41 comments

[ 161 ms ] story [ 490 ms ] thread
That argument would lead to the proposition that any effort to improve cognition is a net zero-sum proposition.
To get a little meta for a second, do we have a universally accepted, value-neutral, standard idea of exactly what we mean when we say cognition? Individual tests might set different parameters and pay attention to different mechanics. Or do they?

To put my question a different way: how confident are we in our current map of the system?

I'm asking as a legitimately curious layperson. Perhaps this comment is naive, and if so, I'm admitting to my naivete in advance.

>do we have a universally accepted, standard idea of exactly what we mean when we say cognition?

no. (sensory neuroscience guy here)

It's awfully tangled up with determinism and incompatibilism, neither of which are universally accepted. This makes it difficult to create a definition that is both meaningful and not controversial.
No, there is no single universal neutral standard of cognition. Memory tasks are different from mental rotation tasks, which are different from emotion-reading tasks, etc.

But, it turns out that performance at a huge variety of sub-tasks is highly correlated. E.g. reaction time (catch a falling ruler as fast as you can) correlates with ability to solve math problems, which correlates with ability to control impulses.

This is why IQ, as a general measure that covers a wide variety of tasks, is so useful. (Of course, that's not to say IQ covers everything about cognition in some absolute way; it doesn't measure ability to balance on a beam, for example, AFAIK).

I think a definition of 'cognition' at the hardware level (wet matter brain) would depend on a conjectured model of 'consciousness' to distinguish it from the more clearly definable 'sensing'.

A simple model of (higher) consciousness could be that it is simply the 'state of the brain'. Sensation would be an almost axiomatic as the 'pattern of state change for specific stimuli' whereas 'thought' is 'pattern of state change invariant of stimuli'.

Cognition could then be defined as 'sensation that [evokes] thought'.

Giving someone pencil and paper or access to the internet will improve their cognition. The key difference is that these improvements happen outside the brain.
Could someone explain this to a layman like myself, please
Also a layman, but from my reading of the abstract, the net zero sum model is saying that ultimately the brain does about as well as it's going to, given its physical constraints. It's the equivalent of "no matter how much you overclock your CPU, you'll still be limited by thermal dissipation."
If it's thermal dissipation (which is believable, as Adderall increases your body temperature), that can be solved artificially: wear a wet towel, increase air conditioning, or wear a cold pack.
In other words, does this amount to abuse of an already maximally efficient organ, resulting in short term gains in exchange for what amounts to injury or perhaps chronic subclinical injuries?

From the abstract:

  based on the assumption that brain 
  resources are subjected to the physical 
  principle of conservation of energy
Which is to say, nothing material is added to the equation, and thus, nothing useful is actually extracted. Conceptually, the idea is to question whether we're really just burning the candle at both ends, to get more light.

Compare to steroid abuse in athletes, resulting in a statistically significant shortening of lifespan.

I doubt this is true. There are smarter people and there are simpler ones. I guess they're in the same ballpark WRT energy usage. Could you turn the latter into the former? Why not?
But that's a different question, isn't it? We are talking about "overclocking" the brain, so to speak, not necessarily making one smarter or dumber. The speculation here (if I understand correctly) is that "overclocking" the brain has a cost somewhere, but that doesn't imply that smarter people are running "overclocked" brains per se.
I suppose anything is possible, though that may well be much more challenging if intelligence is a function of the physical configuration/arrangement of the neurons in the brain. The interesting question if the structure theory is true, is what kinds of effects that would have on the sense of self, and of consciousness as a whole.
That sounds like an assumption, not a conclusion, though.

Also, most of human history (and especially mammalian history) has been relatively food-energy-limited. But we've solved that problem with the agricultural revolution and automation. Does this mean we've removed that barrier?

Hence why it is worded as a question?
> Which is to say, nothing material is added to the equation, and thus, nothing useful is actually extracted. Conceptually, the idea is to question whether we're really just burning the candle at both ends, to get more light.

We could try adding ATP molecules while decreasing cellular decay, I think that would be great, maybe Karma will Punish us but I bet if we fuck around a bit we can do better. But also I think the biggest challenge is political, arguing against people who feel discomfort by the finite and disappointing abilities of their brain organs, projecting their need for comfort to avoid being left behind

It's a model for explaining improvements in cognitive function. The idea is that the brain's energy consumption is fixed, so when you improve its function on something it costs you from elsewhere. So if you take Adderall to get better at school, you're not getting a brain improvement, you're just reallocating resources. It's more like moving money from your Chase account to your Wells Fargo account than it is like getting richer.

During the lead up there are also interesting bits like "if you suppress one part of the brain another does better at its task" and models that explain the interesting behaviour.

So they're saying, they can improve my neural function by also making me burn more calories?

Market cap 1T.

You know what they say about the candle that burns twice as bright, though
The candle that burns twice as bright illuminates things twice as strongly and definitely has no downsides whatsoever which is why it's the top seller in the candle industry. Did I remember that right?
It consumes fuel twice as fast? Fuel is cheap.

