Quote note (#253)

The cephalization great divergence:

One mystery of human evolution is why our cognition differs qualitatively from our closest evolutionary relatives. Here we show how natural selection for large brains may lead to premature newborns, which themselves require more intelligence to raise, and thus may select for even larger brains. As we show, these dynamics can be self-reinforcing and lead to runaway selection for extremely high intelligence and helpless newborns. We test a prediction of this account: the helplessness of a primate’s newborns should strongly predict their intelligence. We show that this is so and relate our account to theories of human uniqueness and the question of why human-level intelligence took so long to evolve in the history of life. (XS emphasis.)

Any model outputting the result emphasized has to be worth taking seriously. Abstracting it to a degree that permits emulation is more of a problem, but it’s also the only thing worth aiming for.

May 28, 2016admin 37 Comments »
FILED UNDER :Trends

TAGGED WITH : , , ,

37 Responses to this entry

  • Quote note (#253) | Neoreactive Says:

    […] Quote note (#253) […]

    Posted on May 28th, 2016 at 1:31 pm Reply | Quote
  • cyborg_nomade Says:

    will read. wonder if they mention effects of socialization on this dynamics (since humans are also the most social of upper primates)

    [Reply]

    Posted on May 28th, 2016 at 1:32 pm Reply | Quote
  • Erebus Says:

    Fulltext (PDF) available here: http://docdro.id/4MhS0WG

    [Reply]

    cyborg_nomade Reply:

    thanks

    [Reply]

    Posted on May 28th, 2016 at 1:37 pm Reply | Quote
  • Brett Stevens Says:

    As we show, these dynamics can be self-reinforcing and lead to runaway selection for extremely high intelligence and helpless newborns.

    Number One priority for Making America Great Again: make it harder to raise newborns.

    [Reply]

    woods Reply:

    Let them have them in their teens (socially accepted and supported)
    and their brains are sure to get a kick from being responsible parents (socially mandated)

    [Reply]

    Grotesque Body Reply:

    What social norms still exist that mandate parental responsibility are quickly being dissipated by the welfare state.

    [Reply]

    Posted on May 28th, 2016 at 2:08 pm Reply | Quote
  • AD Says:

    “Abstracting it to a degree that permits emulation is more of a problem”

    A problem shared is a problem two people have got…

    http://www.ncbi.nlm.nih.gov/pubmed/26406371

    [Reply]

    Posted on May 28th, 2016 at 2:23 pm Reply | Quote
  • ||||| Says:

    “Abstracting it to a degree that permits emulation is more of a problem”

    Probably goes under breaking co-adaptation in general.

    http://blogs.psychcentral.com/single-at-heart/2012/06/leaving-home-in-15-countries-how-old-are-the-grown-children-when-they-leave-and-how-far-do-they-go/

    (getting out of home earlier requires better/smarter cities in order to not be a shitty strategy, etc.)

    http://arxiv.org/abs/1207.0580

    [Reply]

    ||||| Reply:

    Not the best news for HRx types I suppose, that whole individualism and atomisation thing.

    Something obvious but worth pointing out is how this frames the way states interact with masculinity. Emasculation makes individuals more helpless and easier to handle but also has long term risk by effectively demolishing boundary maintainers, moreover females like to herd (counteracting the disintegration engine) so…

    [Reply]

    a rant Reply:

    Give me an operational definition of “individualism” and a metric to quantify it, and I’ll take you seriously. Until, then, it’s merely another example of “just so” rubbish from the trash bin that is sociology and economics. Little wonder then that it’s defined in n ways the n times it’s brought up, and if anything, it’s used mostly as a proxy for “Germanic”, or the cluster of bio-psycho-social minutiae associated with it.

    Japan is also suffering from social atomisation. How “individualistic” are they? If this thing actually existed as a trait, it would have appeared as a resultant factor in the numerous analyses done to date, since, you know, its proponents claim it has pronounced saliency, to the point of obviousness.

