Siyabonga ngokuvakashela i-Nature.com. Inguqulo yesiphequluli oyisebenzisayo inokusekelwa okulinganiselwe kwe-CSS. Ukuze uthole ulwazi olungcono kakhulu, sincoma ukuthi usebenzise isiphequluli esibuyekeziwe (noma ukhubaze i-Compatibility Mode ku-Internet Explorer). Okwamanje, ukuqinisekisa ukwesekwa okuqhubekayo, sizonikeza isayithi ngaphandle kwezitayela kanye ne-JavaScript.
I-carousel ebonisa amaslayidi amathathu ngesikhathi esisodwa. Sebenzisa izinkinobho Zangaphambilini kanye Nezilandelayo ukuze udlule kumaslayidi amathathu ngesikhathi, noma sebenzisa izinkinobho zesilayidi ekugcineni ukuze udlule kumaslayidi amathathu ngesikhathi.
Umbono ophakeme udinga ukusampula okuhle kwe-retina kanye nokuhlanganiswa ukuze kwakhiwe kabusha izakhiwo zento. Kubalulekile ukuqaphela ukuthi lapho kuhlanganiswa amasampula endawo avela ezintweni ezahlukene, ukunemba kuyalahleka. Ngakho-ke, ukuhlukaniswa, ukuqoqwa kwezindawo zesithombe ukuze kucutshungulwe okuhlukile, kubalulekile ekuboneni. Emsebenzini wangaphambilini, izakhiwo ze-lattice ezi-bistable, ezingabhekwa njengendawo eyodwa noma ngaphezulu ehambayo, zasetshenziswa ukutadisha le nqubo. Lapha, sibika ubudlelwano phakathi kwezahlulelo zomsebenzi kanye nokuhlukaniswa kwezindawo ezindaweni eziphakathi kwendlela yokubuka ye-primate. Ikakhulukazi, sithole ukuthi ama-neurons e-median temporal aqondisa ngokukhetha ayezwela izinkomba zokuthungwa ezisetshenziselwa ukuhlanekezela umbono we-bistable gratings futhi abonise ubudlelwano obukhulu phakathi kwezilingo kanye nombono we-subjective wezinto eziqhubekayo. Lokhu kuhlangana kukhulu kumayunithi abonisa ukunyakaza komhlaba wonke ngamaphethini aneziqondiso eziningi zendawo. Ngakho-ke, siphetha ngokuthi isizinda sesikhathi esiphakathi siqukethe izimpawu ezisetshenziselwa ukuhlukanisa izigcawu eziyinkimbinkimbi zibe izinto eziyingxenye kanye nobuso.
Ukubona akuncikile nje kuphela ekuhlukaniseni okunembile kwezimpawu zesithombe eziyisisekelo njengokuqondiswa komphetho kanye nesivinini, kodwa okubaluleke kakhulu ekuhlanganisweni okufanele kwalezi zici ukuze kubalwe izakhiwo zemvelo ezifana nesimo sento kanye nendlela elandelwayo1. Kodwa-ke, izinkinga ziyavela lapho izithombe ze-retina zisekela amaqembu amaningana ezici anengqondo ngokulinganayo 2, 3, 4 (Isithombe 1a). Isibonelo, lapho amasethi amabili ezibonakaliso zesivinini esondelene kakhulu, lokhu kungahunyushwa ngokunengqondo njengento eyodwa ehambayo noma izinto eziningana ezihambayo (Isithombe 1b). Lokhu kubonisa isimo somuntu siqu sokuhlukanisa, okungukuthi akuyona impahla eqondile yesithombe, kodwa inqubo yokuhumusha. Naphezu kokubaluleka kwayo okusobala kokubona okuvamile, ukuqonda kwethu isisekelo se-neural sokuhlukanisa okubonakalayo kusalokhu kungaphelele.
Umfanekiso wekhathuni wenkinga yokuhlukanisa umbono. Umbono wombukeli ngokujula ku-Necker cube (kwesobunxele) uyashintshana phakathi kwezincazelo ezimbili ezingaba khona (kwesokudla). Lokhu kungenxa yokuthi azikho izimpawu esithombeni ezivumela ubuchopho ukuthi bunqume ngokukhethekile ukuma kwe-3D komuntu (okunikezwa yisignali yokuvaleka kwe-monocular ngakwesokudla). b Uma izimpawu eziningi zokunyakaza zethulwa eduze kwendawo, uhlelo lokubona kumele lunqume ukuthi amasampula endawo avela entweni eyodwa noma ngaphezulu. Ukungacaci okungokwemvelo kwezibonakaliso zokunyakaza endawo, okungukuthi ukulandelana kwezinyathelo zezinto kungakhiqiza ukunyakaza okufanayo kwendawo, okuholela ekuchazeni okuningi okunengqondo kokufakwayo okubonakalayo, okungukuthi amasimu e-vector lapha angavela ekunyakazeni okuhlangene kwendawo eyodwa noma ekunyakazeni okusobala kwezindawo ezigqagqene. c (kwesobunxele) Isibonelo sesikhuthazo sethu segridi esinombhalo. Ama-grating angunxande azulazula ngokuqondile esiqondisweni sawo ("iziqondiso zezingxenye" ​​- imicibisholo emhlophe) ayagqagqana ukuze akhe iphethini yokugqagqana. I-lattice ingabonakala njengokunyakaza okukodwa, okuvamile, okuxhunyiwe kweziqondiso (imicibisholo ebomvu) noma ukunyakaza okusobala kweziqondiso ezihlanganisiwe. Umbono we-lattice uphambukiswa ngokufakwa kwezinkomba zokuthungwa kwephuzu okungahleliwe. (Phakathi) Indawo egqanyisiwe ngombala ophuzi iyandiswa futhi iboniswe njengochungechunge lwamafreyimu ezibonakaliso ezihambisanayo nezibonakalayo, ngokulandelana. Ukunyakaza kwechashazi esimweni ngasinye kumelelwa imicibisholo eluhlaza okwesibhakabhaka nebomvu. (Kwesokudla) Igrafu yesikhundla (x, y) sephuzu lokukhetha uma kuqhathaniswa nenani lamafreyimu. Esimweni esihambisanayo, zonke izinhlobo zizulazula ngendlela efanayo. Esimweni sokucaca, ukuthungwa kuhamba ngendlela yengxenye. d Umfanekiso wekhathuni womsebenzi wethu wokuhlukanisa ukunyakaza. Izinkawu zaqala isivivinyo ngasinye ngokulungisa ichashazi elincane. Ngemva kokulibaziseka okufushane, uhlobo oluthile lwephethini yokugoba (ukuhambisana/ukucaca) kanye nosayizi wesignali yokugoba (isib. ukuqhathanisa) kwavela endaweni ye-MT RF. Ngesikhathi sokuhlolwa ngakunye, i-grating ingazulazula kwenye yezindlela ezimbili ezingaba khona zephethini. Ngemva kokuhoxiswa kwe-stimulus, izinhloso zokukhetha zavela ngaphezulu nangaphansi kwe-MT RF. Izinkawu kumele zibonise umbono wazo wegridi kuma-saccades ku-target yokukhetha efanele.
Ukucutshungulwa kokunyakaza okubonakalayo kuchazwe kahle futhi ngaleyo ndlela kunikeza imodeli enhle kakhulu yokufunda izifunda ze-neural zokuhlukanisa ukuqonda. Izifundo eziningana zokubala ziphawule ukusetshenziswa kwamamodeli okucubungula ukunyakaza kwezigaba ezimbili lapho ukulinganisa kokuqala okunesisombululo esiphezulu kulandelwa ukuhlanganiswa okukhethiwe kwamasampula endawo ukususa umsindo nokubuyisela ijubane lezinto7,8. Kubalulekile ukuqaphela ukuthi izinhlelo zokubona kumele ziqaphele ukukhawulela lokhu kuhlanganiswa kulawo masampula endawo avela ezintweni ezijwayelekile. Izifundo ze-Psychophysical zichaze izici zomzimba ezithonya indlela izimpawu zokunyakaza endawo ezihlukaniswa ngayo, kodwa ukuma kwezindlela zokwakheka kanye namakhodi e-neural kuhlala kunemibuzo evulekile. Imibiko eminingi isikisela ukuthi amaseli akhetha ukuqondiswa esifundeni se-temporal (MT) se-primate cortex angabafakizicelo be-neural substrates.
Okubalulekile, kulezi zivivinyo zangaphambilini, izinguquko emisebenzini yezinzwa zihlobene nezinguquko zomzimba ezishukumisweni ezibonakalayo. Kodwa-ke, njengoba kushiwo ngenhla, ukuhlukaniswa kwezingxenye empeleni kuyinqubo yokuqonda. Ngakho-ke, ukutadisha isisekelo sayo sezinzwa kudinga ukuxhumanisa izinguquko emsebenzini wezinzwa nezinguquko ekuboneni izishukumi ezihleliwe. Ngakho-ke, siqeqeshe izinkawu ezimbili ukuthi zibike ukuthi iphethini ye-bistable grating ebonwayo eyakhiwe yi-superimposed drifting rectangular gratings yayiyi-surface eyodwa noma izindawo ezimbili ezizimele. Ukuze sifunde ubudlelwano phakathi komsebenzi wezinzwa kanye nezahlulelo zokuhlukaniswa kwezingxenye, siqophe umsebenzi owodwa ku-MT lapho izinkawu zenza lo msebenzi.
Sithole ubudlelwano obubalulekile phakathi kocwaningo lomsebenzi we-MT kanye nokuqonda. Lokhu kuxhumana kwakukhona noma ngabe izisusa zaziqukethe izinkomba zokuhlukanisa ezisobala noma cha. Ngaphezu kwalokho, amandla alo mphumela ahlobene nokuzwela kwezibonakaliso zokuhlukanisa, kanye nenkomba yephethini. Lokhu kokugcina kulinganisa izinga lapho iyunithi ikhipha khona ukukhanya komhlaba wonke kunokunyakaza kwendawo ngamaphethini ayinkimbinkimbi. Nakuba ukukhetha isiqondiso semfashini sekuyisikhathi eside kuqashelwa njengesici esichazayo se-MT, futhi amaseli akhetha imfashini abonisa ukulungiswa kwezisusa eziyinkimbinkimbi ezihambisana nombono womuntu walezo zisusa, ngokwazi kwethu konke, lobu ubufakazi bokuqala bokuxhumana phakathi kwamaphethini, inkomba kanye nokuhlukanisa okubonakalayo.
Siqeqeshe izinkawu ezimbili ukuze zibonise umbono wazo we-drifting grid stimuli (ukunyakaza okuvumelanayo noma okusobala). Ababukeli abangabantu bavame ukubona lezi stimuli njengokunyakaza okuvumelanayo noma okusobala okucishe kube yimvamisa efanayo. Ukuze sinikeze impendulo efanele kulolu vivinyo futhi sibeke isisekelo somvuzo osebenzayo, sidale izimpawu zokuhlukanisa ngokubhala i-raster yengxenye eyakha i-lattice (Umfanekiso 1c, d). Ngaphansi kwezimo ezihambisanayo, zonke izindwangu zihamba ngendlela yephethini (Umfanekiso 1c, “okuvumelanayo”). Esimweni esibonakalayo, ukuthungwa kuhamba ngendlela eqondile kuya endleleni ye-grating lapho kubekwe khona (Umfanekiso 1c, “okusobala”). Silawula ubunzima bomsebenzi ngokushintsha umehluko waleli lebula lokuthungwa. Ezivivinyweni ezicushiwe, izinkawu zavuzwa ngezimpendulo ezihambisana nezindwangu zokuthungwa, futhi imivuzo yanikezwa ngokungahleliwe (amathuba angu-50/50) ezivivinyweni eziqukethe amaphethini angenazo izindwangu zokuthungwa (isimo sokungafani kwendwangu esingenalutho).
