企业🤖AI智能体构建引擎,智能编排和调试,一键部署,支持私有化部署方案 广告
# 08.教育抵消进化 [TOC=3,5] ## 8 Education counteracts evolution ## 教育抵消进化 ### 8.1 Inefficiency of education ### 教育的低效 > The education system design counters half a billion years of the evolution of the nervous system > > 教育系统的设计违背进化了 5 亿年的神经系统 Ever since the Cambrian Period, animals kept improving their ability to explore and learn from the natural environment. Modern education systems largely toss away all that capital. Very few people responsible for modern schooling are aware of the central problem that makes schools ineffective. Most of kids on this planet are let down or even damaged. If you read those words, shake up everyone around. Time to wake up from that lethargy! 自寒武纪时期以来,动物不断提高其探索和学习自然环境的能力。现代教育系统在很大程度上抛弃了所有老本。负责现代教育的人很少意识到使学校效率低下的核心问题。这个世界上大多数孩子都很失望,甚至受到了伤害。如果你读到这些话,请摇醒周围的每个人。是时候从昏睡中醒来了! If you ask your teacher, headmaster, or even your education minister, chances are pretty high, he or she is unaware of the problem. There is a lot of talk of [reform](https://supermemo.guru/wiki/Reform), emphasis on creativity, independence, [freedom](https://supermemo.guru/wiki/Free_learning), or [problem solving](https://supermemo.guru/wiki/How_to_solve_any_problem%3F). However, the key design flaw puts a spanner in the works even if people involved in running education are preponderantly knowledgeable, and sport good intuition about the ailments of the system. Most educators today score well on knowledge. They score great on their love for children. However, they keep [treading water](https://supermemo.guru/wiki/Why_schools_fail%3F). Every year, schools release masses of kids with knowledge and skills that are a fraction of the human natural potential. 如果你问你的老师、校长,甚至你的教育部长,很可能他或她不知道这个问题。有很多关于[改革](https://supermemo.guru/wiki/Reform)、强调创造力、独立性、[自由](https://supermemo.guru/wiki/Free_learning)或[解决问题](https://supermemo.guru/wiki/How_to_solve_any_problem%3F)的讨论。然而,即使参与运转教育的人员知识渊博,并且对系统的弊病有良好的直觉,关键的设计缺陷也给工作带来了障碍。今天,大多数教育者知识水平很高,他们也很爱孩子。然而,他们仍在[原地踏步](https://supermemo.guru/wiki/Why_schools_fail%3F)。每年,学校都会输出大量的孩子,他们拥有的知识和技能只是人类潜能的一小部分。 ### 8.2 Exploration algorithm ### 探索算法 All effective learning processes in nature are based on exploration. [Exploration](https://en.wikipedia.org/wiki/Exploration_problem) has been modelled accurately in artificial intelligence for decades. It permeates nearly all branches of computer science with various mutations of the search algorithm. For a visualization on a non-linear nature of the [A\* search](https://en.wikipedia.org/wiki/A*_search_algorithm) see this [short video](https://www.youtube.com/watch?v=J-ilgA_XNI0). 自然界中所有有效的学习过程都是基于探索。几十年来,[探索](https://en.wikipedia.org/wiki/Exploration_problem)已在人工智能中准确建模。它通过搜索算法的各种变种渗透到计算机科学的几乎所有分支中。 对于[A\*搜索算法](https://en.wikipedia.org/wiki/A*_search_algorithm)的非线性特性的可视化,请观看这个[短视频](https://www.youtube.com/watch?v=J-ilgA_XNI0)。 However, the [education systems](https://supermemo.guru/wiki/Prussian_education_system) are largely impervious to that realization. Modern schooling does not take into account the well known models. While robotics may use information entropy as guidance to exploration, I propose the concept of [learntropy](https://supermemo.guru/wiki/Learntropy) in reference to human learning. Human brain has evolved mechanisms that make it the best and the most adaptable exploration machine in existence. [Learntropy](https://supermemo.guru/wiki/Learntropy) helps define the optimization criteria that provide guidance in the learning system. Learntropy and the [learn drive](https://supermemo.guru/wiki/Learn_drive) should make it easier to understand effective human learning via exploration. 然而,[教育系统](https://supermemo.guru/wiki/Prussian_education_system)在很大程度上不受这种认识的影响。