That highlighted portion below is like, “Go ahead and surrender. AI’s
incursion into schools is already a given victory to the corporate
interests.” Of course, there was never any doubt. All edtech falls into
this.
 |
| Canva AI Magic Media generated image |
That aside, wouldn’t it be cool to explore AI through the lens of
Hattie’s Visible Learning? I also examine some of my own assertions. Let’s take a look.
The Original Article
Teaching students how to use ChatGPT properly is a critical skill that schools should incorporate into their curriculum. Education should prepare students to function effectively later in their lives by teaching them how to learn and how to use the tools of modern society.
There is no way to undo the technological innovations that ChatGPT has brought, and the engine is certainly not the last AI breakthrough. Students will have to be able to integrate AI into their work throughout their lives. Using ChatGPT takes practice; phrasing prompts and devising creative uses for the AI requires skill.
It is important to acknowledge that overreliance on ChatGPT can harm students’ learning.
Source: https://huntnewsnu.com/75784/editorial/chatgpt-the-write-tool-for-classroom-success/
My first draft of a response via Facebook:
Your original writing contrasts the methodical nature of critical thinking in writing with the potential pitfalls of relying on AI for learning and writing. It suggests that writing is a cognitive exercise that transfers knowledge from short-term to long-term memory and that AI should be used only to reinforce existing writing schemas.
You advocate for the exclusion of technology from K-8 education until critical thinking skills are established, and for teaching AI within a framework that promotes independent thought to avoid indoctrination.
That’s pretty accurate. What bothers me about Facebook pieces is that they lack evidence or citations most of the time. Could AI quickly provide research that supports or negates the assertions I made in my response?
Assertions I Made
Here are the assertions I make, numbered for easier follow up:
- Writing is a cognitive exercise that transfers knowledge from short-term to long-term memory
- AI should be used only to reinforce existing writing schemas
- Exclude AI technology from K-8 education until critical thinking skills are established
- Teach AI within a framework that promotes independent thought to avoid indoctrination
To test these assertions, I will include the following prompt followed by the statement shown above:
What research is there that supports or negates this assertion: [assertion]
What will these AIs say in regards to the information? Since ChatGPT and Claude aren’t connected to the web, I’m going to rely on Microsoft CoPilot and Perplexity.ai with copilot turned on.
Assertion #1: Writing Facilitates Knowledge Transfer
Let’s see…Microsoft CoPilot says:
Writing is indeed a fascinating cognitive process that plays a crucial role in memory formation. Let’s delve into the research to explore both sides of this assertion:
-
Supporting Evidence:
- A study conducted by researchers at the University of Tokyo found that writing on physical paper leads to enhanced brain activity when recalling information an hour later. The unique combination of complex, spatial, and tactile information associated with handwriting on paper appears to contribute to improved memory. In fact, paper contains more one-of-a-kind information that aids in stronger memory recall. Volunteers who used paper for note-taking completed tasks about 25% faster than those using digital tablets or smartphones. The tangible permanence, irregular strokes, and uneven shape of paper seem to enhance memory retention³.
- Multisensory intake (including writing) and positive social interactions play a role in long-term memory. When the brain retrieves stored information, it activates multiple neural networks. Writing engages both motor skills and cognitive processes, reinforcing memory pathways⁴.
-
Contrary Evidence:
- While writing is beneficial, it’s essential to consider the context. Digital tools have their advantages, such as efficiency and convenience. However, they may not provide the same depth of memory enhancement as writing on paper.
- Neuroscientists continue to explore the intricate mechanisms of memory formation. For instance, recent research at Harvard identified a mechanism in adult mouse hippocampal neurons critical for memory consolidation and recall, but it doesn’t specifically focus on writing⁵.
