Jump to content
Lancelot

[Sérieux] Ethno-différencialisme, race-realism, génétique et courbe en cloche

Recommended Posts

Sur les temps de réaction j'avais relevé ce passage rigolo :

Quote

Mackintosh (2011), p. 339 considers reaction time evidence unconvincing and points out that other cognitive tests that also correlate well with IQ show no disparity at all, for example the habituation/dishabituation test. And he points out that studies show that rhesus monkeys have shorter reaction times than American college students, suggesting that different reaction times may not tell us anything useful about intelligence.

 

  • Haha 1

Share this post


Link to post
Share on other sites
Le 31/12/2018 à 11:40, Lancelot a dit :

Sur les temps de réaction j'avais relevé ce passage rigolo :

 

VSauce a récemment posté un episode de sa nouvelle série où il se fait latter sévère par un chimpanzé en vitesse de réaction dans un test de reconnaissance de formes (c'est-à-dire que le chimpanzé est cinq ou dix fois plus rapide, au bas mot).

Share this post


Link to post
Share on other sites

En fait c'est beaucoup moins déséquilibré que je ne le pensais !!! C'est ma vision à moi du classement Élo qui est tronquée (ça reste beaucoup plus asymétrique que le QI mais pas autant que je ne le croyais) :

 

Classement+ELO+Juillet+2009.JPG

Share this post


Link to post
Share on other sites
On 12/29/2018 at 3:56 PM, Mégille said:

Il me semble avoir lu quelque part (chez Lynn, ou de seconde main, peut-être) que l'effet Flynn est principalement dû à une augmentation des plus bas qi grâce à une réduction de la malnutrition. Ce qui aurait caché dès le début un "effet FLynn inversé", plus lent, dû au plus grand succès reproductif des hommes.

 

Dur à trouver ou prouver, mais intuitivement je pense que tu as profondément raison.

 

On doit évidemment "élargir "la malnutrition (réduction des peintures au plomb, des maladies mal soignées, des mauvais traitements à enfants, ...) et  je rajouterais personnellement aussi, de façon plus polémique, la réduction de la taille des fratries, ce qui cadrerait d'ailleurs avec les bonds technologiques fulgurants de la Chine.

Share this post


Link to post
Share on other sites

Si l'effet Flynn s'applique asymétriquement aux deux extrémités de la courbe de QI, alors ça devrait s'observer assez trivialement. Un truc sur lequel Taleb a raison est qu'on ne s'intéresse pas assez à la distribution du QI.

Share this post


Link to post
Share on other sites
6 hours ago, Lancelot said:

Si l'effet Flynn s'applique asymétriquement aux deux extrémités de la courbe de QI, alors ça devrait s'observer assez trivialement. Un truc sur lequel Taleb a raison est qu'on ne s'intéresse pas assez à la distribution du QI.

Résumé de Taleb en un article https://medium.com/incerto/iq-is-largely-a-pseudoscientific-swindle-f131c101ba39

  • Yea 1

Share this post


Link to post
Share on other sites

J'apprends sur le tweeter de NNT que Molyneux lui a fait 50 minutes de commentaire sur son article :icon_sommeil:

Share this post


Link to post
Share on other sites
Il y a 2 heures, Bézoukhov a dit :

J'apprends sur le tweeter de NNT que Molyneux lui a fait 50 minutes de commentaire sur son article :icon_sommeil:

Molyneux est manifestement extrêmement travailleur. Si il n'est pas encore millionnaire, c'est manifestement que son QI est son facteur limitant. :mrgreen:

  • Haha 4

Share this post


Link to post
Share on other sites

Comme son nom l'indique, Moulynex produit de la soupe en quantité industrielle.

  • Yea 1

Share this post


Link to post
Share on other sites
2 hours ago, Eltourist said:

Et comme je suis sympa je viens apporter le commentaire du commentaire :lol:

 

 

 

Un petit topo ? Parce que bon, le commentaire du commentaire de 50 minutes fait quand même plus d'1 heure :ninja:

Share this post


Link to post
Share on other sites
il y a une heure, Extremo a dit :

 

Un petit topo ? Parce que bon, le commentaire du commentaire de 50 minutes fait quand même plus d'1 heure :ninja:

 

 

Blancs > Noirs. Ne me remercie pas.

  • Haha 2

Share this post


Link to post
Share on other sites

Bon par contre y a eu une rapide passe d’armes entre NNT et Garett Jones. 

Ca serait cool si ça pouvait donner lieu à une vraie discussion ; il y a moyen que ce soit intéressant.