现代学校教育没有考虑到众所周知的模式。机器人可以使用信息熵作为探索的指导,同时我提出了关于人类学习的[学习熵](https://supermemo.guru/wiki/Learntropy)。人类大脑已经发展出机制,使其成为现存最好、最具适应性的探索机器。[学习熵](https://supermemo.guru/wiki/Learntropy)帮助定义了在学习系统中提供指导的优化标准。学习熵和[学习内驱力](https://supermemo.guru/wiki/Learn_drive)使得理解借助探索的有效人类学习更加容易。 Exploration is pretty intuitive for everyone, however, its importance remains unnoticed by those who are less computer literate. It is therefore helpful to see exploration as a traversal of a walkthrough tree that takes place while assembling a jigsaw puzzle of knowledge. My [jigsaw puzzle metaphor](https://supermemo.guru/wiki/Jigsaw_puzzle_metaphor) explains why the human brain cannot accept knowledge in an arbitrary sequence, e.g. as dished by the [curriculum](https://supermemo.guru/wiki/Curriculum), teacher in a classroom, or even a good textbook. Exploration can be seen as the tree of available routes for a mini-robot responsible for assembling the jigsaw puzzle. The robot will explore individual branches of the tree and look for pieces of the jigsaw that might fit the current status of the puzzle. 探索对每个人来说都是非常直观的,然而,它的重要性仍然被那些不那么懂电脑的人所忽视。因此,将探索视为在组装知识的拼图游戏时发生的演练树的遍历很有帮助。我的[拼图游戏类比](https://supermemo.guru/wiki/Jigsaw_puzzle_metaphor)解释了为什么人类大脑不能接受任意序列的知识,例如: [课程](https://supermemo.guru/wiki/Curriculum),课堂上的老师,甚至是一本好的教科书。探索可以看作是负责组装拼图游戏的迷你机器人的可达路径树。机器人将探索树的各个分支,并寻找可能适合拼图当前状态的拼图片段。 The exploration can then be visualized as a search tree that needs to be traversed to reach a specific point, e.g. an apple on the tree, or the understanding of multiplication. The evolution has designed simple exploratory algorithms, which all look very similar. The explorer tries out individual branches of the tree and backtracks from dead alleys. In addition, the explorer avoids branches that serve penalties and prefers branches that provide reward. In learning, an episode of _Mythbusters_ may be highly rewarding. Similarly, an authoritarian teacher can serve as a harsh penalty independent of the subject matter presented in the classroom. Instead, a few minutes of Facebook under the desk provide a welcome relief with more rewarding branches of the tree to explore. 然后可以将探索可视化为需要遍历以到达特定点的搜索树,例如,树上的苹果,或对乘法的理解。演化设计了简单的探索算法,它们看起来非常相似。探险家尝试树的各个分支,并从死路回溯。此外,探险家避开提供惩罚的分支并且偏好提供奖励的分支。在学习中,一集_流言终结者_可能会非常有价值。同样,专制教师可以作为一种严厉的惩罚,与课堂上的主题无关。于此相反,在桌子下面花几分钟浏览 Facebook 会让你放松,因为你可以探索更多有价值的分支。 The whole process can be compared to an exploration of a labyrinth in a situation when we do not have access to the birds eye view. Rewards and penalties provide [valuations](https://supermemo.guru/wiki/Knowledge_valuation_network) that [guide](https://supermemo.guru/wiki/Learn_drive) the search. Exploration is also known to single celled organisms. It's simplest variant requires no brain. A paramecium uses a few parameters in the environment to seek food. So does a jelly fish. This type of exploration does not need to involve learning. The exploration algorithm itself might be fixed for good and still provide for good survival. 在我们无法进入鸟瞰图的情况下,可以将整个过程与迷宫的探索进行比较。奖励和惩罚提供[指导](https://supermemo.guru/wiki/Learn_drive)搜索的[评估](https://supermemo.guru/wiki/Knowledge_valuation_network)。单细胞生物也知道探索。它最简单的变种不需要大脑。 草履虫在环境中使用一些参数来寻找食物。水母也是如此。 这种类型的探索不需要涉及学习。探索算法本身可能是固定的,以提供良好的生存。 > Efficient learning is based on exploration. While exploring the labyrinth of life, we look for jigsaw pieces of information that match the puzzle of knowledge. As we keep exploring, we build the jigsaw puzzle. This way a rat may explore an island as depicted [here](https://supermemo.guru/wiki/On_the_superiority_of_a_rat_over_a_schooled_human) > > 有效的学习是基于探索的。 在探索生活的迷宫时,我们会寻找与知识谜团相匹配的拼图信息。在我们不断探索的过程中,我们构建了拼图游戏。 [这里](https://supermemo.guru/wiki/On_the_superiority_of_a_rat_over_a_schooled_human)描绘了老鼠可以探索岛屿的方式 ### 8.3 Evolving imitation ### 进化的模仿 500,000 years ago, on a savannah, early Homo sapiens used exploration to learn about plants, animals, food, dangers, etc. The process was not much different from the one used by a [rat on an island](https://supermemo.guru/wiki/On_the_superiority_of_a_rat_over_a_schooled_human). However, humans developed a strong inclination towards imitation. **Imitation provides a shortcut in the exploration tree**. It makes it possible to jump from one branch to another at little cost. Valid imitation leads to a valid incremental inserts in the [jigsaw puzzle](https://supermemo.guru/wiki/Jigsaw_puzzle) of knowledge. 50 万年前,在大草原上,早期的智人使用探索来了解植物、动物、食物、危险等。这个过程与[岛上的老鼠](https://supermemo.guru/wiki/On_the_superiority_of_a_rat_over_a_schooled_human)使用的过程没有多大差别。然而,人类对模仿产生了强烈的倾向。**模仿在探索树中提供了捷径**。它可以以很低的成本从一个分支跳到另一个分支。有效的模仿引导知识[拼图游戏](https://supermemo.guru/wiki/Jigsaw_puzzle)中的有效的增量嵌入。 Imitation is the first step towards perpetuation of knowledge and speeding up the collective exploration algorithm, in which human brains pull together across generations to accumulate mankind's wisdom. If any individual in the group finds a solution to a problem, the others may imitate and traverse the maze without wasting time on exploration of dead branches. This trick is employed by an ant colony too for maximizing the yield in food gathering. However, imitation has reached its acme in the human species. Back in the day, I was good at pulling up on the rod. I was the best in class. Today, in street workout, I feel like an old klutz. All teens are now educated on YouTube. All forms of unthinkable acrobatics become a commonplace. Kids love to imitate in the quest for their goals. 模仿是迈向知识永久化和加速集体探索算法的第一步,在这种算法中,人类的大脑在几代人之间聚集在一起,以积累人类的智慧。如果群体中的任何一个人找到问题的解决方案,其他人就可以模仿并遍历迷宫,并且不会浪费时间去探索死路。这种技巧也被蚁群用来最大化食物采集的产量。然而,模仿已经在人类这里达到了极致。以前,我很擅长拉杆。我是班上最好的。今天,在街头锻炼时,我觉得自己像个老傻瓜。现在所有青少年都在 YouTube 上接受教育。所有形式的不可思议的杂技都变得司空见惯。 孩子们喜欢模仿他们的目标。 ### 8.4 Imitation vs. novelty seeking ### 模仿与新颖寻求 Imitation is great to speed up learning, however, it also carries a risk of failing to explore the entire search space, or failing to adapt to changes in the environment, or to changes in needs. This is why it is beneficial to occasionally explore unexplored branches even when imitation provides a solution. For a population, it is highly advantageous to have a group of avid explorers who will risk their life to find new lands or planets. Their occasional _"Eureka!"_ may provide a new path for the rest of the group. In the area of imitation, humans have spurted well ahead of the apes. Humans explore less and imitate more. Imitation is the first step to the record of history. However, to survive, we need to retain the quest for novelty. 模仿很有助于加快学习速度,但是,它也存在无法探索整个搜索空间或者无法适应环境变化或需求变化的风险。这就是为什么即使模仿提供解决方案,偶尔探索未探索的分支也是有益的。对于群体而言,拥有一群冒着生命危险寻找新的土地或行星的狂热探险者是非常有利的。他们偶尔会有「_尤里卡!_」可以为该群体的其他人提供新的路径。在模仿领域,人类远远领先于猿类。人类探索更少,模仿更多。模仿是历史记录的第一步。 然而,为了生存,我们需要保持对新奇的追求。 ### 8.5 Language, religions, and war ### 语言、宗教和战争 The real breakthrough came with the development of the language and the development of the print. Now we can perpetuate history of discovery and minimize straying. With the fixed record of historic achievement, we need eager explorers even more. We need a bit of collective [forgetting](https://supermemo.guru/wiki/Forgetting), and a great deal of contrarian exploration. This is where we need [ADHD](https://supermemo.guru/wiki/ADHD) kids. Instead of dumbing them down with Ritalin, we should let them explore \(see: [Caging a puppy](https://supermemo.guru/wiki/Would_you_have_a_heart_to_cage_a_puppy%3F)\). They risk their lives for the collective benefit. 