有益的掙扎,與不該卸載的思考 — Productive Struggle & Cognitive Offloading
繼 HITL、Scaffolding 之後,再分享兩個詞:Productive Struggle(有益的掙扎)與 Cognitive Offloading(認知卸載)。那段孩子卡住的時間,正是大腦真正在學習的時間。而那條「何時該讓孩子自己想、何時才給答案」的線,需要一個人替他拿捏。
繼 HITL、Scaffolding 之後,今天想再分享兩個詞——
**Productive Struggle**,和 **Cognitive Offloading**。
聽起來很學術,但我想說的,其實很日常。
---
## 一、有益的掙扎(Productive Struggle)
大家知道 Productive Struggle 是什麼意思嗎?
翻成中文,大概是「有益的掙扎」。
它指的是——學生在學習時,那段「卡住、想不出來、有點挫折」的時間。
這段時間,看起來很沒效率。孩子皺著眉、停在那裡、試了又錯。
但教育研究發現一件事:
**正是這段掙扎的時間,大腦在真正地學習。**
當一個人卡住、然後努力試著突破,他的大腦在建立新的連結、在把知識變成自己的。這個過程不舒服,但這個不舒服,是有產值的。所以叫「有益的」掙扎。
這就帶出一個,很多人會忽略的問題。
當孩子一卡住,AI 可以在三秒內給出完整答案。孩子拿到答案了,作業完成了,看起來很順利。
但那段「該由他自己掙扎」的時間,被跳過了。那段最有產值的時間,消失了。
這不是說 AI 不好。
是說——有益的掙扎,跟「卡太久而放棄」之間,有一條線。
孩子在線的這一邊,該讓他自己走;快要跨過那條線、要放棄了,才需要支撐。
**而那條線在哪裡,AI 不知道。知道的,是那個正在看著孩子的人。**
---
## 二、認知卸載(Cognitive Offloading)
再來說第二個詞——**Cognitive Offloading**,認知卸載。
意思是——把本來該由大腦做的事,「卸載」給外部的工具去做。
這件事,其實我們一直在做。
以前我們會背電話號碼,現在交給手機;以前心算,現在交給計算機。把這些事卸載出去,不是壞事。大腦的空間有限,把雜事交出去,我們才有餘力想更重要的事。
所以 Cognitive Offloading 本身,是中性的。
**真正的關鍵,在於——你卸載的,是「什麼」。**
你還記得,以前是怎麼找答案的嗎?
更早的年代,遇到不會的,要去圖書館,翻目錄、一頁一頁查。後來有了 Google,我們打字搜尋,跳出一整頁結果——然後得自己判斷:哪個可信?哪個正確?哪個才是我要的?
那段「找」的時間,看起來慢,看起來麻煩。但中間那段篩選、比對、判斷,其實就是大腦在運轉、在練習思考。
教育研究也指出:人在尋找資訊的過程中,會用到一整套的批判性思考能力。
而現在,值得我們留意的是——
當孩子遇到不會的,AI 可以在三秒內,給出一個完整、漂亮的答案。
這時候被卸載出去的,已經不是「背電話號碼」這種雜事了。被卸載出去的,是「判斷、篩選、整合」——**是思考本身。**
這就是 Cognitive Offloading 真正要小心的地方。
卸載雜事,讓我們更有餘力。卸載思考,我們就少了練習的機會。
工具會一直進步,這是好事。圖書館換成 Google,Google 換成 AI,我們不需要抗拒。
但我們得分得清楚——哪些可以交給工具,哪些,要留給自己的大腦練。
**而這條線,什麼時候該讓孩子自己想、什麼時候才給他答案——需要一個人,替他拿捏。**
---
---
## The Struggle That Builds, and the Thinking We Shouldn't Outsource
*Productive Struggle & Cognitive Offloading — What AI Can't Replace*
---
Following our earlier discussions on HITL and Scaffolding, I'd like to share two more concepts today —
**Productive Struggle**, and **Cognitive Offloading**.
They sound academic, but what I want to say is actually quite everyday.
---
### Part One: Productive Struggle
Do you know what Productive Struggle means?
It refers to those moments in learning when a student gets stuck — can't figure it out, feels a little frustrated, tries and fails again.
This period looks inefficient. The child furrows their brow, sits there, tries again and again.