Idatha yokuziphatha evela ezivivinyweni ezimbili ezimele iboniswe kuMfanekiso 2a, futhi izimpendulo zidwetshwa njengengxenye yokwahlulela kokuvumelana uma kuqhathaniswa nokungafani kwezimpawu zokuthungwa (ukungafani kokubonakala kuthathwa njengokubi ngencazelo) kwamaphethini ashintsha phezulu noma phansi, ngokulandelana. Sekukonke, ukuqonda kwezinkawu ngokuhambisana/ukucaca kwathinteka ngokuthembekile kokubili uphawu (olubonakalayo, oluhambisanayo) kanye namandla (umehluko) we-texture cue (ANOVA; inkawu N: isiqondiso – F = 0.58, p = 0.45, uphawu – F = 1248, p < 10−10, umehluko – F = 22.63, p < 10;−10 inkawu S: isiqondiso – F = 0.41, p = 0.52, uphawu – F = 2876.7, p < 10−10, umehluko – F = 36.5, p < 10−10). Sekukonke, umbono wezinkawu ngokuhambisana/ukucaca uthinteke ngokuthembekile kokubili uphawu (olubonakalayo, oluhambisanayo) kanye namandla (umehluko) we-texture cue (ANOVA; inkawu N: isiqondiso – F = 0.58, p = 0.45, uphawu – F = 1248, p < 10−10, umehluko – F = 22.63, p < 10; −10 inkawu S: isiqondiso – F = 0.41, p = 0.52, uphawu – F = 2876.7, p < 10−10, umehluko – F = 36.5, p < 10−10). В целом на восприятие обезьянами когерентности/прозрачности достоверно влияли как знак (прозрачность, когерентность), так истоверн (ANOVA; обезьяна N: направление — F = 0,58, p = 0,45, знак — F = 1248, p < 10−10, контраст – F = 22,63, p < 10; −10 обние, зера = F p = 0,52, признак – F = 2876,7, p <10−10, контраст - F = 36,5, р <10-10). Ngokuvamile, ukuqonda ukuhambisana/ukucaca kwezinkawu kuthinteke kakhulu kokubili uphawu (ukucaca, ukuhambisana) kanye namandla (ukungafani) kwesici sombhalo (i-ANOVA; inkawu N: isiqondiso — F = 0.58, p = 0.45, uphawu — F = 1248, p < 10−10, ukuqhathanisa – F = 22.63, p < 10; -10 inkawu S: isiqondiso – F = 0.41, p = 0.52, uphawu – F = 2876.7, p < 10 −10, ukuqhathanisa – F = 36.5, p < 10-10).总体而言，猴子对连贯性/透明度的感知受到纹理提示（ANOVA）的符号（透明、连贯）和强度（对比度）的可靠影响；猴子N：方向- F = 0.58,p = 0.45,符号- F = 1248, p <10−10, 对比度– F = 22.63, p <10;−10 猴子S: 方向– F = 0.41, p = 0.41, p 0.52, 符号– F = 2876.7, p <10−10, 对比度– F = 36.5,p <10-10）.总体而言，猴子对连贯性/透明度的感知受到纹理提示（ANOVA）的符号（透明、连贯）和强度（对比度）的可靠影响；猴子N：方向- F = 0.58,p = 0.45,符号- F = 1248, p <10−10, 对比度– F = 22.63, p <10;−10 36.5,p <10-10）.Ngokuvamile, umbono wokuhambisana/ukucaca kwenkawu uthintwe kakhulu uphawu (ukucaca, ukubumbana) kanye nokuqina (umehluko) kwezimpawu zokuthungwa (i-ANOVA);обезьяна N: ориентация – F = 0,58, p = 0,45, знак – F = 1248, p < 10−10, Контрастность — F = 22,63, p <10; inkawu N: ukuqondiswa – F = 0.58, p = 0.45, uphawu – F = 1248, p < 10−10, Ukuqhathanisa – F = 22.63, p < 10; −10 Обезьяна S: Ориентация — F = 0,41, p = 0,52, Знак — F = 2876,7, p < 10−10, Контрастность — F = 36,5, p < 10-10). −10 Inkawu S: Ukuqondiswa – F = 0.41, p = 0.52, Uphawu – F = 2876.7, p < 10-10, Ukuqhathanisa – F = 36.5, p < 10-10).Imisebenzi yokuqongelela ye-Gaussian yafakwa kudatha evela kuseshini ngayinye ukuze ichaze izici zengqondo ze-psychic zezinkawu. Ku-fig. 2b kukhombisa ukusatshalaliswa kokuvumelana kwala mamodeli kuzo zonke izikhathi zezinkawu zombili. Sekukonke, izinkawu ziqede umsebenzi ngokunembile nangokungaguquguquki, futhi senqabe ngaphansi kuka-13% wezikhathi zezinkawu ezimbili ngenxa yokungalingani kahle nemodeli ye-Gaussian eqongelela.
a Izibonelo zokuziphatha zezinkawu ezimisweni ezimele (n ​​≥ izilingo ezingu-20 ngesimo ngasinye sokushukumisa). Kumaphaneli angakwesobunxele (kwesokudla), idatha evela kwiseshini eyodwa yezinkawu ze-N(S) idwetshwe njengamaphuzu okukhetha ahambisanayo (ordinate) uma kuqhathaniswa nomehluko wesibonakaliso sezimpawu zokuthungwa (abscissa). Lapha kucatshangwa ukuthi ukuthungwa okusobala (okuhlangene) kunamanani angemahle (amahle). Izimpendulo zakhiwe ngokwehlukana ngokuya ngokuqondisa kokunyakaza kwephethini (phezulu (90°) noma phansi (270°)) esivivinyweni. Kuzo zombili izilwane, ukusebenza, kungakhathaliseki ukuthi impendulo ihlukaniswe ngokuqhathanisa okungu-50/50 (PSE - imicibisholo eqinile) noma inani lokuqhathanisa okubhaliwe okudingekayo ukusekela izinga elithile lokusebenza (umphetho - imicibisholo evulekile), ikulezi zinkomba zokukhukhuleka. b I-histogram efakiwe yamanani e-R2 ​​yomsebenzi wokuqongelela we-Gaussian. Idatha ye-Monkey S(N) iboniswa ngakwesobunxele (kwesokudla). c (Phezulu) I-PSE elinganisiwe yegridi eshintshiwe phansi (i-ordinate) uma iqhathaniswa ne-PSE eshintshiwe phezulu kwegridi (i-abscissa) edwetshwe, enemiphetho emele ukusatshalaliswa kwe-PSE kwesimo ngasinye kanye nemicibisholo ekhombisa isilinganiso sesimo ngasinye. Idatha yazo zonke izikhathi zenkawu ze-N(S) inikezwe kukholomu yesobunxele (kwesokudla). (Ngezansi) Inqubo efanayo nedatha ye-PSE, kodwa yemingcele yesici efakiwe. Kwakungekho mehluko obalulekile emikhawulweni ye-PSE noma ezimfashinini zemfashini (bheka umbhalo). I-d PSE kanye nomthambeka (i-ordinate) zidwetshwe kuye ngokuqondiswa kwe-raster okujwayelekile kwengxenye yokuhlukaniswa kwe-angular ("i-integral grating angle" - i-abscissa). Imibuthano evulekile iyizindlela, umugqa oqinile uyimodeli yokubuyela emuva engcono kakhulu, kanti umugqa onamachashazi uyi-95% yokuqiniseka kwemodeli yokubuyela emuva. Kukhona ubudlelwano obubalulekile phakathi kwe-PSE kanye ne-angle yokuhlanganisa ejwayelekile, kodwa hhayi i-angle yokuhlanganisa ejwayelekile, okuphakamisa ukuthi umsebenzi we-psychometric uyashintsha njengoba i-angle ihlukanisa ama-lattice engxenye, kodwa hhayi ukulola noma ukucaba. (Izinkathi ze-Monkey N, n = 32; I-Monkey S, n = 43). Kuwo wonke amaphaneli, imigoqo yamaphutha imelela iphutha elijwayelekile lesilinganiso. Haha. Ukuhambisana, amaphuzu okulingana angaphansi kwe-PSE, ukumiswa okujwayelekile.
Njengoba kuphawuliwe ngenhla, kokubili umehluko wezimpawu zokuthungwa kanye nesiqondiso sokunyakaza kwephethini kwahlukahluka kuzo zonke izilingo, kanye nezisusa ezikhuphuka noma eziphansi esivivinyweni esithile. Lokhu kwenzelwa ukunciphisa imiphumela ye-psychophysical11 kanye ne-neuronal28 adaptive. Ukuqondiswa kwephethini ngokumelene nokubandlulula (iphuzu lokulingana elizimele noma i-PSE) (ukuhlolwa kwesamba sezinga le-Wilcoxon; inkawu N: z = 0.25, p = 0.8; inkawu S: z = 0.86, p = 0.39) noma umkhawulo womsebenzi ofakiwe (isamba sezinga le-Wilcoxon; inkawu N: z = 0.14, p = 0.89, inkawu S: z = 0.49, p = 0.62) (Isithombe 2c). Ngaphezu kwalokho, bekungekho mehluko obalulekile phakathi kwezinkawu ngezinga lokungafani kokuthungwa okudingekayo ukuze kulondolozwe amazinga omkhawulo wokusebenza (inkawu engu-N = 24.5% ± 3.9%, inkawu engu-S = 18.9% ± 1.9%; isamba sezinga likaWilcoxon, z = 1.01, p = 0.31).
Esifundweni ngasinye, sishintshe i-engeli ye-interlattice ehlukanisa ukuqondiswa kwe-lattice yezingxenye. Izifundo ze-Psychophysical zikhombisile ukuthi abantu banamathuba amaningi okubona iseli 10 njengelixhunyiwe uma le engeli incane. Uma izinkawu bezibika ngokuthembekile umbono wazo wokuhambisana/ukucaca, khona-ke ngokusekelwe kulezi zitholelo, umuntu angalindela i-PSE, umehluko wokuthungwa ohambisana nokuhlukaniswa okufanayo phakathi kokuhambisana nokukhetha kokubonakala, ukuthi ikhuphuke lapho kuxhunyanwa. Lokhu kwakunjalo ngempela (Isithombe 2d; ukugoqeka kuzo zonke izikhombisi-ndlela zephethini, iKruskal–Wallis; inkawu N: χ2 = 23.06, p < 10−3; inkawu S: χ2 = 22.22, p < 10−3; ubudlelwano phakathi kwe-angle ejwayelekile ye-inter-grating kanye ne-PSE – inkawu N: r = 0.67, p < 10−9; inkawu S: r = 0.76, p < 10−13). Lokhu kwakunjalo ngempela (Isithombe 2d; ukugoqeka kuzo zonke izikhombisi-ndlela zephethini, iKruskal–Wallis; inkawu N: χ2 = 23.06, p < 10−3; inkawu S: χ2 = 22.22, p < 10−3; ubudlelwano phakathi kwe-angle ejwayelekile yokuhlanganisa kanye ne-PSE – inkawu N: r = 0.67, p < 10−9; inkawu S: r = 0.76, p < 10−13). Это действительно имело место (рис. 2d; коллапс поперек направления паттерна, Крускал-Уоллис; обезьяна N: χ2 = 23,06, p < 23,06, p <10:–3; 22,22, p < 10–3; корреляция между нормализованными угол решетки и PSE – обезьяна N: r = 0,67, p < 10-9, обезьяна S, 70, 3 = . Lokhu kwenzeke ngempela (Isithombe 2d; ukugoqa ngaphesheya kwephethini, iKruskal-Wallis; inkawu N: χ2 = 23.06, p < 10–3; inkawu S: χ2 = 22.22, p < 10–3; ukuhlangana phakathi kwe-angle ye-lattice ejwayelekile kanye ne-PSE – inkawu N: r = 0.67, p < 10-9, inkawu S: r = 0.76, p < 10-13).情况确实如此（图2d；跨模式方向折叠，Kruskal-Wallis；猴子N:χ2 = 23.06,p <10-3猴子S:χ2=p2. 10-3；标准化间光栅角和PSE – 猴子N：r = 0.67,p <10-9；猴子S：r = 0.76,p <10-13）.情况 确实 如此 （图 图 2D ； 方向 折叠 , kruskal-wallis ; n : :2 = 23.06 , p <10-3 ; <10-3 ; <2.2 - 2 = 2 - 2 = 12 = 2光栅角 和 pse-猴子 猴子 猴子 猴子 猴子N：r = 0.67,p <10-9；猴子S：r = 0.76，p <10-13）. Это действительно имело место (рис. 2d; кратность по оси моды, Крускал-Уоллис; обезьяна N: χ2 = 23,06, p <10-3; обез,2:2, S 10-3 нормализованный межрешеточный угол). Lokhu kwakunjalo ngempela (Isithombe 2d; phinda nge-axis yemodi, iKruskal-Wallis; inkawu N: χ2 = 23.06, p < 10-3; inkawu S: χ2 = 22.22, p < 10-3; ikona elijwayelekile le-interlattice). PSE-обезьяна N: r = 0,67, p < 10–9, обезьяна S: r = 0,76, p < 10–13). Inkawu ye-PSE N: r = 0.67, p < 10–9, inkawu S: r = 0.76, p < 10–13).Ngokuphambene nalokho, ukushintsha i-engeli ye-interlattice akuzange kube nomthelela obalulekile emthambekeni womsebenzi we-psychometric (Isithombe 2d; ukugoqwa kokuqondisa okuphambene, iKruskal-Wallis; inkawu N: χ2 = 8.09, p = 0.23; inkawu S χ2 = 3.18, p = 0.67, ukuhlangana phakathi kwe-engeli ye-interlattice evamile kanye nomthambeko – inkawu N: r = -0.4, p = 0.2, inkawu S: r = 0.03, p = 0.76). Ngakho-ke, ngokuvumelana nedatha ye-psychophysical yomuntu, umphumela ojwayelekile wokushintsha i-engeli phakathi kwama-gratings ukushintsha kwamaphuzu okufuduka, hhayi ukwanda noma ukwehla kokuzwela kumasignali okuhlukaniswa.
