Agency and Sphexishness: A Second Glance
This post is backported from my old personal blog, originally published 11 October, 2014. It’s similar in theme but very different in focus than my recent post, On the Nature of Agency, making for a good complement.
Epistemic Status: Though I wrote this over four years ago and now use different models/lenses/ontologies, what I wrote then still seems largely correct. The exception is the endorsement of noticing which I haven’t thought about much in recent years and therefore can’t endorse, though I do now want to think about it more.
.
Agency
Agency is the property of agents. An agent has explicit goals which they strive to accomplish by planning and executing appropriate actions. Non-agents unreflectively act out default behaviours without considering whether these actions achieve their goals.
Sphexishness
Coined by Douglas Hofstader in reference to the sphex wasp, sphexishness is the execution of seemingly intelligent behaviour by following a rigid algorithm. Sphexish behaviours are repeated automatically, out of habit, without checking for their effectiveness at achieving desired goals.
When I first learnt the concepts of agency and sphexishness, I didn’t give them much thought. I unreflectively assumed that you were either an agent or you weren’t; you were either sphexish or you weren’t. If you were the kind of person who thought about their goals, read Less Wrong, went to CFAR workshops, then clearly you were an agent – you actually cared about your goals. I, of course, was such as person. It’s the general population that is sphexish, they don’t notice when things don’t get them what they wanted. And it all maps well on the controversial metaphor of PCs and NPCs.
But you don’t just get agent status and that’s it. Rather, as I have learnt, there is a constant slide into sphexishness which you must work damn hard to avoid.
Looking at definition given, sphexishness has two components:
Unreflective, automatic, default behaviours.
These behaviours fail to achieve goals.
Really, it’s 2) which matters. Firstly, not only can you have unreflective, automatic behaviours which are not sphexish, you absolutely need unreflective, automatic behaviours. Attention and cognitive resources are limited and you cannot be constantly using System 2 to examine how you do things to achieve your goals. Instead you must have put in place default behaviours which do the job. It works for walking, for what I eat for breakfast, and for how I go about writing code.
Secondly, there is no level of deliberateness or complexity or meta which guarantees that you are not being sphexish. Suppose I do goal factoring weekly, or I set lots of five minute timers, etc., but using these techniques isn’t improving my life that much – then my use of these techniques is sphexish. Even if I periodically review my rationality practices for effectiveness, I could be sphexish is my review routine doesn’t get results.
So wherein lies agency? How does one avoid sphexishness? Noticing. My current hypothesis is that Noticing is a fundamental skill of rationality. One must learn to notice when one’s actions are failing to get the desired result, and then take corrective action. And “failure” is loosely defined here: one could be making progress towards a goal, but at a much slower rate than one could, and this too I would deem sphexish to some extent. And this Noticing goes infinitely meta – you need to be able to notice when your first-order optimisations aren’t working or are too slow, take corrective action on them, and so on.
To avoid sphexishness one needs to notice that the ways one has always done things unthinkingly aren’t great; one needs to notice that unreflective, automatic behaviours which were optimal when one originally implanted them are no longer so, now that circumstances have changed; one needs to notice when one is stuck in a rut and needs to pull themselves out; one needs to notice that they just aren’t making progress and need to do something different.
On this I will say more.
FWIW, I still think ‘noticing’ is a fundamental rationality skill and have used it pretty consistently since Brienne first brought it up.
I do notice that your post seems to focus on the ‘macro level’ (i.e. notice that an overall behavior isn’t working out), whereas the noticing that I’ve made use of is more on the 5-second level – noticing the internal feeling of “hmm, this project feels doomy” or “hmm, I seem to be defensive right now” or “hmm, I just said ‘X is like Y’ – do I actually have examples backing up the claim ‘X is like Y’?”
Noticing seems like a cornerstone skill because you can’t possibly change your behavior strategically if you’re not noticing anything. And, moreover, a lot of the time the mere act of noticing can dissolve whatever maladaptive pattern you’re in.
