It is recommended that we older adults keep our brains active so that people read and play games like Sudoku. That’s a step in the right direction. But I use AI as a learning coach for various subjects, and it doesn’t just keep my brain active. It makes him drink from a fire hose! I use Gemini for my AI stuff and it’s great. I chose it because I bought a Chromebook a little over a year ago, and with it I got a year of free access to all of Google’s advanced AI (this marketing ploy was successful because I got hooked and recently paid for another year). I have friends who like to use other AIs like CoPilot, ChatGpt, Grok, and Claude, and they love them, so I won’t pretend that what I show for Gemini can’t be done with the others as well. But my advice is to pick one and stick to it. The first few years I was into AI, I dabbled in several versions, but didn’t get very deep or skilled at it. But I’ve made a lot of progress over the past year by sticking to just one.
Another caveat I must make is that not all of the features I mention are present in the free version of Gemini that anyone with a Google account can use. As far as I know most of them are, but I’m especially unsure about Deep Research.
I started using Gemini for technical research, both in my own field and for medical-related questions. I have a deep interest in every condition that affects me or my friends! I like to use Deep Research for this. You simply start a Gemini session and choose ‘Deep Research’ from the tools. I’ll go out and find some relevant websites and research papers for your question and then come back with a detailed report. An important tip is to look at it for about a minute. It will then tell you its plan and you have to press start. I’ve made the mistake of going away for a few hours, coming back and expecting my results, but it’s still waiting for you to press start. It may take a while after you start, so it’s a good idea to leave that window and come back later. The first thing I would like to point out is that in “fast” mode you can get very good results faster without doing any in-depth research. But for more detailed results and a nice report that you can save as a document, I prefer the “thinking” mode and in-depth research.
It also helps to explain your level of expertise, to an amateur hobbyist, i.e. me in different areas, or someone with professional expertise and experience. So for a technical problem related to math or engineering, I will say, “I have a PhD in mechanical engineering and experience with numerical methods, but only have a background in computer science and electronics at the undergraduate level.” But recently I asked something related to neuroscience, so I said, “I don’t have a professional background in neuroscience, but I’m pretty into it as a hobby.”
I’ve previously written blog posts about some of my research findings, such as adjusting my training to accommodate my leaky heart valve. But lately I’ve started delving into more ‘hobbyist’ interests, such as neuroscience. I will give a specific result below. But first I want to give an overview of the possibilities. Your first choice is fast or thinking mode. Fast works great for most things and produces faster results (hence the name). Thinking is still quite fast, but goes a little deeper. After setting up one of these, which will remain the default for future chats, you can simply ask a question from the Gemini window and get great results, even if you use it to learn a new topic. That’s option 1. It will find a lot of information for you to answer your question, and sometimes create things like spreadsheets to explain concepts.
It then suggests follow-up questions: ‘Would you like me to investigate xxx’, and all you have to do is say ‘yes’ if you’re interested. I learned a lot chasing this one. Option 2 is to specifically put it in learning mode by choosing the ‘guided learning’ tool, which I love. This behaves the same as a normal Gemini query, but adds images and quizzes to help you learn. Here’s how it describes itself: “A specialized interactive mode/tool within the app. Modular; it breaks topics down into a structured ‘learning path’. Automatically integrates diagrams, videos, and images. Includes Canvas-based quizzes, flashcards, and study guides. Built on LearnLM, a model specifically trained in learning science.” The third option, which I mentioned above, is to enable the deep research tool. But I don’t recommend that for a topic you’re trying to learn about because it’s a lot slower. Use deep research for things you are already familiar with, but to delve deeper into a specific area. Sometimes I follow up a learning session with an in-depth research session and keep the in-depth research report.
A simple example: the health benefits of pickleball
I was worried that the neuroplasticity example I give below would turn people off because it still got technical even though I said I was an amateur in that field. After all, neuroscience is quite a technical field. So let’s start with a simple example where I said to keep it non-technical. This is a question in Gemini with guided learning enabled: “Assuming I am a mature layman in this field, please keep the answer less technical. Is it true that pickleball has some health benefits for seniors.” The answer is still very informative, including a quiz and suggestions for further research. Note: I kept this chat short, but I can come back to it later and continue learning. Results are here.
Another example: the history of neuroplasticity
This question started with a fun anecdote I had read about. In the last 1990s, the Dalai Lama asked visiting scientists an interesting question along these lines: Western science is based on consciousness and our mind comes from our brain, which can be considered as the mind running on the brain, the hardware. Can something the mind does change the brain? He specifically thought about whether meditation can change the brain. The existing paradigm in neuroscience was that our brains are only malleable in childhood and that there is very little neuroplasticity in adulthood. So the neuroscientists all said the answer was no. But we now know they were wrong, spectacularly wrong, and neuroplasticity and how consciousness can influence it is now a hotly debated topic in neuroscience. And some of the best evidence comes from imaging the brains of advanced meditators, including the Dalai Lama’s Tibetan colleagues.
So I wondered, “I don’t have a professional background in neuroscience, but am quite well versed in it as a hobby. I would like to know the history of developments in the understanding of neuroplasticity since the 1990s: This led to a wonderful exchange in which I learned quite a bit, which I shared here.
I will post weekly on areas relevant to healthy aging, both physical and mental, that I have been exploring with Gemini. I will always make sure to give credit to the AI. I don’t like it when people make blog posts or YouTube videos with AI and don’t acknowledge it. I guess that’s like cheating on your homework!
Share this:
Related
Published by
I am a mechanical engineer, born in 1953, Ph. D, Stanford, 1980. I have been active in the field of mechanical CAE for decades. I also have a lifelong interest in outdoor activities and fitness. I had both hips replaced and a heart valve replacement due to a genetic condition. This blog chronicles my adventures in staying active despite these bumps in the road. View all posts by BionicOldGuy
Published
#learning #coach