The MIT Professor's Guide to Speaking That Actually Works

2025/09/01 13:44

One day Youtube algorithms showed me the lecture "How to Speak" from MIT channel. Turned out the lecture is well-know and became a local tradition: MIT freshmen had it for over 40 years. It struck me as something rare. Not another “10 tricks in 10 minutes” video, but a patient, deep reflection on how to communicate.

The speaker was Professor Patrick Winston, famous due to his work on AI: he authored several influential textbooks such as "Artificial Intelligence" and "The Psychology of Computer Vision". He approached the subject of speaking with the same passion, and put effort into making the lecture a memorable experience.

And as I later checked, many of his ideas were backed by research, so let me share the concepts that stayed with me.

Opening a talk

Ask everyone to turn off their phones and laptops. Professor Winston claims what many suspect: humans process information through a single cognitive thread, meaning any distraction – even seemingly minor ones – significantly reduces information retention rates. Multiple studies, including this comprehensive analysis, demonstrate measurable drops in learning when attention is divided

For your opening, skip the joke unless you're already known for charisma – with an unfamiliar audience, failed humor creates immediate distance rather than connection. Instead, lead with a clear promise about what attendees will gain by the end of your talk. Many effective speakers from my experience enhance this by asking the audience about their expectations, then explicitly addressing each one throughout the presentation, creating a natural checkpoint system that keeps everyone engaged.

Keeping your audience engaged

The professor gives 4 techniques (he calls them Samples or Heuristics) to maintain contact with the audience.

Circle back to previously covered topics. Referencing a thought multiple times during your talk creates a web of interconnected ideas. He mentions that an idea should be referenced at least 3 times to imprint in memory. Recent studies confirm that spaced learning – which is exactly how the professor recommends approaching your talk – significantly increases both short and long-term information recognition.

Build "fences" around your concepts. Comparing and contrasting your ideas with others helps the audience better understand and memorise them. The iconic example is Steve Jobs dubbing the iPod as "1000 songs in your pocket" – this puts the device clearly in the musical player category while helping it stand out.

Create "seams" with verbal punctuation. First, enumerate the topics you'll cover during your talk. Then clearly mark transitions between those enumerated topics. This makes your talk easier to follow and allows those who drifted to jump back in.

Ask questions the right way. Asking questions isn't just a good way to wake the audience up – they need to be the right questions. Asking something complex that people surely can't answer looks arrogant. Conversely, asking something trivial makes people feel like they're being treated as fools. Look for a sweet spot in between.

Choosing time and place

Since talks are often planned ahead, it's worth ensuring the time and place will allow you to best engage with the audience. This applies to any important meeting: if you can control these factors, don't skip the preparation.

The meeting room should be well-lit so the audience doesn't doze off. The ideal time is 11am, when everyone is awake enough to engage but not yet tired or thinking about lunch. At least one study confirms this empirical observation: peak cognitive performance occurs between 10:30 and 11:00am.

Using props

Some things are better seen than told. With a few examples, the professor shows how physical objects make any talk memorable. He advocates for old-school versus digital props: chalk, board, and a pointer allow both the lecturer and the audience to move at the same pace, matching the talk flow to learning speed.

However, there's a study that shows a hybrid approach yields the best academic results, so speakers need to think about and understand their audience and their preferences. Gen Z might not appreciate the long and steady talk.

Preparing slides

Ideally, your talk should not rely on slides, but nowadays this isn't quite feasible. Hence, P. Winston gives a few good tips on how to make your slides bearable:

  • The font size should be a minimum of 40-50 points, to be readable even by those who sit far from the screen
  • All non-essential elements should be ruthlessly eliminated: copy should be simplified, images and decoration stripped away
  • Text should supplement, not duplicate your words, otherwise your speech starts competing with your slides

As an exception, the Professor shows when slides are actually useful. He displays an extremely crowded slide (you can check it here) with lots of callouts and arrows showing an ecosystem of funds moving through Afghanistan. The purpose of such slides is to function as a complete object, not something that needs to be read, but understood as a whole concept to convey the message (in this case, complexity itself is the message).

Giving specific cases of talks

The Professor then moves from generic tips to specific types of talks: sharing information and pitching to be hired.

Information Transfer

Obviously, if your goal by the end of the talk is to share information and leave people to draw conclusions, it would be great to state this upfront. As already covered in “Opening the talk” paragraph, this is what the professor recommends: giving a promise of what the audience will get at the end.

He then shares his thoughts on what makes these types of talks stand out, and that is passion. The example he gives of students sharing that passionate lecturers influenced them most resonates with me deeply. Any subject can be made interesting if a person cares enough.

