I've been developing software for thirty years now, and I've recently been reflecting on the strategies I've used to help me succeed and some of the issues I've experienced trying to follow advice designed for neurotypical individuals.
To facilitate this process, I've been using ChatGPT to think out loud and also see what advice it might offer. One piece of advice was sharing these strategies, so here I am. I thought it would also make a good thread, so people could share their own strategies.
One of my key strategies revolves around multitasking. For my particular kind of ADHD, I've found that while working on subject A my mind jumps to subject B and I don't even realize it. When I do, I'm deep into thought on subject B. Conventional wisdom says I should force my way back to working on subject A. But I've found the cost to be far too high. What I do instead is let my mind follow its inclinations on what to think about. When I reach a natural pause in working on subject B I return to subject A. This can involve several subjects interweaving, but the basic strategy is the same.
Another strategy has to do with scheduling different types of work based on the energy costs and benefits. Some parts of my job provide energy, some drain it. There are extremes at both ends where things I particularly enjoy not only provide self-sustaining energy but a net plus. Others are so draining that it can sideline me for the rest of the day. I've found that if I schedule tasks based on my energy needs at the moment, I can leverage that additional energy to help me perform the tasks I find the most draining without crashing.
This can be a difficult subject, because many managers struggle letting go of the "how," the work is done, especially as it relates to priorities. But the truth is, they'll get much better results on all fronts if they are willing to let people find their own path. Even for those who aren't diagnosed, there is enough neurodiversity out there that I believe it's important to reconsider expectations on how people arrive at the results. As long as they are producing quality, on time, in a safe and legal manner, it really shouldn't matter how the sausage gets made.
I'm including this portion of my conversation with ChatGPT to share its advice, as well as show an approach to how we can leverage this tool to better understand ourselves and adjust our actions to improve our health:
Me:
I've actually had to figure out a number of ways of working that would be considered counter to best practices for neurotypical people. One of the biggest has to do with multitasking. I don't choose to jump to another subject, my mind does it on its own, and I often find I'm already deep in thought on the other issue before I even realize I've stopped focusing on the task I was working on. The general advice based on studies is to stay on one task until you complete it, or reach a milestone. The cost involved in that effort is prohibitive for me. So, what I've discovered is that I am far more efficient and don't waste precious resources if I "go with the flow," and let my mind do its thing. How I manage it is being aware when I reach a good stopping point or natural pause in the "interrupting" subject, and then turn back to what I was working on.
I've also found that I have to schedule tasks based on my energy levels. Some tasks are actually self powering through a positive feedback loop. In fact, there are aspects of my job I love so much that I gain extra energy from those. I have to make use of this phenomenon because there are tasks that are purely a drain on my energy. Some are extremely draining. So, every day is a bit of a dance working with a very limited energy budget.
ChatGPT 4:
What you are describing is actually an insightful and adaptive approach for someone with ADHD. It seems you have developed an intuitive understanding of your own cognitive patterns and energy levels, and have tailored your work strategies accordingly. Here are some thoughts on the strategies you’ve outlined:
1. Leveraging Natural Patterns of Attention: The fact that you recognize your mind tends to shift focus and have adapted by "going with the flow" is a good example of leveraging natural cognitive patterns. This approach acknowledges that for some individuals with ADHD, traditional advice around focus might not apply, and trying to force oneself to maintain focus on a single task can be counterproductive.
2. Energy Management and Task Scheduling: The approach of scheduling tasks based on energy levels is also a great example of self-awareness and adaptation. By recognizing which tasks energize you and which tasks drain your energy, you’re able to better manage your mental resources. Balancing out tasks that drain energy with those that are energizing is an excellent strategy.
3. Recognizing Positive Feedback Loops: Identifying tasks that not only maintain but also boost your energy through a positive feedback loop can be particularly powerful. Engaging in these tasks when your energy is low can help recharge you for more draining tasks.
