What is so special about the human brain?

What is so special about the human brain? Why is it that we study other animals instead of them studying us? What does a human brain have or do that no other brain does? When I became interested in these questions about 10 years ago, scientists thought they knew what different brains were made of. Though it was based on very little evidence, many scientists thought that all mammalian brains, including the human brain, were made in the same way, with a number of neurons that was always proportional to the size of the brain. This means that two brains of the same size, like these two, with a respectable 400 grams, should have similar numbers of neurons. Now, if neurons are the functional information processing units of the brain, then the owners of these two brains should have similar cognitive abilities. And yet, one is a chimp, and the other is a cow. Now maybe cows have a really rich internal mental life and are so smart that they choose not to let us realize it, but we eat them. I think most people will agree that chimps are capable of much more complex, elaborate and flexible behaviors than cows are. So this is a first indication that the "all brains are made the same way" scenario is not quite right.

But let's play along. If all brains were made the same way and you were to compare animals with brains of different sizes, larger brains should always have more neurons than smaller brains, and the larger the brain, the more cognitively able its owner should be. So the largest brain around should also be the most cognitively able. And here comes the bad news: Our brain, not the largest one around. It seems quite vexing. Our brain weighs between 1.2 and 1.5 kilos, but elephant brains weigh between four and five kilos, and whale brains can weigh up to nine kilos, which is why scientists used to resort to saying that our brain must be special to explain our cognitive abilities. It must be really extraordinary, an exception to the rule. Theirs may be bigger, but ours is better, and it could be better, for example, in that it seems larger than it should be, with a much larger cerebral cortex than we should have for the size of our bodies. So that would give us extra cortex to do more interesting things than just operating the body. That's because the size of the brain usually follows the size of the body. So the main reason for saying that our brain is larger than it should be actually comes from comparing ourselves to great apes. Gorillas can be two to three times larger than we are, so their brains should also be larger than ours, but instead it's the other way around. Our brain is three times larger than a gorilla brain.

The human brain also seems special in the amount of energy that it uses. Although it weighs only two percent of the body, it alone uses 25 percent of all the energy that your body requires to run per day. That's 500 calories out of a total of 2,000 calories, just to keep your brain working.

So the human brain is larger than it should be, it uses much more energy than it should, so it's special. And this is where the story started to bother me. In biology, we look for rules that apply to all animals and to life in general, so why should the rules of evolution apply to everybody else but not to us? Maybe the problem was with the basic assumption that all brains are made in the same way. Maybe two brains of a similar size can actually be made of very different numbers of neurons. Maybe a very large brain does not necessarily have more neurons than a more modest-sized brain. Maybe the human brain actually has the most neurons of any brain, regardless of its size, especially in the cerebral cortex. So this to me became the important question to answer: how many neurons does the human brain have, and how does that compare to other animals?

Now, you may have heard or read somewhere that we have 100 billion neurons, so 10 years ago, I asked my colleagues if they knew where this number came from. But nobody did. I've been digging through the literature for the original reference for that number, and I could never find it. It seems that nobody had actually ever counted the number of neurons in the human brain, or in any other brain for that matter.

So I came up with my own way to count cells in the brain, and it essentially consists of dissolving that brain into soup. It works like this: You take a brain, or parts of that brain, and you dissolve it in detergent, which destroys the cell membranes but keeps the cell nuclei intact, so you end up with a suspension of free nuclei that looks like this, like a clear soup. This soup contains all the nuclei that once were a mouse brain. Now, the beauty of a soup is that because it is soup, you can agitate it and make those nuclei be distributed homogeneously in the liquid, so that now by looking under the microscope at just four or five samples of this homogeneous solution, you can count nuclei, and therefore tell how many cells that brain had. It's simple, it's straightforward, and it's really fast. So we've used that method to count neurons in dozens of different species so far, and it turns out that all brains are not made the same way. Take rodents and primates, for instance: In larger rodent brains, the average size of the neuron increases, so the brain inflates very rapidly and gains size much faster than it gains neurons. But primate brains gain neurons without the average neuron becoming any larger, which is a very economical way to add neurons to your brain. The result is that a primate brain will always have more neurons than a rodent brain of the same size, and the larger the brain, the larger this difference will be. Well, what about our brain then? We found that we have, on average, 86 billion neurons, 16 billion of which are in the cerebral cortex, and if you consider that the cerebral cortex is the seat of functions like awareness and logical and abstract reasoning, and that 16 billion is the most neurons that any cortex has, I think this is the simplest explanation for our remarkable cognitive abilities. But just as important is what the 86 billion neurons mean. Because we found that the relationship between the size of the brain and its number of neurons could be described mathematically, we could calculate what a human brain would look like if it was made like a rodent brain. So, a rodent brain with 86 billion neurons would weigh 36 kilos. That's not possible. A brain that huge would be crushed by its own weight, and this impossible brain would go in the body of 89 tons. I don't think it looks like us.