This part concludes with advice on thinking frameworks: no matter what your talk is about, your audience needs to understand how to think about it. The Professor briefly outlines what "learning how to think" means:

  • Give your audience stories (as we're story-telling animals)
  • Provide them with questions to ask about those stories
  • Give them tools to analyze and evaluate those stories
  • Show them the way to synthesize insights

Here again we can find connections to other theories. For instance, Idries Shah in his "Learning How to Learn" book outlines the same approach: learners are expected to interrogate each tale, extract patterns, and apply them to new situations.

Career Pitch

The second specific type of talk is, plainly speaking, about how to get hired. It's really short and boils down to 2 things:

Vision: you need to show that you understand where to go.

Track record: you need to provide evidence of actually being able to deliver on your vision in the past.

Just vision without execution makes you an empty-talker, and just execution makes you a follower. Not a bad thing, but leading the way feels much more rewarding.

Winston’s Star: The 5S Framework

In the final part of the lecture, the professor touches on the topic of being famous. He raises a good point: Why does anyone need to be famous at all? And gives an answer that seems obvious, but is powerful: Ideas are like children. Famous parents get them better opportunities.

The Professor shares very briefly his 5S framework. Each "S" plays a distinct role in making your message stick with the audience:

Symbol: A visual or conceptual anchor people can latch onto. Think of it as a mental image that represents your idea.

Slogan: A concise, memorable phrase or tagline that captures your idea's essence in a punchy form.

Surprise: Something unexpected – an insight or twist that jolts your audience's attention and makes the message buzzworthy.

Salient idea: The standout notion in your talk – not necessarily the most important fact, but the one that truly lodges in memory.

Story: Narrative structure. As we saw above, humans are story-telling animals, and wrapping your ideas in a story makes them more appealing.

This framework is referred to as Winston's Star or the "5 S's" of Getting Famous, and it helps you go beyond ideas just being useful to making them unforgettable. The Professor tested it on his early PhD thesis and only later realised that work scored all 5 points.

How to end

Unsurprisingly, the lecture ends with ways to end a lecture 😄. The Professor emphasises that the ending is what will stick with the audience, and you should not waste this opportunity (hence he does not recommend ending with questions or simply reiterating the talk).

Ideally, you want to show what contribution you made with your talk – what you gave to the audience and to the field.

Saying "Thank you," though acceptable, implies people stayed mostly because they're polite. So, the good options, according to the professor, are:

  • A memorable, distinguishable phrase
  • Salute: personal respect to the audience with an explanation
  • Parting wish: something like "I hope you will bring your friends next time"

I hope this summary helps you craft talks that don't just inform, but truly connect with and inspire your audience. Watch the original, and drop in comments what stood out the most for you!

ข้อจำกัดความรับผิดชอบ: บทความที่โพสต์ซ้ำในไซต์นี้มาจากแพลตฟอร์มสาธารณะและมีไว้เพื่อจุดประสงค์ในการให้ข้อมูลเท่านั้น ซึ่งไม่ได้สะท้อนถึงมุมมองของ MEXC แต่อย่างใด ลิขสิทธิ์ทั้งหมดยังคงเป็นของผู้เขียนดั้งเดิม หากคุณเชื่อว่าเนื้อหาใดละเมิดสิทธิของบุคคลที่สาม โปรดติดต่อ service@mexc.com เพื่อลบออก MEXC ไม่รับประกันความถูกต้อง ความสมบูรณ์ หรือความทันเวลาของเนื้อหาใดๆ และไม่รับผิดชอบต่อการดำเนินการใดๆ ที่เกิดขึ้นตามข้อมูลที่ให้มา เนื้อหานี้ไม่ถือเป็นคำแนะนำทางการเงิน กฎหมาย หรือคำแนะนำจากผู้เชี่ยวชาญอื่นๆ และไม่ถือว่าเป็นคำแนะนำหรือการรับรองจาก MEXC
แชร์ข้อมูลเชิงลึก