But educational research has found something important:
**It is precisely during this struggle that the brain is truly learning.**
When someone gets stuck and pushes through, the brain is building new connections — turning knowledge into something that belongs to them. The process is uncomfortable, but that discomfort is productive. That's why it's called *productive* struggle.
This raises a question many people overlook.
The moment a child gets stuck, AI can deliver a complete answer in three seconds. The child gets the answer, the homework is done, everything looks smooth.
But that window of time — the one where the child was supposed to struggle on their own — has been skipped. The most productive window, gone.
This isn't saying AI is bad.
It's saying — between productive struggle and "stuck so long they give up," there is a line.
On this side of the line, the child should walk on their own. Only when they're about to cross it — about to give up — do they need support.
**Where that line is, AI doesn't know. The person who knows is the one watching the child.**
---
### Part Two: Cognitive Offloading
Now for the second concept — **Cognitive Offloading**.
It means outsourcing what the brain would normally do to an external tool.
We've always done this.
We used to memorize phone numbers; now we let our phones do it. We used to do mental math; now we use calculators. Offloading these things isn't bad. The brain has limited capacity — handing off the small stuff frees us to think about what matters more.
So Cognitive Offloading itself is neutral.
**The real question is: what are you offloading?**
Do you remember how we used to find answers?
In earlier times, when you didn't know something, you went to the library — flipped through the catalog, searched page by page. Then came Google — you typed a query, got a page of results, and had to decide for yourself: which source is reliable? Which answer is correct? Which one is actually what I need?
That process of searching looked slow, looked tedious. But the filtering, comparing, and judging in between — that was the brain working, practicing how to think.
Educational research confirms: the act of searching for information engages a full set of critical thinking skills.
And now, here's what's worth paying attention to —
When a child doesn't know something, AI can produce a complete, polished answer in three seconds.
What's being offloaded now isn't something trivial like memorizing a phone number. What's being offloaded is judgment, filtering, synthesis — **thinking itself.**
That's the real concern with Cognitive Offloading.
Offloading the small stuff gives us more capacity. Offloading thinking takes away our practice.
Tools will keep improving — that's a good thing. Libraries gave way to Google, Google is giving way to AI. We don't need to resist this.
But we do need to be clear about one thing: what can we hand to the tools, and what should we keep for our own minds to work through?
**And when it comes to children — when to let them think it through, and when to step in with an answer — that judgment belongs to a person. Someone who is there, watching, and knows.**
---
*cramschool.ai — 為教育現場設計的 AI 教學平台 | contact@cramschool.ai*
**Productive Struggle**,和 **Cognitive Offloading**。
聽起來很學術,但我想說的,其實很日常。
---
## 一、有益的掙扎(Productive Struggle)
大家知道 Productive Struggle 是什麼意思嗎?
翻成中文,大概是「有益的掙扎」。
它指的是——學生在學習時,那段「卡住、想不出來、有點挫折」的時間。
這段時間,看起來很沒效率。孩子皺著眉、停在那裡、試了又錯。
但教育研究發現一件事:
**正是這段掙扎的時間,大腦在真正地學習。**
當一個人卡住、然後努力試著突破,他的大腦在建立新的連結、在把知識變成自己的。這個過程不舒服,但這個不舒服,是有產值的。所以叫「有益的」掙扎。
這就帶出一個,很多人會忽略的問題。
當孩子一卡住,AI 可以在三秒內給出完整答案。孩子拿到答案了,作業完成了,看起來很順利。
但那段「該由他自己掙扎」的時間,被跳過了。那段最有產值的時間,消失了。
這不是說 AI 不好。
是說——有益的掙扎,跟「卡太久而放棄」之間,有一條線。
孩子在線的這一邊,該讓他自己走;快要跨過那條線、要放棄了,才需要支撐。
**而那條線在哪裡,AI 不知道。知道的,是那個正在看著孩子的人。**
---
## 二、認知卸載(Cognitive Offloading)
再來說第二個詞——**Cognitive Offloading**,認知卸載。
意思是——把本來該由大腦做的事,「卸載」給外部的工具去做。
這件事,其實我們一直在做。
以前我們會背電話號碼,現在交給手機;以前心算,現在交給計算機。把這些事卸載出去,不是壞事。大腦的空間有限,把雜事交出去,我們才有餘力想更重要的事。
所以 Cognitive Offloading 本身,是中性的。
**真正的關鍵,在於——你卸載的,是「什麼」。**
你還記得,以前是怎麼找答案的嗎?