So this brings us to a very important conclusion already, which is that we are not rodents. The human brain is not a large rat brain. Compared to a rat, we might seem special, yes, but that's not a fair comparison to make, given that we know that we are not rodents. We are primates, so the correct comparison is to other primates. And there, if you do the math, you find that a generic primate with 86 billion neurons would have a brain of about 1.2 kilos, which seems just right, in a body of some 66 kilos, which in my case is exactly right, which brings us to a very unsurprising but still incredibly important conclusion: I am a primate. And all of you are primates.

And so was Darwin. I love to think that Darwin would have really appreciated this. His brain, like ours, was made in the image of other primate brains.

So the human brain may be remarkable, yes, but it is not special in its number of neurons. It is just a large primate brain. I think that's a very humbling and sobering thought that should remind us of our place in nature.

Why does it cost so much energy, then? Well, other people have figured out how much energy the human brain and that of other species costs, and now that we knew how many neurons each brain was made of, we could do the math. And it turns out that both human and other brains cost about the same, an average of six calories per billion neurons per day. So the total energetic cost of a brain is a simple, linear function of its number of neurons, and it turns out that the human brain costs just as much energy as you would expect. So the reason why the human brain costs so much energy is simply because it has a huge number of neurons, and because we are primates with many more neurons for a given body size than any other animal, the relative cost of our brain is large, but just because we're primates, not because we're special.

Last question, then: how did we come by this remarkable number of neurons, and in particular, if great apes are larger than we are, why don't they have a larger brain than we do, with more neurons? When we realized how much expensive it is to have a lot of neurons in the brain, I figured, maybe there's a simple reason. They just can't afford the energy for both a large body and a large number of neurons. So we did the math. We calculated on the one hand how much energy a primate gets per day from eating raw foods, and on the other hand, how much energy a body of a certain size costs and how much energy a brain of a certain number of neurons costs, and we looked for the combinations of body size and number of brain neurons that a primate could afford if it ate a certain number of hours per day.

And what we found is that because neurons are so expensive, there is a tradeoff between body size and number of neurons. So a primate that eats eight hours per day can afford at most 53 billion neurons, but then its body cannot be any bigger than 25 kilos. To weigh any more than that, it has to give up neurons. So it's either a large body or a large number of neurons. When you eat like a primate, you can't afford both.

One way out of this metabolic limitation would be to spend even more hours per day eating, but that gets dangerous, and past a certain point, it's just not possible. Gorillas and orangutans, for instance, afford about 30 billion neurons by spending eight and a half hours per day eating, and that seems to be about as much as they can do. Nine hours of feeding per day seems to be the practical limit for a primate.

What about us? With our 86 billion neurons and 60 to 70 kilos of body mass, we should have to spend over nine hours per day every single day feeding, which is just not feasible. If we ate like a primate, we should not be here.

How did we get here, then? Well, if our brain costs just as much energy as it should, and if we can't spend every waking hour of the day feeding, then the only alternative, really, is to somehow get more energy out of the same foods. And remarkably, that matches exactly what our ancestors are believed to have invented one and a half million years ago, when they invented cooking. To cook is to use fire to pre-digest foods outside of your body. Cooked foods are softer, so they're easier to chew and to turn completely into mush in your mouth, so that allows them to be completely digested and absorbed in your gut, which makes them yield much more energy in much less time. So cooking frees time for us to do much more interesting things with our day and with our neurons than just thinking about food, looking for food, and gobbling down food all day long.

So because of cooking, what once was a major liability, this large, dangerously expensive brain with a lot of neurons, could now become a major asset, now that we could both afford the energy for a lot of neurons and the time to do interesting things with them. So I think this explains why the human brain grew to become so large so fast in evolution, all of the while remaining just a primate brain. With this large brain now affordable by cooking, we went rapidly from raw foods to culture, agriculture, civilization, grocery stores, electricity, refrigerators, all of those things that nowadays allow us to get all the energy we need for the whole day in a single sitting at your favorite fast food joint. So what once was a solution now became the problem, and ironically, we look for the solution in raw food.