真正的突破来自于语言的发展和印刷品的发展。现在,我们可以延续发现的历史,并尽量减少偏离。凭借历史性成就的固定记录,我们需要更多的热切探索者。我们需要一些集体[遗忘](https://supermemo.guru/wiki/Forgetting),以及大量的逆向探索。这是我们需要 [ADHD](https://supermemo.guru/wiki/ADHD) 孩子的地方。我们应该让他们探索(参见:[驯养小狗](https://supermemo.guru/wiki/Would_you_have_a_heart_to_cage_a_puppy%3F)),而不是用利他林让他们变笨。为了集体利益,他们冒着生命危险。 With the arrival of well-established religions, and the advent of massive armies, we have been tempted to improve upon imitation by "rational" design. Religions provide a code of conduct that makes sure the followers imitate the prescription from the gods. Military leaders build armies and drill rules of behaviors that deprive the individual of the self to the benefit of the massive movement of the cannon fodder. A Prussian army behaves like a perfectly-programmed robot in which one soldiers receives the program, and others imitate. 随着成熟宗教的到来,以及大规模军队的出现,我们一直试图通过「理性」设计来改进模仿。宗教提供行为准则,确保追随者模仿神的旨意。军事领导人建立军队并制定行为规则,剥夺个人的自我利益,以利于炮弹的大规模运动。普鲁士军队的行为就像一个完美编程的机器人,其中一名士兵接受该程序,其他人则模仿。 ### 8.6 The advent of mass education ### 大众教育的出现 It is only natural that imitation is used massively in education. The teacher makes the show, and the kids repeat. By mastering calculus, they do not need to retrace the footsteps of Newton. This is fantastically effective. New generations can now focus on solving new problems using previously discovered tools. With the success of early schools, with an increase in literacy, with an increase in numeracy, we could see that education can solve many problems facing humanity. We fell in love with imitation and rushed ahead too far. You can stroke a cat to death, and we nearly did! 在教育中大量使用模仿是很自然的。老师制作演示,孩子们重复。通过掌握微积分,他们不需要追溯牛顿的脚步。这非常有效。新一代现在可以专注于使用以前发现的工具解决新问题。随着早期学校的成功,随着识字率的提高,计算能力的提高,我们可以看到教育可以解决人类面临的许多问题。我们爱上了模仿,并向前冲得太远。你可以抚摸一只猫到死,我们几乎做到了! ### 8.7 Coercive immitation ### 强制的模仿 The next seemingly logical step in improving education is to introduce coercion. If imitation is so effective, why not make it compulsory to the benefit of society? The right to education became universal. Overtime, the right became a duty. The reasoning is similar to the rationale behind vaccination. If we educate most of the citizenry, we can run a new kind of society with a myriad of new rules and new solution that advance the civilization. Importantly, good education serves the individual. If this is good for a citizen and for society, making it compulsory cannot be considered a violation of human rights. In the course of evolution, coercion has also proven efficient and popular. In a shoal of fish, ironically called _"a school"_, not a single one can make a wrong move, or the whole school will turn into a massive freeway pileup. 改善教育的下一个看似合乎逻辑的步骤是引入强制。如果模仿是如此有效,为什么不为了社会利益而强制要求?受教育的权利变得普遍。久而久之,权利成为一种责任。推理类似于疫苗接种背后的理由。 如果我们教育大多数公民,我们就可以通过无数新规则和推进文明的新解决方案来建立一种新型社会。 重要的是,良好的教育服务于个人。如果这对公民和社会都有好处,那么强制它不能被视为侵犯人权。在进化过程中,强制也被证明是有效和受欢迎的。在一群被讽刺地称为「_学校_」的鱼群中,没有一条鱼能走错一步,否则整个鱼群就会变成高速公路上的连环相撞。 ![File:Multi-seat toddler desk as perfect preschool torture device.jpg](https://box.kancloud.cn/6ac84a1fb68a947ef0c42c3960c0f9bf_784x482.jpg) > **Figure:** Multi-seat toddler desk is a preschool torture device. [The brain says "keep exploring"](https://supermemo.guru/wiki/Education_counteracts_evolution). The authoritarian teacher says "Stop moving! Read my lips!". For more see: [Would you have a heart to cage a puppy?](https://supermemo.guru/wiki/Would_you_have_a_heart_to_cage_a_puppy%3F) > > **图:** 多座位幼儿桌是学龄前的折磨设备。 [大脑说「继续探索」](https://supermemo.guru/wiki/Education_counteracts_evolution)。 专制的老师说「别动!注意听!」。 