  • Yea 1

Share this post


Link to post
Share on other sites
Il y a 1 heure, Extremo a dit :

 

Un petit topo ? Parce que bon, le commentaire du commentaire de 50 minutes fait quand même plus d'1 heure :ninja:

 

Résumé écrit

Citation
While I share some of your views about the rhetoric, the fundamental points made by Nassim are largely ignored in this video.

A few notes:
1. On the question of fat tails and skewed non-normal distributions. Nassim's point is that the presence of such non-normal characteristics in the distribution of IQ render the statistical tests made by psychologists invalid for mathematical assumption reasons. You provide an explanation for why the tails could be skewed non-normally; a hypothesis whereby most intelligent people have more diverse and therefore less predictable lives. His point remains; he's saying the truths established by these statistical tests are untrue; you're saying you may have some explanation to explain how these truths came to be if they happen to be true.

2. You claim that North and South Korea are a form of on-going experiment showing that environment and genes can be separated, and we can assume that the differences between the two countries relate only or mostly to the environment. While this does not conflict with any point of Nassim, it remains incorrect. Tomorrow morning, if the east or west coasts were to separate from the US, the genes of the people in the east or west would be quite different from those who remain in the US. There are such bottleneck effects at all sizes of population, from small tribes of a hundred people separating up to nations forming like South and North Korea.

3. At 32:00 you say "I don't know what that means. High IQ is not always good for survival." But that is precisely Nassim's point. He's saying in this paragraph that there are potentially measures of success in future complex environments (what he calls intelligence), but that IQ is an inappropriate measure to represent this skill.

4. At 35:43 you say "He says that the only reason you'd want to focus on these issues is because you're a eugenecist and a racist." This is a slight strawman of the article since Nassim only mentions that some racists might be interested in using that data, not that it's 'the only' possibility.

5. At 35:52 you reject as stupid the idea that intentional perception (perceiving thoughts, intentions and goals in other people's actions) is important. I have to believe that you must have accidentally overstated your case since these features of the human mind are obviously important in strategic social decision-making. An entire branch of science studies this under the label 'theory of mind' and while we could argue over the science, I would hope that you don't think that such social perceptions are completely non-existent and unimportant.

6. At 36:20, you are attacking a strawman where you seem to believe that the use of 'crisp' by Nassim refers to the blurriness or non-blurriness of taking IQ measures from an average from a group and making conclusions about individuals. Nassim is not talking about that at all, he is presenting a view of the world in which the decisions that humans must face are blurred both in the clarity of the problem and the clarity of the proper response. In other words, he's framing the IQ test as relying on problems with complete information when the real world is filled with problems of incomplete information and we face an extremely noisy reality.

7. Between 37:00 and 37:20, you present a hypothetical explanation in which a brain drain would explain the lower IQ of certain countries. This is certainly not in line with the migration of the blacks of America, which was forced due to slavery and it doesn't seem that we know whether the IQ of slaves was lower or higher than their country of origins.

8. At 40:50 you seem to be confused by the point Nassim is making. He's essentially saying that recognizing all arbitrary patterns in real life would be extremely mal-adaptative as the key to success is in ignoring the uninteresting patterns. His critique is that the IQ test fails to detect this ability to ignore bad problems and rather forces the brain to solve them all (or get a reduced score).

At many points in the video, you agree with the essential statements of Nassim. For instance, you both agree that 1) IQ is not a measure of intelligence, it is only a measure of IQ. 2) IQ must be shaved off at the higher-skill range, which is equivalent to his statement that IQ is uninformative at high levels of intelligence. 3) That IQ can be useful in the low range (you provide the example of the army refusing people below 83, he says IQ is a measure of extreme unintelligence).

You do succeed at reframing the discussion to what matters, which in your view is immigration and the reduction of IQ in the west in general, and in that sense, you do differ from him at the pragmatic and political implementation levels, but fundamentally you two are agreeing.

 

  • Yea 1

Share this post


Link to post
Share on other sites
Le 07/01/2019 à 14:13, Bézoukhov a dit :

Bon par contre y a eu une rapide passe d’armes entre NNT et Garett Jones. 

Ca serait cool si ça pouvait donner lieu à une vraie discussion ; il y a moyen que ce soit intéressant.

 

Bon le problème de Taleb c’est qu’il a tendance à prendre les gens pour des cons :

 

 

 

 

  • Haha 1

Share this post


Link to post
Share on other sites

Wow. That's escalated quickly.

 

 

 

Bon, en substance, NNT n'a pas forcément tort ; toute façon, son argument mathématique principal est que l'on sous-estime la volatilité pour une raison X ou Y (X étant souvent "fat tail", Y souvent "faut tout décrire comme une v.a. dans votre modèle"), et que donc on ne peut rien conclure de rien.