If there any evidence that intense mental loads somehow wear down the brain, or the body as a whole, any faster? If anything, I only know that intense mental loads stave off the onset of Alzheimer's.

Animals up to rhesus monkeys which reduce their caloric intake have reduced incidence of age-related and of all-cause mortality [1], and this is conserved - at least - in rats, mice, fish, flies, worms and yeast [2]. While the fuel itself may not be expensive, there is evidence that consuming/using it is anything but. These studies are for obvious reasons difficult to conduct in humans, though I would be very surprised if they weren't under way, and time will tell.

[1] https://www.nature.com/articles/ncomms4557

[2] http://ajcn.nutrition.org/content/78/3/361.full

Running also increases caloric and definitely doesn't reduce lifespan in humans. So the abstraction "increase energy causes bodies to wear out faster" is very leaky, and probably so leaky that it's not super useful.
Depends how you count it. Humans are quite effective engines - number of calories burned per hour of running is quite low, and you only do it for limited number of hours per week.

At the same time runners tend to be the kind of people that are very concerned about their health, so they eat healthy food, with moderation.

So overall caloric intake of a runner can be much less than that of obese couch potato.

> definitely doesn't reduce lifespan in humans

Compared to what though? The average sugar-addicted citizen? Someone who does no exercise?

I remember metaphors developed before calorie/$ dropped so much
Dated: "2014 Jan 15"
it's a paper, not a javascript framework
This is a good and funny reply. Also, it accidentally led me down a fun rabbit hole and I ended up trying to learn about the oldest scientific works are that are still regularly cited today.

Answers: https://scholar.google.com/scholar?hl=en&as_sdt=0%2C48&as_yl... is a good starting point. Euler's paper demonstrating that there is no solution to the Königsberg bridge problem and inventing graph theory in the process is one of the more recognizable (to me) papers, but there's a lot of truly interesting, foundational work there.

My wife guessed that Newton would be on there, but I couldn't easily turn him up. I'm probably just bad with Google scholar, or maybe it's so foundational that nobody bothers to provide a citation for e.g the very existence of _calculus_ or _gravity_

My original question was "the oldest still-cited scientific work") but I relaxed my criteria due to poorly defined definitions on 'still-cited' and 'scientific'. Still an interesting way to spend some time, so thanks for the digression!

I got lost in that paper trying to find out the reason that lead to asking if neuroenhancement is a zero-sum proposition. Can anyone point it out to me.
See also: "Popular electric brain stimulation method used to boost brainpower is detrimental to IQ scores"

https://www.sciencedaily.com/releases/2015/05/150505152140.h...

I guess this tech will only be used by a strange niche group that wants the average intelligence of the world to decrease then, huh. Oh well!
Direct current across the brain. Really. I mean, in my younger or attractive days, I fell in for a while with a bunch of kink players who for safety reasons wouldn't even put DC across their et cetera, but other people consider it just fine for their actual brains. Sure.

Ever get that odd morning moment where you're not sure you woke up in the same worldline where you fell asleep? I think this one might be a Cronenberg film, but hey, at least I'm probably not actually in it.

Yeah but you know where a man's brain is.
Isn't it false to assume the human body (and the human brain) is a 'closed system'? This is going to sound mighty stupid, but doesn't the fact that we ingest variable amounts of food (i.e. energy) suggest that in surplus situations this energy can be used 'on the spot', rather than converted to long-term storage (i.e. fat)?
Why run electrical current through your brain (which seems like a bad idea on general principle)? Audio waves are demonstrably effective - not just for stimulation / focus, but adjustable for other cogntitive states, like relaxation and sleep. See [Brain.FM](https://brain.fm).
I remember playing games like Falcons and Hard Hat Mack on my Apple II+ as a kid. Those two games in particular were cracked copies. As was fairly common in the Apple II warez scene, pressing the reset button during gameplay would dump you into the Monitor prompt. You could access locations $C050 and $C057 to turn fullscreen graphics mode back on, and type G to resume execution from the point of interruption. The details were more complex than my recollection at this late date, but that was the basic idea.

What was cool about this was that you could also alter random values in memory from the Monitor before resuming the game. Of course, without the source code it was tough to say what effects a given change might have. Usually nothing would happen after modifying a few arbitrary memory values, but sometimes I'd get lucky and gameplay would resume in an interesting, undocumented way. Fewer enemies, more powerful weapons, access to an otherwise-unreachable level, wacky graphical effects. Occasionally I'd get really lucky and stumble upon a way to jack up my score to 999,999,999 or whatever. When that happened, I'd hit reset again, reboot, and dump the graphics RAM to my printer for guaranteed bragging rights on the school bus the next day.

Frequently, though, after altering something at random, the game would just crash.

And that seems a lot like what you're doing by running electric current through your brain.