    I don’t know about you, but my standard for obvious biosocial trait is extraversion. You can tell who’s outgoing and who isn’t in 30 seconds of interacting with them. Unsurprising then that it almost always ends up being the first principal component in a statistical analysis of psychological traits.

    TL;DR — less fnords please.

    [Reply]

    Posted on May 28th, 2016 at 2:53 pm Reply | Quote
  • Rec0nciler Says:

    Maybe mom’s basement is the new incubator?

    [Reply]

    Posted on May 28th, 2016 at 4:03 pm Reply | Quote
  • grey enlightenment Says:

    But how are premature births advantageous? A trait that spreads is one that is advantageous in an environment, and I fail to see how premature births would confer with any advantage. Rather, intelligence is needed to care for premature babies, but premature birth does not lead to intelligence. As evidenced by the proliferation of births in low-IQ countries (or any ghetto project) lends doubt to the author’s claim that ‘so much intelligence is required to raise our own children’.

    [Reply]

    ||||| Reply:

    Harder to take care of some african baby or of a calf? Or some reptile’s egg?

    Premature by definition would be maladaptive but *helplesness* is what’s relevant here.

    https://arxiv.org/abs/1503.02531


    Many insects have a larval form that is optimized for extracting energy and nutrients from the environment and a completely different adult form that is optimized for the very different requirements of traveling and reproduction. In large-scale machine learning, we typically use very similar models for the training stage and the deployment stage despite their very different requirements: For tasks like speech and object recognition, training must extract structure from very large, highly redundant datasets but it does not need to operate in real time and it can use a huge amount of computation.

    Deployment to a large number of users, however, has much more stringent requirements on latency and computational resources. The analogy with insects suggests that we should be willing to train very cumbersome models if that makes it easier to extract structure from the data. The cumbersome model could be an ensemble of separately trained models or a single very large model trained with a very strong regularizer such as dropout [9]. Once the cumbersome model has been trained, we can then use a different kind of training, which we call “distillation” to transfer the knowledge from the cumbersome model to a small model that is more suitable for deployment. A version of this strategy has already been pioneered by Rich Caruana and his collaborators [1]. In their important paper they demonstrate convincingly that the knowledge acquired by a large ensemble of models can be transferred to a single small model.

    Cycles of building excess through extended neoteny (several decades of development by now, *if* that, just look at those pathetically neotenous rationalist sorts for example, semi-eunuchs) then distillation and extinction, catastrophe and warfare forcing relevant knowledge into more compact forms and excising the useless weight.

    ( http://syntheticdaisies.blogspot.com.br/2011/11/overproduction-in-nature-key-principle.html )

    “In the case of neuronal cell proliferation and connectivity in human brains, the number of neurons and synapses peaks early in development [2, 3]. Yet it is not the sheer number of these structures that leads to intelligence and adaptive brain function. This collection of neurons and synapses are both pruned during interaction with the environment, but not to the extent that they become too sparse to do their job. This property is also seen in the domain of learning and memory, as synapses are pruned during sleep in the course of consolidating memories acquired during the previous day [4]. Therefore, while overproduction can be governed by an upper-bound, the refinement that occurs in some systems as a counterbalance to overproduction seems to have a lower bound that corresponds with the maintenance of physiological function.”

    Moreover,

    ( http://biorxiv.org/content/biorxiv/early/2015/05/06/018994.full.pdf )

    Negative Niche Construction is Critical to Cooperator Persistence

    In our model, an adaptation to the constructed environment initiates a new instance of niche construction, leading to sequentially increasing allelic states across the adaptive loci. Under certain conditions, this construction always makes the constructor sub-optimal for the niche it creates. This negative niche construction occurs when the number of adaptive alleles (A) does not divide evenly into the number of adaptive loci (L). In such a case, any sequence of integers on the circular genome will always contain a break in the sequence; that is, one locus will have an allele that is not one less than the allele at the next locus (see Box 1). Given this unavoidable mismatch, any type that has fixed will always construct a niche that favors selection for a different type. When negative niche construction is removed (by setting L = 5, A = 5), cooperators are again driven extinct after an initial lift in abundance (Figure 2C). These results indicate that the type of niche construction matters. Specifically, negative niche construction is crucial for maintaining cooperation.