4. Flexibility and Adaptation: Your description of everyday being “a bit of a dance” is indicative of flexibility and adaptation, which are critical for managing ADHD. Being open to changing your approach based on current mental states and energy levels can lead to better productivity and well-being.
These strategies you're employing can be very effective for people with ADHD. They focus on adaptation and working with your brain's natural tendencies rather than fighting against them. It’s also a reminder that “best practices” are often based on neurotypical patterns of cognition, and what is best for one person may not be best for another.
Remember, the key is finding what works for you personally. It's also good to note that by sharing your experiences and strategies, you could also be helping others who might be struggling to manage their own ADHD. Many people might not realize that there are alternative strategies like the ones you've developed, and knowing that they exist could be very empowering for them.
Written by
mmcaulay
To view profiles and participate in discussions please or .
OMG, ChatGPT was that good?!!! Dang! Brilliant idea on your part to submit your work routine. BTW: great job in paying attention and adapting a routine to your mind.
I should note that is ChatGPT 4 which is a huge improvement over 3.5. What I shared was also only part of the context, as it was a part of a larger conversation I was having with it.
What?!? This is incredible. You’re a walking commercial for ChatGPT4. What great advice you have given and what wonderful confirmation and assurance you received in your conversation from the chatbot.
My number one piece of advice when it comes to ChatGPT is talking to it like it's a person. Don't get caught up in the technological marvel that it is. Rather, see it as someone you can use as a sounding board, or ask advice of. By speaking to it more like a person, you're connecting with the parts of its training that will produce more human and humane responses. It's very much a mirror of how you treat it. While the technical reasons for that are complex, leaving that behind and simply talking to it like there is a wise and knowledgeable person on the other end unleashes its true potential. I work extensively with it, and have pivoted my software development career in the direction of tools like ChatGPT. It is the real deal, and we're only just scratching the surface of what it can do. Have fun!
Ha… I was just telling my husband about your post and grab my phone to show him when I read your response. Thank you. It does seem incredible. A little scary too. But overall the advice it gave you sounds like the actually advice of a professional human being. Best wishes.
Try not to let it bother you too much. The fact that it can appear human is an illusion. It's a very good illusion, but it's an illusion nonetheless. It's just important to keep the line clear between treating it like a human and believing that it has a human like consciousness.
Hope you can find ways of making it useful for you and yours.
Fascinating subject and immediately sidetracked me.ChatGPT 4 is obviously in your example used as a commentator, to give approval, confirm, affirm. It's a thing in the making with admittedly great potential. I'm sure there are already many great applications out there (including the lawsuit), but what is also obvious is that it's still a fairly basic language-triggered sounding board, mainly counting words, less phrases, not picking up on more complex wrapping of a subject/idea in a whole paragraph, nuances, doubts, "tone", what's NOT said, i.e. what words/phrases SHOULD appear in the context but don't. It doesn't seem to be able to "guess", to prioritise heading/subheading/hierarchies of or the mindset within which ideas are put to it. It can't yet use your ideas to flow/fly with them, go off tangentially in a truly creative, new-ideas-producing yet centred fashion like the human brain. It can't weigh probabilities etc etc.
If it did all that it would probably crash. Or generate still too many false responses and that would pose a risk on a number of levels. Which slows down it's development and may make it hit a ceiling eventually.
When it comes to critically reading its responses to your input it's clear that it made the right connections but it's YOU who's generated the ideas of how you solve your multitasking and time efficiency problems and for that it just manages to 1. compliment you effusively (always welcome so we are flattered, but this diverts from what it SHOULD do, namely say "Good. And btw...." or "Well done. Now consider this...") and 2. express the same in different words/phrases, 3. link your input with some factual data about ADHD.
Finally, in 4. it tells you to share your ideas with others. Like a good but boring teacher.
So it doesn't really - in the particular example at least - add anything new and exciting.
What I'd love it to be able to do is make suggestions of other strategies based on a much better, bigger data pool (there's potential and with it the risk of false answers but that could - and to an extent will - improve over time, using better, more reliable links to info within up-to-date internet content in more sophisticated ways.