Okokugcina, imivuzo inikezwa ngokungahleliwe ngamathuba angu-0.5 ezivivinyweni ezine-zero texture contrast. Uma zonke izinkawu bezikwazi lokhu kungahleliwe okuyingqayizivele futhi zikwazi ukuhlukanisa phakathi kwe-zero texture contrast kanye ne-cue stimuli, zingathuthukisa amasu ahlukene ezinhlobo ezimbili zezilingo. Ukubuka okubili kusikisela ngokuqinile ukuthi akunjalo. Okokuqala, ukushintsha i-grating angle kube nemiphumela efanayo ngokwekhwalithi kumaphuzu e-cue kanye ne-zero texture contrast (Isithombe 2d kanye ne-Supplementary Fig. 1). Okwesibili, kuzo zombili izinkawu, ukukhetha kwe-bistable trial akunakwenzeka ukuthi kube ukuphindaphindwa kokukhethwa kwemivuzo kwakamuva (kwangaphambilini) (ukuhlolwa kwe-binomial, izinkawu ze-N: 0.52, z = 0.74, p = 0.22; Izinkawu ze-S: 0.51, r = 0.9, p = 0.18).
Ekuphetheni, ukuziphatha kwezinkawu emsebenzini wethu wokuhlukanisa kwakungaphansi kokulawulwa okuhle kokukhuthaza. Ukuthembela kwezahlulelo zokuqonda esibonakalisweni nasesayizini wezimpawu zokuthungwa, kanye nezinguquko ku-PSE nge-grating angle, kubonisa ukuthi izinkawu zibike umbono wazo wokuqondana/ukucaca kwezimoto. Okokugcina, izimpendulo zezinkawu ezivivinyweni zokungafani kokuthungwa eziyi-zero azizange zithinteke emlandweni womvuzo wezilingo zangaphambilini futhi zathinteka kakhulu yizinguquko ze-angular eziphakathi kwe-raster. Lokhu kusikisela ukuthi izinkawu ziyaqhubeka nokubika umbono wazo wokuqonda ngokucushwa kobuso be-lattice ngaphansi kwalesi simo esibalulekile.
Njengoba kushiwo ngenhla, ukuguquka kokungafani kokwakheka kusuka kokubi kuya kokuhle kufana nokuguquka kokubona kwezinto ezishukumisayo kusuka kokusobala kuya kokuvumelana. Ngokuvamile, kuseli elithile, impendulo ye-MT ivame ukwanda noma ukwehla njengoba ukungafani kokwakheka kushintsha kusuka kokubi kuya kokuhle, futhi isiqondiso salo mphumela ngokuvamile sincike ekuqondeni kokunyakaza kwephethini/ingxenye. Isibonelo, ama-directional tuning curves amaseli amabili e-MT amele aboniswe kuMfanekiso 3 kanye nezimpendulo zala maseli kuma-grating aqukethe izimpawu zokuthungwa ezicacile noma ezicacile eziphansi noma eziphakeme. Sizame ngandlela thile ukulinganisa kangcono lezi zimpendulo zegridi, ezingase zihlobane nokusebenza kwengqondo kwezinkawu zethu.
Isakhiwo se-polar sejika lokujika eliqondisayo leseli le-MT lenkawu elimele i-S ngokuphendula kuhlu olulodwa lwe-sinusoidal. I-engeli ikhombisa indlela yokunyakaza kwe-grating, ubukhulu bubonisa ukuphuma, kanti indlela yeseli ekhethwayo idlula cishe ngo-90° (phezulu) nesiqondiso senye yezingxenye ohlangothini lwephethini ye-grating. b I-histogram yesikhathi se-stimulus-time (PSTH) yamasonto onke yegridi yokuphendula, ishintshelwe ohlangothini lwethempulethi ngo-90° (eboniswe ngokwesikimu kwesobunxele) yeseli eboniswe ku-a. Izimpendulo zihlelwa ngohlobo lwe-texture hint (iphaneli ehlangene/ebonakalayo - ephakathi/kwesokudla ngokulandelana) kanye ne-Michelson contrast (i-PSTH color hint). Kuboniswa imizamo efanele kuphela yohlobo ngalunye lwezimpawu zokuthungwa eziphansi neziphezulu. Amaseli asabela kangcono kumaphethini e-lattice akhuphukayo anezinkomba zokuthungwa ezisobala, futhi impendulo kula maphethini yanda ngokwanda kokungafani kokuthungwa. c, d yizimiso ezifanayo njengaku-a no-b, kodwa kumaseli e-MT ngaphandle kwenkawu S, ukuqondiswa kwawo okukhethwayo cishe kudlula okwegridi yokuthungwa ehlayo. Le yunithi ikhetha ama-grating adonsela phansi anezimpawu zokuthungwa ezihambisanayo, futhi impendulo kula maphethini iyanda ngokwanda kokungafani kokuthungwa. Kuwo wonke amaphaneli, indawo enomthunzi imelela iphutha elijwayelekile lesilinganiso. Ama-Spokes. Ama-Spikes, amasekhondi. okwesibili.
Ukuze sihlole ubudlelwano phakathi kokucushwa kobuso be-lattice (okubumbene noma okusobala) njengoba kuboniswe yizimpawu zethu zokuthungwa kanye nomsebenzi we-MT, siqale sahlehlisa ubudlelwano phakathi kwamaseli ukuze kube nokunyakaza okubumbene (ukuthambeka okuhle) noma ukunyakaza okusobala (ukuthambeka okungekuhle) ngokuhlehla. okunikezwe ukuhlukanisa amaseli ngesilinganiso sokuphendula kwesibonakaliso uma kuqhathaniswa nokuqhathanisa (ngokwahlukana kwesiqondiso ngasinye semodi). Izibonelo zala ma-curve okulungisa i-lattice avela kuseli lesibonelo elifanayo kuMfanekiso 3 ziboniswe kuMfanekiso 4a. Ngemuva kokuhlukanisa, sisebenzise ukuhlaziywa kokusebenza kokwamukela (ROC) ukuze silinganise ukuzwela kweseli ngalinye ekuguqulweni kwezibonakaliso zokuthungwa (bheka Izindlela). Imisebenzi ye-neurometric etholwe ngale ndlela ingaqhathaniswa ngqo nezici zengqondo zenkawu esimisweni esifanayo ukuze siqhathanise ngqo ukuzwela kwengqondo kwe-neurons nezimpawu ze-lattice. Senze ukuhlaziya okubili kokuthola izimpawu kuwo wonke amayunithi kusampula, sibala izici ze-neurometric ezihlukene zesiqondiso ngasinye sephethini (futhi, phezulu noma phansi). Kubalulekile ukuqaphela ukuthi, kulokhu kuhlaziya, sifake kuphela izilingo lapho (i) izinto ezishukumisayo zaziqukethe i-texture cue futhi (ii) izinkawu zasabela ngokuvumelana naleyo quote (okungukuthi, izilingo "ezilungile").
Amazinga omlilo adwetshwa ngokuqhathanisa uphawu lokuthungwa, ngokulandelana, ukuze kube nokuguguleka okushintshela phezulu (kwesobunxele) noma phansi (kwesokudla), umugqa oqinile umelela ukuhlehla okuqondile okuhambisana kahle, futhi idatha emugqeni ophezulu (ophansi) ithathwe kulokho okuboniswe ku-Fig. . Rice. 3a Cell, b (Fig. 3c, d). Izici ze-Regression slope zisetshenziswe ukunikeza izinkomba zokuthungwa ezikhethwayo (ezihambisanayo/ezibonakalayo) kunhlanganisela ngayinye yokuqondiswa kweseli/i-lattice (n ≥ izivivinyo ezingu-20 ngesimo ngasinye sokushukumisa). Amabha ephutha amelela ukuphambuka okujwayelekile kwesilinganiso. Imisebenzi ye-neurometric yamayunithi aboniswe ku-ba ichazwe kanye nemisebenzi ye-psychometric eqoqwe ngesikhathi esifanayo. Manje, ngesici ngasinye, sihlela ukukhetha kwe-tooltip ekhethwayo (ordinate) (bheka umbhalo) njengephesenti lokwehluka kwe-signip ye-texture (abscissa). Ukuhluka kwe-texture kushintshiwe ukuze i-tooltip ekhethwayo ibe mihle futhi i-tooltip engenalutho ibe mibi. Idatha evela kumagridi akhuphukayo akhuphukayo (aya phansi) iboniswa kumaphaneli angakwesobunxele (kwesokudla), emigqeni ephezulu (engezansi) - idatha evela kumaseli aboniswe ku-Fig. 3a,b (Fig. 3c,d). Izilinganiso zomkhawulo we-neurometric kanye ne-psychometric (N/P) ziboniswa kuphaneli ngayinye. Ama-Spokes. Ama-Spikes, amasekhondi. isekhondi, isiqondisi. isiqondiso, isifundazwe esithandwayo, i-psi. I-Psychometry, i-Neurology.
