How to build Intrinsic Motivation: a review of the science
erringtowardsanswers.substack.com
erringtowardsanswers.substack.com
Unfortunately I've only experienced this three times in my life; typically around major life events (once when starting a new job in a new industry, once when quitting that job to make my own stuff, and once in grade school: the summer between 10th and 11th grade, for some reason). I look forward to seeing more research, and hopefully one day can apply these learnings to manually trigger this intense focus and motivation.
As i get older, this happens less and less – which is a massive shame.
I wanted to understand whether there was any good evidence as to what intrinsic motivation is and how i might be able to cultivate it in my adult life. To do this, i did a massive deep dive of the scientific literature surrounding intrinsic motivation. This is the outcome of that research.
Anecdotally (because I’m not going to syringe my brain) I am feeling a lot more enduring wakefulness and motivation than when I skip them in my morning routine.
I did a chatGPT dive to validate this but that’s not exactly a biochemical lit review.
NAC to clean free-radical oxygen damage to dna.
There is a wide list of other maneuvers to try first.
- Time in bed does not equate to "great sleep." Make sure you don't have sleep apnea. Practice good sleep hygiene.
- Delivery of oxygen to your brain is just as important as mitochondrial aging, if not more. Get some aerobic exercise (even walking is fine) -- it wakes me up and maybe will do the same for you. Heed your vascular risk factors, because crap in your cerebral vessels will not help.
- The one supplement exception is vitamin B12, which neurons must have. Deficiency can be very hard to judge by symptoms, so I'd get it measured and act accordingly.
One of my other favorite theories is HEXACO. And personality does play into intrinsic motivation, to some extent.
Disclaimer: I skimmed the article.
Fun autonomy hacks:
1. Reframe the narrative. For example, when I studied CS at school, I didn't study CS. I studied how to learn as fast as possible. I happened to have studied CS.
2. Listen to Spotify to get into a solo task. I usually turn it down if I happen to get focused.
Also a note: intrinsic motivation is tough when you're sleep deprived. I've had moments where I was motivated and sleep deprived but they often don't coincide.
This is all to say that stuff like this go onto a fundamental layer of physical health. Something I dind't quite get when I was younger.
That's an example
As for the Spotify example. I just like listening to my playlists, every task becomes more chill. Also, I like working on a Mac more than a Windows laptop. I've had one company restricting my choice there to Windows. Me sort of hacking their company policies such that I could work on a Mac made me feel a lot better.
Simple example: if you believe an action you did was a really bad thing, you will most likely feel negative emotions about it. However, if you can figure out a perspective that will reframe the information you have in a different light and therefore you now believe it was a positive thing, you will likely feel good about what you did.
Example (I'm improvizing so not fully according to the sketch outlined above):
Negative: I don't dare to talk to that person because they don't know me and it is not done to talk to someone you don't know without a context.
Positive: While it is unusual to talk to someone you don't know without a context, I give that person a chance to meet me. If I tend to do this often enough, then there will be people that are open to this.
I can see approaching things in a different way. I was fond of a more Socratic approach for a while, as an example. But that is more than just reframing, that is using a different approach.
For your example, it looks like you are making sure to consider things in a way that does not assume the outcome?
It's not wishful thinking if that's what you're alluding to. The idea of reframing isn't to tell yourself positive things so that you feel good. It's to highlight positive things that you'd otherwise not have seen and to factor them into the full equation of what it is that you're doing.
It's also more about the following: with many situations it's hard to objectively evaluate the truth to such an extent that different perspectives seem equally true given the knowledge you have. If that's the case, then pick the most optimistic, hopeful and motivating one.
It's not just "the power of positive thinking". There is truth in the fact that a person that doesn't know you hasn't had the chance of meeting you. It's up to you to decide whether that's a shame or not. I personally find it a shame, because the world is a more beautiful place when people are more well-connected in a way that they're happy with. The only way to do that is to reach out to people. Hence, the perspective holds value. Missing out on that perspective would be a shame, but that is what would happen to me if all I can do is having a knee jerk level of a reaction.
Another example is: studying math will teach you discipline. If you hate studying math but you want to learn discipline, there is some solace in that. It's true that many things will teach you discipline, so that doesn't mean that all of a sudden it's all okay now. However, excluding that fact does give a negative bias that doesn't need to be there. Removing that bias gives someone a bit more autonomy, especially when they also have some control over their schedule so they can study when they feel sharp.
Now, it could be that someone then would say "yea but the issue actually is that I'm doing this all by myself". Now you're touching upon relatedness. Relatedness is not something to reframe, it's something to seek out. So you could search for a math community. Now, one could hear "I don't like these people" (or they do and then it's problem solved). If that's the case, it could very well mean they're just not feeling fully socially settled in at school and that's actually interferring with them from having a calm mind.
That last example is semi-auto biographic. What eventually happened was that I got good at socializing and then studying for geeky things became easier again.
Now and then, evaluation points in life emerge, where you question the 'why'. Those periods can be quite loaded with emotions of feeling lost or being insecure of where to go next. They might greatly shift perspective and hence your course of life.
To me, everything is feedback loops: you pour in energy and you get positive energy back. In that sense, the system is self sustained. However it is fragile as well, because over time, you tend to need more and more back to provide feelings of contentment.
Motivation is like love and relationships, you need to work, sometimes very hard, to sustain them.
This is why I am good at those things — I spent a lot of my time doing them and I did so because I enjoy it.
Let yourself play, find the aspects that interest you about a topic and go wild. No need for a goal, thiose will come eventually. First you need to find joy in what you're doing.
And if you don't find any of that, maybe your motivation is just money, fame or whatever and thst might be ok as well, if you're happy with doing a thing you don't like.
