Check this out: Here's a grid, nothing special, just a basic grid, very grid-y. But look closer, into this white spot at the center where the two central vertical and horizontal lines intersect. Look very closely. Notice anything funny about this spot? Yeah, nothing. But keep looking. Get weird and stare at it. Now, keeping your gaze fixed on this white spot, check what's happening in your peripheral vision. The other spots, are they still white? Or do they show weird flashes of grey? Now look at this pan for baking muffins. Oh, sorry, one of the cups is inverted. It pops up instead of dipping down. Wait, no spin the pan. The other five are domed now? Whichever it is, this pan's defective. Here's a photo of Abraham Lincoln, and here's one upside down. Nothing weird going on here. Wait, turn that upside down one right side up. What have they done to Abe? Those are just three optical illusions, images that seem to trick us. How do they work? Are magical things happening in the images themselves? While we could certainly be sneaking flashes of grey into the peripheral white spots of our animated grid, first off, we promise we aren't. You'll see the same effect with a grid printed on a plain old piece of paper. In reality, this grid really is just a grid. But not to your brain's visual system. Here's how it interprets the light information you call this grid. The white intersections are surrounded by relatively more white on all four sides than any white point along a line segment. Your retinal ganglion cells notice that there is more white around the intersections because they are organized to increase contrast with lateral inhibition. Better contrast means it's easier to see the edge of something. And things are what your eyes and brain have evolved to see. Your retinal ganglion cells don't respond as much at the crossings because there is more lateral inhibition for more white spots nearby compared to the lines, which are surrounded by black. This isn't just a defect in your eyes; if you can see, then optical illusions can trick you with your glasses on or with this paper or computer screen right up in your face. What optical illusions show us is the way your photo receptors and brain assemble visual information into the three-dimensional world you see around you, where edges should get extra attention because things with edges can help you or kill you. Look at that muffin pan again. You know what causes confusion here? Your brain's visual cortex operates on assumptions about the lighting of this image. It expects light to come from a single source, shining down from above. And so these shading patterns could only have been caused by light shining down on the sloping sides of a dome, or the bottom of a hole. If we carefully recreate these clues by drawing shading patterns, even on a flat piece of paper, our brain reflexively creates the 3D concave or convex shape. Now for that creepy Lincoln upside down face. Faces trigger activity in areas of the brain that have specifically evolved to help us recognize faces. Like the fusiform face area and others in the occipital and temporal lobes. It makes sense, too, we're very social animals with highly complex ways of interacting with each other. When we see faces, we have to recognize they are faces and figure out what they're expressing very quickly. And what we focus on most are the eyes and mouth. That's how we figure out if someone is mad at us or wants to be our friend. In the upside down Lincoln face, the eyes and mouth were actually right side up, so you didn't notice anything was off. But when we flipped the whole image over, the most important parts of the face, the eyes and mouth, were now upside down, and you realized something fishy was up. You realized your brain had taken a short cut and missed something. But your brain wasn't really being lazy, it's just very busy. So it spends cognitive energy as efficiently as possible, using assumptions about visual information to create a tailored, edited vision of the world. Imagine your brain calling out these edits on the fly: "Okay, those squares could be objects. Let's enhance that black-white contrast on the sides with lateral inhibition. Darken those corners! Dark grey fading into light grey? Assume overhead sunlight falling on a sloping curve. Next! Those eyes look like most eyes I've seen before, nothing weird going on here." See? Our visual tricks have revealed your brain's job as a busy director of 3D animation in a studio inside your skull, allocating cognitive energy and constructing a world on the fly with tried and mostly -- but not always -- true tricks of its own.
看過來! 這有一張網格圖 沒什麼特別就是普通的格子 但仔細看中心的白點 就是兩條平行和垂直線相交的點 再靠近一點看 這個點有比較特別嗎? 呃...沒有 但繼續盯著它 開始有點怪怪的 現在請把目光定在這個點上 用餘光看看附近的點 其他的點還是白的嗎? 還是閃著詭異的灰色 再看這個烤瑪芬的烤盤 喔有其中一個好像反了 應該要凹下去結果它凸起來 但把它轉過來看 其他五個才是反的吧? 反正就是 這個烤盤有問題 這裡有林肯的照片 還有另一張是反著放的 應該沒什麼奇怪的吧 但把反的照片擺正 他們對林肯幹了什麼好事? 上述只是三個「錯視」的例子 我們會被所見欺騙 這是怎麼回事 是這幾張圖有什麼古怪嗎? 我們是可以把影片裡的白點附近 偷偷混進灰點 但我們先澄清絕對沒有 你自己用紙印一樣的圖 出來也會看到一樣的效果 事實上它只是普通的格子 但大腦的視覺系統可不這麼認為 他們是這麼解讀 網格上的光學訊息: 交叉上的白點 被相對較白的四條線包圍 其他的交叉點也是如此 視網膜神經節細胞發現 交叉點附近有更白的部分 因為神經的「側抑制」會增加對比 增加對比意味著 能更清楚發現界線 眼睛和腦的進化 就是為了看見這些差異 視網膜神經節細胞 不把焦點放在交叉點上 因為會受到周圍白點 更多的「側抑制」影響 相較於被黑色包圍的白線 這也不能說是眼睛的問題 因為就算你戴著眼鏡、 用紙本或電腦螢幕看圖 都有這樣的錯視效果 錯視讓我們發現 我們的感光受器和大腦 如何把生活周遭的光學訊息 整合在一起 通常這些邊緣或界線 需要格外留意 因為它們要不能幫你 就是會害死你 再重新看看瑪芬烤盤 知道哪裡怪怪的了嗎? 大腦的視覺皮層 會自己假想一個光源 它假設光是單一方向、 從圖的上方往下照 所以陰影會在突起的下方 或是凹陷的上方 只要我們畫好陰影的位置 就算是在平面上 我們的腦會自然地 製造出凸起或凹陷的錯覺 至於詭異的林肯顛倒照片 「臉」的影像會啟動大腦中 特殊進化來辨識臉部的區域 例如梭狀臉區(FFA) 或枕葉和顳葉的某些區域 這也很合理 因為我們是社會性的動物 具有高度複雜化的互動模式 當我們看到臉 必須要認出「那是臉」 然後趕快理解 他們的表情在說什麼 最主要先看眼睛和嘴巴 我們藉此理解人家在不爽我們 還是想跟我們當朋友 在反過來的林肯中 眼睛和嘴巴其實是被擺正的 所以你不會覺得有什麼問題 但當我們把照片轉正 臉上最重要的部份 也就是眼睛和嘴巴反過來了 你就會發現哪裡有問題 你發現你的腦抄捷徑 結果漏掉一些細節 但你的腦不是想偷懶 而是因為他太忙了 所以要儘可能有效率地耗費能量 運用理所當然的假設 理解、創造出完美的世界 想像你的腦有以下的飛快對話: 「嗯~這些方格可能蠻重要的」 「那我們用『側抑制』加強黑白的對比」 「把其他角落變黑」 「深灰漸淡成淺灰?」 「假裝光從上面來製造出陰影, 下一個!」 「這些眼睛跟平常一樣 沒什麼問題」 看到了嗎? 這些錯視的把戲 揭露大腦在你頭殼下的工作室 像3D動畫導演的忙碌工作 要有效的分配精力 迅速地建構出周遭的世界 用那些屢試不爽-- 雖然有時候會出錯的方法