欲了解更多信息:[你会忍心把小狗关在笼子里吗](https://supermemo.guru/wiki/Would_you_have_a_heart_to_cage_a_puppy%3F)? ### 8.8 Resistance to social pressure ### 反抗社会压力 The problem with the transition from imitation to coercive imitation is that it destroys the exploratory algorithm designed by millions of years of evolution. The evolution delivered the most efficient brain machinery responsible for learning. Sometime during the evolution of birds and mammals, populations developed more advanced forms of social life. [Social intelligence is based on learning](https://supermemo.guru/wiki/Optimal_socialization), which can be accelerated with imitation. However, the purpose of social groups is to co-operate and act in harmony. For social life to evolve efficiently, individuals needed to develop mechanisms that would efficiently interweave coercion in the learning system. 从模仿到强制模仿的过渡问题在于它破坏了数百万年进化所设计的探索算法。 进化提供了最有效的学习机器。 在鸟类和哺乳动物的进化过程中,人们开发出更先进的社会生活形式。 [社交智能基于学习](https://supermemo.guru/wiki/Optimal_socialization),可以通过模仿加速学习。 然而,社会团体的目的是合作和协调行动。 为了使社会生活有效地发展,个人需要发展一种机制,使强制与学习系统有效地交织在一起。 If social coercion worked perfectly, it would be like a perfect imitation. It would lead to inhibited exploration that could backfire in the long run. For that, a degree of **resistance to coercion** need to be injected into the efficient algorithm. Resistance occurs when exploration is curbed by social pressure. Social pressure may add to the penalty system, but is treated differently than other penalties such as pain. Non-social penalties may stop an exploration of a branch. Resistance to social penalties will result in backtracking on the exploration tree. **Resistance results in forgoing fruits of exploration**. Fruits are forgone in the name of independence. This gave the biological origin to Zapata's: _"I'd rather die on my feet, than live on my knees"_ 如果社会强制运作得很好,就像一个完美的模仿。 这将导致抑制探索,从长远来看可能适得其反。为此,需要对强制算法注入一定程度的**强制反抗**。 当探索受到社会压力的抑制时,就会出现阻力。社会压力可能会增加惩罚系统,但与其他惩罚如疼痛的处理方式不同。非社会处罚可能会停止对分支的探索。 抵制社会处罚将导致探索树的回溯。**反抗导致放弃探索的成果**。这是以独立的名义放弃的。 这给了 Zapata 的生物学起源:「_我宁愿站着死,也不愿跪着生_」 ### 8.9 Resistance to extrinsic valuation ### 反抗外部评估 In addition to the exploratory value of resistance, the goals of a community and the goals of an individual may often be contradictory. A well-exploring brain cannot accept the situation in which his learning capacity is hijacked to fulfill the goals of another individual: [the bully](https://supermemo.guru/wiki/Bullying). The key to social intelligence is an intelligent resistance mechanism that will prevent extrinsic valuations in the goal tree \(see: [Knowledge valuation network](https://supermemo.guru/wiki/Knowledge_valuation_network)\). The teacher is powerless. She cannot motivate kids [extrinsically](https://supermemo.guru/wiki/Extrinsic_motivation). Persuasion has its limits. 除了反抗的探索价值之外,社区的目标和个人的目标可能经常是矛盾的。 一个训练有素的大脑无法接受他的学习能力被劫持以实现另一个人目标的情况:[欺凌](https://supermemo.guru/wiki/Bullying)。 社交智能的关键是一种智能反抗机制,可以防止目标树中的外部评估(参见:[知识评估网络](https://supermemo.guru/wiki/Knowledge_valuation_network))。老师无能为力。她无法在从[外部](https://supermemo.guru/wiki/Extrinsic_motivation)激励孩子。劝说有其局限性。 ### 8.10 Variations of resistance in a population ### 群体中的反抗变化 For exploratory reasons, and for independence reasons, the resistance is highest for young risk-taking individuals. If a parent tried to force a child to follow a specific exploratory path, she will meet with a degree of resistance. Very often, the child will attempt to do the opposite. This is the evolution's way of saying: "_you guys explored that path, I will gladly risk my life to try something new_". The same resistance algorithm is the key to the failure of schooling. [Coercive learning](https://supermemo.guru/wiki/Coercive_learning) does not work. 出于探索性原因,并且出于独立性原因,年轻冒险者的反抗最高。 如果父母试图强迫孩子遵循特定的探索路径,她会遇到一定程度的反抗。 很多时候,孩子会尝试做相反的事情。这就是进化论的说法:「_你们探索了这条道路,我很乐意冒着生命危险尝试新事物_」。