In summary, writing—especially on physical paper—can indeed facilitate knowledge transfer from short-term to long-term memory. So, the next time you’re jotting down notes, consider reaching for that trusty pen and paper! 📝🧠
Source: Conversation with Bing, 2/25/2024 (1) Study shows stronger brain activity after writing on paper than on …. https://www.sciencedaily.com/releases/2021/03/210319080820.htm. (2) Writing and the Brain: Neuroscience Shows the Pathways to Learning. https://archive.nwp.org/cs/public/print/resource/3555. (3) How neurons form long-term memories — Harvard Gazette. https://news.harvard.edu/gazette/story/2020/12/how-neurons-form-long-term-memories/. (4) undefined. https://www.foxnews.com/politics/nikki-haley-says-texas-can-secede-from-the-united-states-thats-their-decision-to-make. (5) undefined. https://flip.it/oYRDBO. (6) Elon Statement on Writing Transfer – Center for Engaged Learning. https://www.centerforengagedlearning.org/publications/elon-statements/elon-statement-on-writing-transfer/
So is my assertion accurate? Yes.
Assertion #2: Use AI to Reinforce Existing Writing Schemas
Again using Microsoft CoPilot:
The role of AI in writing is a topic of growing interest, and it’s essential to consider both perspectives. Let’s explore the research on this assertion:
Supporting Evidence:
- AI-powered writing tools have the potential to significantly reduce manuscript preparation time for researchers. These tools leverage deep learning to assist with tasks such as spellchecks, chatbots, and even auto-generated tweets. By automating certain aspects of writing, researchers can focus more on content creation and analysis².
- Schema theory, rooted in cognitive science, emphasizes the importance of organized mental structures (schemata) for knowledge retention. External cognitive strategies, including writing, play a role in developing and modifying schemata. Writing, whether by hand or digitally, contributes to memory consolidation and understanding of new information. AI tools that aid in writing can align with this schema-based approach¹.
Contrary Evidence:
- AI tools are not limited to reinforcing existing writing schemas. They can also generate novel content, provide suggestions, and enhance creativity. For instance, AI-driven language models like GPT-4 can produce original text, simulate different writing styles, and offer fresh perspectives.
- Researchers and writers should strike a balance between using AI to reinforce existing knowledge structures and allowing it to contribute new ideas. Overreliance on existing schemas may hinder creativity and limit exploration of alternative viewpoints.
In summary, while AI can certainly reinforce existing writing patterns, it also has the potential to expand horizons and inspire innovative content. Researchers should embrace AI as a versatile tool that complements human creativity rather than merely reinforcing established norms⁴.
Source: Conversation with Bing, 2/25/2024 (1) AI writing tools promise faster manuscripts for researchers. https://www.nature.com/nature-index/news/artificial-intelligence-writing-tools-promise-faster-manuscripts-for-researchers. (2) The Use of Schema Theory in Learning, Design, and Technology. https://link.springer.com/article/10.1007/s11528-018-0319-0. (3) The best AI tools for research papers and academic research (Literature …. https://academiainsider.com/ai-tools-for-research-papers-and-academic-research/. (4) Best AI Tools for Research Paper Writing – SciSpace by Typeset. https://typeset.io/resources/ai-tools-for-research-paper-writing/. (5) The Future of Search: AI, Machine Learning, & Schema Markup. https://www.schemaapp.com/schema-markup/the-future-of-search-ai-machine-learning-schema-markup/.
Ok, for this one, it seems the matter is up for debate. I decided to run it through my ChatGPT Critical Analysis Tool (requires ChatGPT Plus), which relies on Melanie Trecek-King’s FLOATER) to see what it would say:
| FLOATER Criterion | Score (1-5) | Comments |
|---|---|---|
| Falsifiability | 4 | The claim is structured for potential falsification but creativity aspects may be hard to quantify. |
| Logic | 4 | Logical structure is solid, but relies on some assumptions not fully examined. |
| Objectivity | 3 | Needs more critical assessment of sources and potential biases. |
| Alternative Explanations | 3 | Briefly considers other factors but doesn’t fully explore them. |
| Tentative Conclusions | 4 | Presents conclusions as tentative, aligning with skepticism but could explicitly state conditions. |
| Evidence | 3 | Includes relevant sources but lacks deep evaluation of evidence quality. |
| Replicability | 3 | Implies benefits are widespread without sufficient emphasis on consistent replicability. |
It’s obvious then that this assertion is half-baked. More evidence is required.
