Laura Schulz: The surprisingly logical minds of babies

Prevodilac: Ivana Krivokuća Lektor: Ivana Korom

Mark Twain summed up what I take to be

Mark Tven je sumirao ono što smatram

one of the fundamental problems of cognitive science

jednim od temeljnih problema kognitivne nauke

with a single witticism.

samo jednom dosetkom.

He said, "There's something fascinating about science.

Rekao je: "Postoji nešto fascinantno u vezi sa naukom.

One gets such wholesale returns of conjecture

Dobija se veliki obrt pretpostavki

out of such a trifling investment in fact."

od tako sitnog ulaganja u činjenice."

(Laughter)

(Smeh)

Twain meant it as a joke, of course, but he's right:

Tven je mislio to kao šalu, naravno, ali u pravu je,

There's something fascinating about science.

postoji nešto fascinantno u vezi sa naukom.

From a few bones, we infer the existence of dinosuars.

Na osnovu nekoliko kostiju, zaključujemo o postojanju dinosaurusa.

From spectral lines, the composition of nebulae.

Na osnovu spektralnih linija, o sastavu nebula.

From fruit flies,

Od voćne mušice,

the mechanisms of heredity,

o mehanizmima nasleđivanja,

and from reconstructed images of blood flowing through the brain,

a na osnovu rekonstruisanih snimaka protoka krvi kroz mozak,

or in my case, from the behavior of very young children,

ili u mom slučaju, na osnovu ponašanja veoma male dece,

we try to say something about the fundamental mechanisms

pokušavamo da kažemo nešto o osnovnim mehanizmima

of human cognition.

ljudske kognicije.

In particular, in my lab in the Department of Brain and Cognitive Sciences at MIT,

Konkretno, u mojoj laboratoriji

na Odeljenju za mozak i kognitivne nauke na Masačusetskom tehnološkom institutu,

I have spent the past decade trying to understand the mystery

provela sam proteklu deceniju pokušavajući da razumem misteriju

of how children learn so much from so little so quickly.

kako deca uče tako mnogo iz tako malo tako brzo.

Because, it turns out that the fascinating thing about science

Jer, ispostavlja se da je fascinantna stvar u vezi sa naukom

is also a fascinating thing about children,

takođe i fascinantna stvar u vezi sa decom,

which, to put a gentler spin on Mark Twain,

a to je, da ublažim verziju Marka Tvena,

is precisely their ability to draw rich, abstract inferences

upravo njihova sposobnost da izvuku bujne, apstraktne zaključke

rapidly and accurately from sparse, noisy data.

brzo i tačno iz oskudnih, izmešanih podataka.

I'm going to give you just two examples today.

Daću vam dva primera.

One is about a problem of generalization,

Jedan je o problemu generalizacije,

and the other is about a problem of causal reasoning.

a drugi je o problemu uzročnog rezonovanja.

And although I'm going to talk about work in my lab,

I mada ću govoriti o radu u mojoj laboratoriji,

this work is inspired by and indebted to a field.

ovaj rad je imao inspiraciju na terenu i njemu ga dugujem.

I'm grateful to mentors, colleagues, and collaborators around the world.

Zahvalna sam mentorima, kolegama i saradnicima širom sveta.

Let me start with the problem of generalization.

Počeću problemom generalizacije.

Generalizing from small samples of data is the bread and butter of science.

Uopštavanje na osnovu malih uzoraka podataka

je osnovni izvor nauke.

We poll a tiny fraction of the electorate

Izbrojimo mali deo izbornog tela

and we predict the outcome of national elections.

i predviđamo ishod nacionalnih izbora.

We see how a handful of patients responds to treatment in a clinical trial,

Vidimo kako nekolicina pacijenata reaguje na tretman u kliničkom ispitivanju,

and we bring drugs to a national market.

i donosimo lekove na domaće tržište.

But this only works if our sample is randomly drawn from the population.

Ali ovo funkcioniše samo ako je naš uzorak nasumično izvučen iz populacije.

