Putting the "akrasia" in ADHD

From the extended mind to the extended will

My writing output has been going towards work as of late, but I hope to get back into the swing of Substacking now that my adderall prescription is being filled reliably.

ADHD is often treated like a BS disorder, and I have no doubt that there’s rampant over-diagnosis. For me, however, it’s been a lifelong struggle, though one I only became self-conscious of in adulthood. As a kid, I’d forget to bring my lunch box home thirty days in a row despite daily reminders. I just figured I was a disorganized procrastinator; the type who’d cram weeks worth of homework into single, late-night productivity sprint out of necessity — not because that’s when my brain’s dopamine finally ebbs enough for me to enter a state of blissful focus.

While “attention” is in the name, I think ADHD is better thought of as a disorder of executive function. For as long as I can remember, I have struggled with planning and organization, vacillating between bursts of time-blind hyper-focus and long stretches of paralytic mental mayhem. At its worst, my internal monologue feels like the driver of a car whose wheel was disconnected from its steering column. I can articulate what it is I need to do and how to do it with precision, but the rest of brain somehow doesn’t get the message.

In retrospect, I think this is why I developed an early interest in cognitive psychology and philosophy of mind before turning to economics. The phenomenon of “weakness of will” fascinates me in particular — what Aristotle called akrasia. Socrates denied that akrasia could even exist, as surely one cannot both know the good and fail to act on it. If you know you shouldn’t eat another donut but do so anyway, an economist would call that your “demonstrated preference.” You only think you have a preference for calorie cutting, but actions speak louder than words.

Focusing on demonstrated rather than stated preferences is a good rule of thumb for sniffing out bullshitters, but the Socratic / economistic view still leaves something out. Higher and lower order preferences really do exist. And it’s also not solely for lack of willpower that some people seem worse at aligning their actions to their aspirations. They often simply lack the social scaffolding and “choice architectures” that makes doing the good thing easy. This is why people join spin classes or set out their running clothes before bed. Having peers that egg you on and an environment structured to reduce frictions to good behavior are ways your brain, in a period of high willpower, conspires against your brain in its periods of low willpower.

Andy Clark and David Chalmers famously called all the ways we embed our cognitive competencies into the external environment “the extended mind.” In “Procrastination and the Extended Will,” philosophers Joseph Heath and Joel Anderson elaborate on the extended mind thesis to the context of self-control:

When it comes to tasks such as arithmetic, the limitations of our cognitive system are so obvious and solved in such familiar ways that no one dwells on the remarkable fact that most of us are unable to remember a set of four double-digit numbers for long enough to carry out long multiplication. Our dependence on environmental prosthetics is self-evident. When it comes to practical rationality, on the other hand, many theorists have either ignored or denied the presence of any systematic defects. Rational-choice theorists have been front and center in this campaign. Gary Becker’s analysis of addiction as a consequence of straightforward, unproblematic utility maximization represents perhaps the apogee of this tendency. This is partly a result of the noncognitive (or subjectivist) conception of preference that typically informs rational-choice theories, which results in practical rationality being held to a lower standard than theoretical rationality. After all, if one puts no constraints at all on the content of preferences, then it becomes possible to construe any action as rational (in a rather undemanding sense of the term) simply by positing a preference for doing it.

Heath and Anderson go on to show how their analysis of “the extended will” can be deployed in social criticism:

Our externalist account of procrastination suggests that this has potentially important political implications. To the extent to which we can assume (1) that certain social and economic developments in modern societies place increasingly high demands on the capacities that make up our ability to avoid procrastination, (2) that current trends toward individualization and liberalization involve the dismantling or abandonment of traditional forms of unchosen scaffolding, and (3) that the impact of the two preceding effects will be especially devastating for those who are already most vulnerable (think of the difficulties that the homeless have in taking advantage of environmental supports), then procrastination becomes not just an issue of individual psychology but also an issue of social justice. If people’s life chances are significantly shaped (perhaps along lines of class, race, degree of disability, and so forth) by their access to scaffolding and if many forms of scaffolding are being dismantled or rendered inadequate as the result of social processes that could be addressed (at least to some extent) by public policy, then the negative consequences of procrastination are not just the result of people failing to cognize appropriately as individuals but are partly the result of decisions and dynamics over which individuals have little control.

It seems to me that many of America’s contemporary social problems can be cast in these terms, from the opioid crisis to collapsing fertility rates. It also lays bare the hollowness of the “freedom from” conception of personal autonomy. True autonomy lies in aligning one’s higher and lower preferences, i.e. one’s theoretical and practical rationality. Traditional cultures, social scripts, and seemingly paternalistic public policies (like forced-savings plans) may seem like they reduce autonomy at first blush, but often represent social technologies for individuals to offload, or externalize, their willpower onto the collective. This may even be the mechanism behind why belonging to a religion confers so many benefits to mental health.

Yet to the extent that intrinsic willpower still matters, it is far from uniformly distributed. As Heath has elsewhere noted, our elites are largely selected for their self-control ability. If you’ve written a PhD dissertation or passed the LSAT, you most likely have above-average self-control or are otherwise embedded in a supportive environment. This has given rise to a “self-control aristocracy,” Heath argues. In turn, right-wing elites oppose restrictions on tobacco and alcohol primarily meant for chain smokers and alcoholics so they can sip their IPAs and indulge in the occassional cigar. Meanwhile, left-wing elites downplay the consequences of sexual liberalization and other forms of destigmatization from the stability of their high-functioning, two-parent households. On this account, these parallel forms of neoliberalism — economic on the right, cultural on the left — are an ideological guise for class domination.

My hope is that AI can help lean against some of these trends. Autistic people have been using ChatGPT to rehearse their communication skills, for example. Longer-term, perhaps those who find themselves bereft of the social scaffolding provided by strong communities, a supportive faith, and other forms of extended cognition will bootstrap their autonomy through a society of AIs, solving the problem of akrasia outright. At the very least, a Jarvis-like executive assistant will be a game-changer for my ADHD.