Assertion #3:Exclude AI technology from K-8 education until critical thinking skills are established
What research is there that supports or negates this assertion: Exclude AI technology from K-8 education until critical thinking skills are established
CoPilot says:
The integration of AI technology in K-8 education is a topic that warrants thoughtful consideration. Let’s explore the research to understand the implications of introducing AI to young learners:
Supporting Evidence:
- Generative AI tools, such as chatbots, can serve as valuable pedagogical aids. Rather than fearing plagiarism, educators should harness AI to enhance student learning. These tools provide immediate and nonjudgmental responses, allowing learners to reflect freely on their thoughts and concepts. AI doesn’t replace instructors but becomes an additional tool for teaching, inspiring, and guiding students. By using AI, educators can focus on developing critical thinking processes¹.
- Ethical dilemmas are a crucial aspect of critical thinking. AI can be employed to present scenarios, engage students in thought challenges, and encourage them to analyze and discuss complex issues. The combination of AI-generated content and classroom discussions fosters deeper understanding and critical thinking skills¹.
Contrary Evidence:
- Overreliance on AI tools might hinder the development of independent critical thinking. If students become accustomed to immediate AI-generated answers, they may rely less on their own reasoning and problem-solving abilities.
- Balancing AI integration with traditional teaching methods is essential. Educators should ensure that AI doesn’t replace opportunities for students to think critically, analyze evidence, and form their own conclusions.
In summary, rather than excluding AI technology outright, educators should thoughtfully incorporate it into K-8 education. By using AI as a complementary tool, we can nurture critical thinking skills while maintaining a balance between technological assistance and independent thought¹.
Source: Conversation with Bing, 2/25/2024 (1) How to Evaluate Critical Thinking in the Age of AI | AACSB. https://www.aacsb.edu/insights/articles/2024/02/how-to-evaluate-critical-thinking-in-the-age-of-ai. (2) Critical Imaginaries and Reflections on Artificial Intelligence and …. https://link.springer.com/article/10.1007/s42438-019-00046-x. (3) Artificial intelligence (AI) learning tools in K-12 education: A …. https://link.springer.com/article/10.1007/s40692-023-00304-9. (4) Enhancing Computational Thinking Skills Through Artificial Intelligence …. https://link.springer.com/article/10.1007/s11191-022-00392-6.
NO,
AI should not necessarily be excluded from K-8 education until critical
thinking skills are established, but it should be integrated with
careful consideration of the developmental stages and educational goals
for students. Here’s the reasoning:
Supporting Development:
Research in developmental psychology indicates that critical thinking
skills develop over time and can be nurtured through various educational
interventions. Thoughtfully designed AI applications can potentially
support and enhance the development of these skills rather than hinder
them.
Pedagogical Integration:
Educational technology research shows that when technology is
integrated into the curriculum to support pedagogical goals, it can have
a positive impact on learning outcomes. AI can be a tool to foster
critical thinking if it is used in a way that encourages students to
question, analyze, and evaluate information.
Digital Literacy:
Digital literacy is an essential skill in the modern world. Introducing
AI technology in a controlled and educational manner can help students
understand the workings and implications of AI, preparing them for a
future where AI will be ubiquitous.
Personalized Learning:
AI has the potential to personalize learning experiences, providing
students with individualized challenges and support. This can help
students at various stages of critical thinking development to progress
at their own pace and in ways that are most effective for them.
Ethical and Data Literacy:
By introducing AI in K-8 education, educators have the opportunity to
teach students about ethical considerations, data privacy, and the
societal impact of technology. This can be an important part of
developing critical thinking skills, as students learn to consider the
broader implications of technology use.