If our sample is cherry-picked in some way --

Ako je naš uzorak biran na neki način -

say, we poll only urban voters,

recimo, ispitamo samo gradske birače,

or say, in our clinical trials for treatments for heart disease,

ili recimo, u kliničkim ispitivanjima tretmana bolesti srca

we include only men --

uključimo samo muškarce -

the results may not generalize to the broader population.

rezultati se možda neće generalizovati na širu populaciju.

So scientists care whether evidence is randomly sampled or not,

Dakle, naučnike zanima da li su dokazi slučajno uzorkovani ili ne,

but what does that have to do with babies?

ali kakve to ima veze sa bebama?

Well, babies have to generalize from small samples of data all the time.

Pa, bebe moraju stalno da generalizuju na osnovu malih uzoraka podataka.

They see a few rubber ducks and learn that they float,

Vide nekoliko gumenih pataka i nauče da one plutaju,

or a few balls and learn that they bounce.

ili nekoliko lopti i nauče da one odskaču.

And they develop expectations about ducks and balls

I razvijaju očekivanja u vezi sa patkama i loptama

that they're going to extend to rubber ducks and balls

koje će proširiti na gumene patke i lopte

for the rest of their lives.

do kraja njihovih života.

And the kinds of generalizations babies have to make about ducks and balls

A vrste generalizacija koje bebe prave o patkama i loptama

they have to make about almost everything:

moraju da prave o gotovo svemu:

shoes and ships and sealing wax and cabbages and kings.

cipelama, brodovima, vosku za pečaćenje, kupusu i kraljevima.

So do babies care whether the tiny bit of evidence they see

Da li bebe zanima da li delić dokaza koji one vide

is plausibly representative of a larger population?

verodostojno predstavlja veću populaciju?

Let's find out.

Hajde da to otkrijemo.

I'm going to show you two movies,

Pokazaću vam dva filma,

one from each of two conditions of an experiment,

jedan iz svake od situacija u eksperimentu,

and because you're going to see just two movies,

i pošto ćete videti samo dva filma,

you're going to see just two babies,

videćete samo dve bebe,

and any two babies differ from each other in innumerable ways.

a bilo koje dve bebe se razlikuju međusobno na bezbroj načina.

But these babies, of course, here stand in for groups of babies,

Ali ove bebe, naravno, ovde zastupaju grupe beba,

and the differences you're going to see

i razlike koje ćete videti

represent average group differences in babies' behavior across conditions.

predstavljaju prosečne grupne razlike u ponašanju beba kroz različite uslove.

In each movie, you're going to see a baby doing maybe

U svakom filmu ćete videti kako beba radi

just exactly what you might expect a baby to do,

možda baš ono što biste očekivali da će beba uraditi,

and we can hardly make babies more magical than they already are.

a teško da možemo da učinimo bebe čarobnijim nego što već jesu.

But to my mind the magical thing,

Ali za mene je čarobna stvar,

and what I want you to pay attention to,

i ono na šta želim da obratite pažnju,

is the contrast between these two conditions,

kontrast između ova dva uslova,

because the only thing that differs between these two movies

jer jedino što razlikuje ova dva filma

is the statistical evidence the babies are going to observe.

je statistički dokaz koji će bebe primetiti.

We're going to show babies a box of blue and yellow balls,

Pokazaćemo bebama kutiju plavih i žutih lopti,

and my then-graduate student, now colleague at Stanford, Hyowon Gweon,

a moja tadašnja studentkinja, sada koleginica na Stenfordu, Jouon Gvon,

is going to pull three blue balls in a row out of this box,

izvući će tri plave lopte zaredom iz ove kutije,

and when she pulls those balls out, she's going to squeeze them,

i kada izvuče te lopte, stisnuće ih,

and the balls are going to squeak.

a lopte će zapištati.

And if you're a baby, that's like a TED Talk.

Ako ste beba, to je kao TED govor.

It doesn't get better than that.

Ne može biti bolje od toga.

(Laughter)

(Smeh)

But the important point is it's really easy to pull three blue balls in a row

Ali bitna poenta je da je veoma lako izvući tri plave loptice zaredom

out of a box of mostly blue balls.

iz kutije sa pretežno plavim lopticama.

You could do that with your eyes closed.

Možete to da uradite sa zatvorenim očima.

It's plausibly a random sample from this population.

To je verovatno slučajni uzorak iz ove populacije.