So what is the human advantage? What is it that we have that no other animal has? My answer is that we have the largest number of neurons in the cerebral cortex, and I think that's the simplest explanation for our remarkable cognitive abilities. And what is it that we do that no other animal does, and which I believe was fundamental to allow us to reach that large, largest number of neurons in the cortex? In two words, we cook. No other animal cooks its food. Only humans do. And I think that's how we got to become human.

Studying the human brain changed the way I think about food. I now look at my kitchen, and I bow to it, and I thank my ancestors for coming up with the invention that probably made us humans. Thank you very much. (Applause)

Translated by Dasheng Chen

Reviewed by Zhiting Chen

对于人类的大脑来说，什么是最特别的？ 我们为什么要研究其他的动物？ 而不是他们研究我们？ 有没有什么是人类大脑可以做的 而其他的大脑不能做的？ 十年前，我对这些问题产生兴趣 科学家认为他们知道大脑不同的组成部分 即使那些都是基于一些非常小的证据， 很多科学界认为所有的哺乳动物的大脑， 包括人类的大脑 是由同样的东西组成， 一系列的神经元和大脑的大小 总是成为成正比 这个以为着，两个同样大小的大脑 就看这两个，有大概400克， 应该有同样数量的神经元 如果我们说神经元是大脑 功能信息处理的单位 然后，这些两个大脑各自的拥有者 应该有的相近的认知能力。 然而，其中一个是黑猩猩， 和另一种是一头母牛。 现在也许牛有很丰富的 精神生活，并且很聪明 他们选择不会让我们意识到这一点， 但我们吃他们。 我想大多数人会同意 相比于牛来说，黑猩猩具有更加复杂 和细节，以及灵活的行为。 所以这是第一个征兆显示 "所有的大脑的构成方式都是一样的" 不是特别正确的。

我们不妨继续这个比喻 如果所有的大脑由相同的方式组成 你要将大脑大小不同的动物拿来做比较, 大一点的大脑对于小一点的大脑来说 应该具有等多的神经元， 并且越大的大脑， 它的主人应该更有认知能力。 所以最大的脑应该具有 最好的认知能力。 这里有个坏消息要说： 我们的大脑，不是最大的一个。 看起来相当棘手。 我们的大脑重量 1.2 和 1.5 公斤之间 但大象脑子称之间四、 五公斤 鲸鱼大脑可重达 9 公斤 这就是为什么科学家们用来诉诸说 我们的大脑肯定是非常特别地 来解释我们的认知能力。 它必须是真的非同寻常， 一个例外情况。 他们的大脑可能会更大，但我们的更好， 并且我们的可以变得更加好，例如， 在这似乎比它应该更大 与很多大大脑皮层比我们应该有的 对于我们的身体的大小。 因此，这样将会给我们额外皮质 去做更多有趣的事情，而不是只主管我的身体。 这是因为大脑的大小 通常遵循身体的大小。 所以说这句话的主要原因就是 我们的大脑已经比它应有的大小大出好多了 其实是来自于比较我们自己 和巨猿。 大猩猩可以是比我们大两到三倍 因此，他们的大脑也应该比我们的大 但相反，它的大脑比我们的小。 我们的大脑是比大猩猩脑大三倍。

人类的大脑似乎也很特别 在于它使用的能源的量。 虽然它的重量只有全身的2%， 但是它单独使用 你身体每天所需能量的25% 2000卡路里里面的500卡路里 仅仅只是为了让你的大脑的工作。

所以，人类的大脑已经比应有的大小大出很多了 它消耗了更加多的能量，对于他应该消耗的来说 所以它很特殊。 这是这个故事开始让我感到疑惑。 在生物学中，我们发现规则 一般情况下，适用于所有的动物和生活 那么为什么进化的规则 应用到了其他生物，但不是我们？ 也许这个问题是一个基础的假设 所有的大脑都是由相同的方式组成。 也许两个大小相似的大脑 实际上可以由数目不同的神经元组成。 也许是一个非常大的大脑 不一定需要具有更多的神经元 对于一个更具规模适度的大脑。 也许，相比于其他大脑来说，人类的大脑具有 最多的神经元，无论其规模大小， 尤其是在大脑皮质中。 所以这对我来说成为了 对于这个答案来说最重要的问题 人类大脑有多少神经元 和其他动物比较起来怎样？