Ama-curve okulungisa i-lattice kanye nemisebenzi ye-neurometric yamaseli amabili e-MT amele kanye nemisebenzi yawo ye-psychometric ehlobene, ehlanganiswe kanye nalezi zimpendulo, aboniswa kumaphaneli aphezulu naphansi e-Figure 4a, b, ngokulandelana. Lawa maseli abonisa ukwanda noma ukwehla okucishe kube yinto eyisicefe njengoba i-tint yokuthungwa isuka ekukhanyeni iye ekuhlanganeni. Ngaphezu kwalokho, isiqondiso namandla alesi sibopho kuncike esiqondisweni sokunyakaza kwe-lattice. Okokugcina, imisebenzi ye-neurometric ebalwe kusukela ezimpendulweni zala maseli isondele kuphela (kodwa ayizange ihambisane) nezakhiwo zengqondo zokunyakaza kwegridi okuqondiswe ngakunye. Yomibili imisebenzi ye-neurometric kanye neye-psychometric ifingqiwe ngemingcele, okungukuthi ihambelana cishe no-84% wokungafani okukhethwe kahle (okuhambisana nesilinganiso + 1 sd somsebenzi we-Gaussian ohlanganisiwe ofakiwe). Kuyo yonke isampula, isilinganiso se-N/P, isilinganiso somkhawulo we-neurometric kuya kulowo we-psychometric, isilinganiso esingu-12.4 ± 1.2 ku-monkey N kanye no-15.9 ± 1.8 ku-monkey S, kanye nokuthi i-lattice ihambe okungenani ohlangothini olulodwa, kuphela Kumayunithi angu-~16% (18). %) avela ku-monkey N (monkey S) (Isithombe 5a). Kusukela esibonelweni seseli esiboniswe esithombeni. Njengoba kubonwe kuZithombe 3 no-4, ukuzwela kwama-neurons kungathinteka ubuhlobo phakathi kokuqondiswa okuthandwayo kweseli kanye nokuqondiswa kokunyakaza kwe-lattice okusetshenziswe ekuhlolweni. Ikakhulukazi, ama-curve okulungiswa kokuqondiswa ku-Fig. 3a, c abonisa ubudlelwano phakathi kokusethwa kokuqondiswa kwe-neuron kwe-sinusoidal array eyodwa kanye nokuzwela kwayo ekunyakazeni okusobala/okuhambisanayo ku-array yethu e-textured. Lokhu bekunjalo kuzo zombili izinkawu (i-ANOVA; izikhombisi-ndlela ezikhethwayo ezifakwe ku-10° resolution; inkawu N: F = 2.12, p < 0.01; inkawu S: F = 2.01, p < 0.01). Lokhu bekunjalo kuzo zombili izinkawu (i-ANOVA; izikhombisi-ndlela ezikhethwayo ezifakwe ku-10° resolution; inkawu N: F = 2.12, p < 0.01; inkawu S: F = 2.01, p < 0.01). Это имело место для обеих обезьян (ANOVA; относительные предпочтительные направления объединены в группы с разрешением, 10°; p <0,01; обезьяна S: F = 2,01, p <0,01). Lokhu bekunjalo kuzo zombili izinkawu (i-ANOVA; izikhombisi-ndlela ezikhethwayo eziqoqwe ngesisombululo esingu-10°; inkawu N: F=2.12, p<0.01; inkawu S: F=2.01, p<0.01).两只猴子都是這种情况（方差分析；以10° 分辨率合并的相对首选方向；猴子N：F = 2.10,S = 2.10,p = 10; 0.01).两 只 猴子 都 是 這 种 （方差 分析 以 以 10 ° 分辨率 合并 的 相对 方向 ； 猴子 n: : f = 2 ； f = 2 ; f = 20. 2.01, p <0.01.....".. Это имело место для обеих обезьян (ANOVA; относительная предпочтительная ориентация объединена при разрешении 10°; обезрешении 10°; обезREшении 10°; обезREшении 10° обезьяна S: F = 2,01, p <0,01). Lokhu bekunjalo kuzo zombili izinkawu (i-ANOVA; ukuqondiswa okuthandwayo okuhlobene okuhlanganiswe kusisombululo esingu-10°; inkawu N: F=2.12, p<0.01; inkawu S: F=2.01, p<0.01).Njengoba kunezinga elikhulu lokuguquguquka kokuzwela kwama-neuron (Isithombe 5a), ukuze sibone ngeso lengqondo ukuncika kokuzwela kwama-neuron ekuqondisweni okuthandwayo okuhlobene, siqale salungisa ukuqondiswa okuthandwayo kweseli ngalinye kube ukuqondiswa "okungcono kakhulu" kokunyakaza kwephethini yegridi (okungukuthi isiqondiso). lapho i-grating yakha i-engeli encane kakhulu phakathi kokuqondiswa kweseli okuthandwayo kanye nokuqondiswa kwephethini ye-grating). Sithole ukuthi imingcele ehlobene yama-neurons (umkhawulo wokuqondiswa kwe-lattice/umkhawulo “okubi kakhulu” wokuqondiswa kwe-lattice “okungcono kakhulu”) yahlukahluka ngalokhu kuqondiswa okuthandwayo okujwayelekile, kanye neziqongo kulesi silinganiso somkhawulo ezenzeka eduze kwamaphethini noma ukuqondiswa kwezingxenye (Isithombe 5b). )). Lo mphumela awukwazanga ukuchazwa ngokubandlulula ekusatshalalisweni kweziqondiso ezikhethwayo kumayunithi kusampula ngayinye ukuya kwenye yezindlela zephethini ye-plaid noma iziqondiso zezingxenye (Isithombe 5c; Ukuhlolwa kwe-Rayleigh; inkawu N: z = 8.33, p < 10−3, isilinganiso esiyindilinga = 190.13 deg ± 9.83 deg; inkawu S: z = 0.79, p = 0.45) futhi yayihambisana kuzo zonke izingqimba ze-plaid inter-grating (Isithombe Esingeziwe 2). Lo mphumela awukwazanga ukuchazwa ngokubandlulula ekusatshalalisweni kweziqondiso ezikhethwayo kumayunithi kusampula ngayinye ukuya kwenye yephethini ye-plaid noma iziqondiso zezingxenye (Isithombe 5c; Ukuhlolwa kwe-Rayleigh; inkawu N: z = 8.33, p < 10−3, isilinganiso esiyindilinga = 190.13 deg ± 9.83 deg; inkawu S: z = 0.79, p = 0.45) futhi yayihambisana kuzo zonke izingqimba ze-plaid inter-grating (Isithombe Esingeziwe 2). Этот эффект нельзя было объяснить смещением распределения предпочтительных направлений в единицах в каждой выборке здорке направлений или направлений компонентов (рис. 5в; критерий Рэлея; обезьяна N: z = 8,33, p < 10–3). Lo mphumela awukwazanga ukuchazwa ngokushintsha kokusatshalaliswa kweziqondiso ezikhethwayo kumayunithi kusampula ngayinye kuya kwenye yeziqondiso ezi-checkered noma iziqondiso zezingxenye (Isithombe 5c; ukuhlolwa kwe-Rayleigh; inkawu N: z = 8.33, p < 10–3)., isilinganiso esiyindilinga = 190.13 degrees ± 9.83 degrees; inkawu S: z = 0.79, p = 0.45) futhi yayifana kuwo wonke amakhona egridi ye-plaid (Isithombe Esingeziwe 2).這种效应不能通过每个样本中单元中的优选方向分布偏向格子图案或组件方向之一來解释（图5c；瑞利测试；猴子N：z = 8.33,p <10-3 ，圆形平均值= 190.13 ± 9.83 度；猴子S：z = 0.79,p = 0.45）并且在格子间光栅角上是一致的（补图2）.這 种 效应 不 能 通过 每 來 樣本 中 单元 中 优选 方向 分布 偏向 偏向 图案 或 元（图 图 图 瑞利测试 ；猴子 n : z = 8.33 , p <10-3 , 平均值 平均值 圆形 圆墢 圆墢圆形 圆形 圆形 圆形 圆形 圆形z Этот эффект не может быть объяснен тем, что распределение предпочтительных ориентаций в клетках в каждом образце смеще решетки, либо в сторону одной из ориентаций компонентов (рис. 5в; критерий Рэлея; обезьяна N: z = 8,33, p <10–3). Lo mphumela awunakuchazwa iqiniso lokuthi ukusatshalaliswa kwezindlela ezikhethwayo kumaseli kusampula ngayinye kuthambekele ekwakhiweni kwe-lattice noma kwenye yezindlela ezisetshenziswayo (Isithombe 5c; ukuhlolwa kukaRayleigh; inkawu N: z = 8.33, p < 10–3)., isilinganiso esiyindilinga) = 190.13 degrees ± 9.83 degrees; inkawu S: z = 0.79, p = 0.45) futhi zazilingana ngama-engeli e-lattice phakathi kwamagridi (Isithombe Esingeziwe 2).Ngakho-ke, ukuzwela kwama-neurons kuma-grid ane-texture kuncike, okungenani ngokwengxenye, ezicini eziyisisekelo zokulungisa i-MT.
Iphaneli yesobunxele ikhombisa ukusatshalaliswa kwezilinganiso ze-N/P (umkhawulo we-neuron/psychophysiological); iseli ngalinye linikeza amaphuzu amabili edatha, elilodwa lesiqondiso ngasinye lapho iphethini ihamba khona. Iphaneli yesokudla idweba imingcele ye-psychophysical (ordinate) uma kuqhathaniswa nemingcele ye-neuronal (abscissa) yawo wonke amayunithi kusampula. Idatha emugqeni ophezulu (ophansi) ivela ku-N (S). b Izilinganiso zomkhawulo ezijwayelekile zidwetshwa ngokuhambisana nobukhulu bomehluko phakathi kokuqondiswa kwe-lattice okulungile kanye nokuqondiswa kweseli okuthandwayo. Isiqondiso "esingcono kakhulu" sichazwa njengokuqondiswa kwesakhiwo se-grating (kulinganiswe nge-sinusoidal grating eyodwa) esiseduze nesiqondiso seseli esithandwayo. Idatha yaqale yafakwa ku-bin ngokuqondiswa okuthandwayo okujwayelekile (ama-bins angu-10°), bese izilinganiso zomkhawulo zabekwa ku-value ephezulu futhi zalinganiswa ngaphakathi kwe-bin ngayinye. Amaseli anokuqondiswa okuthandwayo okukhulu kancane noma okuncane kunokuqondiswa kwezingxenye ze-lattice ayenomehluko omkhulu ekuzweleni kokuqondiswa kwephethini ye-lattice. c I-histogram epinki yokusatshalaliswa kokuqondiswa okuthandwayo kwawo wonke amayunithi e-MT aqoshwe ku-monkey ngayinye.
Okokugcina, impendulo ye-MT ilawulwa yisiqondiso sokunyakaza kwe-grating kanye nemininingwane yezibonakaliso zethu zokuhlukanisa (ukuthungwa). Ukuqhathaniswa kokuzwela kwe-neuronal kanye ne-psychophysical kubonise ukuthi, ngokuvamile, amayunithi e-MT ayengazweli kakhulu ezimpawini zokuthungwa eziphikisanayo kunezinkawu. Kodwa-ke, ukuzwela kwe-neuron kwashintsha kuye ngomehluko phakathi kokuqondiswa okuthandwayo kweyunithi kanye nesiqondiso sokunyakaza kwegridi. Amaseli azwela kakhulu ayevame ukuba nezintandokazi zokuqondiswa ezazicishe zimboze iphethini ye-lattice noma enye yezindlela ezisetshenziswayo, futhi ingxenye encane yamasampula ethu yayizwela noma izwela kakhulu kunokuqonda kwezinkawu umehluko wokuqhathanisa. Ukuze sinqume ukuthi izimpawu ezivela kula mayunithi azwela kakhulu zihlobene kakhulu nokuqonda ezinkawini, sihlole ukuhlobana phakathi kokuqonda kanye nezimpendulo ze-neuronal.
Isinyathelo esibalulekile ekusunguleni uxhumano phakathi komsebenzi we-neural nokuziphatha ukusungula ubudlelwano phakathi kwama-neurons nezimpendulo zokuziphatha ezishukumisweni eziqhubekayo. Ukuze kuhlotshaniswe izimpendulo ze-neural nezahlulelo zokuhlukanisa, kubalulekile ukudala isishukumi, naphezu kokuba sifana, esibonakala ngendlela ehlukile ezivivinyweni ezahlukene. Kulesi sifundo samanje, lokhu kuboniswa ngokucacile yi-grating yokungafani kokuthungwa okungenalutho. Nakuba sigcizelela ukuthi, ngokusekelwe emisebenzini ye-psychometric yezilwane, ama-grating anokungafani okuncane (ngaphansi kuka-~20%) okuthungwa ngokuvamile abhekwa njengokuhambisana noma okusobala.
Ukuze silinganise izinga izimpendulo ze-MT ezihlobana ngalo nemibiko yokuqonda, senze ukuhlaziywa kwamathuba okukhetha (CP) kwedatha yethu yegridi (bheka 3). Ngamafuphi, i-CP iyisilinganiso esingeyona i-parametric, esingeyona ejwayelekile esilinganisa ubudlelwano phakathi kwezimpendulo ze-spike kanye nezahlulelo zokuqonda30. Sikhawulela ukuhlaziywa ezivivinyweni sisebenzisa amagridi angenawo umehluko wombhalo kanye nezikhathi lapho izinkawu zenze okungenani izinketho ezinhlanu zohlobo ngalunye lwalezi zivivinyo, sibale i-SR ngokwehlukana kwesiqondiso ngasinye sokunyakaza kwegridi. Kuzo zonke izinkawu, sibone inani eliphakathi le-CP likhulu kakhulu kunalokho ebesingakulindela ngengozi (Isithombe 6a, d; inkawu N: isilinganiso se-CP: 0.54, 95% CI: (0.53, 0.56), ukuhlolwa kwe-t okunezinhlangothi ezimbili ngokumelene ne-null ye-CP = 0.5, t = 6.7, p < 10−9; inkawu S: isilinganiso se-CP: 0.55, 95% CI: (0.54, 0.57), ukuhlolwa kwe-t okunezinhlangothi ezimbili, t = 9.4, p < 10−13). Kuzo zonke izinkawu, sibone inani eliphakathi le-CP likhulu kakhulu kunalokho ebesingakulindela ngengozi (Isithombe 6a, d; inkawu N: isilinganiso se-CP: 0.54, 95% CI: (0.53, 0.56), ukuhlolwa kwe-t okunezinhlangothi ezimbili ngokumelene ne-null ye-CP = 0.5, t = 6.7, p < 10−9; inkawu S: isilinganiso se-CP: 0.55, 95% CI: (0.54, 0.57), ukuhlolwa kwe-t okunezinhlangothi ezimbili, t = 9.4, p < 10−13).Ezinkawini, sibone isilinganiso se-CP siphezulu kakhulu kunalokho obekulindelekile ngokungahleliwe (Isithombe 6a, d; inkawu N: isilinganiso se-CP: 0.54, 95% CI: (0.53, 0.56), i-t-test enemisila emibili vs. amanani angenalutho).CP = 0,5, t = 6,7, p <10–9; I-CP = 0.5, t = 6.7, p <10–9; обезьяна S: среднее CP: 0,55, 95% ДИ: (0,54, 0,57), двусторонний t-критерий, t = 9,4, p <10–13) . inkawu S: isilinganiso se-CP: 0.55, 95% CI: (0.54, 0.57), ukuhlolwa kwe-t okunemisila emibili, t = 9.4, p < 10–13).在猴子中，我們观察到平均CP 值显着大于我們偶然预期的值（图6a，d；猴子N：平均CP:0.54,95% CI: (0.53, 0.56)双边t 检验, t = 9.4, p <10−13) .在 猴子 中 , 我們 观察 平均 平均 值 显着 大于 我們 偶然 的 值 图图 6a , d: : ; 坳 5: 5 0.53,0.56） , 空值 检验 CP = 0.5, t = 6.7, p <10−9; 猴子S: 平均CP: 0.55, 95% CI: (0.54, 0.57), 双边t检验, t=9.4, p <10−13) У обезьян мы наблюдали средние значения CP, значительно превышающие то, что мы могли бы ожидать случайно (рис. 6a, зье CP: d: d 0,54, 95% ДИ: (0,53, 0,56), двусторонний t- тест CP против нуля = 0,5, t = 6,7, p < 10-9, обезьяна S: средний CP, И5, 5:0,5 0,57), двусторонний t-критерий, t = 9,4, p < 10- 13) . Ezinkawini, sibone amanani aphakathi kwe-CP ngaphezu kwalokho esingakulindela ngengozi (Isithombe 6a, d; inkawu N: isilinganiso se-CP: 0.54, 95% CI: (0.53, 0.56), i-t-test CP enemisila emibili vs. zero = 0.5, t = 6.7, p < 10-9, inkawu S: isilinganiso se-CP: 0.55, 95% CI: (0.54, 0.57), i-t-standard enemisila emibili, t = 9.4, p < 10-13).Ngakho-ke, ama-neuron e-MT avame ukudubula kakhulu ngisho noma kungekho zimpawu zokuhlukaniswa okusobala, lapho umbono wesilwane ngokunyakaza kwe-lattice uhambisana nezintandokazi zeseli.