And if you can reach into a box at random and pull out things that squeak,

A ako možete posegnuti u kutiju nasumice i izvaditi stvari koje pište,

then maybe everything in the box squeaks.

onda možda sve u toj kutiji pišti.

So maybe babies should expect those yellow balls to squeak as well.

Možda bebe očekuju da žute lopte takođe pište.

Now, those yellow balls have funny sticks on the end,

Te žute lopte imaju zabavne štapiće na kraju,

so babies could do other things with them if they wanted to.

tako da bebe mogu da rade druge stvari sa njima ako hoće.

They could pound them or whack them.

Mogu da ih lupaju ili udaraju.

But let's see what the baby does.

Ali hajde da vidimo šta beba radi.

(Video) Hyowon Gweon: See this? (Ball squeaks)

(Video) Jouon Gvon: Vidiš ovo? (Lopta pišti)

Did you see that? (Ball squeaks)

Jesi li videla to? (Lopta pišti)

Cool.

Kul.

See this one?

Vidiš ovu?

(Ball squeaks)

(Lopta pišti)

Wow.

Opa!

Laura Schulz: Told you. (Laughs)

Lora Šulc: Rekla sam vam. (Smeh)

(Video) HG: See this one? (Ball squeaks)

(Video) JG: Vidiš ovu? (Lopta pišti)

Hey Clara, this one's for you. You can go ahead and play.

Hej Klara, ova je za tebe. Možeš da se igraš.

(Laughter)

(Smeh)

LS: I don't even have to talk, right?

LŠ: Ne moram ni da pričam, zar ne?

All right, it's nice that babies will generalize properties

U redu, lepo je to što će bebe generalizovati osobine

of blue balls to yellow balls,

plavih loptica na žute loptice,

and it's impressive that babies can learn from imitating us,

i impresivno je to što bebe mogu da uče imitirajući nas,

but we've known those things about babies for a very long time.

ali to sve znamo o bebama još odavno.

The really interesting question

Zaista zanimljivo pitanje

is what happens when we show babies exactly the same thing,

je šta se dešava kada pokažemo bebama isto to,

and we can ensure it's exactly the same because we have a secret compartment

a možemo da obezbedimo da bude baš isto jer imamo tajnu pregradu

and we actually pull the balls from there,

i izvlačimo lopte odatle,

but this time, all we change is the apparent population

ali ovog puta menjamo samo vidljivu populaciju

from which that evidence was drawn.

iz koje se izvlači dokaz.

This time, we're going to show babies three blue balls

Ovoga puta ćemo bebama pokazati tri plave loptice

pulled out of a box of mostly yellow balls,

izvučene iz kutije sa pretežno žutim lopticama,

and guess what?

i pogodite šta?

You [probably won't] randomly draw three blue balls in a row

Verovatno nećete nasumično izvući tri loptice zaredom

out of a box of mostly yellow balls.

iz kutije sa većinom žutim lopticama.

That is not plausibly randomly sampled evidence.

To nije verovatan slučajno uzorkovani dokaz.

That evidence suggests that maybe Hyowon was deliberately sampling the blue balls.

Taj dokazi ukazuje da je možda Jouon namerno uzorkovala plave loptice.

Maybe there's something special about the blue balls.

Možda postoji nešto posebno u vezi sa plavim lopticama.

Maybe only the blue balls squeak.

Možda samo plave loptice pište.

Let's see what the baby does.

Hajde da vidimo šta beba radi.

(Video) HG: See this? (Ball squeaks)

(Video) JG: Vidiš ovo? (Lopta pišti)

See this toy? (Ball squeaks)

Vidiš ovu igračku? (Lopta pišti)

Oh, that was cool. See? (Ball squeaks)

O, to je bilo kul. Vidiš? (Lopta pišti)

Now this one's for you to play. You can go ahead and play.

Ova je za tebe da se igraš. Možeš da se igraš.

(Fussing) (Laughter)

(Beba negoduje) (Smeh)

LS: So you just saw two 15-month-old babies

LŠ: Upravo ste videli dve bebe stare 15 meseci

do entirely different things

koje rade potpuno različite stvari

based only on the probability of the sample they observed.

samo na osnovu verovatnoće uzorka koji su zapazile.

