Thursday, 31 March 2016

Heartbeats, crime, genetics: tudo bem

In the last post I suggested that although low heart rates might have an influence on violent crime, other biological measures needed to be considered.

Quick as a flash, reader East Coast came up with a relevant reference, and lest this languish in the comment columns I thought I would highlight the main findings, which are at the end of the abstract.!po=1.04167

The genetic and environmental basis of a well-replicated association between antisocial behavior (ASB) and resting heart rate was investigated in a longitudinal twin study, based on two measurements between the ages of 9 and 14 years. ASB was defined as a broad continuum of externalizing behavior problems, assessed at each occasion through a composite measure based on parent ratings of trait aggression, delinquent behaviors, and psychopathic traits in their children. Parent ratings of ASB significantly decreased across age from childhood to early adolescence, although latent growth models indicated significant variation and twin similarity in the growth patterns, which were explained almost entirely by genetic influences. Resting heart rate at age 9–10 years old was inversely related to levels of ASB but not change patterns of ASB across age or occasions. Biometrical analyses indicated significant genetic influences on heart rate during childhood, as well as ASB throughout development from age 9 to 14. Both level and slope variation were significantly influenced by genetic factors. Of importance, the low resting heart rate and ASB association was significantly and entirely explained by their genetic covariation, although the heritable component of heart rate explained only a small portion (1–4%) of the substantial genetic variance in ASB. Although the effect size is small, children with low resting heart rate appear to be genetically predisposed toward externalizing behavior problems as early as age 9 years old

Prof Joe Murray, lead author of the heart rate paper I reviewed has sent in some further material:

Childhood behaviour problems predict crime and violence in late adolescence: Brazilian and British birth cohort studies

Crime and violence in Brazil: Systematic review of time trends, prevalence rates and risk factors

Additionally, Prof Murray says “Yes, it was the gap in knowledge about offenders identified in the 2013 review that motivates work in this in the Pelotas cohorts” so we may get better details about the ethnic composition of offenders in due course.

Wednesday, 30 March 2016

The heartbeat of crime

My eye was caught by a paper, summarised in the BPS Research Digest of 17th March, on the relationship between low resting heart rate and crime. The conclusion was that adolescents with low resting heart rate were more likely to go on to commit crimes, including violent crimes, possibly because they did not get too fussed when attacking others. Cool, calm, collected and lethal. All this takes place in the city of Pelotas, Rio Grande do Sul, Brazil.

Low resting heart rate is associated with violence in late adolescence: a prospective birth cohort study in Brazil. Joseph Murray, Pedro C. Hallal, Gregore I. Mielke, Adrian Raine, Fernando C. Wehrmeister, Luciana Anselmi and Fernando C. Barros. International Journal of Epidemiology, 2015, 1–10 doi: 10.1093/ije/dyv340

Abstract Background: Youth violence is a major global public health problem. Three UK and Swedish studies suggest that low resting heart rate predicts male youth violence, but this has not been tested in other social settings nor for females. Methods: A prospective, population-based birth cohort study was conducted in Pelotas, Brazil. Heart rate was measured using a wrist monitor at ages 11, 15 and 18 years. Violent crime and non-violent crime were measured at age 18 in self-reports and official records (N=3618). Confounding variables were assessed in the perinatal period and at age 11, in interviews with mothers and children. Logistic regression was used to estimate associations between quartiles of heart rate at each age, and violent and non-violent crime at age 18, separately for males and females. Results: Lower resting heart rate was a robust correlate of violent and non-violent crime for males. Comparing males in the lowest and top quartiles of heart rate at age 15 years, adjusted odds ratios were 1.9 for violent crime [95% confidence interval (CI) 1.4–2.7] and 1.7 for non-violent crime (95% CI 1.1–2.6). For females, crime outcomes were associated only with low resting heart rate at age 18. Associations were generally linear across the four heart rate quartiles. There was no evidence that associations differed according to socioeconomic status at age 15. Conclusions: Low resting heart rate predicted violent and non-violent crime for males, and was cross-sectionally associated with crime for females. Biological factors may contribute to individual propensity to commit crime, even in a middle-income setting with high rates of violence.