a Ukusatshalaliswa kwamathuba okukhetha kwamagridi ngaphandle kwezibonakaliso zokuthungwa kwamasampula aqoshwe ku-monkey N. Iseli ngalinye linganikela ngamaphuzu amabili edatha (elilodwa ohlangothini ngalunye lokunyakaza kwegridi). Inani elimaphakathi le-CP elingaphezu kwe-random (imicibisholo emhlophe) libonisa ukuthi sekukonke kunobudlelwano obubalulekile phakathi komsebenzi we-MT kanye nokuqonda. b Ukuze sihlole umthelela wanoma yikuphi ukukhetha okungenzeka, sibale i-CP ngokwehlukana nganoma yisiphi isikhuthazo lapho izinkawu zenze khona iphutha elilodwa okungenani. Amathuba okukhetha adwetshwe njengomsebenzi wesilinganiso sokukhetha (pref/null) sazo zonke izikhuthazo (kwesobunxele) kanye namanani aphelele okungafani kophawu lokuthungwa (kwesokudla, idatha evela kumaseli ayi-120 ngamanye). Umugqa oqinile nendawo enomthunzi kuphaneli yesobunxele imelela isilinganiso ± sem sesilinganiso sokuhamba samaphuzu angama-20. Amathuba okukhetha abalwe izikhuthazo ngezilinganiso zokukhetha ezingalingani, njengegridi ezinezibonisi eziphezulu zokuqhathanisa, ahluke kakhulu futhi aqoqwe ngamathuba. Indawo enomthunzi ompunga kuphaneli yesokudla igcizelela umehluko wezici ezifakiwe ekubalweni kwamathuba okukhetha aphezulu. c Amathuba okukhetha okukhulu (i-ordinate) adwetshwe ngokumelene nomkhawulo we-neuron (i-abscissa). Amathuba okukhetha ahlotshaniswe kabi kakhulu nomkhawulo. I-convention df ifana ne-ac kodwa isebenza kudatha eyodwa engu-157 evela ku-monkey S ngaphandle kokuthi kuphawulwe ngenye indlela. g Amathuba okukhetha aphezulu kakhulu (i-ordinate) adwetshwe ngokumelene nesiqondiso esithandwayo esijwayelekile (i-abscissa) senkawu ngayinye kwezimbili. Iseli ngalinye le-MT linikele ngamaphuzu amabili edatha (elilodwa ohlangothini ngalunye lwesakhiwo se-lattice). h Ama-box plots amakhulu amathuba okukhetha e-engeli ngayinye ye-inter-raster. Umugqa oqinile uphawula i-median, imiphetho engezansi nengaphezulu yebhokisi imelela ama-percentile angu-25 no-75 ngokulandelana, ama-whiskers anwetshwa izikhathi ezingu-1.5 kunebanga le-interquartile, kanti ama-outliers angaphezu kwalomkhawulo ayaphawulwa. Idatha kumaphaneli esobunxele (kwesokudla) ivela kumaseli enkawu angu-120 (157) e-N(S) ngamanye. i Amathuba aphezulu okukhetha (i-ordinate) adwetshwe ngokumelene nesikhathi sokuqala kwe-stimulus (i-abscissa). I-CP enkulu ibalwa ngama-rectangles ashelelayo (ububanzi obuyi-100 ms, isinyathelo esingu-10 ms) kulo lonke isivivinyo bese kubalwa isilinganiso ngamayunithi.
Ezinye izifundo zangaphambilini zibike ukuthi i-CP incike enanini elilinganiselwe lezilingo ekusabalalisweni kwesilinganiso se-basal, okusho ukuthi lesi silinganiso asithembeki kangako ezishukumisi ezibangela umehluko omkhulu ngesilinganiso senketho ngayinye. Ukuze sihlole lo mphumela kudatha yethu, sibale i-CP ngokwehlukana kuzo zonke izishukumisi, kungakhathaliseki ukuthi uphawu luhluke kanjani, futhi izinkawu zenze okungenani isilingo esisodwa esingamanga. I-CP idwetshwe ngokuhambisana nesilinganiso sokukhetha (pref/null) sesilwane ngasinye ku-Figure 6b kanye no-e (iphaneli yesobunxele), ngokulandelana. Uma sibheka izilinganiso ezihambayo, kusobala ukuthi i-CP ihlala ingaphezu kwamathuba phezu kobubanzi bezinkinga zokukhetha, yehla kuphela lapho amathuba ehla (enyuka) ngaphansi (ngenhla) 0.2 (0.8). Ngokusekelwe ezicini ze-psychometric zezilwane, singalindela ama-coefficients okukhetha alolu hlobo lobukhulu ukuthi asebenze kuphela ezishukumisini ezinezinkomba zokuthungwa eziphikisana kakhulu (ezihambisanayo noma ezisobala) (bheka izibonelo zezici ze-psychometric ku-Fig. 2a, b). Ukuze sinqume ukuthi ngabe lokhu kunjalo nokuthi ngabe i-PC ebalulekile iyaqhubeka yini ngisho nasezintweni ezishukumisayo ezinezimpawu ezicacile zokuhlukanisa, sihlole umphumela wamanani aphelele okungafani kombhalo ku-PC (Umfanekiso 6b, e-right). Njengoba bekulindelekile, i-CP yayiphakeme kakhulu kunethuba lokuthi izitho ezishukumisayo zibe nezinkomba zokuhlukanisa ezifika kumaphakathi (~20% ukungafani noma ngaphansi).
Emisebenzini yokuqondisa, isivinini, kanye nokuqaphela okungafani, i-MT CP ivame ukuba phezulu kakhulu kuma-neuron azwela kakhulu, mhlawumbe ngoba la ma-neuron aphethe izimpawu ezifundisa kakhulu30,32,33,34.Ngokuhambisana nalezi zitholelo sibone ubudlelwano obuncane kodwa obubalulekile phakathi kwe-grand CP, obubalwe kusukela kumazinga okudubula anikezwe amaphuzu angu-z kulo lonke umehluko we-texture cue oqokonyiswe kuphaneli engakwesokudla kakhulu ye-Fig.6b, e, kanye nomkhawulo we-neuronal (Isithombe 6c, f; i-geometric mean regression; inkawu N: r = −0.12, p = 0.07 inkawu S: r = −0.18, p < 10−3). 6b, e, kanye nomkhawulo we-neuronal (Isithombe 6c, f; i-geometric mean regression; inkawu N: r = −0.12, p = 0.07 inkawu S: r = −0.18, p < 10−3).Ngokuhambisana nalezi zitholelo, sibone ubudlelwano obuncane kodwa obubalulekile phakathi kwe-CP enkulu ebalwe kusukela ku-z-score yemvamisa yokuvuselela kusukela ekuqhathanisweni kwesignali yokuthungwa okugqanyiswe kuphaneli engakwesokudla ye-Fig. 6b, e, kanye nomkhawulo we-neuronal (Fig. 6c, f; geometric). i-geometric mean regression;обезьяна N: r = -0,12, p = 0,07 обезьяна S: r = -0,18, p < 10-3). inkawu N: r = -0.12, p = 0.07 inkawu S: r = -0.18, p < 10-3).与這些发现一致，我們观察到大CP 之间存在适度但显着的相关性，這是根据图6b, e和神经元阈值（图6c、f；几何平均回归；猴子N：r = -0.12,p = 0.07 amayunithiS:r = -0.18,p <10-3）.与 這些 发现 一致 ，我們到 大 大 之间 存在 适度 但 显着 的 相关性這 是根据 图 6b , e 和图 6c , f ; 回归 ； 猴子 n : r = -0.12 , p = 0.07 子S:r = -0.18,p <10-3）.Ngokuhambisana nalezi zitholelo, sibone ubudlelwano obuncane kodwa obubalulekile phakathi kwama-CV amakhulu njengoba kuboniswe ku-Fig. 6b,e kanye nemingcele ye-neuron (Isithombe 6c,f; i-geometric mean regression; inkawu N: r = -0.12, p = 0.07).Обезьяна S: г = -0,18, р < 10-3). Inkawu S: r = -0.18, p < 10-3).Ngakho-ke, izinkomba ezivela kumayunithi anolwazi kakhulu zivame ukubonisa ukuguquguquka okukhulu ngezigwebo zokuhlukanisa ezisekelweni, okubalulekile kungakhathaliseki ukuthi yiziphi izinkomba zombhalo ezingezwe ekubandlululweni kokubona.
Njengoba besisungule ubudlelwano ngaphambilini phakathi kokuzwela kwezimpawu zokuthungwa kwegridi kanye nokuqondiswa kwe-neuronal okukhethwayo, sizibuze ukuthi ngabe kukhona yini ubudlelwano obufanayo phakathi kwe-CP kanye nokuqondiswa okukhethwayo (Isithombe 6g). Lokhu kuhlangana kwakubaluleke kancane kuphela ku-monkey S (ANOVA; monkey N: 1.03, p=0.46; monkey S: F=1.73, p=0.04). Asibonanga mehluko ku-CP wama-engeli e-lattice phakathi kwama-lattice kunoma yisiphi isilwane (Isithombe 6h; ANOVA; monkey N: F = 1.8, p = 0.11; monkey S: F = 0.32, p = 0. 9).
Ekugcineni, umsebenzi wangaphambilini ukhombisile ukuthi i-CP iyashintsha kulo lonke uvivinyo. Ezinye izifundo zibike ukwanda okukhulu okulandelwa umphumela wokukhetha obushelelezi,30 kanti ezinye zibike ukwanda okuqhubekayo kwesignali yokukhetha phakathi novivinyo31. Ngenkawu ngayinye, sibale i-CP yeyunithi ngayinye ezivivinyweni ezingenayo umehluko wombhalo (ngokulandelana, ngokuya ngokuqondisa kwephethini) kumaseli ayi-100 ms sihamba njalo nge-20 ms kusukela ekuqaleni kwe-pre-stimulus kuya ekulinganisweni kwe-post mean pre-stimulus. Isilinganiso se-CP dynamics sezinkawu ezimbili siboniswe ku-Fig. 6i. Kuzo zombili izimo, i-CP yahlala ezingeni elingahleliwe noma eduze kakhulu nayo kwaze kwaba cishe yi-500 ms ngemva kokuqala kwe-stimulus, ngemva kwalokho i-CP yanda kakhulu.