Let me show you the experimental results.

Dozvolite da vam pokažem eksperimentalne rezultate.

On the vertical axis, you'll see the percentage of babies

Na vertikalnoj osi ćete videti procenat beba

who squeezed the ball in each condition,

koje su stiskale loptu u svakoj situaciji, i kao što ćete videti,

and as you'll see, babies are much more likely to generalize the evidence

mnogo je verovatnije da će bebe generalizovati dokaz

when it's plausibly representative of the population

kada verodostojnije predstavlja populaciju

than when the evidence is clearly cherry-picked.

nego kada je očigledno probran.

And this leads to a fun prediction:

A to navodi na zabavno predviđanje:

Suppose you pulled just one blue ball out of the mostly yellow box.

recimo da ste izvukli samo jednu plavu loptu

iz uglavnom žute kutije.

You [probably won't] pull three blue balls in a row at random out of a yellow box,

Verovatno nećete izvući tri plave lopte zaredom iz žute kutije,

but you could randomly sample just one blue ball.

ali biste mogli nasumice uzeti samo jednu plavu loptu.

That's not an improbable sample.

To nije neverovatan uzorak.

And if you could reach into a box at random

A ako posegnete u kutiju nasumice

and pull out something that squeaks, maybe everything in the box squeaks.

i izvučete nešto što pišti, možda sve u kutiji pišti.

So even though babies are going to see much less evidence for squeaking,

Dakle, iako će bebe videti mnogo manje dokaza za pištanje,

and have many fewer actions to imitate

i imati mnogo manje radnji za oponašanje

in this one ball condition than in the condition you just saw,

u situaciji sa jednom loptom nego u situaciji koju ste upravo videli,

we predicted that babies themselves would squeeze more,

predvideli smo da će bebe stiskati više,

and that's exactly what we found.

i to je upravo ono što smo pronašli.

So 15-month-old babies, in this respect, like scientists,

Dakle, bebama od 15 meseci, u ovom pogledu, kao i naučnicima,

care whether evidence is randomly sampled or not,

je bitno da li je dokaz nasumično uzorkovan ili ne,

and they use this to develop expectations about the world:

i one koriste to da stvore očekivanja o svetu:

what squeaks and what doesn't,

šta pišti, a šta ne,

what to explore and what to ignore.

šta istražiti, a šta ignorisati.

Let me show you another example now,

Dozvolite mi da vam sada pokažem još jedan primer,

this time about a problem of causal reasoning.

ovog puta o problemu uzročnog rasuđivanja.

And it starts with a problem of confounded evidence

Počinje problemom zbunjujućeg dokaza,

that all of us have,

koji postoji kod svih nas,

which is that we are part of the world.

a to je da smo deo sveta.

And this might not seem like a problem to you, but like most problems,

I to vam možda ne deluje kao problem ali, kao i većina problema,

it's only a problem when things go wrong.

postaje problem tek kada stvari krenu naopako.

Take this baby, for instance.

Uzmite ovu bebu, na primer.

Things are going wrong for him.

Stvari mu ne polaze za rukom.

He would like to make this toy go, and he can't.

Želeo bi da pokrene ovu igračku, ali ne može.

I'll show you a few-second clip.

Pokazaću vam snimak od nekoliko sekundi.

And there's two possibilities, broadly:

Postoje dve mogućnosti, uglavnom.

Maybe he's doing something wrong,

Možda radi nešto pogrešno,

or maybe there's something wrong with the toy.

ili možda nešto nije u redu sa igračkom.

So in this next experiment,

Dakle, u sledećem eksperimentu,

we're going to give babies just a tiny bit of statistical data

daćemo bebama samo delić statističkih podataka

supporting one hypothesis over the other,

koji podržavaju jednu od hipoteza,

and we're going to see if babies can use that to make different decisions

i videćemo da li bebe mogu to da koriste kako bi donosile različite odluke

about what to do.

o onome što će činiti.

Here's the setup.

Evo postavke.

Hyowon is going to try to make the toy go and succeed.

Jouon će pokušati da pokrene igračku i uspeti u tome.