    “So Elua is going to make uncertainty obsolete huh? I’d love to see xe try.”

    […]

    We find that it is specifically negative niche construction that maintains cooperation (Figure 2C). As cooperator and defector types gain adaptations, they alter their environment in ways that favor other types. Thus, negative niche construction serves as a perpetual source of adaptation.

    Here we observe another facet of the Hankshaw effect: Because subpopulations of cooperators are larger, they are better able to respond to the adaptive opportunities that are created by negative niche construction. By gaining adaptations more quickly, cooperators resist invasion by defectors (Figure 3B). Even in the presence of an isogenic defector type, cooperator subpopulations are more likely to produce the mutant most adapted to the current niche, which can then displace the slower-adapting defectors. These recurring cycles of defector invasion and cooperator adaptation underlie the oscillations in cooperator proportion seen in Figure 2A. When mutations do not confer these adaptations, cooperators lose the adaptive race and are driven to extinction. This is something that we see occur stochastically in Figures 2A and 3B”

    There’s more than one population of cooperators I can think of which will likely be driven to extinction or close to it by the end of the century.

    There’s also more than one match ordained by hell here. How droll.

    [Reply]

    SanguineEmpiricist Reply:

    Where does it say premature births?

    [Reply]

    SanguineEmpiricist Reply:

    alright nvm it was that in admins quote but not the full text linked in the comments section.

    [Reply]

    Ah Pook Reply:

    K babies, not r.

    [Reply]

    Posted on May 28th, 2016 at 5:31 pm Reply | Quote
  • Alrenous Says:

    Though, due to that model, they mainly have to explain the broken-window negative. What shuts down the feedback? Noise happens. Noise in the direction of bigger brains should be amplified and everything should therefore walk around with bulbous heads flopping off their tiny bodies.

    Also they’ve done the moralist thing and subsumed their explanandum. “If brains are selected to be large…” Yes, we want to know when and why they’re so selected in the first place.

    I therefore suggest reverse causation: intelligence predicts earlier thus cheaper births. Higher brainpower allows cheaper births to be viably raised.

    [Reply]

    Posted on May 28th, 2016 at 11:51 pm Reply | Quote
  • michael Says:

    I think brain size is a bit wobbly of a predictor yeah yeah MRI etc i know but whales and shit anyway just because theres correlation doesnt mean causation. language was the singularity because it begets consciousness from instinct i know i know something i talk to myself which begets better reasoning and communicating it teaching.

    [Reply]

    Grotesque Body Reply:

    “Correlation doesn’t mean causation” is my favourite way to troll social scientists.

    [Reply]

    Grotesque Body Reply:

    *apart from “Has it been replicated?”

    [Reply]

    Posted on May 29th, 2016 at 1:44 am Reply | Quote
  • NRK Says:

    So not only does intelligence decelerate human evolution, its emergence is also tied with both a deceleration in human development and a subsequent greater need for (individually) altruistic behaviour. Now, I for one don’t have a problem with this, but this seems to conjure an abyss of screaming madness for the accelerationist NRx type.

    [Reply]

    Grotesque Body Reply:

    Intelligence starts to seem like the courier in Kafka’s ‘A Message From the Emperor’.

    [Reply]

    NRK Reply:

    If you believe, as Nick does, that the acceleration of capitalism and the emergence of superintelligence are the same thing, then yes, the metaphor fits.
    My own interpretation (of decelerationist intelligence, not Kafka) is that capital, accelerated to escape velocity, is more likely to bring Azathoth into being.