现在，您可能会听到或读到过 我们有 1000 亿个神经元 10 年前，我就问我的同事们 他们是否知道这个数字是从哪里来的。 但没人知道。 我通过深入细致的研究文献 试图找出该数字的原始参考 但我找不到它。 这似乎其实没有人去真正的数过 人类大脑里面的神经元数量， 或者其他任何的大脑里面神经元的数量。

所以我想出了我自己的方式来统计大脑里面的细胞， 它最重要的一个步骤是 那大脑溶解在试剂里。 它的工作方式： 你带一个大脑或部分的大脑， 让它溶化在洗涤剂里， 洗涤剂会破坏细胞膜 但保持细胞核不变， 所以你最终会看到细胞核悬浮 看起来像这样， 像一碗清汤。 这汤包含所有细胞核 曾经是一个老鼠的大脑。 现在，这个液体的魅力在于因为它是液体， 可以摇晃它，并使这些细胞核 在液体中均匀分布 所以，现在通过在显微镜下观察 只需四个或五个 这种细胞核均匀分布的样本 你可以数数原子核，并能够知道 有多少脑细胞了。 这样很简单，非常直接， 并且，这样的速度也很快。 所以我们使用了这种方法来计数神经元 目前为止计数了几十个不同的物种， 结果发现，所有的大脑 是由不同的方式组成的。 以啮齿类动物和灵长类动物为例： 在大一些的啮齿类动物大脑里， 神经元的平均尺寸在增加， 所以，大脑非常迅速地膨胀 尺寸增长比获得神经元更加快 但灵长类动物的大脑得到神经元 而不需要神经元平均尺寸增大 这是一种非常经济的方法 在你大脑里添加神经元 结果是灵长类动物的大脑 和啮齿动物相同大小的脑相比, 总有更多的神经元 和大脑越大 这种差异性將越來越大 那么,我们的大脑又是如何呢? 我们发现我们平均有 860 亿个神经元 其中有 160 亿个存在于大脑皮质里 而且如果你认为大脑皮质 是下列功能的基础， 意识,逻辑和抽象推理， 那么，这160 亿个神经元 对大脑皮质来说是最重要的 我想这是最简单的說明 对于我们卓越的认知能力。 但,就和 860 亿神经元的意义一样重要 因为我们发现了 大脑的大小和其神经元数目之间的关联 是可以用数学模型来计算的 我们可以计算出人类大脑 如果被作成像是一个啮齿类动物的大脑会如何。 所以，有着 860 亿神经元的啮齿类动物大脑 将重 36 公斤。 那是不可能的。 巨大的大脑將会被 它自己本身的重量所挤压， 这个不可能的大脑將会适用于 一个89吨的身体 。 我不认为它看起来跟我们一样。

因此，这已经带給我们的一个非常重要的结论， 我们不是啮齿类动物。 人类的大脑不是一个大老鼠的大脑。 和一只老鼠相比，我们看起来可能特别些，是的 但这样的比较不公平， 我们知道我们不是啮齿类动物。 我们是灵长类动物， 因此和其他的灵长类动物做正确的比较。 這里还有，如果你自己来算一下， 你会发现一般的的灵长类动物 有着860 亿神经元 将会有一个约 1.2 公斤的大脑 这似乎刚刚好， 在大约66 公斤身体中， 在我的研究是完全正确的 我们对此一点儿也不惊讶 但这仍是令人难以置信的重要结论： 我是灵长类动物。 而且,你们全部都是灵长类动物。

达尔文也是。 我认为达尔文会很欣赏这个研究结论。 他的大脑，像我们一样， 和其它灵长类动物的大脑一样。

所以，人类的大脑可能是显著的,是的 但它特别之处不在于其神经元的数目。 它只是一个大的灵长类动物的大脑。 我想这是个非常谦卑和发人深省的思考 提醒了我们在自然界的位置。

为什么我们的大脑消耗这么大的能量，然后呢？ 有些人已经想通了， 人类的大脑能消耗多少能量 和其他物种的大脑消耗多少能量 而现在,我们知道 每个大脑中有多少神经元，我们可以算一算。 结果是,人类 和其他物种的大脑消耗相同的能量 平均每日，每十亿个神经元消耗六个卡路里的热量。 所以一个大脑全部精力的消耗 是简单的，线性函数 可计算其神经元的数目 结果发现，人类的大脑 消耗的能量就跟你预期的一样 所以，为什么人类的大脑 消耗如此多的能量的原因容易理解， 因为它有大量的神经元， 还有,因为我们是灵长类动物 在一个特定大小的身体里, 人类比任何其他的动物，有着更多的神经元 我们的大脑的相对消耗量是很大的， 但只是因为我们是灵长类动物，不是因为我们是特別的。