I am then going to try twice and fail both times,

Ja ću potom pokušati dva puta i oba puta neću uspeti,

and then Hyowon is going to try again and succeed,

zatim će Jouon pokušati ponovo i uspeti,

and this roughly sums up my relationship to my graduate students

i to otprilike rezimira odnos koji imam sa mojim studentima

in technology across the board.

po pitanju svih vrsta tehnologija.

But the important point here is it provides a little bit of evidence

Ali, ono što je ovde važno jeste to da se pruža malo dokaza

that the problem isn't with the toy, it's with the person.

da problem nije sa igračkom, već sa osobom.

Some people can make this toy go,

Neki ljudi mogu da pokrenu ovu igračku,

and some can't.

a neki ne mogu.

Now, when the baby gets the toy, he's going to have a choice.

Sad, kada beba dobije igračku, imaće izbor.

His mom is right there,

Njegova mama je tu pored,

so he can go ahead and hand off the toy and change the person,

tako da može da joj priđe, preda igračku i promeni osobu,

but there's also going to be another toy at the end of that cloth,

ali na kraju te krpe će biti još jedna igračka,

and he can pull the cloth towards him and change the toy.

i on može da povuče krpu ka sebi i promeni igračku.

So let's see what the baby does.

Hajde da vidimo šta će beba uraditi.

(Video) HG: Two, three. Go! (Music)

(Video) JG: Dva, tri. Sad! (Muzika)

LS: One, two, three, go!

LS: Jedan, dva, tri, sad!

Arthur, I'm going to try again. One, two, three, go!

Arture, pokušaću ponovo. Jedan, dva, tri, sad!

YG: Arthur, let me try again, okay?

JG: Arture, dopusti da ja pokušam ponovo, okej?

One, two, three, go! (Music)

Jedan, dva, tri, sad! (Muzika)

Look at that. Remember these toys?

Pogledaj. Sećaš li se tih igračaka?

See these toys? Yeah, I'm going to put this one over here,

Vidiš te igračke? Da, staviću ovu ovde,

and I'm going to give this one to you.

a ovu ću ti dati.

You can go ahead and play.

Možeš da se igraš.

LS: Okay, Laura, but of course, babies love their mommies.

LŠ: Okej, Lora, ali naravno, bebe vole svoje mame.

Of course babies give toys to their mommies

Naravno da bebe daju igračke svojim mamama

when they can't make them work.

kada ne mogu da učine da prorade.

So again, the really important question is what happens when we change

Još jednom, zaista bitno pitanje je šta se dešava kada promenimo

the statistical data ever so slightly.

statističke podatke neznatno.

This time, babies are going to see the toy work and fail in exactly the same order,

Ovog puta, bebe će videti

kako igračka radi i ne radi potpuno istim redosledom,

but we're changing the distribution of evidence.

ali ćemo izmeniti raspodelu dokaza.

This time, Hyowon is going to succeed once and fail once, and so am I.

Ovog puta će Jouon uspeti jednom i neće uspeti jednom, a isto tako ću i ja.

And this suggests it doesn't matter who tries this toy, the toy is broken.

Ovo ukazuje da nije bitno ko isprobava igračku, igračka je pokvarena.

It doesn't work all the time.

Ne radi uvek.

Again, the baby's going to have a choice.

Još jednom, beba će imati izbor.

Her mom is right next to her, so she can change the person,

Njena mama je tu pored, tako da može da promeni osobu,

and there's going to be another toy at the end of the cloth.

i biće tu još jedna igračka na kraju krpe.

Let's watch what she does.

Hajde da vidimo šta će uraditi.

(Video) HG: Two, three, go! (Music)

(Video) JG: Dva, tri, sad! (Muzika)

Let me try one more time. One, two, three, go!

Daj da probam još jednom. Jedan, dva, tri, sad!

Hmm.

Hmmm.

LS: Let me try, Clara.

LŠ: Daj da ja probam, Klara.

One, two, three, go!

Jedan, dva, tri, sad!

Hmm, let me try again.

Hmmm, daj da probam još jednom.

One, two, three, go! (Music)

Jedan, dva, tri, sad! (Muzika)

JG: Staviću ovu ovde,

HG: I'm going to put this one over here,

and I'm going to give this one to you.

a ovu ću ti dati.