Share this post


Link to post
Share on other sites

Il fait très one-trick pony oui. Après on peut lire le même genre de trucs ici concernant par exemple la pertinence de regarder une "température moyenne" sur le thread du réchauffement climatique. Ce qui est ironique quand on sait que Taleb est réchauffiste. Science is hard.

Share this post


Link to post
Share on other sites

Taleb n’est pas rechauffiste. Je pense que tu le mets dans une pièce avec un mec du GIEC y a des morceaux qui ressortent.

Par contre il considère que le risque est trop élevé quand on a qu’une planète parce que Faaaat tails. En fait c’est sensiblement le même argument qu’il a quand il explique pourquoi tout le monde ne devrait pas placer ses sous sur les marchés action.

 

Au fond, faut pas débattre avec lui de sujets X ou Y. Il faut allier philosophie et physique pour lui poser la vieille question « Que puis-je savoir ? » Chez Taleb, la réponse est presque toujours « rien ».

Share this post


Link to post
Share on other sites

Il est quand même bien fétichiste de la fat tail pour un gars qui critique les fétichistes de la gaussienne. Et il est quand même bien sûr de lui pour un gars qui pense ne rien savoir.

  • Yea 2

Share this post


Link to post
Share on other sites

Tu vois ton pote philosopho-geek du lycée "tout ce que je sais c'est que je ne sais rien" ? Ben v'la ce que ça donne 40 ans et 10m$ plus tard :D .

  • Haha 1

Share this post


Link to post
Share on other sites

 En parallèle du débat sur la baisse du QI moyen, j'aimerais savoir si quelqu'un a des infos (ou carrément une source) qui montreraient que le nombre de haut QI augmente malgré tout (en valeurs et non en volumes donc) ? Du fait de l'augmentation de la population par exemple...

 

 Quelqu'un ? 

Share this post


Link to post
Share on other sites
16 hours ago, Nigel said:

 En parallèle du débat sur la baisse du QI moyen

En admettant que l'effet Flynn a une limite d'une part on est loin de l'avoir atteinte pour tout le monde et d'autre part je suis bien plus convaincu par une stagnation à terme qu'une inversion.

 

16 hours ago, Nigel said:

le nombre de haut QI augmente malgré tout (en valeurs et non en volumes donc) ? Du fait de l'augmentation de la population par exemple...

Quel est le taux de baisse proposé pour le QI ? Quel est le taux d'augmentation de la population ? À partir de ces éléments le calcul est trivial.

Share this post


Link to post
Share on other sites

Je remonte le thread parce que je suis tombé en cherchant des trucs sur le système nerveux des homards (ça m'arrive) sur cet article qui me fait forte impression parce que je n'y connais rien : https://www.mdpi.com/2624-8611/1/1/1/htm

Il donne une explication climatologique des différences de niveaux de "groupes d'ancêtres biogéographiques" (je crois que @Mégille en parlait) :

Citation

The cause(s) of inter-BGA differences in cognitive ability specifically is the subject of several evolutionary theories, which decompose broadly into two categories: (1) Pleistocene-selection models and (2) Holocene-selection models. The former posit the action of evolutionarily novel factors that differed systematically in the regions to which human groups migrated after leaving Africa around 60 to 100 kya (thousand years ago), and which specifically selected for increased cognitive ability as a mechanism for enhancing survival [15]. Salient evolutionarily novel challenges likely included the presence of seasonality, specifically cold winters during the main Würm glaciation event 60 kya, when temperatures in Europe and (especially) Northeast Asia were considerably lower on average than they are today. Extreme cold, coupled with the challenge of provisioning for the future such as to anticipate seasonality (e.g., food storage) have been proposed as sources of selection for higher cognitive ability and other somatic traits related to innovation and productivity [16,17]. These climatic models predict that North East Asian and European BGA groups will have higher levels of cognitive ability than Pacific Islanders, sub-Saharan Africans, and Amerindians (at least from the tropical and subtropical regions of the Americas).

 

La corrélation me semble poser deux problèmes : l'utilisation de la taille du cerveau comme critère pour mesurer l'intelligence et la controverse pour savoir si the species already possessed larger brains when they invaded more seasonal regions. Je ne sais pas si ça peut t'intéresser, @Lancelot.