This looks good. A birth cohort of good size, repeated measures, and a prospective basis for judging an important behaviour: violence. In addition to repeatedly measuring heart rate and checking public records for offences (and also looking at self-reports) the authors measured the following:

Confounding variables. The following variables were measured in the perinatal period in interviews with mothers: unplanned pregnancy (yes/no), mother smoked in pregnancy (yes/no), maternal alcohol use in pregnancy (yes/no), maternal age (years), number of siblings, maternal education (years of schooling) and family income (minimum wages per month). All were previously found to be associated with violence at age 18 years for men or women in this sample. Self-reported skin colour was measured and categorized as white, black, mullato/brown, yellow and indigenous. The following participant characteristics were measured at age 11 years: smoking (yes/no), drinking (yes/no), physical activity (min per week in leisure time physical activity and active transportation to/from school), height (centimetres), weight (kilograms), and blood pressure (mmHg continuous). Maternal mental health was also measured when participants were aged 11, using the Self Report Questionnaire (SRQ), previously validated in a Brazilian sample of 485 subjects.22 The continuous SRQ score from 0 to 20 was used in the analyses.

The results: The prevalence of violent crime at age 18 was 26.6% among men and 11.3% among women (P < 0.001). Equivalent figures for non-violent crime were 14.8% and 5.8%, respectively (P < 0.001). Among those who committed violence (N= 679), less than half (44%) also committed a non-violent crime; by contrast, among those who committed non-violent crime (N ¼ 367) the vast majority (82%) also committed violence.

The prevalence of violent crime was 34% among men in the first quartile of heart rate (low heart rate) at 15 years, whereas it was 20% among those in the fourth quartile (high heart rate). Equivalent figures for non-violent crime were 19% and 12%, respectively.

This is an enormously high rate of violence, which needs some explanation. Within that violent population, those with low heart rates are even more violent, a significant increase in proneness to violence.

The authors go on to observe in their discussions: The key finding of this study is that lower heart rate was a robust predictor of male violent and non-violent crime. Although only cross-sectional associations were found for females, prospective and concurrent associations were observed for males after adjusting for a range of confounding variables. The main strengths of the study were: (I) the use of a large prospective community cohort; (ii) the repeated measures of heart rate; (iii) the multiple measures of crime; (iv) the inclusion of females as well as males; and (v) the wide range of confounding variables included.

They add: It should also be considered that the current study was conducted in one Brazilian city, and results should not be generalized to the rest of the country. Pelotas is a relatively poor city in a relatively rich southern state of Brazil. When crime data were collected for this study in 2011, there were 18.9 homicides in Pelotas per 100 000 population, lower than the national rate of 27.1 but considerably higher than in England and Wales (1.3) and Sweden (0.9) where previous studies of heart rate and violence have been conducted.

So, something raises the level of violence in Brazil 30 times above Sweden, and 21 times above England and Wales. In the grizzly list of the 50 most dangerous cities in the world, Brazil accounts for 21 of them and Pelotas, violent as it is at 19 homicides per 10,000, is not one of them. Here they are, so have a quick scan and see if any patterns present themselves:

The 50 Most Violent Cities in the World 2015

Rank  City Country            Homicide Rate/10,000

1 Caracas Venezuela                      120

2 San Pedro Sula Honduras             111

3 San Salvador El Salvador             109

4 Acapulco Mexico                        105

5 Maturín Venezuela                       86

6 Distrito Central Honduras             74

7 Valencia Venezuela                     72

8 Palmira Colombia                        71

9 Cape Town South Africa             66

10 Cali Colombia                            64

11 Ciudad Guayana Venezuela       62

12 Fortaleza Brazil                        60

13 Natal Brazil                              61

14 Salvador (y RMS) Brazil           60

15 St. Louis U.S.A.                      59

16 João Pessoa (conurb)Brazil      58

17 Culiacán Mexico                      56

18 Maceió Brazil                          56

19 Baltimore U.S.A.                    55

20 Barquisimeto Venezuela          55

21 São Luís Brazil                       53

22 Cuiabá Brazil                         49

23 Manaus Brazil                       48

24 Cumaná Venezuela               48

25 Guatemala Guatemala           48

26 Belém Brazil                        46

27 Feira de Santana Brazil        46

28 Detroit U.S.A.                    43

29 Goiânia Brazil 847              43

30 Teresina Brazil                  42

31 Vitoria Brazil                    42

32 New Orleans U.S.A.        41

33 Kingston Jamaica             41

34 Gran Barcelona Venezuela 40

35 Tijuana Mexico                   39

36 Vitória da Conquista, Brazil  38

37 Recife Brazil                        38

38 Aracaju Brazil                      38

39 Campos dos Goytacazes Brazil 36

40 Campina Grande Brazil             36

41 Durban South Africa                 35

42 Nelson Mandela Bay South Africa 36

43 Porto Alegre Brazil                         35

44 Curitiba Brazil                                35

45 Pereira Colombia                            33

46 Victoria Mexico                              31

47 Johannesburg South Africa               30

48 Macapá Brazil                                 30

49 Maracaibo Venezuela                     29

50 Obregón Mexico                             28

(The usual disclaimers apply: in places which are really, really violent there are few people left standing to stick around and count the bodies. That is done later by other techniques, if ever at all. Over lunch a few years ago John Sloboda explained to me how he calculated the body count in Iraq.)

However, within bodily error, the list gives pause for thought. Central America, South Africa, Brazil and USA all figure prominently. In that latter democracy  St. Louis, Baltimore, Detroit, and New Orleans are very violent. Whatever the reason, it would be prudent to check whether the same reason may apply in Pelotas.

The authors conclude: It is a striking conclusion that an individual-level biological characteristic, such as heart rate, is associated with crime in a Brazilian sample, given the high levels of serious violence in Brazil. One might speculate that individual level factors would be irrelevant in this social context, because of major socio-cultural drivers of crime and violence, including poverty, inequality, gangs, drug trafficking and corrupt and under-resourced criminal justice systems. However, the current study demonstrates that, even in this setting, a fully integrated biopsychosocial understanding of violence is required.

This paper does many good things, as far as it goes, so I wanted to push it further. The authors are willing to consider biological causes of crime, which is a good thing. What would an integrated bio-psychological understanding of violence involve? When reading a paper I ask myself whether the authors been psychologists, and whether they tested people for intelligence and personality, and if not, why not? These measures are not mentioned in the paper. Perhaps they were not included in the original study. If so, it is a great pity, because we psychologists must get ourselves together. Doctors measure blood pressure: can’t psychologists measure intelligence and personality, the most psychological aspects of people?

Also, I think one needs to sort out the genetics. I have looked in the supplementary tables, and also contacted the lead author to see if there may be further analyses either available or planned on the racial composition of these adolescents. Skin reflectance measures were taken, but appear to have contributed nothing to crime rates. This is worthy of publication itself. In the absence of data and discussion, here are a few pointers.

First, there is some evidence that heart rates vary by racial background.

Emergence of Ethnic Differences in Blood Pressure in Adolescence The Determinants of Adolescent Social Well-Being and Health Study. Seeromanie Harding, Melissa Whitrow, Erik Lenguerrand, Maria Maynard, Alison Teyhan, J. Kennedy Cruickshank, Geoff Der

Abstract—The cause of ethnic differences in cardiovascular disease remains a scientific challenge. Blood pressure tracks from late childhood to adulthood. We examined ethnic differences in changes in blood pressure between early and late adolescence in the United Kingdom. Longitudinal measures of blood pressure, height, weight, leg length, smoking, and socioeconomic circumstances were obtained from London, United Kingdom, schoolchildren of White British (n692), Black Caribbean (n670), Black African (n772), Indian (n384), and Pakistani and Bangladeshi (n402) ethnicity at 11 to 13 years and 14 to 16 years. Predicted age- and ethnic-specific means of blood pressure, adjusted for anthropometry and social exposures, were derived using mixed models. Among boys, systolic blood pressure did not differ by ethnicity at 12 years, but the greater increase among Black Africans than Whites led to higher systolic blood pressure at 16 years (2.9 mm Hg). Among girls, ethnic differences in mean systolic blood pressure were not significant at any age, but while systolic blood pressure hardly changed with age among White girls, it increased among Black Caribbeans and Black Africans. Ethnic differences in diastolic blood pressure were more marked than those for systolic blood pressure. Body mass index, height, and leg length were independent predictors of blood pressure, with few ethnic-specific effects. Socioeconomic disadvantage had a disproportionate effect on blood pressure for girls in minority groups. The findings suggest that ethnic divergences in blood pressure begin in adolescence and are particularly striking for boys. They signal the need for early prevention of adverse cardiovascular disease risks in later life. (Hypertension. 2010;55:1063-1069.)