更早的年代,遇到不會的,要去圖書館,翻目錄、一頁一頁查。後來有了 Google,我們打字搜尋,跳出一整頁結果——然後得自己判斷:哪個可信?哪個正確?哪個才是我要的?
那段「找」的時間,看起來慢,看起來麻煩。但中間那段篩選、比對、判斷,其實就是大腦在運轉、在練習思考。
教育研究也指出:人在尋找資訊的過程中,會用到一整套的批判性思考能力。
而現在,值得我們留意的是——
當孩子遇到不會的,AI 可以在三秒內,給出一個完整、漂亮的答案。
這時候被卸載出去的,已經不是「背電話號碼」這種雜事了。被卸載出去的,是「判斷、篩選、整合」——**是思考本身。**
這就是 Cognitive Offloading 真正要小心的地方。
卸載雜事,讓我們更有餘力。卸載思考,我們就少了練習的機會。
工具會一直進步,這是好事。圖書館換成 Google,Google 換成 AI,我們不需要抗拒。
但我們得分得清楚——哪些可以交給工具,哪些,要留給自己的大腦練。
**而這條線,什麼時候該讓孩子自己想、什麼時候才給他答案——需要一個人,替他拿捏。**
---
---
## The Struggle That Builds, and the Thinking We Shouldn't Outsource
*Productive Struggle & Cognitive Offloading — What AI Can't Replace*
---
Following our earlier discussions on HITL and Scaffolding, I'd like to share two more concepts today —
**Productive Struggle**, and **Cognitive Offloading**.
They sound academic, but what I want to say is actually quite everyday.
---
### Part One: Productive Struggle
Do you know what Productive Struggle means?
It refers to those moments in learning when a student gets stuck — can't figure it out, feels a little frustrated, tries and fails again.
This period looks inefficient. The child furrows their brow, sits there, tries again and again.
But educational research has found something important:
**It is precisely during this struggle that the brain is truly learning.**
When someone gets stuck and pushes through, the brain is building new connections — turning knowledge into something that belongs to them. The process is uncomfortable, but that discomfort is productive. That's why it's called *productive* struggle.
This raises a question many people overlook.
The moment a child gets stuck, AI can deliver a complete answer in three seconds. The child gets the answer, the homework is done, everything looks smooth.
But that window of time — the one where the child was supposed to struggle on their own — has been skipped. The most productive window, gone.
This isn't saying AI is bad.
It's saying — between productive struggle and "stuck so long they give up," there is a line.
On this side of the line, the child should walk on their own. Only when they're about to cross it — about to give up — do they need support.
**Where that line is, AI doesn't know. The person who knows is the one watching the child.**
---
### Part Two: Cognitive Offloading
Now for the second concept — **Cognitive Offloading**.
It means outsourcing what the brain would normally do to an external tool.
We've always done this.
We used to memorize phone numbers; now we let our phones do it. We used to do mental math; now we use calculators. Offloading these things isn't bad. The brain has limited capacity — handing off the small stuff frees us to think about what matters more.
So Cognitive Offloading itself is neutral.
**The real question is: what are you offloading?**
Do you remember how we used to find answers?
In earlier times, when you didn't know something, you went to the library — flipped through the catalog, searched page by page. Then came Google — you typed a query, got a page of results, and had to decide for yourself: which source is reliable? Which answer is correct? Which one is actually what I need?
That process of searching looked slow, looked tedious. But the filtering, comparing, and judging in between — that was the brain working, practicing how to think.
Educational research confirms: the act of searching for information engages a full set of critical thinking skills.
And now, here's what's worth paying attention to —
When a child doesn't know something, AI can produce a complete, polished answer in three seconds.
What's being offloaded now isn't something trivial like memorizing a phone number. What's being offloaded is judgment, filtering, synthesis — **thinking itself.**
That's the real concern with Cognitive Offloading.
Offloading the small stuff gives us more capacity. Offloading thinking takes away our practice.
Tools will keep improving — that's a good thing. Libraries gave way to Google, Google is giving way to AI. We don't need to resist this.
But we do need to be clear about one thing: what can we hand to the tools, and what should we keep for our own minds to work through?
**And when it comes to children — when to let them think it through, and when to step in with an answer — that judgment belongs to a person. Someone who is there, watching, and knows.**
---
*cramschool.ai — 為教育現場設計的 AI 教學平台 | contact@cramschool.ai*