Teacher and Student Preparedness:
The effectiveness of AI in education depends on the preparedness of
teachers to integrate it into their teaching and the readiness of
students to engage with it. With proper training and resources, teachers
can use AI as a tool to enhance critical thinking rather than replace
it.
Equity in Education:
Excluding AI from K-8 education could widen the digital divide, as
students without exposure to AI may be less prepared for a future that
will be heavily influenced by AI technologies. Introducing AI in a way
that is accessible and equitable can help mitigate this risk.
In conclusion, the key is not to exclude AI from K-8
education but to use it judiciously and as part of a broader educational
strategy that prioritizes the development of critical thinking skills.
The focus should be on how AI is implemented, ensuring that it is a
complement to education that engages students actively and thoughtfully
rather than a passive tool that could potentially lead to over-reliance
or hinder the development of independent problem-solving abilities.
What would using AI in a way that complements education, engaging students actively and thoughtfully?
To align the strategies for integrating AI into K-8 education with the high-effect size instructional strategies from the Visible Learning MetaX database, we can create a table that matches each AI integration strategy with relevant instructional strategies that have been shown to have a high effect on student learning.
Below is a table that makes these connections [NOTE: Miguel here. I went in and added the up to date effect size and link from the Visible-Learning MetaX database since the cited text is from 2009 and lots has changed since then).
| AI Integration Strategy in K-8 Education | Visible Learning MetaX Teaching Strategies | Citation |
|---|---|---|
| Personalized Learning | Feedback (Tasks & Processes) d=0.63 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Interactive Learning Environments | Inquiry-based teaching d=0.53 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Support for Diverse Learning Needs | Response to Intervention d=0.73 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Enhanced Assessment Tools | Effects of testing d=0.63 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Development of Digital Literacy | Explicit teaching strategies d=0.63 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Facilitation of Project-Based Learning | Problem-solving teaching d=0.61 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Language Learning | Direct instruction d=0.56 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| AR and VR | Constructivist teaching d=0.92 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Teacher Assistance | Scaffolding and situated learning d=0.52 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Encouraging Creativity | Philosophy in schools d=0.54 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Collaborative Learning | Cooperative learning d=0.53 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
| Critical Thinking and Problem-Solving | Epistemic cognitive training d=0.51 | Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. |
Please note that the citations provided are based on the seminal work of John Hattie, who compiled the Visible Learning MetaX database. The specific effect sizes and more detailed information about each strategy can be found in Hattie’s work and the accompanying database. The strategies listed in the table are aligned with the AI integration strategies based on their descriptions and the outcomes they aim to achieve in the context of K-8 education.
| AI Integration Strategy in K-8 Education | Visible Learning MetaX Student Learning Strategy |
|---|---|
| Personalized Learning | Self-directed learning |
| Interactive Learning Environments | Elaborative interrogation |
| Support for Diverse Learning Needs | Effort management |
| Enhanced Assessment Tools | Self-judgement and reflection |
| Development of Digital Literacy | Strategy to integrate with prior knowledge |
| Facilitation of Project-Based Learning | Self-regulation strategies |
| Language Learning | Imagery |
| AR and VR | Elaboration and organisation |
| Teacher Assistance | Help seeking |
| Encouraging Creativity | Self-verbalization / self-questioning |
| Collaborative Learning | Peer tutoring on Tutee |
| Critical Thinking and Problem-Solving | Meta-cognition strategies |
And, if those two tables are combined:
| AI Integration Strategy in K-8 Education | Student Learning Strategy | Teacher Learning Strategy |
|---|---|---|
| Personalized Learning | Self-directed learning | Feedback (Tasks & Processes) |
| Interactive Learning Environments | Elaborative interrogation | Inquiry-based teaching |
| Support for Diverse Learning Needs | Effort management | Response to Intervention |
| Enhanced Assessment Tools | Self-judgement and reflection | Effects of testing |
| Development of Digital Literacy | Strategy to integrate with prior knowledge | Explicit teaching strategies |
| Facilitation of Project-Based Learning | Self-regulation strategies | Problem-solving teaching |
| Language Learning | Imagery | Direct instruction |
| AR and VR | Elaboration and organisation | Constructivist teaching |
| Teacher Assistance | Help seeking | Scaffolding and situated learning |
| Encouraging Creativity | Self-verbalization / self-questioning | Philosophy in schools |
| Collaborative Learning | Peer tutoring on Tutee | Cooperative learning |
| Critical Thinking and Problem-Solving | Meta-cognition strategies | Epistemic cognitive training |
This table presents a comprehensive overview of how AI can be integrated into K-8 education, aligning each AI strategy with both student and teacher learning strategies that have been identified as having a high effect on learning outcomes. The strategies are categorized to reflect the dual focus on enhancing student learning directly and supporting teachers in their instructional roles.