I wonder what distinguishes sphexishness from a simple habit. They’re both unreflective, automatic, default behaviors, and “bad habits” are just habits that fail to achieve goals. But they feel different to me. The best I can come up with is something like: habits are in theory changeable, whereas an actual sphex wasp will never change its behavior based on experience. Habits are acting sphexish.
But we need habits. I’m reminded of this:
So I think I agree with you about noticing and agency. Agency isn’t the opposite of sphexishness. But it does seem to require choosing when to act so, and that requires noticing when you’re doing it.
(somewhere in my unposted-blog-notes folder is something about noticing that horrible mental loop where I click random links all over the web, no matter how much I’m not-enjoying-myself, because I can’t seem to context-switch. I titled it “Noticing Boredom.”)
Perhaps there is an optimal balance between habits and deliberation.
Too much on the side of habits, and you just keep doing the same behavior over and over again. Not necessarily a bad thing; sometimes you get lucky and the strategy you started with is actually a good one, and can bring you success in life. But you need the luck.
Too much on the side of deliberation, and your clever ideas get undermined by lack of “automated operations” that would keep you moving forward. The result is procrastination; well known among the readers of this website.
And the optimal balance probably depends on your current situation in life. After you achieve some success, you have more choices, and now deliberation probably becomes more useful. But again, there is such thing as too much meta-deliberation; obsessing “exactly how much time should I spend thinking and how much time should I spend working” generates neither useful work nor useful directions for work.
I guess, the more meta, the less time you should give it, unless you already have evidence that the previous level of meta was useful to you. (When you notice that spending some time thinking increases the productivity of the time when you are working, that is the right moment to think about how much time do you actually want to spend planning.) Also, meta decisions take time to bring fruit at the object level, so when you make plans, you should spend the following days executing the plans instead of adjusting them; otherwise you decide without feedback.
Execution is Actual Work, though! Noooooooooooooooo!
(I’m adding that to my fortune file. I could use the reminder from time to time.)
Thank you for bringing this up—it’s a comparison that doesn’t resonate with me. I suspect that “sphexishness” is a different modeling layer than “agency”, so a direct comparison is confusing. More importantly, it’s assumed without explanation that one is bad and one is good.
For some reason, nobody’s talking about the amazing success of the Sphex wasps, and looking for ways to ensure successful behavior without everyone having to model reality individually. And we don’t talk (much) about the horror of bad choices, and how all suffering is caused by agency.
A bit tangential to the main point, but Woodridge is a misleading tertiary source and while some sphex species display looping behavior under some circumstances it is not as universal or irrational as the Hofstader makes it sound.
I had not heard of “sphexishness” before. But my fledgling blog seems to be hell-bent on respectfully prying apart what I there termed “automatic” thoughts/behaviours.
I’m not sure about component #2 of sphexishness there—it appears the wasp achieves its goal of nurturing very well! But, I’m just going off the info referenced here, for its definition.
Lacking neo-cortices and executive function, wasps obviously wouldn’t have the power (and the stress) of conscious goal-setting and future-planning. Its behavioural routines are innate, dictated by its genes, fixed throughout its life; (though they may be activated/inactivated at certain life stages or in response to trigger events. I’m no entomologist, I don’t know the specifics.)
To me, this is worlds apart from the myriad habits and routines that humans learn to integrate into their lives. The fact that different humans have different habits indicates it is something learnt individually, be it through conscious or through unconscious intent. This is quite unlike the described wasp behaviour, which presumably applies to all Sphex ichneumoneus indiscriminately.
As such, it seems to me foolhardy for a human to wage war on one’s own behavioural presentation of phenotype, which is what I understand sphexishness to be. However, auditing one’s habits (or for a real challenge, try objectively examining one’s own biases) - that which you call “noticing”—is surely invaluable for any adult human who values life beyond being a proverbial drone within the hive.