You can go ahead and play.

Možeš da se igraš.

(Applause)

(Aplauz)

LS: Let me show you the experimental results.

LŠ: Dozvolite da vam pokažem rezultate eksperimenta.

On the vertical axis, you'll see the distribution

Na vertikalnoj osi ćete videti raspodelu

of children's choices in each condition,

izbora dece u svakoj od situacija,

and you'll see that the distribution of the choices children make

i videćete da raspodela izbora koji deca donose

depends on the evidence they observe.

zavisi od dokaza koje posmatraju.

So in the second year of life,

U drugoj godini života

babies can use a tiny bit of statistical data

bebe mogu da koriste malo statističkih podataka

to decide between two fundamentally different strategies

da bi odabrali između dve fundamentalno različite strategije

for acting in the world:

za postupanje u svetu:

asking for help and exploring.

pitati za pomoć i istraživati.

I've just shown you two laboratory experiments

Upravo sam vam pokazala dva laboratorijska eksperimenta

out of literally hundreds in the field that make similar points,

od bukvalno stotina u ovoj oblasti koji imaju sličnu poentu,

because the really critical point

jer je presudna poenta

is that children's ability to make rich inferences from sparse data

da se sposobnost dece da donose bogate zaključke iz oskudnih podataka

underlies all the species-specific cultural learning that we do.

nalazi u osnovi svakog specifičnog kulturnog učenja.

Children learn about new tools from just a few examples.

Deca uče o novim alatkama na osnovu samo nekoliko primera.

They learn new causal relationships from just a few examples.

Uče nove uzročno-posledične veze iz samo nekoliko primera.

They even learn new words, in this case in American Sign Language.

Čak uče i nove reči, u ovom slučaju američki znakovni jezik.

I want to close with just two points.

Želim da završim sa samo dve poente.

If you've been following my world, the field of brain and cognitive sciences,

Ako ste pratili moj svet, oblast mozga i kognitivne nauke,

for the past few years,

poslednjih nekoliko godina,

three big ideas will have come to your attention.

tri ideje su vam privukle pažnju.

The first is that this is the era of the brain.

Prva je da je ovo era mozga.

And indeed, there have been staggering discoveries in neuroscience:

I zaista, bilo je neverovatnih otkrića u neuronaukama:

localizing functionally specialized regions of cortex,

lokalizacija funkcionalno specijalizovanih regija korteksa,

turning mouse brains transparent,

dovođenje mišjeg mozga u transparentno stanje,

activating neurons with light.

aktiviranje neurona svetlošću.

A second big idea

Druga velika ideja

is that this is the era of big data and machine learning,

je da je ovo era velikih podataka i mašinskog učenja,

and machine learning promises to revolutionize our understanding

a mašinsko učenje obećava revoluciju u našem razumevanju

of everything from social networks to epidemiology.

svega, od društvenih mreža do epidemiologije.

And maybe, as it tackles problems of scene understanding

I možda će nam, kako se bavi problemima razumevanja scene

and natural language processing,

i obrade prirodnog jezika,

to tell us something about human cognition.

reći nešto o ljudskoj kogniciji.

And the final big idea you'll have heard

A poslednja velika ideja koju ćete čuti

is that maybe it's a good idea we're going to know so much about brains

je da je možda dobra ideja da ćemo toliko znati o mozgu

and have so much access to big data,

i imati toliko pristupa velikim podacima,

because left to our own devices,

jer prepušteni sami sebi,

humans are fallible, we take shortcuts,

ljudi su skloni greškama, koristimo prečice,

we err, we make mistakes,

grešimo, pravimo pogreške,

we're biased, and in innumerable ways,

imamo predrasude, i na bezbroj načina,

we get the world wrong.

shvatamo svet pogrešno.

I think these are all important stories,

Mislim da su ovo sve važne priče,

and they have a lot to tell us about what it means to be human,

i imaju mnogo toga da nam kažu o tome šta znači biti čovek,

but I want you to note that today I told you a very different story.

ali želim da primite k znanju da sam vam danas ispričala veoma drugačiju priču.