最后一个问题， 我们为什么有数量巨大的神经元， 特别是，如果巨猿 比我们更大， 那为什么他们沒有一个有着更多神经细胞,比我们还大的大脑呢? 当我们意识到这个有多珍貴, 在大脑中有大量的神经元，我想， 可能有一个简单的原因。 他们只是不能负担 一个大型的身体以及多量的神经细胞共存所需的能量。 所以我们算过 一方面，我们计算着 一个灵长类动物每一天获取多少能量 借由吃生的食物， 另一方面，有多少能量 在一个特定大小的身体中消耗 和有多少能量在含有特定数量神经元的大脑中消耗 然后我们找到一个结合 身体尺寸和大脑神经元数目 灵长类动物能夠负担得起 如果它每天花一定数量的时间在吃上面。

然后, 我们发现到 因为神经细胞太珍貴了， 有一个权衡出先在身体大小和神经元的數量之间。 所以, 一个灵长类动物每天吃八个小时 能够负担最多 530 亿神经元， 但是，它的身体不能 超过25 公斤。 任何比这更重的重量 它必须放弃一些神经细胞。 所以,它不是一个庞大身体 就是一个大数量的神经元。 当你吃的像灵长类动物时， 你不能负担任何一方。

一个方法超出此种代谢限制是 每一天花更多的时间在吃饭上面， 但是，这样是很危险的 某种程度上，它是不可能发生的。 大猩猩和红毛猩猩，例如， 能夠负担约 300 亿个神经元 它们每天花八个半小时吃饭， 这似乎是它们所能做的一切。 9 个小时的每日饲喂 是灵长类动物的实际限制。

那我们呢? 根据我们的860 亿神经元 和 60 到 70 公斤的体重， 我们应该要花超过九个小时 每天在吃饭上 这是不可行的。 如果我们像灵长类动物一样吃， 我们不应该在这里。

我们怎么在这里呢？ 好吧，如果我们的大脑消耗的能量跟他 应该所消耗的一致，并且如果我们不能把每天 醒着的时间都花在吃饭上， 然后这个唯一的选择，真的， 是以某种方式 从相同的食品上获取更多的能量。 引人注目的是，这和我们的祖先 早已发现的完全匹配 在一百五十万年前， 他们发明了烹饪。 要做饭就得使用火 在食物进入你身体前预先消化食物。 煮熟的食物比较软，所以他们更容易嚼烂 在你的嘴里完全粉碎 这就使他们能够完全被消化 并且被你的肠道所吸收， 这使得它们在更少的时间内带来更多能源。 所以烹饪让我们有更多的时间， 去用我们的神经元 做更多有趣的事情 而不是只想着食物， 寻找食物，并吞噬下食物 花去整整一天。

所以由于烹饪，曾是 一个沉重的负担，这个大， 且极其昂贵的，有着一大堆神经元的大脑， 现在能够成为一项重要资产， 现在，我们可以为给予神经元提供能量 并且把时间用在去做一些有趣的事情。 因此，我认为这就解释了为什么人类的大脑 在进化过程中增长如此之快， 当所有剩余的只是一个灵长类动物的大脑。 与这么大的大脑现在可以负担地起烹饪， 我们去迅速地进步着，从生食到文化， 农业、 文明、 杂货店、 电力、 冰箱、 所有这些事物，现今 使我们能够得到我们需要的所有能量 一整天所需的所有能量 来自于你最喜欢的快餐食品。 所以曾是一种解决办法 现在成为了这一问题， 讽刺的是，我们在生食中寻找解决方案。

那么人类的优势是什么？ 我们有什么东西 是其他动物没有的吗？ 我的答案是我们有最大数量 神经元存在于大脑皮层中， 并且我认为这能最简单地解释 为什么我们有卓越的认知能力。 还有就是我们做了什么是没有其他动物做的， 我认为 使我们能够达到这么多的， 最多数量的神经元存在于大脑皮质中的基本原因是 两个词，我们做饭。 没有其他动物烹饪它们的食物。只有人类才做。 而且我认为，这就是我们是怎么成为人类的。

研究人类大脑改变我对食物的想法。 现在我看看自己的厨房， 然后我向它鞠躬， 然后我谢谢我的祖先想到 这个有可能使我们成为人类的发明。 谢谢。 （掌声）