相同的反抗算法是学校教育失败的关键。[强制性学习](https://supermemo.guru/wiki/Coercive_learning)不起作用。 As much as it is the case with imitation, it is beneficial to have a group of individuals in a population that will always rebel and look for their own ways. These are the leaders of revolutions. They will often be expelled from a population. They will often die young. They may be [ostracized like president Trump](https://supermemo.guru/wiki/Mystery_of_Donald_Trump%27s_brain). However, they provide a vital ingredient in the survival of populations and societies. 与模仿的情况一样,在群体中总是反叛并寻找自己的方式是有益的。 这些是革命的领导者。他们经常被驱逐出群体。他们经常英年早逝。 他们可能[像特朗普总统一样被排斥](https://supermemo.guru/wiki/Mystery_of_Donald_Trump%27s_brain)。然而,它们是群体和社会生存的重要组成部分。 Via personality traits, genes have an important impact on exploration, imitation, and resistance. If you hug a puppy, one will cuddle, another will wriggle out to run away. The two puppies may even come from the same litter. 通过个性特征,基因对探索,模仿和反抗具有重要影响。如果你拥抱一只小狗,一个可能会抱抱,另一个可能会蠕动逃跑。这两只小狗甚至可能来自同一窝。 Similarly, children differ by their resistance to coercion and to their sensitivity to peer pressure. Some will always look up to their parents for approval. Each new move they make, they look up. Others will keep exploring without heeding the presence of others. Those unruly kids will react with violent resistance to all forms of imposition. If you tell them, "go right", they will always turn left. This is how they can help the population discover new pathways. 同样,儿童对强制的抵制和对同侪压力的敏感性也不同。 有些人总是希望他们的父母批准。 他们所做的每一个新动作都会请示。 其他人将继续探索,而不会留意他人的存在。 那些不守规矩的孩子会对所有形式的强加做出暴力反抗。 如果你告诉他们,「向右走」,他们将一直向左转。 这就是他们如何帮助群体发现新的途径。 ### 8.11 Wieman drafts the future of exploratory learning ### Wieman 起草了探索性学习的未来 Compulsory schooling tramples upon the natural exploratory algorithm based on the [learn drive](https://supermemo.guru/wiki/Learn_drive). Future education will take a full advantage of exploration. [Carl Wieman](https://supermemo.guru/wiki/Carl_Wieman) is one of those who could see the problem and who devoted his Nobel Prize winnings to provide a light in the tunnel. His [science simulations](https://supermemo.guru/wiki/PhET_simulations) provide a natural exploratory environment for children and scientists alike. 义务教育践踏了基于学习内驱力的内在探索算法。未来的教育将充分利用探索。 [Carl Wieman](https://supermemo.guru/wiki/Carl_Wieman) 是能够看到这个问题的人之一,他将诺贝尔奖奖金用于为这个问题提供线索。他的[科学模拟](https://supermemo.guru/wiki/PhET_simulations)为儿童和科学家提供了一个自然的探索环境。 Wieman noticed that imitation via linear presentation often breaks on a single bad label, or a single gap in knowledge. If the branch reward keeps dropping, the urge to explore decreases. Wieman simulations require few concept labels and minimum prior knowledge. Each simulation is an independent capsule of reality. Wieman simulations are like little virtual worlds of science to explore. This is the best way of exploring scientific models in existence \(in addition to the life in real\). Wieman 注意到,通过线性呈现进行模仿经常会打破一个不好的标签,或者知识中的单一空白。如果分支奖励不断下降,探索的欲望就会降低。Wieman 模拟需要很少的概念标签和最少的先验知识。每个模拟都是现实的独立囊。Wieman 模拟就像探索的小科学虚拟世界。这是探索现存科学模型的最佳方式(除了真实的生活)。 ### 8.12 Incremental reading is based on exploration ### 渐进阅读基于探索 In the domain of written records, the same exploration can be undertaken with [incremental reading](https://supermemo.guru/wiki/Incremental_reading). 在书面记录领域,可以通过[渐进阅读](https://supermemo.guru/wiki/Incremental_reading)进行相同的探索。 > SuperMemo insert. [What is SuperMemo?](https://supermemo.guru/wiki/What_is_SuperMemo%3F) > > SuperMemo 插入。 [什么是 SuperMemo?](https://supermemo.guru/wiki/What_is_SuperMemo%3F) > > In [incremental reading](https://supermemo.guru/wiki/Incremental_reading), a student collects materials for a body of knowledge that serves like a virtual island of knowledge to explore. The exploration then proceeds unhindered with the guidance of the [learn drive](https://supermemo.guru/wiki/Learn_drive). It provides best forms of imitation, and unlimited creative branching. Wieman simulations provide scientific models to explore. [Incremental reading](https://supermemo.guru/wiki/Incremental_reading) makes it easy to assemble the [jigsaw puzzle](https://supermemo.guru/wiki/Jigsaw_puzzle) of knowledge from pieces made ready for consumption via imitation. Via set of priorities and the appropriate timing, well-managed [incremental reading](https://supermemo.guru/wiki/Incremental_reading) can maximize the [reward in learning](https://supermemo.guru/wiki/Pleasure_of_learning). Boring branches can always be deprioritized. They are simply unproductive at the moment. Unproductive branches can wait their turn in exploration > > 在[渐进阅读](https://supermemo.guru/wiki/Incremental_reading)中,学生收集知识主干的材料,这些材料就像一个虚拟的知识岛,可供探索。然后,在[学习内驱力](https://supermemo.guru/wiki/Learn_drive)的指导下,探索不受阻碍地进行。 它提供了最好的模仿形式,以及无限的创意分支。Wieman 模拟提供了探索的科学模型。[渐进阅读](https://supermemo.guru/wiki/Incremental_reading)使我们可以很容易地将知识的[拼图](https://supermemo.guru/wiki/Jigsaw_puzzle)拼装起来,而这些知识片段都是通过模仿准备用于消费。通过优先级和适当的时间安排,良好管理的[渐进阅读](https://supermemo.guru/wiki/Incremental_reading)可以最大限度地提高[学习的奖励](https://supermemo.guru/wiki/Pleasure_of_learning)。无聊的分支总是可以优先考虑。它们目前只是非生产性的。非生产性的分支可以等待他们的探索 The bigger the [knowledge tree](https://supermemo.guru/wiki/Knowledge_tree) the wider the range of reward options. [Curiosity](https://supermemo.guru/wiki/Learn_drive) should be cultivated. Comparing pleasurable learning to a dopamine drip is a highly harmful metaphor. If social media provide the best reward, it only comes from the crippling effects of the environment on the other competing branches of knowledge. If school makes one hate mathematics, it may make one love Instagram that much more. [知识树](https://supermemo.guru/wiki/Knowledge_tree)越大,奖励选项的范围越广。应该培养[好奇心](https://supermemo.guru/wiki/Learn_drive)。把愉快的学习比作多巴胺的滴注是一个非常有害的类比。如果社交媒体提供最好的奖励,那么它只会来自环境对其他竞争知识分支的严重影响。如果学校讨厌数学,那么可能会让人更喜欢 Instagram。 ### 8.13 Two key errors of education systems ### 教育系统中的两个关键错误 **Imitation** and **resistance** are vital parts of the **exploratory algorithm** hard-coded in the human brain. The **first error of education** is to provide a linear [curriculum](https://supermemo.guru/wiki/Curriculum), which is the supposed straight pathway to wisdom \(as opposed to a **set** of **recommendations**\). Curriculum provides imitation with curbs on exploration. However, this is a misguided imitation as it does not take into account the status of the [jigsaw puzzle](https://supermemo.guru/wiki/Jigsaw_puzzle_metaphor) that is being built. It is too easy to make a bridge to nowhere. All kids are to imitate at the same time, at the same pace, and all mutations in the trajectory, instead of improving upon exploration, will gradually lead to the [collapse of imitation](https://supermemo.guru/wiki/Knowledge_crystallization). Your typical school will not let you leave the first grade by taking a kid chemistry exam. You need to do your alphabet and numbers first. The **second error of education** is to make curriculum compulsory. Compulsory education triggers resistance that interferes with imitation and seed chaos into the process. The imitation trajectory is no longer predictable. Pre-designed imitation makes it fall apart. Triggered resistance makes imitation fall apart faster. At first, kids stop exploring, and then they stop imitating. In the end, they may hate school and learning in equal measure. [War of the networks](https://supermemo.guru/wiki/War_of_the_networks) leads to[ learned helplessness](https://supermemo.guru/wiki/Learned_helplessness), apathy and depression. By trying too hard, we are building an unhappy society. **模仿**和**反抗**是人类大脑中硬编码的**探索算法**的重要组成部分。