It's a story about minds and not brains,

To je priča o umu, a ne o mozgu,

and in particular, it's a story about the kinds of computations

a naročito, to je priča o vrstama proračuna

that uniquely human minds can perform,

koje jedino ljudski um može da vrši,

which involve rich, structured knowledge and the ability to learn

što podrazumeva bogato, strukturirano znanje i sposobnost učenja

from small amounts of data, the evidence of just a few examples.

iz malih količina podataka, dokaz samo na osnovu nekoliko primera.

And fundamentally, it's a story about how starting as very small children

I u osnovi, to je priča o tome kako počevši kao veoma mala deca

and continuing out all the way to the greatest accomplishments

i nastavljajući sve do najvećih dostignuća

of our culture,

naše kulture,

we get the world right.

shvatamo svet na pravi način.

Folks, human minds do not only learn from small amounts of data.

Narode, ljudski um ne uči samo iz malih količina podataka.

Human minds think of altogether new ideas.

Ljudski umovi smišljaju potpuno nove ideje.

Human minds generate research and discovery,

Ljudski umovi rađaju istraživanja i otkrića,

and human minds generate art and literature and poetry and theater,

rađaju umetnost i književnost, poeziju i pozorište,

and human minds take care of other humans:

i ljudski umovi se brinu o drugim ljudima:

our old, our young, our sick.

našim starima, mladima, bolesnima.

We even heal them.

Čak ih i lečimo.

In the years to come, we're going to see technological innovations

U godinama koje su pred nama, videćemo tehnološke inovacije

beyond anything I can even envision,

kakve ne mogu ni da zamislim,

but we are very unlikely

ali je veoma malo verovatno

to see anything even approximating the computational power of a human child

da ćemo videti bilo šta čak ni približno moći proračuna ljudskog deteta

in my lifetime or in yours.

tokom mog života ili vašeg.

If we invest in these most powerful learners and their development,

Ako ulažemo u te najmoćnije učenike i njihov razvoj,

in babies and children

u bebe i decu

and mothers and fathers

i majke i očeve

and caregivers and teachers

i staratelje i učitelje

the ways we invest in our other most powerful and elegant forms

onako kako ulažemo u druge naše najmoćnije i najelegantnije oblike

of technology, engineering and design,

tehnologije, inženjeringa i dizajna,

we will not just be dreaming of a better future,

nećemo samo sanjati o boljoj budućnosti,

we will be planning for one.

već ćemo je planirati.

Thank you very much.

Mnogo vam hvala.

(Applause)

(Aplauz)

Chris Anderson: Laura, thank you. I do actually have a question for you.

Kris Anderson: Lora, hvala. Ja zapravo imam jedno pitanje za tebe.

First of all, the research is insane.

Pre svega, istraživanje je suludo.

I mean, who would design an experiment like that? (Laughter)

Mislim, ko bi osmislio takav eksperiment? (Smeh)

I've seen that a couple of times,

Video sam to par puta,

and I still don't honestly believe that that can truly be happening,

i još uvek iskreno ne verujem da se to stvarno dešava,

but other people have done similar experiments; it checks out.

ali i drugi su uradili slične eksperimente; provereno je.

The babies really are that genius.

Bebe su stvarno toliko genijalne.

LS: You know, they look really impressive in our experiments,

LŠ: Znaš, izgledaju stvarno impresivno u našim eksperimentima,

but think about what they look like in real life, right?

ali pomisli na to kako izgledaju u stvarnom životu.

It starts out as a baby.

Počinje kao beba.

Eighteen months later, it's talking to you,

Osamnaest meseci kasnije priča sa vama,

and babies' first words aren't just things like balls and ducks,

a bebine prve reči nisu samo one poput lopte i patke,

they're things like "all gone," which refer to disappearance,

to su i "nema", što se odnosi na nestajanje,

or "uh-oh," which refer to unintentional actions.

ili "o-o", što se odnosi na nenamerne postupke.

It has to be that powerful.

To mora da je toliko moćno.

It has to be much more powerful than anything I showed you.

To mora da je mnogo moćnije od svega što sam vam pokazala.

They're figuring out the entire world.

Oni otkrivaju ceo svet.

A four-year-old can talk to you about almost anything.