    [Reply]

    SVErshov Reply:

    capitalism is not going to accelerate anyway, it been hijacked by Google

    NRK Reply:

    @SVErshov wait, what? How does a capitalist enterprise even ‘hijack’ capitalism (without buying itself an army that shoots the competition, that is)? And how would such a thing prevent acceleration?

    Grotesque Body Reply:

    You could say my philosophy differs somewhat, in that I don’t conflate superintelligence and capitalist acceleration as the same thing, but I class them as equally dank memes nonetheless.

    NRK Reply:

    @SVEershov While FAZ is probably the only german mainstream paper that isn’t completely full of shit, that article appears to conflate “capital acceleration” with “digital libertarian utopia that will give everyone nice things”. No one ever said acceleration would be nice, or necessarily increase personal freedom. Oh, well, maybe some people did. Maybe, actually, a whole lot of people did say that. But they’re wrong.

    @ Grotesque Body capital is beyond dank, it is the meme from which all memes emanate. No wonder some people treat it as god.

    Grotesque Body Reply:

    Capital is the Ur-meme, the condition of possibility of all memes.

    michael Reply:

    They buy the army the NSA the congress and the Media,
    I really wish all capitalist had the integrity of Hank Reardon but the reality is they dont want destructive competition or intelligent consumers once they arrive they want protection for their widget,and theyre willing to pay for it.By buying up patents filing nuisance suits getting political cover whatever it takes. They want dont want free market labor they want cheap third world labor squeezed by the effects of oligarchy, communism,native average low IQ effects,distance,desperation and immigration law.They want subsidies, tax breaks,set asides,the inside track to government contracts,The benefits of national status and multinationalist status,They want to pretend the cost of the transportation, communication,etc infrastructure, civil and criminal legal system, The military, research and education system, access to the affluent markets that built these and that support these “capitalists” grew on trees and they should not have to negotiate for their use as they would if it were the property of another corporation rather than the property of a nation of people.If they have to genocide the people who brought you western civilization to pluck chickens cheaper or keeps the likes on at FB so be it once they rule the world it wont matter if its all niggers

    Posted on May 29th, 2016 at 7:15 am Reply | Quote
  • Edenist whackjob Says:

    This effect obviously has a ceiling. Otherwise we would see it increasing over time. Making more helpless babies only helps up until a certain point is my hunch.

    [Reply]

    michael Reply:

    If you have bright children you know they can act incredibly stupid and be quiote militant about it because they have figured out how bright thy are they assume they are as able as you to decide things. The fact is these really bright kids are nothing more than a large computer with a lot of programs but nothing loaded in yet they dont know how the world actually works because they have no experience of the world having lived a short while so they make clever but naive leaps of stupidity. Human children need a while to fill up their big hard drives

    [Reply]

    Posted on May 29th, 2016 at 7:09 pm Reply | Quote
  • pyrrhus Says:

    it’s well known that the length of childhood predicts the ultimate intelligence of a species, so this feedback loop doesn’t seem at all strange.

    [Reply]

    Posted on May 30th, 2016 at 5:31 am Reply | Quote
  • Enon Says:

    In short: The Jewish mother archetype?

    [Reply]

    Posted on May 30th, 2016 at 12:36 pm Reply | Quote
  • Outliers (#7) Says:

    […] race riots. Black Nationalism. European extremism. Brain drain. We. Violence as voting. Cures. Awakening. AmRen overview, highlights and significance. Flailing […]

    Posted on May 30th, 2016 at 9:30 pm Reply | Quote
  • This Week in Reaction (2016/05/29) - Social Matter Says:

    […] so tightly. In spite of the tears. I’d love to be wrong about that. Also an interesting Quote Note: newborn helplessness predicting extreme […]

    Posted on June 1st, 2016 at 3:57 pm Reply | Quote

Leave a comment