In the light of these findings, it would be good to see the blood pressure measures in this paper broken out by ethnic group. To my eye the differences aren’t very big, but it is worth checking. Also, there are skin luminosity findings and self reports of racial identification, but those are not shown in the paper, nor in the supplementary tables. This is a significant omission, but easily rectified. Null findings should get publicity.

Second, it would be relevant to look at the rates of violence by ethnic group. In other parts of the world these vary widely, at least six-fold. It would be important to see whether Brazil conforms to this pattern, or is an exception. The best validated data comes from the US, and currently I do not know of comparable Brazilian data.

Since the US and Brazil have had different histories as regards race, with the latter being far more relaxed about inter-marriage, so the contrast is potentially informative. As far as I can see, the outcomes for Africans should be better in Brazil, because they were subjected to less racial discrimination. However, the violence rates in the above table suggest that my hypothesis about the effects of racial discrimination is wrong.

As to what it is like to live in Pelotas, a lady friend writes:

Yes, there is violence here in Pelotas and has been for years. I was assaulted at the front gate of my condominium by 2 young boys on a motorbike with a gun. Luckily one of my neighbours had just come in from his farm and the kids saw him, took my briefcase and left. From then on violence has grown to frightening levels, although one learns to live with it. To what level? Well, there is a restaurant a block away from my building and when I go there at night, I take my car as there is no way I would walk one block in the dark back home. When I go out at night to friend's or restaurants I usually go in a taxi and come home in one as the driver waits for me to get into the condominium. If I go by car, on getting home I'm a sitting duck as I wait for the gate to open and close.

As for your question about "which adolescent boys/men are violent to whom", I don't have precise research data, but there are a lot of men being violent to women and kids, although we have had the "Maria da Penha law" for years (My note: legislation seeking to prevent domestic violence). Other laywoman remarks are:

- poor black boys are violent to just about anyone, mostly involved with drugs (I think).

- poor men black and white ... To whom? As an exact thing, I don't know... Just about anyone.

I contacted this friend without knowing this part of her history, and the above remarks are shown verbatim.

A 2005 survey gives the Pelotas population figures as 280,897 whites, 34,172 blacks, 25,395 of mixed ethnicities, 998 native Brazilians, 498 Asians, and 998 of unknown ethnicity. The 1993 birth cohort sample probably conforms to these proportions. These are sufficiently large for violence rates to be compared, particularly when the overall rate is so high.

Happily, we have other estimates of the European fractions of Brazilian states, and those also have intelligence estimates (Fuerst and Kirkegaard 2016). As a general rule, lower intelligence is usually associated with higher crime. This is not just because they get caught, because it also shows up on self-report.

The State of Rio Grande do Sul, in which the city of Pelotas is situated, is 79% European 9.5% African and 11% American Indian. The estimated IQ is 87.5 which is high for Brazil, but low in international terms, and consistent with poorly organised states and higher crime rates.  Since Pelotas is described as a poor city in a rich state, one might assume that the ability levels were even lower than the state average.

So, one might hazard a guess that some mixture of intelligence and personality is causally involved in crime and violence. Whether heart rate is the most powerful of the biological measures remains to be seen, and is worth testing against broader biological measures like population stratification.

Sunday, 27 March 2016

What intelligence researchers think about heritability


I discussed this survey of intelligence researchers in December 2013

Asked: Is there sufficient evidence to arrive at a reasonable estimate of the
heritability of intelligence in populations of developed countries?” 73% said Yes.

Asked: What are the sources of U.S. black-white differences in IQ?

