & o

GA3#ma -etGPc j; o a „ 1) #8iJ(D4.%6f. ^sKA&5%a

-e^^o A;i/7-A-h7h>^m^6AA7y-A(c^3#6o

2) #^d<#A*fif:y=*X ^m#m

GAir^mefF'jT&^x h v >ya<##rc&aAax -3/:o GA^m^mmum^mo#^ 8mi:GA&#m-#Afa#gd<&cao C^#^L^0^GP-eab^6U3 3) gfmcimm^^wr# &„ GATtJu fFffi(evaluate)©fc^6iifi tnessPHSf^-ffl© -S^A fitnessl9l^©iS5^GA© GP©#A, fitnessMm<^

&U:©###. WmT&5 3o f%?#Al§AT#GP6©#&##

6 B©^tAJl/^6©AA%M%Al^T©77'D-At©jtg^gT&aAA

L A# A b A A 9 o GPCJ; &##%#&© @#^8% Professor John Koza® ~Cli PGP(C A 6##%#i»© t LTs 7 A □ 711

Cfi# EHW^^L

©#%a^rGPT#@'76#n\ ffl(>\07 — &0&&Wtf~:&o Professor John Koza©

182 Power PC3fnty-f64&;(n&%:a#?ij3>t:"jL-f&mVT#LaLTV6o Forrest H Bennett III 6. 'N — K1)7 xT&MBM& LT L b ofc#*. C ©M?iJ ^ > h° x-^£ l^li7-nf->3 >6#& b&^tl&o (^ilbE%ffl©n >h0jL-^SSttStanford*Kr-td:^< > WlDg^fcEg

) V6V6#^f 6C6TGP(:"3VT#g^-WrTmL^^. rGPz)#im^#ALTV6©(j:.

AITDomain Knowledge^SA"^"-§©

©#A(ift^t LTDomain Independent£V9 -5J LTV 6d'6T&&o

l#3W:RmT6-3f:;6<, (@^#^0 6) GPK-oVT©# mt%i# (2) < 16 LT#V/cProfessor John Kozafc b0^c4‘lll©SAtiffit@$©T;£8l;£:S(;V

(jE#A# N a: m)

(1) H?I& ; VLSI-^-un>t^-^. (37A8 ) Genetic Algorithm, pp. 188-190. AMM1991) (2) J.Koza; Genetic Programming II: Automatic Discovery of Reusable Programs, MIT Press (1994)

-183- %-n 0

8 8 (±)

9 0 (0)

150 (±) -^>77 > v%3

160 (0)

-184-