Ngaphezu kokushintsha kokuzwela, i-CP iboniswe ukuthi ithinteka yizimfanelo ezithile zezimpawu zokulungisa amaseli. Isibonelo, i-Uka ne-DeAngelis34 bathole ukuthi i-CP emsebenzini wokuqaphela ukungafani kwe-binocular incike ekulinganisweni kwejika lokulungisa ukungafani kwe-binocular yedivayisi. Kulesi simo, umbuzo ohlobene nawo ukuthi ngabe amaseli akhetha isiqondiso sephethini (i-PDS) azwela kakhulu kunamaseli akhetha isiqondiso sezingxenye (i-CDS). Amaseli e-PDS afaka ikhodi yokuqondiswa okujwayelekile kwamaphethini aqukethe ukuqondiswa okuningi kwendawo, kuyilapho amaseli e-CDS esabela ekunyakazeni kwezingxenye zephethini eziqondisayo (Isithombe 7a).
ukumelwa kwe-Schematic kwesisusa sokulungisa izingxenye zemodi kanye ne-grating ecatshangelwayo (kwesobunxele) kanye nama-curve okuqondisa ukujiya kwe-grating (kwesokudla) (bheka Izinto Nezindlela). Ngamafuphi, uma iseli lihlangana kuzo zonke izingxenye zegridi ukuze lihambise iphethini yesignali, umuntu angalindela ama-curve afanayo okulungisa igridi ngayinye kanye ne-grid stimuli (ikholomu yokugcina, i-curve eqinile). Ngokuphambene nalokho, uma iseli lingahlanganisi iziqondiso zezingxenye ekunyakazeni kwephethini yesignali, umuntu angalindela i-curve yokulungisa izingxenye ezimbili ene-peak ohlangothini ngalunye lokunyakaza kwe-grating ehumusha ingxenye eyodwa ibe yisiqondiso esithandwayo seseli (ikholomu yokugcina, i-curve eqhekeziwe). . ama-curve b (kwesobunxele) okulungisa ukuqondiswa kwe-array ye-sinusoidal yamaseli aboniswe kuZithombe 1 no-2. 3 no-4 (umugqa ophezulu - amaseli kuZithombe 3a,b no-4a,b (phezulu); iphaneli engezansi - amaseli e-Fig. 3c, d no-4a, b (phansi)). (Phakathi) Ukubikezela kwephethini kanye nezingxenye kubalwa kusukela kumaphrofayili okulungisa i-lattice. (Kwesokudla) Ukulungisa igridi yala maseli. Amaseli ephaneli ephezulu (ezansi) ahlukaniswa njengamaseli ethempulethi (ingxenye). Qaphela ukuthi akukho ukuxhumana phakathi kokuhlukaniswa kwezingxenye zephethini kanye nezintandokazi zokuhamba kweseli okuqondile/okusobala (bheka izimpendulo ze-texture lattice zala maseli ku-Fig. 4a). c I-Coefficient yokuxhumana okungaphelele kwemodi ye-z-score (i-ordinate) edwetshwe ngokumelene ne-coefficient yokuxhumana okungaphelele kwengxenye ye-z-score (i-abscissa) yawo wonke amaseli aqoshwe ezinkawini ze-N (kwesobunxele) kanye ne-S (kwesokudla). Imigqa emincane ikhombisa izindlela zokubaluleka ezisetshenziselwa ukuhlukanisa amaseli. d Isakhiwo samathuba aphezulu okukhetha (i-ordinate) ngokumelene nenkomba yemodi (Zp - Zc) (abscissa). Idatha kumaphaneli esobunxele (kwesokudla) ibhekisela ku-N(S) yenkawu. Imibuthano emnyama ikhombisa idatha kumayunithi acishe abe yiyo. Kuzo zombili izilwane, kwakukhona ukuxhumana okuphawulekayo phakathi kwamathuba aphezulu okukhetha kanye nenkomba yephethini, okuphakamisa ukuxhumana okukhulu kokuqonda kwamaseli anokuqondisa kwephethini yesignali ezishukumisweni ezinokuqondisa okuningi kwezingxenye.
Ngakho-ke, kusethi yokuhlolwa ehlukile, silinganise izimpendulo kumagridi e-sinusoidal namagridi ukuze sihlukanise ama-neurons kumasampula ethu njenge-PDS noma i-CDS (bheka Izindlela). Ama-curve okulungisa i-Lattice, izibikezelo zezingxenye zethempulethi ezakhiwe kusukela kule datha yokulungisa, kanye nama-curve okulungisa i-lattice amaseli aboniswe kuZithombe 1 no-3. Izibalo 3 no-4 kanye neSithombe Esingeziwe 3 ziboniswe kuZithombe 7b. Ukusatshalaliswa kwamaphethini kanye nokukhetha izingxenye, kanye nokuqondiswa kweseli okuthandwayo esigabeni ngasinye, kuboniswa inkawu ngayinye kuZithombe 7c kanye nesithombe esingeziwe 4 ngokulandelana.
Ukuze sihlole ukuncika kwe-CP ekulungisweni kwezingxenye zephethini, siqale sabala inkomba yephethini 35 (PI), amanani amakhulu (amancane) abonisa ukuziphatha okufanayo kwe-PDS (CDS) enkulu. Njengoba kuboniswe ngenhla ukuthi: (i) ukuzwela kwe-neuronal kuyahlukahluka ngomehluko phakathi kokuqondiswa kweseli okuthandwayo kanye nesiqondiso sokunyakaza kwesikhuthazo, futhi (ii) kukhona ukuhlobana okubalulekile phakathi kokuzwela kwe-neuronal kanye namathuba okukhetha kusampula yethu, sithole ubudlelwano phakathi kwe-PI kanye ne-CP iyonke efundwe ukuze kutholakale isiqondiso "esingcono kakhulu" sokunyakaza kweseli ngalinye (bheka ngenhla). Sithole ukuthi i-CP yayihlobene kakhulu ne-PI (Isithombe 7d; i-geometric mean regression; i-grand CP monkey N: r = 0.23, p < 0.01; i-bi-stable CP monkey N = 0.21, p = 0.013; i-grand CP monkey S: r = 0.30, p < 10−4; i-bi-stable CP monkey S: r = 0.29, p < 10−3), okubonisa ukuthi amaseli ahlukaniswe njenge-PDS abonise umsebenzi omkhulu ohlobene nokukhetha kune-CDS namaseli angahlukaniswanga. Sithole ukuthi i-CP yayihlobene kakhulu ne-PI (Isithombe 7d; i-geometric mean regression; i-grand CP monkey N: r = 0.23, p < 0.01; i-bi-stable CP monkey N = 0.21, p = 0.013; i-grand CP monkey S: r = 0.30, p < 10−4; i-bi-stable CP monkey S: r = 0.29, p < 10−3), okubonisa ukuthi amaseli ahlukaniswe njenge-PDS abonise umsebenzi omkhulu ohlobene nokukhetha kune-CDS namaseli angahlukaniswanga. Мы обнаружили, что CP значительно коррелирует с PI (рис. 7d; регрессия среднего геометрического; большая обезьяна CP N:3, r = 0,2, p; бистабильная обезьяна CP N r = 0,21, p = 0,013 большая обезьяна CP S: r = 0,30, p < 10-4; 10-3), ngenza into ethile, futhi, классифицированные как PDS, проявляли большую активность, связанную с выбором, чем CDS ne-неклассифицированные клетки. Sithole ukuthi i-CP yayihlobene kakhulu ne-PI (Isithombe 7d; i-geometric mean regression; i-great monkey CP N: r = 0.23, p < 0.01; i-bistable monkey CP N r = 0.21, p = 0.013; i-great monkey CP S: r = 0.30, p < 10-4; i-monkey S bistable CP: r = 0.29, p < 10-3), okubonisa ukuthi amaseli ahlukaniswe njenge-PDS abonise umsebenzi omningi, ohlotshaniswa nokukhetha kune-CDS namaseli angahlelwanga.我們发现CP 与PI 显着相关（图7d；几何平均回归；大CP 猴N:r = 0.23，p <0.01；双稳态CP = 猴0 CP = 3 0.2 CP = 3 0.1 CP = 3 0.2r. S: r = 0.30,p <10-4;双稳态CP 猴S:r = 0.29,p < 10-3），表明分类为PDS的细胞比CDS 和未分类的细胞表现出更大的选择相关活性. CP 与PI 显着相关（图7d；几何平均回归；大CP猴N:r = 0.23,p <0.01；双稳态CP 猴N r = 0,20 ; Мы обнаружили, что CP был значительно связан с PI (рис. 7d; регрессия среднего геометрического; большая обезьяна CP N:3, r = 0,20 бистабильная обезьяна CP N r = 0,21, p = 0,013 большая обезьяна CP S: r = 0,013) 0,30, p < 10-4; Sithole ukuthi i-CP yayihlotshaniswa kakhulu ne-PI (Isithombe 7d; i-geometric mean regression; inkawu enkulu CP N: r = 0.23, p < 0.01; inkawu e-bistable CP N r = 0.21, p = 0.013; inkawu enkulu CP S: r = 0.013) 0.30, p < 10-4; бистабильный CP обезьяны S: r = 0,29, p < 10-3), что указывает на то, что клетки, классифицированные как PDS, прояоюнцульклишу активность, чем клетки, классифицированные как CDS and неклассифицированные. inkawu S bistable CP: r = 0.29, p < 10-3), okubonisa ukuthi amaseli ahlukaniswe njenge-PDS abonise umsebenzi omkhulu wokukhetha kunamaseli ahlukaniswe njenge-CDS futhi angahlukaniswanga.Ngenxa yokuthi kokubili ukuzwela kwe-PI kanye ne-neuron kuhlobene ne-CP, senze ukuhlaziya okuningi kokuhlehla (ngokuqonda kwe-PI kanye ne-neuron njengezinto eziguquguqukayo ezizimele kanye ne-CP enkulu njengezinto ezixhomeke kuyo) ukuze sinqume ukuhlobana phakathi kwezilinganiso ezimbili eziholela ekutheni kube nomphumela. Womabili ama-coefficients okuxhumana okungaphelele ayebalulekile (inkawu N: umkhawulo vs. CP: r = −0.13, p = 0.04, PI vs. CP: r = 0.23, p < 0.01; inkawu S: umkhawulo vs. CP: r = −0.16, p = 0.03, PI vs CP: 0.29, p < 10−3), okuphakamisa ukuthi i-CP iyanda ngokuzwela futhi ngendlela ezimele iyanda nge-PI. Womabili ama-coefficients okuxhumana okungaphelele ayebalulekile (inkawu N: umkhawulo vs. CP: r = −0.13, p = 0.04, PI vs. CP: r = 0.23, p < 0.01; inkawu S: umkhawulo vs. CP: r = −0.16, p = 0.03, PI vs CP: 0.29, p < 10−3), okuphakamisa ukuthi i-CP iyanda ngokuzwela futhi ngendlela ezimele iyanda nge-PI. Оба частных коэффициента корреляции были значимыми (обезьяна N: порог против CP: r = -0,13, p = 0,04, PI против CP: 3, r0 = 0, p: 0, 2; порог против CP: r = -0,16, p = 0,03, PI iqhudelana ne-CP: 0,29, p < 10-3), предполагая, что CP увеличивается с чувствительностью и неваобличай невамбесть PI. Womabili ama-coefficients okuxhumana okungaphelele ayebalulekile (inkawu N: umkhawulo vs. CP: r=-0.13, p=0.04, PI vs. CP: r=0.23, p<0.01; inkawu S: umkhawulo vs. CP: r = -0.16, p = 0.03, PI vs CP: 0.29, p < 10-3), okuphakamisa ukuthi i-CP iyanda ngokuzwela futhi iyanda ngokuzimela nge-PI.两个偏相关系数均显着（猴子N：阈值与CP：r = -0.13,p = 0.04,PI 与CP:r = 0.23,p <0.01 = 阈值与CP = 阈值与CP = 0.03,PI vs CP:0.29,p <10-3），表明CP 随灵敏度增加而增加，并且以独立方式随PI 增加。两个偏相关系数均显着（猴子N：阈值与CP:r = -0.13,p = 0.04,PI = 0.03,PI vs CP:0.29，p <10-3漉， Оба частных коэффициента корреляции были значимыми (обезьяна N: порог против CP: r = -0,13, p = 0,04, PI против CP: 3, r0 = 0, p: 0, 2; порог против CP: r = -0,16, p = 0,03 , PI против CP: 0,29, p < 10-3), что указывает на то, что CP увеличивалась с чувстевительвелисть независимым образом. Womabili ama-coefficients okuxhumana okungaphelele ayebalulekile (inkawu N: umkhawulo vs. CP: r=-0.13, p=0.04, PI vs. CP: r=0.23, p<0.01; inkawu S: umkhawulo vs. CP: r = -0.16, p = 0.03, PI vs CP: 0.29, p < 10-3), okubonisa ukuthi i-CP yanda ngokuzwela futhi yanda nge-PI ngendlela ezimele.
Siqophe umsebenzi owodwa endaweni ye-MT, futhi izinkawu zibike umbono wazo ngamaphethini angabonakala njengokunyakaza okuqondile noma okusobala. Ukuzwela kwama-neurons ezindleleni zokuhlukanisa okungeziwe emibonweni ebandlululayo kuyahlukahluka kakhulu futhi kunqunywa, okungenani ngokwengxenye, ngobudlelwano phakathi kokuqondiswa okukhethwa yiyunithi kanye nesiqondiso sokunyakaza kwesikhuthazo. Kubo bonke abantu, ukuzwela kwama-neuron kwakuphansi kakhulu kunokuzwela kwengqondo, yize amayunithi abucayi kakhulu afanisa noma adlula ukuzwela kokuziphatha kumasignali okuhlukanisa. Ngaphezu kwalokho, kukhona ukuhlangana okuphawulekayo phakathi kwemvamisa yokudubula kanye nokubona, okuphakamisa ukuthi ukusayina kwe-MT kudlala indima ekuhlukaniseni. Amaseli anokuqondiswa okukhethwayo athuthukise ukuzwela kwawo kokungafani kwesignali yokuhlukanisa i-lattice futhi athambekele ekuboniseni ukunyakaza komhlaba wonke kuma-stimuli ngezinkomba eziningi zendawo, okubonisa ukuhlangana okuphezulu kakhulu kokubona. Lapha sicabangela ezinye izinkinga ezingaba khona ngaphambi kokuqhathanisa le miphumela nomsebenzi wangaphambilini.