Assertion #4: Teach AI within a framework that promotes independent thought to avoid indoctrination
Prompt:What research is there that supports or negates this assertion: “Teach AI within a framework that promotes independent thought to avoid indoctrination”
Here’s what MS CoPilot had to say. I also share Perplexity.AI’s summary:
The topic of AI and independent thought is both intriguing and critical. Let’s explore the research to understand the implications of teaching AI within such a framework:
Supporting Evidence:
- Chatbots and other AI systems have the potential to influence people’s beliefs and behaviors. By leveraging data-driven insights, AI can deliver personalized content that exploits individual vulnerabilities and predispositions. This customization enhances the persuasive power of AI-generated content, making it susceptible to indoctrination, especially in vulnerable individuals¹.
- Confirmation bias, where people seek out information confirming their existing beliefs, can be exploited by AI systems. Selectively presenting content that aligns with users’ unformed beliefs reinforces convictions and may lead to radicalization¹.
- Investing in research and development of AI technologies specifically designed to identify and counteract extremist content and radicalization efforts is crucial. Such systems can monitor online platforms and flag potential instances of indoctrination, helping prevent the spread of extremist ideologies¹.
Contrary Evidence:
- Biased training data can hinder AI’s objectivity. If an AI is trained on biased information or censors scientific research, it may not produce objective results. Large language models often ascribe confidence to answers based on internet prevalence, which can perpetuate biases².
- Independent thought is essential for creativity and critical thinking. While AI can reinforce existing knowledge structures, it should also encourage exploration of alternative viewpoints and novel ideas¹.
In summary, teaching AI within a framework that promotes independent thought is crucial to avoid indoctrination. Balancing customization with openness to diverse perspectives ensures AI benefits society without causing harm¹².
Source: Conversation with Bing, 2/25/2024 (1) Beware the AI indoctrination engines – I by IMD. https://www.imd.org/ibyimd/innovation/beware-the-ai-indoctrination-engines/. (2) Generative AI and the Importance of Independent Thought. https://www.brownstoneresearch.com/bleeding-edge/generative-ai-and-the-importance-of-independent-thought/. (3) Artificial Intelligence: A help or threat to contemporary education …. https://link.springer.com/content/pdf/10.1007/s10639-023-11947-7.pdf. (4) The moral weight of AI consciousness | MIT Technology Review. https://www.technologyreview.com/2023/10/16/1081149/ai-consciousness-conundrum/.
And, here is Perplexity.ai’s perspective:
In summary, while there may not be direct research that either supports or negates the assertion in question, there is a body of work that relates to the principles behind it. This research suggests that developing AI within a framework that promotes fairness, transparency, and accountability is important to avoid the risks of indoctrination and bias. However, the concept of independent thought in AI is complex and is subject to ongoing debate and research within the fields of AI ethics, machine learning, and cognitive science.
Assertion is accurate.
Discover more from Another Think Coming
Subscribe to get the latest posts sent to your email.
Another Think Coming 