Dete od četiri godine može da priča sa vama o gotovo svemu.

(Applause)

(Aplauz)

CA: And if I understand you right, the other key point you're making is,

KA: I ako sam te dobro razumeo, druga tvoja ključna poenta je,

we've been through these years where there's all this talk

protekle su godine sa tom pričom

of how quirky and buggy our minds are,

o tome kako je um uvrnut i blesav,

that behavioral economics and the whole theories behind that

bihejvioralna ekonomija i čitave teorije

that we're not rational agents.

o tome kako nismo razumni izvršioci.

You're really saying that the bigger story is how extraordinary,

Ti u stvari govoriš da je veća priča kako je izvanredan,

and there really is genius there that is underappreciated.

i da je tu zapravo genije koji se potcenjuje.

LS: One of my favorite quotes in psychology

LŠ: Jedan od mojih omiljenih citata u psihologiji

comes from the social psychologist Solomon Asch,

potiče od socijalnog psihologa Solomona Eša,

and he said the fundamental task of psychology is to remove

a on je rekao da je osnovni zadatak psihologije da ukloni

the veil of self-evidence from things.

zavesu samodokazivanja.

There are orders of magnitude more decisions you make every day

Postoji milion redova veličine više odluka koje donosite svakodnevno

koje pravilno shvataju svet.

that get the world right.

You know about objects and their properties.

Imate znanje o predmetima i njihovim osobinama.

You know them when they're occluded. You know them in the dark.

Prepoznajete ih kada su zaklonjeni. Prepoznajete ih u mraku.

You can walk through rooms.

Možete da se krećete kroz prostorije.

You can figure out what other people are thinking. You can talk to them.

Možete da shvatite šta drugi ljudi misle. Možete da razgovarate sa njima.

You can navigate space. You know about numbers.

Možete se kretati u prostoru. Razumete brojeve.

You know causal relationships. You know about moral reasoning.

Razumete uzročno-posledične veze. Razumete moralno rasuđivanje.

You do this effortlessly, so we don't see it,

Radite to bez napora, tako da se ne vidi,

ali to je način na koji poimamo svet, a to je neverovatno dostignuće

but that is how we get the world right, and it's a remarkable

and very difficult-to-understand accomplishment.

i veoma teško za razumevanje.

CA: I suspect there are people in the audience who have

KA: Pretpostavljam da postoje ljudi u publici koji imaju

this view of accelerating technological power

gledište o ubrzanoj tehnološkoj moći

koji bi mogli da ospore tvoju izjavu da nikada za vreme našeg života

who might dispute your statement that never in our lifetimes

will a computer do what a three-year-old child can do,

računar neće uraditi ono što može trogodišnje dete,

but what's clear is that in any scenario,

ali ono što je jasno jeste da u bilo kom scenariju,

our machines have so much to learn from our toddlers.

naše mašine mogu mnogo toga da nauče od naših beba.

LS: I think so. You'll have some machine learning folks up here.

LŠ: Mislim da je tako. Tu su neki ljudi koji se bave mašinama koje uče.

I mean, you should never bet against babies or chimpanzees

Mislim, nikada se ne treba kladiti protiv beba ili šimpanzi

or technology as a matter of practice,

ili tehnologije tek tako,

but it's not just a difference in quantity,

ali nije u pitanju samo razlika u količini,

it's a difference in kind.

već razlika u vrsti.

We have incredibly powerful computers,

Imamo neverovatno moćne kompjutere,

and they do do amazingly sophisticated things,

i oni stvarno obavljaju neverovatno sofisticirane stvari,

often with very big amounts of data.

često sa veoma velikom količinom podataka.

Human minds do, I think, something quite different,

Ljudski um čini, po meni, nešto sasvim drugačije,

and I think it's the structured, hierarchical nature of human knowledge

a mislim da je strukturirana, hijerarhijska priroda ljudskog znanja

that remains a real challenge.

ono što ostaje pravi izazov.

CA: Laura Schulz, wonderful food for thought. Thank you so much.

KA: Lora Šulc, sjajna hrana za misli. Mnogo ti hvala.

LS: Thank you. (Applause)

LŠ: Hvala. (Aplauz)