Inkinga enkulu ngocwaningo olusebenzisa izisusa ezibizwa ngokuthi i-bistable kumamodeli ezilwane ukuthi izimpendulo zokuziphatha kungenzeka zingasekelwe kubukhulu besithakazelo. Isibonelo, izinkawu zethu zingabika umbono wazo wokuqondiswa kokuthungwa ngaphandle kombono wazo wokuvumelana kwe-lattice. Izici ezimbili zedatha ziphakamisa ukuthi akunjalo. Okokuqala, ngokuhambisana nemibiko yangaphambilini, ukushintsha i-angle yokuqondiswa okuhlobene kwezingxenye ze-array ezihlukanisayo kushintshe ngokuhlelekile amathuba okuqonda okuqondile. Okwesibili, ngokwesilinganiso, umphumela uyafana kumaphethini aqukethe noma angenawo amasignali okuthungwa. Uma kubhekwa konke, lokhu okubonwayo kusikisela ukuthi izimpendulo zezinkawu zibonisa njalo umbono wazo wokuxhumana/ukucaca.
Enye inkinga engaba khona ukuthi asikalungiseleli kahle amapharamitha okunyakaza kwe-grating ngesimo esithile. Emisebenzini eminingi yangaphambilini eqhathanisa ukuzwela kwe-neuronal kanye ne-psychophysical, izisusa zakhethwa ngazinye kuyunithi ngayinye ebhalisiwe [31, 32, 34, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45]. Lapha sisebenzise izinkomba ezimbili ezifanayo zokuhamba kwephethini ye-lattice kungakhathaliseki ukulungiswa kokuqondiswa kweseli ngalinye. Lo mklamo usivumele ukuthi sifunde ukuthi ukuzwela kushintshe kanjani ngokuhambisana phakathi kokunyakaza kwe-lattice kanye nokuqondiswa okuthandwayo, nokho, akuzange kunikeze isisekelo sokuqala sokunquma ukuthi amaseli akhetha ama-lattice ahambisanayo noma asobala. Ngakho-ke, sithembele ezindinganisweni zokuhlola, sisebenzisa impendulo yeseli ngalinye ku-mesh ethambile, ukuze sinikeze ukukhetha kanye namalebula angekho esigabeni ngasinye sokunyakaza kwe-mesh. Nakuba kungenakwenzeka, lokhu kungaphazamisa ngokuhlelekile imiphumela yokuhlaziywa kwethu kokuzwela kanye nokutholwa kwesignali ye-CP, okungenzeka kudlulele noma yisiphi isilinganiso. Kodwa-ke, izici eziningana zokuhlaziywa kanye nedatha okuxoxwe ngazo ngezansi ziphakamisa ukuthi akunjalo.
Okokuqala, ukwabela amagama akhethwayo (angenalutho) ezishukumisweni ezibangele umsebenzi omningi (ongaphansi) akuzange kuthinte ukuhlukaniswa kwalokhu kusatshalaliswa kwempendulo. Esikhundleni salokho, kuqinisekisa kuphela ukuthi imisebenzi ye-neurometric neye-psychometric inesibonakaliso esifanayo, ngakho-ke ingaqhathaniswa ngqo. Okwesibili, izimpendulo ezisetshenziselwa ukubala i-CP (izivivinyo "ezingalungile" zama-gratings ane-texture kanye nazo zonke izivivinyo zama-gratings ngaphandle kokungafani kwe-texture) azifakwanga ekuhlaziyweni kokuhlehla okunqume ukuthi iseli ngalinye "lithanda" yini imidlalo exhunyiwe noma esobala. Lokhu kuqinisekisa ukuthi imiphumela yokukhetha ayithambekele ezigabeni ezikhethwayo/ezingavumelekile, okuholela emathubeni amakhulu okukhetha.
Izifundo zikaNewsom kanye nozakwabo [36, 39, 46, 47] zaba ngezokuqala ukunquma indima ye-MT ekulinganisweni okulinganiselwe kwendlela yokunyakaza. Imibiko elandelayo iqoqe idatha ngokubamba iqhaza kwe-MT ngokujulile34,44,48,49,50,51 kanye nesivinini32,52, ukuqondiswa okuhle33 kanye nokuqonda isakhiwo se-3D kusukela ekunyakazeni31,53,54 (amahlathi azinzile e-3D). Sinweba le miphumela ngezindlela ezimbili ezibalulekile. Okokuqala, sinikeza ubufakazi bokuthi izimpendulo ze-MT zinegalelo ekuhlukanisweni kokubona kwezimpawu ze-visuomotor. Okwesibili, sibone ubudlelwano phakathi kokukhetha kokuqondiswa kwemodi ye-MT kanye nalesi signali sokukhetha.
Ngomqondo, imiphumela yamanje ifana kakhulu nomsebenzi we-3-D SFM, njengoba kokubili kuwukubona okuyinkimbinkimbi kwe-bistable okubandakanya ukunyakaza kanye nokuhleleka kokujula. UDodd nabanye.31 bathole ithuba elikhulu lokukhetha (0.67) emsebenzini wenkawu bebika ukuqondiswa kokujikeleza kwesilinda se-bistable 3D SFM. Sithole umphumela omncane kakhulu wokukhetha we-bistable grid stimuli (cishe u-0.55 kuzo zombili izinkawu). Njengoba ukuhlolwa kwe-CP kuncike ku-coefficient yokukhetha, kunzima ukuhumusha i-CP etholwe ngaphansi kwezimo ezahlukene emisebenzini ehlukene. Kodwa-ke, ubukhulu bomphumela wokukhetha esiwubonile babufana kuma-gratings angenalutho kanye nama-low texture contrast, kanye nalapho sihlanganisa i-low/no texture contrast stimuli ukwandisa amandla. Ngakho-ke, lo mehluko ku-CP cishe awubangelwa umehluko emazingeni okukhetha phakathi kwamasethi edatha.
Izinguquko ezincane ezingeni lokudubula le-MT ezihambisana nokuqonda esimweni sokugcina zibonakala zixaka uma kuqhathaniswa nezimo zokuqonda eziqinile nezihlukile ngokwekhwalithi ezibangelwa ukukhuthazwa kwe-3-D SFM kanye nezakhiwo zegridi eziqinile. Okunye okungenzeka ukuthi sinciphise umphumela wokukhetha ngokubala izinga lokudubula phakathi nesikhathi sonke sokukhuthazwa. Ngokuphambene necala le-31 3-D SFM, lapho umehluko emsebenzini we-MT wakhula cishe ngama-250 ms ezivivinyweni bese wanda kancane kancane kuzo zonke izivivinyo, ukuhlaziywa kwethu kokuguquguquka kwesikhashana kwezimpawu zokukhetha (bheka ama-500 ms ngemuva kokuqala kwesikhuthazo kuzo zombili izinkawu. Ngaphezu kwalokho, ngemuva kokunyuka okukhulu phakathi nalesi sikhathi, sabona ukuguquguquka ku-CP ngesikhathi esisele sesivivinyo. UHupe noRubin55 babika ukuthi umbono womuntu ngama-array angunxande aguquguqukayo uvame ukushintsha phakathi nezivivinyo ezinde. Nakuba izikhuthazo zethu zethulwe imizuzwana eyi-1.5 kuphela, umbono wezinkawu zethu ungahluka kusukela ekuhambisaneni kuya ekukhanyeni ngesikhathi sesivivinyo (izimpendulo zabo zibonise umbono wazo wokugcina ekukhetheni i-cue.) Ngakho-ke, inguqulo yesikhathi sokuphendula yomsebenzi wethu, noma uhlelo lapho izinkawu zingabika khona ngokuqhubekayo umbono wazo, kulindeleke ukuthi ibe nomphumela omkhulu wokukhetha. Ithuba lokugcina ukuthi isignali ye-MT ifundwa ngendlela ehlukile emisebenzini emibili. Nakuba sekuyisikhathi eside kucatshangwa ukuthi izimpawu ze-CPU zivela ekuchazeni imizwa nomsindo ohlobene,56 UGu nozakwabo57 bathole ukuthi kumamodeli wokubala, amasu ahlukene okuhlanganisa, kunamazinga okuxhumana ukuhlukahluka, kungachaza kangcono i-CPU kuma-neurons e-dorsal media-temporal superior. umsebenzi wokuqaphela ukushintsha kweshidi (MSTd). Umphumela omncane wokukhetha esiwubonile ku-MT cishe ubonisa ukuhlanganiswa okubanzi kwama-neurons amaningi angenalo ulwazi oluningi ukudala imibono yokuhambisana noma ukucaca. Kunoma yikuphi lapho izinkomba zokunyakaza kwendawo kwakufanele ziqoqwe zibe yinto eyodwa noma ezimbili (ama-grating ahlanganisiwe) noma izindawo ezihlukene zezinto ezivamile (3-D SFM), ubufakazi obuzimele bokuthi izimpendulo ze-MT zazihlotshaniswa kakhulu nezahlulelo zokuqonda, kwakukhona izimpendulo ze-MT ezinamandla. kuphakanyiswa ukuthi kudlale indima ekuhlukanisweni kwezithombe eziyinkimbinkimbi zibe yizigcawu zezinto eziningi kusetshenziswa ulwazi lokunyakaza okubonakalayo.
Njengoba kushiwo ngenhla, besingabaqala ukubika ubudlelwano phakathi komsebenzi weseli wephethini le-MT kanye nokuqonda. Njengoba kwakhiwe kumodeli yokuqala yezigaba ezimbili nguMovshon kanye nozakwabo, iyunithi yemodi iyisigaba sokukhipha se-MT. Kodwa-ke, umsebenzi wakamuva ukhombisile ukuthi amaseli emodi kanye nezingxenye amelela iziphetho ezahlukene ze-continuum nokuthi umehluko we-parametric esakhiweni sensimu yokwamukela unesibopho se-tuning spectrum yezingxenye zemodi. Ngakho-ke, sithole ubudlelwano obubalulekile phakathi kwe-CP ne-PI, obufana nobudlelwano phakathi kokuvumelanisa kokulungiswa kwe-binocular mismatch kanye ne-CP emsebenzini wokuqaphela ukujula noma ukucushwa kokusetha kokuqondisa emsebenzini wokuhlukaniswa kokuqondisa okuhle. Ubudlelwano phakathi kwamadokhumenti kanye ne-CP 33. UWang noMovshon62 bahlaziye inani elikhulu lamaseli anokukhetha kokuqondisa kwe-MT futhi bathola ukuthi, ngokwesilinganiso, inkomba yemodi yayihlotshaniswa nezakhiwo eziningi zokulungisa, okuphakamisa ukuthi ukukhetha kwemodi kukhona kwezinye izinhlobo eziningi zezimpawu ezingafundwa kubantu be-MT. Ngakho-ke, ezifundweni zesikhathi esizayo zobudlelwano phakathi komsebenzi we-MT kanye nokuqonda okuqondene nomuntu, kuzobaluleka ukunquma ukuthi inkomba yephethini ihlobene yini nezinye izimpawu zokukhetha umsebenzi kanye nezikhuthazo, noma ukuthi lobu budlelwano buqondene ngqo nendaba yokuhlukaniswa kokuqonda.
Ngokufanayo, uNienborg noCumming 42 bathole ukuthi yize amaseli aseduze nakude akhetha ukungafani kwe-binocular ku-V2 ayenozwelo olufanayo emsebenzini wokuhlukanisa ngokujulile, inani lamaseli athanda kakhulu kuphela abonise i-CP ebalulekile. Kodwa-ke, ukuqeqesha kabusha izinkawu ukuze zilinganisele umehluko okude ngokukhethekile kwaholela kuma-CP abalulekile ezindlini ezithandwa kakhulu. Ezinye izifundo zibike nokuthi umlando wokuqeqesha uncike ekuxhumaneni kokuqonda34,40,63 noma ubudlelwano obuyimbangela phakathi komsebenzi we-MT kanye nokwahlukana okuhlukile48. Ubudlelwano esibubonile phakathi kwe-CP kanye nokukhetha isiqondiso serejimeni cishe bubonisa isu elithile elisetshenziswe izinkawu ukuxazulula inkinga yethu, hhayi indima ethile yezimpawu zokukhetha imodi ekuboneni kwe-visual-motor. Emsebenzini wesikhathi esizayo, kuzobaluleka ukunquma ukuthi umlando wokufunda unomthelela omkhulu yini ekunqumeni ukuthi yiziphi izimpawu ze-MT ezilinganiswa ngokukhethekile nangokuguquguqukayo ukuze kwenziwe izahlulelo zokuhlukanisa.