Share this post


Link to post
Share on other sites
17 hours ago, Vilfredo Pareto said:

l'utilisation de la taille du cerveau comme critère pour mesurer l'intelligence

Vaste question. Pour la faire très courte (sinon il faut que je fasse une review avec une petite dizaine d'articles au minimum) la taille du cerveau dans l'absolu n'est pas un bon critère pour comparer l'intelligence (quoi que ça veuille dire) entre espèces, sinon les éléphants et les baleines sont plus intelligents que les humains. Par contre intra-espèce ça peut se défendre, et particulièrement intra-l'espèce qu'on connait le mieux et dont on peut le mieux mesurer l'intelligence, l'humain, aux dernières nouvelles c'est corrélé de manière robuste. Maintenant corrélation ou causalité, inné ou acquis, tout ça tout ça...

  • Yea 2

Share this post


Link to post
Share on other sites
il y a une heure, Lancelot a dit :

intra-espèce ça peut se défendre

Parfait merci. D'ailleurs c'est aussi utilisé pour défendre les différences biologiques innées entre les sexes.

Citation

They note that others have found, for example, that female infants have significantly greater brain volume in the prefrontal cortex compared with males. They conclude that “It seems likely that these volumetric differences [found after birth] are mirrored by [the] differences observed in the present study.”

Faut-il en conclure que les femmes sont plus intelligentes que les hommes ? Je crois qu'il n'y a pas de différences significatives en termes de QI... Après ça sort un peu du sujet ("ethno-différentialisme").

Share this post


Link to post
Share on other sites
1 hour ago, Vilfredo Pareto said:

Parfait merci. D'ailleurs c'est aussi utilisé pour défendre les différences biologiques innées entre les sexes.

Faut-il en conclure que les femmes sont plus intelligentes que les hommes ? Je crois qu'il n'y a pas de différences significatives en termes de QI... Après ça sort un peu du sujet ("ethno-différentialisme").

Autre énorme champ de mine (oui il y en a plein, c'est le sujet qui veut ça).

 

En vérité on observe plutôt la tendance inverse, par exemple :

Quote
van der Linden, D., Dunkel, C. S., & Madison, G. (2017). Sex differences in brain size and general intelligence (g). Intelligence, 63, 78-88.

 

Utilizing MRI and cognitive tests data from the Human Connectome project (N = 900), sex differences in general
intelligence (g) and molar brain characteristics were examined. Total brain volume, cortical surface area, and
white and gray matter correlated 0.1–0.3 with g for both sexes, whereas cortical thickness and gray/white matter
ratio showed less consistent associations with g. Males displayed higher scores on most of the brain
characteristics, even after correcting for body size, and also scored approximately one fourth of a standard
deviation higher on g. Mediation analyses and the Method of Correlated Vectors both indicated that the sex
difference in g is mediated by general brain characteristics. Selecting a subsample of males and females who were
matched on g further suggest that larger brains, on average, lead to higher g, whereas similar levels of g do not
necessarily imply equal brain sizes.

 

Avec la figure qui va bien (j'entends d'ici Taleb s'arracher le peu de cheveux qui lui reste) :

1-s2.0-S0160289616302975-fx1a.jpg

 

Mais bien souvent aucune différence n'est trouvée et la question est directement adressée par Lynn :

Quote
Lynn, R. (2017). Sex differences in intelligence: The developmental theory. Mankind Quarterly, 58(1), 58-1.

 

It is a paradox that males have a larger average brain size than females, that brain size is positively associated with intelligence, and yet numerous experts have asserted that there is no sex difference in intelligence. This paper presents the developmental theory of sex differences in intelligence as a solution to this problem. This states that boys and girls have about the same IQ up to the age of 15 years but from the age of 16 the average IQ of males becomes higher than that of females with an advantage increasing to approximately 4 IQ points in adulthood.

 

Un autre gros problème dont j'ai dit il y a quelques mois que je voulais l'aborder (mais je n'ai pas le temps d'avoir le temps...) est que toutes ces études (y compris celles sur la relation entre QI et taille du cerveau) tendent à utiliser la "method of correlated vectors" introduite par Jensen (Jensen, A. R. 1998. The factor. Westport CT. Praeger), c'est-à-dire à utiliser les g-loadings pour calculer g. Sans rentrer dans les détails c'est une bidouille statistique un peu mystique qui a pour résultat de booster beaucoup de corrélations... et qui ne manque pas de critiques virulents :

Quote

Wicherts, J. M. (2017). Psychometric problems with the method of correlated vectors applied to item scores (including some nonsensical results). Intelligence, 60, 26-38.