คุณอาจชอบเช่นกัน

AI and Writing on Medium: A Mix of Possibilities and Worries

AI and Writing on Medium: A Mix of Possibilities and Worries

This document explores the diverse perspectives of writers on the Medium platform regarding the integration and impact of Artificial Intelligence (AI) tools in their creative process and the broader writing landscape. It delves into the opportunities AI presents for enhancing productivity, generating ideas, and improving writing quality, while also addressing the anxieties surrounding originality, job security, and the potential devaluation of human creativity.Photo by BoliviaInteligente on Unsplash The Allure of AI: Efficiency and Inspiration Many Medium writers acknowledge the potential benefits of AI tools, particularly in areas where efficiency and inspiration are paramount. Boosting Productivity: AI-powered writing assistants can significantly speed up the writing process. Tools like Grammarly and ProWritingAid, which have been around for a while, are now being augmented with AI features that offer more sophisticated grammar and style suggestions. AI can also assist with tasks such as: Generating outlines: Quickly creating structured outlines for articles, saving time on initial planning Research and summarization: Efficiently gathering information and summarizing lengthy documents, allowing writers to focus on analysis and interpretation. Drafting initial content: Producing initial drafts of sections or entire articles, providing a starting point for writers to refine and personalize. Sparking Creativity: AI can serve as a source of inspiration and help writers overcome writer’s block. Generating ideas: AI tools can suggest potential topics, angles, and keywords based on trending themes and audience interests. Exploring different writing styles: Experimenting with AI to generate content in various tones and styles, broadening a writer’s creative range. Overcoming writer’s block: Using AI to generate initial text or prompts to jumpstart the writing process when facing creative stagnation. Improving Writing Quality: AI tools can help writers refine their work and ensure clarity and accuracy. Grammar and style checking: Identifying and correcting grammatical errors, stylistic inconsistencies, and awkward phrasing. Readability analysis: Assessing the readability of text and suggesting improvements to make it more accessible to a wider audience. SEO optimization: Optimizing content for search engines by suggesting relevant keywords and improving overall SEO performance. The Shadow of AI: Concerns and Ethical Considerations Despite the potential benefits, many Medium writers express concerns about the ethical implications and potential negative consequences of AI in writing. Originality and Authenticity: A primary concern is the potential for AI to diminish originality and authenticity in writing. Content mills and plagiarism: The fear that AI will be used to generate low-quality, generic content that floods the internet and potentially infringes on copyright. Loss of unique voice: The concern that relying too heavily on AI will lead to a homogenization of writing styles and a loss of individual voice. Defining authorship: Questions arise about who is the true author of AI-generated content — the user who prompts the AI or the AI itself? Job Security and the Value of Human Writers: Many writers worry about the potential for AI to displace human writers and devalue their skills. Automation of writing tasks: The fear that AI will automate many of the tasks currently performed by writers, leading to job losses. Lowering rates for writing services: The concern that the availability of AI-generated content will drive down rates for human writers, making it difficult to earn a living. The evolving role of the writer: The need for writers to adapt their skills and focus on areas where human creativity and expertise are essential, such as critical thinking, emotional intelligence, and storytelling. Ethical Considerations: The use of AI in writing raises several ethical questions. Transparency and disclosure: The importance of being transparent about the use of AI in content creation and disclosing when AI has been used to generate or assist with writing. Bias and fairness: The potential for AI algorithms to perpetuate biases and stereotypes, leading to unfair or discriminatory content. Misinformation and manipulation: The risk that AI will be used to generate and spread misinformation or manipulate public opinion. Navigating the Future: A Balanced Approach The consensus among many Medium writers is that AI is a tool that should be used responsibly and ethically. A balanced approach is needed, one that leverages the benefits of AI while mitigating the risks. Embrace AI as a Tool, Not a Replacement: View AI as a tool to enhance writing skills and productivity, rather than a replacement for human creativity and expertise. Focus on Human Strengths: Emphasize the unique qualities that human writers bring to the table, such as critical thinking, emotional intelligence, storytelling, and the ability to connect with audiences on a personal level. Develop New Skills: Acquire new skills in areas such as AI prompt engineering, content strategy, and data analysis to remain competitive in the evolving writing landscape. Advocate for Ethical Guidelines: Support the development of ethical guidelines and best practices for the use of AI in writing, promoting transparency, fairness, and accountability. Foster a Community of Learning: Engage in discussions and share knowledge about AI tools and their impact on the writing profession, fostering a community of learning and collaboration. In conclusion, the Medium writing community holds a complex and nuanced view of AI tools. While recognizing the potential for increased efficiency and creative inspiration, writers also express legitimate concerns about originality, job security, and ethical considerations. The key to navigating this evolving landscape lies in embracing AI as a tool, focusing on human strengths, developing new skills, and advocating for ethical guidelines. By adopting a balanced and responsible approach, writers can harness the power of AI to enhance their craft while preserving the value of human creativity and storytelling. AI and Writing on Medium: A Mix of Possibilities and Worries was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story
แชร์
Medium2025/09/04 15:06
แชร์