But even ask it to show you its scratch pad work.

而是能够让它向你展示它的工作流程。

OK, so we have scratch pads for the brain. Now we need a scratch pad for AI.

现在我们有了大脑的草稿纸，我们需要拿到 AI 的草稿纸。

But like scratch pads in brains, that's not something we can actually see, right?

但是，大脑中的“草稿本”不是我们能随便看到的东西，对吧？

Well, it turns out that we can kind of see scratch pads in brains.

实际上，我们可以在一定程度上“看到”大脑中的草稿。

The scratch pad tells us a little bit about how the brain represents the language.

这些草稿本向我们透露了一些大脑表示语言的方法。

And those brain images are like fuzzy, out-of-focus pictures of the scratch pad of the brain.

这些脑成像的图片就像是模糊、失焦的草稿本照片。

And we would like to know what is the scratch pad of a neural network?

我们想要知道神经网络的草稿纸长什么样？

And everything they have to remember in order to do this task has to be written on that little scratch pad.

完成这项任务需要的所有记忆都必须记在小草稿本上。

Now we need a scratch pad for AI.

现在我们需要 AI 的便笺簿。

Again, this scratch pad prediction task works.

这个草稿预测任务得到了同样的结果。

OK, so we have scratch pads for the brain.

好的，所以我们有大脑的便笺簿。

Now we have two scratch pads, one from the brain and one from AI.

现在我们有了两张草稿纸，一张来自大脑，一张来自人工智能。

To illustrate that, let's say inside the minds of our two people, inside of our two rooms, is a little scratch pad.

为了说明这一点，假设在两个房间里的两个人，两个人的头脑中各有一个小草稿本。

We're able to predict the neural network scratch pad from the brain and vice versa.

我们可以从大脑图像预测神经网络，反过来也可以。

We need to see its scratch pad, and we need to be able to compare it to the scratch pad of somebody who actually understands language.

我们需要看到它的草稿本，并将其与真正理解语言的人类的草稿本进行比较。

Though these scratch pad prediction tasks show above-chance accuracy, the underlying correlations are still pretty weak.

尽管这些草稿预测任务表现出高于随机的准确率，两者底层的相关性仍然非常弱。

Tasks that challenge working memory, the mental " scratch pad" we use to temporarily store phone numbers and grocery lists, are especially vulnerable to pressure.

工作记忆是我们头脑中的“速写板”， 我们用它短暂存储电话号码和购物清单等。对于高度要求工作记忆的任务来说，压力对其影响尤为严重。

Still, these scratch pad prediction experiments have held up — multiple brain imaging experiments, multiple neural networks.

尽管如此，这些草稿预测实验仍然站得住脚——多个大脑成像结果，多个神经网络模型。

If we could see what was written on that scratch pad, we would be able to tell how different their approach to the task is.

如果我们能看到写在草稿本上的东西，我们就能知道他们完成任务的方法有什么不同。

It's going to look at the neural network scratch pad for a particular word and try to predict the brain scratchpad.

它会检查神经网络对特定单词的草稿，并预测大脑的草稿。