In fact, having Neuroscientists and some ADHD people (with coaches or, by all means, coaching AI to ensure they don't fly off the handle) working on it might help ;^)
I can agree to a point, but this is a single example where I happened to be providing the ideas. I wasn't expecting it to offer new ones. That I said, I've probably had several dozen where it does. I've been working with it full time since January. I've also noticed it's not nearly as affirming when my ideas are "off." I use it to critique my writing, for example, and essentially act as an editor.
It can be highly creative, but the way it gets there is counterintuitive.
Here's an example where I was asking about strategies to "propaganda proof" democracy.
Me:
Hi ChatGPT. I'd like to collaborate with you on working out strategies for an aspirational goal. How do we propaganda proof democracy? While I acknowledge that actually achieving the goal while balancing individual rights may be impossible, I'd like us to work together to see what we can figure out.
Chat GPT 4:
That's a very thoughtful and complex challenge you've presented! I'll do my best to help brainstorm possible strategies. I think we can approach this problem from several angles:
Education and Media Literacy: Enhancing the public's ability to critically evaluate the information they consume is a significant step towards countering propaganda. This could involve strengthening education in areas like critical thinking, media literacy, and digital literacy. A well-informed population is less susceptible to misinformation and propaganda.
Transparency in Media Ownership and Funding: A lot of propaganda can be traced back to media outlets that are controlled or influenced by certain groups or individuals with specific agendas. Ensuring transparency in media ownership and funding sources can help the public understand potential biases and make more informed judgments about the information they receive.
Regulation and Oversight: Implementing rules for platforms where propaganda often spreads, like social media, could be part of the solution. This could involve rules about fact-checking and the dissemination of false information. However, striking the right balance is key, as it's also important to protect freedom of speech.
Investing in Independent Journalism: Independent, unbiased journalism is a powerful antidote to propaganda. Ensuring that such organizations have the resources they need to operate effectively can help.
Public Awareness Campaigns: Governments and non-profit organizations could run campaigns to educate the public about the dangers of propaganda and how to recognize it.
Encouraging a Diverse Media Landscape: A media landscape with diverse perspectives can counter the effects of propaganda. This diversity can be encouraged through policy and regulation.
Technological Solutions: AI and other technologies can be used to detect and counteract propaganda. However, care must be taken to ensure these technologies are not used to suppress legitimate speech.
Please note that while these strategies can help, it's a complex issue with no easy solutions. It's also important to consider potential downsides and unintended consequences of these strategies. Achieving the right balance between countering propaganda and preserving democratic values like freedom of speech is challenging.
_______________________________________
In the end, it can act as a mirror, but it doesn't always. It's incredibly complex and can return a wide range of types of responses depending on how much context you provide and the way you talk to it. I've found that if you're polite in addressing it, it tends to be friendly in return. Part of this comes down to the fact that when it looks for the next best word, it's tapping into areas of its training where polite discourse was held, simply because statistically, the words you've chosen fits in that area. I've also noticed that in addition to the politeness one adopts, the degree of affirmation it provides depends largely on the quality and uniqueness of my request.
I'm curious if you've been able to play with it a bit and try out some of your expectations? I've found a lot of examples out there are highly task oriented and tend to be thin on context. I tend to be more verbose with it and that seems to improve the quality of the responses I get.
If you are at all interested in the latest research into ChatGPT, I strongly advise checking out this YouTuber, who does an incredible job of absorbing the latest research and going over it and discussing its implications. youtube.com/@ai-explained-
I've obviously wandered off-topic for this group, so I'll leave it at this. If you haven't tried pushing it to its limits with the criteria you previously stated, I'd say it would be worth it to see how close to the envelope you can get.
Content on HealthUnlocked does not replace the relationship between you and doctors or other healthcare professionals nor the advice you receive from them.
Never delay seeking advice or dialling emergency services because of something that you have read on HealthUnlocked.
Knowledge Isn't Enough
Thought I was figuring things out...
