It is recommended that we as aging adults keep our brains active, so people read and play games like Sudoku. This is a step in the right direction. But I use AI as a learning coach on various topics and it’s not just about keeping my brain active. It’s making him drink with a fire hose! I use Gemini for my AI work and it’s amazing. I chose it because I bought a Chromebook a little over a year ago and it came with a free year of all of Google’s advanced AI (this marketing ploy was successful because I got hooked and recently paid for an additional year). I have friends who enjoy using other AIs like CoPilot, ChatGpt, Grok and Claude and they love them so I won’t pretend that what I show for Gemini couldn’t also be done with the others. But my advice is to pick one and stick with it. During my first two years of studying AI, I tried several versions and did not gain in-depth or very proficient knowledge. But I’ve made a lot of progress this past year by sticking to just one.
Another caveat I have to make is that not all of the features I mention are in the free version of Gemini that anyone with a Google account can use. As far as I know, most are, but I’m particularly unsure about Deep Research.
I started using Gemini for technical research, either in my own field or for medical questions. I have a deep interest in any condition that I or my friends are afflicted with! For this, I like to use Deep Research. You have just started a Gemini session and chosen “Deep Search” from the tools. I go out and find a bunch of websites and research papers relevant to your query, then come back with a detailed report. An important tip is to watch it for about a minute. It will then tell you its plan and you will have to press Start. I made the mistake of leaving for several hours, coming back and waiting for my results, but it still waits for you to press start. After starting, it may take a while, so it’s a good idea to exit this window and come back later. The first thing I would like to point out is that you can get very good results faster in “quick” mode without resorting to extensive research. But for more detailed results and a nice report that you can save as a document, I prefer “think” mode and deep search.
It also helps explain what your level of expertise is, from an amateur hobbyist, meaning me in various fields, or someone with professional expertise and experience. So, for a technical problem related to math or engineering, I would say “I have a PhD in mechanical engineering and experience in numerical methods, but only undergraduate training in computer science and electronics.” But recently I was asking a question related to neuroscience, so I responded: “I have no professional training in neuroscience, but I am quite well read in this field as a hobby.”
I’ve previously written blog posts about some of my research findings, such as changing my training to accommodate my leaky heart valve. But recently, I have moved towards more “amateur” interests like neuroscience. I will give a specific result about this below. But first I want to give an overview of the options. Your first choice is quick or reflection mode. Fast works great for most things and gets faster results (hence the name). The reflection is still quite quick, but goes a little deeper. After setting one of them, which will remain the default for future discussions, from the Gemini window you can simply perform a query and get great results even if you use it to learn a new topic. This is option 1. It will find lots of information to help you answer your question and will sometimes create things like worksheets to explain concepts.
It will then offer you follow-up queries “Do you want me to investigate xxx” and you just have to say yes if you are interested. I learned a lot from hunting them. Option 2 is to specifically put it in learning mode by choosing the “guided learning” tool, which I love. This will behave the same as a normal Gemini query, but will add images and quizzes to help you learn. Here’s how it describes itself: “A specialized interactive mode/tool within the app. Modular; it divides topics 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 for science learning. The third option, which I mentioned above, is to enable the Deep Search tool. But I don’t recommend it for a topic you’re trying to learn, because it’s much slower Use in-depth research on things you already know, but to delve deeper into a specific area Sometimes I follow up a learning session with an in-depth research session and then back up the report with an in-depth research.
A simple example: the health benefits of pickleball
I was afraid that the example of neuroplasticity I give below would scare people away because it was still technical even though I said I was an amateur in this field. After all, neuroscience is a fairly technical field. So let’s start with a simple example where I told him to stay non-technical. This is a Gemini query with guided learning enabled: “Suppose I am a lay adult in this field, please keep the answer less technical. Is it true that pickleball has some health benefits for older adults?” The answer is always very informative, including a quiz and suggestions for further study. Note that I kept this discussion short, but I can come back to it later and continue learning. The results are here.
Another example: the history of neuroplasticity
This request started from a funny anecdote that I had read. In the late 1990s, the Dalai Lama asked visiting scientists an interesting question along the lines of: Western science assumes that consciousness and our mind emerge from our brain, which can be seen as the software running on the brain, the hardware. Can something the mind does change the brain? He specifically wondered if meditation could change the brain. The existing paradigm in neuroscience was that our brain is only malleable in childhood, while very little neuroplasticity exists in adulthood. Neuroscientists therefore all answered no. But we now know they were wrong, spectacularly so, and neuroplasticity and how consciousness can influence it is now a hot 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 asked: “I have no professional training in neuroscience but I am quite well read in this field as a hobby. I would like to know the history of developments in the understanding of neuroplasticity since the 1990s:. This led to a pleasant exchange in which I learned a lot, which I shared. here.
I will be posting weekly articles on areas related to healthy aging, both physical and mental, that I have studied with Gemini. I will always make sure to give credit to AI. I don’t like it when people create blog posts or YouTube videos with AI and don’t recognize it. I think it’s like cheating on your homework!
Share this:
Related
Published by
I am a mechanical engineer born in 1953, PhD, 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 a hip replacement and a heart valve 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