**教育的第一个错误**是提供线性[课程](https://supermemo.guru/wiki/Curriculum),这被认为是智慧的直接途径(与**一系列建议**相对)。课程提供模仿与探索的限制。然而,这是一种被误导的模仿,因为它没有考虑正在构建的[拼图游戏](https://supermemo.guru/wiki/Jigsaw_puzzle_metaphor)的状态。建造一座无处可去的桥太容易了。所有的孩子都要以同样的速度同时模仿,并且轨道中的所有突变,而不是在探索上的改进,将逐渐导致[模仿的崩溃](https://supermemo.guru/wiki/Knowledge_crystallization)。你认知中的典型学校不会让你通过化学考试后就离开一年级。你需要先学你的字母和数字。**教育的第二个错误**是使课程成为必修课。义务教育会引发反抗,干扰模仿,并在过程中制造混乱。模仿轨道不再可预测。预先设计的模仿使它崩溃。触发反抗使模仿更快地崩溃。一开始,孩子们停止探索,然后他们停止模仿。最后,他们可能会同等地讨厌学校和学习。[网络之战](https://supermemo.guru/wiki/War_of_the_networks)导致[习得性无助](https://supermemo.guru/wiki/Learned_helplessness)、冷漠和沮丧。通过过头的努力,我们正在建设一个不幸的社会。 ![](https://box.kancloud.cn/ef3e42ebe26d71d50381c2c270e17419_392x428.jpg) > **Figure:** Even if children [hate school](https://supermemo.guru/wiki/Why_kids_hate_school) they accept their fate via [learned helplessness](https://supermemo.guru/wiki/Learned_helplessness). This destroys their [natural love of learning](https://supermemo.guru/wiki/Pleasure_of_learning), suppresses natural creativity, deprives of independence, and may lead to mental disorders later in life \(picture source: Wikipedia\) > > **图:** 即使儿童[讨厌学校](https://supermemo.guru/wiki/Why_kids_hate_school),他们也会通过[习得性无助](https://supermemo.guru/wiki/Learned_helplessness)来接受自己的命运。这破坏了他们[天生的求知欲](https://supermemo.guru/wiki/Pleasure_of_learning),压制了天生的创造力,剥夺了他们的独立性,并可能在以后的生活中导致精神障碍(图片来源:维基百科) ### 8.14 Overoptimization ### 过度优化 The education systems employ the tools that are basically functional. However, in the never ending quest for better performance, we [arrived to the point of absurdity](https://supermemo.guru/wiki/PISA). What used to work in the past has now turned into a monster that keeps the kids unhappy and poorly educated. Many educators seem to follow linear reasoning: if a glass of water rejuvenates, drinking a tub might perhaps make someone immortal. This is a maximization based on wrong criteria. Instead, we need to [optimize education](https://supermemo.guru/wiki/Optimization_of_education). [Free learning](https://supermemo.guru/wiki/Free_learning) is the simplest approach to optimizing education. It costs nothing and comes naturally. All healthy kids are born with a powerful exploratory algorithm hard-coded in their brain. If schools [damage the control systems](https://supermemo.guru/wiki/War_of_the_networks) involved in that algorithm, they harm societies, and lead to a multitude of aberration that torment modern populations. 教育系统采用基本功能的工具。然而,在永无止境的追求更好表现的过程中,我们[达到了荒谬的程度](https://supermemo.guru/wiki/PISA)。过去曾经起作用的东西现在变成了一个怪物,让孩子们不高兴,受到不充分的教育。许多教育工作者似乎都遵循线性推理:如果喝一杯水能恢复活力,那么喝一个浴缸可能会让某人不朽。这是基于错误标准的最大化。相反,我们需要[优化教育](https://supermemo.guru/wiki/Optimization_of_education)。[自由学习](https://supermemo.guru/wiki/Free_learning)是优化教育的最简单方法。它没有任何成本,而且很自然。所有健康的孩子天生就有一种强大的探索算法,在他们的大脑中进行了硬编码。如果学校**破坏**了该算法中的[控制系统](https://supermemo.guru/wiki/War_of_the_networks),它们就会危害社会,并导致折磨现代人的大量精神失常。 > Compulsory schooling triggers mechanisms that block exploratory learning that underlies efficient acquisition of knowledge and skills > > 义务教育触发了阻碍探索性学习的机制,而探索性学习是有效获取知识和技能的基础 ### 8.15 The Grand Education Reform ### 教育大改革 The solutions to the woes of education have been known for decades now. Solutions are cheap and simple. However, the set of entrenched myths of education still keeps the reformers at bay. 数十年来,人们已经知道解决教育困境的方法。 解决方案既便宜又简单。然而,一系列根深蒂固的错误教育观念仍然让改革者陷入困境。 Here is my proposition: [Grand Education Reform](https://supermemo.guru/wiki/Grand_Education_Reform). 这是我的主张:[教育大改革](https://supermemo.guru/wiki/Grand_Education_Reform)。