Some games are made to burn time, like Thumper.
Some games are made to burn you neurons like Baba is You.
Minecraft has 2 modes. Creative and Zombie. Both equally powerful incentives.
I try to keep the plasticity of my brain. Not to let it crust and crumble like Play Doh left outside the tub.
https://www.danpink.com/books/drive/
Presumably built off the same research.
Consciousness is a consensus mechanism by which the self is constructed. It is a recursive loop where the self emerges, experiences, and folds in the next experience to create an evolving, expanding self. With language we have the ability to freeze many of these ideas and we are able to go much further. "I can think, feel, experience and reflect on this"
And why a consensus mechanism? Because "you" are actually a constellation of cells and experiences that needs to be sufficiently decentralized but also be able to act and plan in the very short and long term. How do you get 87 billion cells (in our case) to decide as one? That is actually a pretty difficult engineering problem where you have to think about both compute (all the different data streams coming from different sources need to be digested and acted on) and commute (one cell group in the prefrontal cortex needs to immediately broadcast a danger message to other corners of the brain, and we dont have direct wiring)
Now the natural question to ask is, what do synthetic beings need to develop both? If you are interested you might want to read our book Journey of the Mind
Here is a short read on the idea of consciousness as a consensus mechanism https://saigaddam.medium.com/consciousness-is-a-consensus-me...
I have an interesting and truly successful experience related to one "discussion" point:
> For another example, consider the Marinek and Cambrell (2008) experiment where they compared the effect of token-rewards, e.g. a gold star, with task-related rewards, e.g. a book, on reading motivation.
> Would the reward of a book really be experienced as less controlling than the reward of a gold star?
> Instead, I think the token reward was probably more distracting than the task-related reward – which makes sense, since the task-related reward was really just a means of spending more time doing the task at hand anyway.
My experience would suggest another effect is happening. A reward of a book aligns with the activity more than just being less distracting. It makes the activity less of an event, and more of a path. I.e. reading gets framed as not something they did, but something they are beginning, that they can look forward to.
I think we are attracted to learning things more, if the learning has a forward path, if the forward path is made more visible, or the forward path is more enabled. A book reward emphasizes those intangibles, while tangibly enabling another step.
Path continuation/enablement rewards are higher level increased autonomy rewards.
Contrast that to giving a book reward and also being told they were going to be required to read it the next day! It would suddenly represent an anticipated continuation of control instead of an anticipated path of more autonomy and competence. I would suspect it would has an even greater inhibiting effect, by projecting control instead of potential autonomy/competency into the future.
--
Ok, here is the related "psychological experiment" which worked out really well with my young children. (Grown up now.)
Every night, I or my partner would read them a bedtime story. We usually read one or two short stories, after which we asked them to get ready for bed. Invariably, they would beg for another story, which we would read, then put them to bed.
But children don't give up, so they of course begged for another story. No matter how many stories you read, their is the inevitable disappointment of story time ending. Getting them into bed represented the enforcement of that ending, and often required some degree of "control" to get them settled. No time for autonomy! And settled again and again! Until they gave up their valiant struggle against the night and actually settled down.
So we tried something, by random instinct one night, and it worked so well it became a staple of their young lives.
Instead of ending the book focus abruptly, we told them that stories were over, but they could pick a book to sleep with. It was the funniest thing. They would get quite excited and enjoy choosing "a favorite book" from their selection. We would say just pick one, but then "give in" and let them pick two if they couldn't decide.
All this autonomy of what books they were going to take to bed, including talking us into letting them have more than one, really motivated them into bed.
Then lights went out. They couldn't read the books. They couldn't look at the pictures. Nevertheless, they could feel them and it made them very happy going to sleep!
Other factors played a part, but I am convinced that part of the reason they are lifelong habitual book readers, something already getting rare in their generation, is how much they fell in love with the literal physical feeling of books, for eight hours a night.
And their parents have heartwarming and still funny memories of checking on them, seeing them sleeping happily with favorite books in their arms, under pillows, tucked into the covers with them, or on top of the covers with one of their hands or feet sticking out still touching a book's cover.
People with high intrinsic motivation and agency will rule the world of tomorrow, weilding AI to acheive their personal visions. Everyone else will be weilded by AI.
You may well be right. Interesting to think about the relationship between agency & intrinsic motivation...
To allow an embodied agent to perform actions within an environment that would generally be considered positive, without the definition of an objective function.
To break that down, to be embodied in this case is to act, sense and have some internal model that can be adapted, all operating within an environment that can be considered external to the agent.
An objective function is where there is some external push towards optimality that requires knowledge of the sensors, actuators, environment, etc. A good test for whether you accidentally baked in system knowledge is if you change the rules considerably and the agent will not operate.
Whether or not an agent acts positively can itself be measured by an environment specific objective function. A properly operating intrinsically motivated agent may perform well on some metrics, i.e. long time lived, reduced search time, etc.
Why do you want an intrinsically motivated agent? Almost all reward/objective functions are somewhat flawed, even if the problem is simple. I am reminded of a group training a robot to walk fast, measured by speed over time with a cut off. Simple enough? Well, they reviewed the trained agent and they immediately feel to the ground to be reset far away. In another test, the agents would purposely break the simulation environment, causing the agents to glitch and be launched far. One thing to note is that in each of those scenarios, the agent optimised for the reward, but made themselves "useless" after doing so.
For AI I have found Empowerment an interesting solution to intrinsic motivation [1]. Essentially agents choose actions to "keep their options open", and try to avoid actions that would reduce the action state space. The actual environment itself is not encoded into the algorithm and the state spaces are arbitrary and could be replaced with any symbol. As a result, you can make large changes to the environment and use the same motivation algorithm.