 

Highlights

• The method of correlated vectors (MCV) is often used to study group differences in performance on cognitive test items

• It is shown here that this method does not convey much information about the role of g in the ethnic groups being compared

• MCV was applied to scenarios in which cognitive tests were replaced by measures of mood or personality in one of the two groups in the analysis

• MCV incorrectly suggested group differences in g (a Jensen Effect) even though group differences could not reasonably be on g

 

Abstract

Spearman's hypothesis stating that ethnic group differences on cognitive tests are most pronounced on the most highly g loaded tests has been commonly tested with Jensen's method of correlated vectors (MCV). This paper illustrates and explains why MCV applied to item-level data does not provide a test of measurement invariance and fails to provide accurate information about the role of g in group differences in test scores. I focus on studies that applied MCV to study group differences on items of Raven's Standard Progressive Matrices (SPM). In an empirical illustration of the psychometric problems with this method, I show that MCV applied to 60 SPM items incorrectly yields support for Spearman's hypothesis (so-called Jensen Effects suggesting that the group difference is on g) even when the items in the second group are not from the SPM but rather from a test composed of 60 items measuring either anxiety and anger or the big five personality traits. This shows that MCV applied to item level data does not accurately reflect the degree to which item bias or g plays a role in group differences. I conclude that MCV applied to items lacks both sensitivity and specificity.

 

Pour le plaisir des deux du fond qui sont toujours réveillés, je ne résiste pas à vous proposer une "critique féministe" à la prose pitoresque :

Spoiler
Bryant, K. L., Grossi, G., & Kaiser, A. (2019). Feminist interventions on the sex/gender question in neuroimaging research.
 

[Je saute le début de l'abstract qui est une longue enfilade de platitudes]

This debate raises many questions. Broadly, why is there a renewed interest in sex/gender difference in neuroscience research? Why might a scientist consider it riskier to report a failure to find differences than to disseminate findings that have poor chances of replication? Why might a journal editor require nearly all neuroscientific papers to measure “sex as a biological variable,” and so implicitly privilege sex over other biological variables, regardless of the goal of the study in question?

Feminist scientists have already started to tackle these and other questions pertaining to the neuroscience of sex/gender (e.g., Fausto-Sterling 2000; Kaiser et al. 2009; Roy 2012; Fine 2010a, 2010b; Jordan-Young 2010; Kuria 2012; Joel 2012; Rippon et al. 2014; Bluhm 2013; Schmitz and Höppner 2014; Rippon et al. 2017; Maney 2015). Their approaches have been inspired by feminist science and technology studies, which aim to identify and interrogate pretheoretical assumptions regarding sex/gender (e.g., equation of femaleness with passivity, assumption of male-typical traits and behaviors as normative or superior, the notion that biological sex determines social expressions and institutional arrangements of gender, and the assumption that sex/gender generally sorts bodies and behavior into only two dimorphic groups). The work of feminist neuroscientists is driven by a commitment to scientific rigor, not a commitment to a particular ideology. Their critique has focused on issues of statistics, measurement, methodology, and replication; it has offered alternative explanations and contributed to new interpretations of empirical neuroscientific data (e.g., Joel et al. 2015; Joel and Fausto-Sterling 2016; Kaiser et al. 2009; Fausto-Sterling 2005a, 2005b; Fine 2013; van Anders and Watson 2006; Joel et al. 2018; Rippon et al. 2017). In a nutshell, by critically analyzing the field’s dominant assumptions, theories, and practices, feminist scientists call for more rigorous science. We, as feminist scientists, see our work situated in this tradition.

This paper contributes to this field first by reviewing some outstanding issues pertaining to the question of sex/gender in the human brain, with a special focus on neuroimaging data; second by proposing several methodological interventions for more accurately interrogating sex/gender differences; and third by applying these interventions to an examination of similarities and differences between women and men in major white matter tracts using diffusion MRI (dMRI) data from a publicly available neuroimaging dataset. The interventions we describe and apply here include running empirical tests of randomization control and permutation analyses to protect against false positives, calculating measures of both difference (i.e., effect size) and overlap/similarity between groups, and reporting null results. Lastly, we discuss our findings and implications for study design with the aim to create a template for greater transparency in sex/gender neuroscience.

 

  • Yea 1

Share this post


Link to post
Share on other sites
il y a 35 minutes, Lancelot a dit :

la figure qui va bien

Justement à propos de la figure, j'ai l'impression que pour toutes les mesures sauf le GM/WM ratio et la "cortical thickness", les scores féminins et masculins tendent à se confondre pour des niveaux de g plus élevés et (donc ?) de plus volumineux cerveaux. La distribution serait donc plus égale pour des niveaux de g plus importants ou je lis de travers ? Merci d'avoir pris le temps de faire une réponse si détaillée.

 

il y a 40 minutes, Lancelot a dit :

j'entends d'ici Taleb s'arracher le peu de cheveux qui lui reste

Why though ?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...