UStoner kanye nozakwabo14,23 babe ngabokuqala ukubika ukuthi ukushintsha ukukhanya kwezifunda zegridi ezidlulayo kuthinte ukuhambisana nokucaca kwemibiko yabantu ababukayo kanye nokulungiswa kwesiqondiso kuma-neuron e-MT aqinile. Abalobi bathole ukuthi lapho ukukhanya kwezifunda ezidlulayo kuhambisana nokucaca ngokomzimba, ababukeli babike ukuqonda okusobala kakhulu, kuyilapho ama-neurons e-MT ekhombisa ukunyakaza kwezingxenye ze-raster. Ngokuphambene nalokho, lapho ukukhanya okudlulayo nokugqagqana okusobala kungahambisani ngokomzimba, umbukeli ubona ukunyakaza okuvumelanayo, kanti ama-neurons e-MT akhombisa ukunyakaza komhlaba wonke kwephethini. Ngakho-ke, lezi zifundo zibonisa ukuthi izinguquko zomzimba ezishukumisweni ezibonakalayo ezithinta ngokuthembekile imibiko yokuhlukanisa nazo zibangela izinguquko ezibikezelwayo ekuvukeni kwe-MT. Umsebenzi wakamuva kule ndawo uhlole ukuthi yiziphi izimpawu ze-MT ezilandelela ukubonakala kokubona kwezishukumi eziyinkimbinkimbi18,24,64. Isibonelo, i-subset yama-neurons e-MT iboniswe ukuthi ibonisa ukulungiswa kwe-bimodal kumephu yokunyakaza kwephuzu elingahleliwe (RDK) enezinkomba ezimbili eziqhelelene kakhulu kune-RDK eqondisiwe. I-bandwidth yokulungiswa kweselula 19, 25. Abaqapheli bahlala bebona iphethini yokuqala njengokunyakaza okusobala, yize iningi lama-neuron e-MT libonisa ukuzivumelanisa kwe-unimodal ekuphenduleni lezi zinto ezishukumisayo, futhi isilinganiso esilula sawo wonke amaseli e-MT sinikeza impendulo ye-unimodal yabantu. Ngakho-ke, i-subset yamaseli abonisa ukulungiswa kwe-bimodal ingakha i-neural substrate yalokhu kuqonda. Ngokuthakazelisayo, kuma-marmosets, leli nani lifaniswe namaseli e-PDS lapho lihlolwa kusetshenziswa i-grid evamile kanye ne-grid stimuli.
Imiphumela yethu ihamba phambili kuneyangenhla, okubalulekile ekusunguleni indima ye-MT ekuhlukaniseni ukuqonda. Eqinisweni, ukuhlukaniswa kuyinto eyenzeka ngeso lengqondo. Izibonisi eziningi ezibonakalayo eziguquguqukayo zibonisa ikhono lesistimu yokubona ukuhlela nokuhumusha izinto eziqhubekayo ngezindlela ezingaphezu kweyodwa. Ukuqoqa izimpendulo zemizwa kanye nemibiko yokubona ngesikhathi esisodwa esifundweni sethu kusivumele ukuthi sihlole ukuhlangana phakathi kwesilinganiso sokudubula kwe-MT kanye nokuchazwa kokucabanga kokukhuthazwa okuqhubekayo. Njengoba sesibonise lobu budlelwano, siyavuma ukuthi isiqondiso sembangela asikasungulwa, okungukuthi, kudingeka ukuhlolwa okwengeziwe ukuze kunqunywe ukuthi isignali yokuhlukaniswa kokuqonda esibonwe yithi, njengoba abanye bephikisana [65, 66, 67], iyazenzakalela yini. Inqubo iphinde imele izimpawu zokwehla ezibuyela ku-cortex yezinzwa ezivela ezindaweni eziphakeme 68, 69, 70 (Umfanekiso 8). Imibiko yengxenye enkulu yamaseli akhetha amaphethini ku-MSTd71, enye yezinhloso eziyinhloko ze-MT, isikisela ukuthi ukwandisa lokhu kuhlolwa ukuze kufakwe ukuqoshwa ngasikhathi sinye kwe-MT kanye ne-MSTd kungaba isinyathelo sokuqala esihle sokuqonda kabanzi izindlela zokuqonda ze-neural.
Imodeli yezigaba ezimbili yokukhetha ukuqondiswa kwengxenye kanye nemodi kanye nomthelela ongaba khona wempendulo evela phezulu emsebenzini ohlobene nokukhetha ekuhumusheni komshini. Lapha, ukukhetha ukuqondiswa kwemodi (PDS – “P”) esinyathelweni se-MT kudalwe (i) isampula enkulu yedatha yokufaka ekhethiwe ehambisana nesivinini esithile semodi, kanye (ii) nokucindezela okuqinile kokuhlela. Ingxenye ekhethiwe (CDS) yesigaba se-MT (“C”) inobubanzi obuncane bokusampula ohlangothini lokufaka futhi ayinakho ukucindezelwa okuningi kokuhlela. Ukuvinjelwa okungalungiswanga kunikeza ukulawula kuzo zombili izakhamuzi. Imicibisholo enemibala ikhombisa ukuqondiswa kwedivayisi okuthandwayo. Ukuze kucace, kuboniswa ingxenye encane kuphela yokuxhumeka kwe-V1-MT kanye nebhokisi elilodwa lemodi yengxenye kanye nokukhethwa kokuqondiswa. Kumongo wokuhumusha imiphumela yethu yokuphakela phambili (FF), ukusethwa kokufaka okubanzi kanye nokuvinjelwa okuqinile kokuhlela (okugqanyisiwe ngombala obomvu) kumaseli e-PDS kubangele umehluko omkhulu emsebenzini ekuphenduleni amaphethini amaningi okunyakaza. Enkingeni yethu yokuhlukaniswa, leli qembu liqhuba amaketanga ezinqumo futhi lihlanekezela ukuqonda. Ngokuphambene nalokho, esimweni sempendulo (i-FB), izinqumo zokuqonda zenziwa kumasekethe aphezulu ngedatha yezinzwa kanye nokubandlulula kwengqondo, futhi umphumela omkhulu we-FB engezansi kumaseli e-PDS (imigqa ejiyile) ukhiqiza izimpawu zokukhetha. b Ukumelwa kwe-schematic kwamamodeli ahlukile amadivayisi e-CDS kanye ne-PDS. Lapha izimpawu ze-PDS ku-MT azikhiqizwa kuphela ngokufaka okuqondile kwe-V1, kodwa futhi nangokufaka okungaqondile kwendlela ye-V1-V2-MT. Izindlela ezingaqondile zemodeli zilungiswa ukuze zinikeze ukukhetha emingceleni yokuthungwa (izindawo ezidlulayo zegridi). Imojula ye-CDS yesendlalelo se-MT yenza isamba esilinganiselwe sokufaka okuqondile nokungaqondile futhi ithumela umphumela kumojula we-PDS. I-PDS ilawulwa ukuvinjelwa okuncintisanayo. Futhi, kuboniswa kuphela lezo zixhumanisi ezidingekayo ukudweba ukwakheka okuyisisekelo kwemodeli. Lapha, indlela ehlukile ye-FF kunaleyo ephakanyisiwe ku-a ingaholela ekuhlukeni okukhulu ekuphenduleni kwe-lattice yeselula ku-PDS, futhi kuholele ekubandlululeni emaphethini ezinqumo. Ngaphandle kwalokho, i-CP enkulu kumaseli e-PDS ingase ibe umphumela wokubandlulula emandleni noma ekusebenzeni kahle kokunamathiselwa kwe-FB kumaseli e-PDS. Ubufakazi busekela amamodeli e-MT PDS anezigaba ezimbili nezintathu kanye nokuhumusha kwe-CP FF kanye ne-FB.
Ama-macaque amabili amadala (i-macaca mulatta), elilodwa leduna nelinye lensikazi (elineminyaka engu-7 nengu-5 ngokulandelana), elinesisindo esisukela ku-4.5 kuya ku-9.0 kg, asetshenziswa njengezinto zokutadisha. Ngaphambi kwazo zonke izivivinyo zokuhlinzwa okungenamafutha, izilwane zazifakwe igumbi lokuqopha elenziwe ngokwezifiso lama-electrode aqondile asondela endaweni ye-MT, isitendi se-headrest sensimbi engagqwali (i-Crist Instruments, i-Hagerstown, MD), kanye nendawo yamehlo ene-coil yokusesha ye-scleral elinganisiwe. (i-Cooner Wire, e-San Diego, e-California). Zonke izinqubo zihambisana nemithethonqubo yoMnyango Wezolimo wase-United States (i-USDA) kanye ne-National Institutes of Health (NIH) Guidelines for the Humane Care and Use of Laboratory Animals futhi zivunyiwe yi-University of Chicago Institutional Animal Care and Use Committee (IAUKC).
Zonke izinto ezishukumisayo ezibonakalayo zethulwe ngendlela evulekile ebhekene nesizinda esimnyama noma esimpunga. Ngesikhathi sokuqopha, indawo kanye nobubanzi balo mgodi kwalungiswa ngokuhambisana nensimu yokwamukela yama-neurons ephepheni le-electrode. Sisebenzise izigaba ezimbili ezibanzi zezinto ezishukumisayo ezibonakalayo: izinto ezishukumisayo zengqondo kanye nezinto ezishukumisayo zokulungisa.
Isikhuthazo se-psychometric siyiphethini ye-grating (20 cd/m2, umehluko ongu-50%, umjikelezo womsebenzi ongu-50%, ama-degrees/sec angu-5) edalwe ngokubeka ama-grating amabili angunxande azulazula ngendlela eqondile eya endaweni yawo (Isithombe 1b). Kuye kwaboniswa ngaphambilini ukuthi ababukeli abangabantu babona la maphethini e-grid njengezinto ezishukumisayo eziguquguqukayo, ngezinye izikhathi njengephethini eyodwa ehamba ngendlela efanayo (ukunyakaza okuhlangene) futhi ngezinye izikhathi njengezindawo ezimbili ezihlukene ezihamba ngezindlela ezahlukene (ukunyakaza okusobala). izingxenye zephethini ye-lattice, eziqondiswe ngokulinganayo - i-engeli ephakathi kwama-lattice isukela ku-95° kuya ku-130° (ethathwe kusethi: 95°, 100°, 105°, 115°, 120°, 125°, 130° °, kuyo yonke iseshini Ama-neurons e-Isolation angle ku-115° awazange agcinwe, kodwa sifaka idatha ye-psychophysical lapha) - cishe ama-90° noma ama-270° (ukuqondiswa kwephethini). Esifundweni ngasinye, kwasetshenziswa ikona elilodwa kuphela le-lattice interlattice; ngesikhathi sesifundo ngasinye, ukuqondiswa kwephethini yesilingo ngasinye kwakhethwa ngokungahleliwe ezintweni ezimbili ezingenzeka.
Ukuze sihlukanise ukuqonda kwegridi futhi sinikeze isisekelo esibonakalayo somvuzo wesenzo, sethula ukuthungwa kwamaphuzu okungahleliwe esinyathelweni sesibani sokukhanya 72 sengxenye ngayinye yegridi. Lokhu kufezwa ngokwandisa noma ukunciphisa (ngenani elinqunyiwe) ukukhanya kwengxenye encane yamaphikseli akhethwe ngokungahleliwe (Umfanekiso 1c). Isiqondiso sokunyakaza kokuthungwa sinikeza isignali eqinile eshintsha umbono wombukeli uye ekunyakazeni okuqondile noma okusobala (Umfanekiso 1c). Ngaphansi kwezimo eziqondile, zonke ukuthungwa, kungakhathaliseki ukuthi iyiphi ingxenye ye-texture lattice cover, zihunyushwa ngendlela yephethini (Umfanekiso 1c, okuqondile). Esimweni esicacile, ukuthungwa kuhamba ngokuqondile kuya endleleni ye-grating embozwayo (Umfanekiso 1c, okusobala) (Ifilimu Engeziwe 1). Ukuze kulawulwe ubunzima bomsebenzi, ezinhlanganweni eziningi ukuqhathanisa kweMichelson (Lmax-Lmin/Lmax+Lmin) kwalolu phawu lokuthungwa kwahlukahluka kusuka kusethi ka-(-80, -40, -20, -10, -5, 0, 5). , 10, 20, 40, 80). Ukungafani kuchazwa njengokukhanya okuhlobene kwe-raster (ngakho inani lokungafani elingu-80% lizoholela ekuthungeni okungu-36 noma okungu-6 cd/m2). Kumaseshini angu-6 kumaseshini angu-N enkawu kanye namaseshini angu-5 kumaseshini angu-S enkawu, sisebenzise amabanga okungafani okulinganiselwe kombhalo (-30, -20, -15, -10, -5, 0, 5, 10, 15, 20, 30), lapho izici zengqondo zilandela iphethini efanayo nokungafani okugcwele, kodwa ngaphandle kokugcwala.
Izisusa zokulungisa ziyizigridi ze-sinusoidal (umehluko ongu-50%, umjikelezo/idigri eli-1, amadigri ama-5/umzuzwana) ezihamba kwenye yeziqondiso eziyi-16 ezilinganiselwe ngokulinganayo, noma izigridi ze-sinusoidal ezihamba kulezi zindlela (eziqukethe ama-engeli amabili aphambene angu-135° ama-grating e-sinusoidal abekwe ngaphezulu komunye nomunye) ohlangothini olufanayo lwephethini.