0% of differences due to genes: (17% of our experts)
0-40% of differences due to genes: 42% of our experts
50% of differences due to genes: 18% of our experts
60-100% of differences due to genes: 39% of our experts
100% of differences due to genes: (5% of our experts)
M=47% of differences due to genes (SD=31%)

Now we have the first publication on this survey:

Survey of Expert Opinion on Intelligence: Causes of International Differences in Cognitive Ability Tests Heiner Rindermann David Becker and Thomas R. Coyle

Intelligence March 2016 | Volume 7 | Article 39

Following Snyderman and Rothman(1987,1988), we surveyed expert opinions on the current state of intelligence research. This report examines expert opinions on causes of international differences in student assessment and psychometric IQ test results. Experts were surveyed about the importance of culture, genes, education(quantity and quality), wealth, health, geography, climate, politics, modernization, sampling error, test knowledge, discrimination, test bias, and migration.The importance of these factors was evaluated for diverse countries, regions, and groups including Finland, East Asia, sub-Saharan Africa, Southern Europe, the Arabian-Muslim world, Latin America, Israel, Jews in the West, Roma (gypsies), and Muslim immigrants. Education was rated by N=71 experts as the most important cause of international ability differences. Genes were rated as the second most relevant factor but also had the highest variability in ratings. Culture, health, wealth, modernization, and politics were the next most important factors, whereas other factors such as geography, climate, test bias, and sampling error were less important. The paper concludes with a discussion of limitations of the survey (e.g.,response rates and validity of expert opinions).

Differences between countries with the lowest and highest ability levels are large. For example, in TIMSS 2011, 4th grade Yemeni pupils achieved 209 student assessment study (SAS) points, whereas South Korean pupils achieved 587 SAS points. If SAS points are converted to IQ points, the Yemeni would have an IQ of 56 and the Koreans would have an IQ of 113, a difference of 11 years of schooling. Psychometric IQ studies show similar results. For example, Malawi has an estimated IQ of 60, whereas Singapore and Hong Kong have estimated IQs around 108, a difference that translates into SAS≈233 and 555 or 16 years of schooling.

When authors give ability differences as “years of schooling” this always provokes the response that countries with lower levels of ability need more schooling. It really means “despite schooling, as if they permanently required 16 more years of schooling”. The main reason that intelligence is not a popular subject is that it has been shown not to be very malleable. At a rough estimate, anyone of IQ 93 and below finds it difficult to earn good wages.

The authors explain that: the conversion transforms the SAS-scale (M = 500, SD = 100) in developed countries to an IQ scale (M = 100, SD = 15). We assumed an increase of 35 SAS points, or 3 IQ points per year of schooling (Winship and Korenman, 1997; Rindermann, 2011).

In the current study, data collection procedures were designed to ensure anonymity. The anonymity was implemented to reduce pressure for socially desirable responses, and to increase the likelihood of obtaining honest opinions. Opinions made in anonymity (without fear of retribution) may differ from public appraisals such as those reported in Gottfredson’s (1994) “Mainstream Science on Intelligence,” which was signed by more than 50 researchers.



Notice of the study was emailed to experts who published articles on or after 2010 in journals on intelligence, cognitive abilities, and student achievement. The journals included Intelligence, Cognitive Psychology, Contemporary Educational Psychology, New Ideas in Psychology, and Learning and Individual Differences. Notice of the study was also emailed to members of the International Society for Intelligence Research (ISIR), and posted to the web site for the International Society for the Study of Individual Differences (ISSID). ISIR and ISSID support intelligence research and host professional conferences with intelligence researchers. Finally, the study was announced at the 2013 ISIR conference in Melbourne, Australia. A total of 1345 people received an email invitation. An expert was defined as a person who had published on cognitive ability or who had attended intelligence conferences and presented research. Compared to Snyderman and Rothman (1988, pp. 46–49), our selection criteria were based more on publications in specific scientific journals and less on membership in scientific organizations. In addition, we used email and a web based survey rather than traditional mail and paper-pencil surveys.

The low response rates may be attributed to the length of the survey (which took about 40–90 min to complete), self-censorship, or fear of addressing a controversial subject (despite assurances of anonymity). The low response rates may also reflect a paucity of experts on intelligence and international differences in cognitive ability. There may be 20–50 scientists who study international differences in intelligence. Based on this estimate, the number of respondents (71 people) may exceed the number of scientists who study the topic!

Speaking personally, I found the survey too long, and too interested in subtle points of distinction between various ways of describing intelligence. However, I am easily bored.

Around 90% of experts believed that genes had at least some influence on cross-national differences in cognitive ability.

National IQ causes

As you can see, experts back genes, education quality and quantity, culture and health, plus a scattering of other causes as being the reasons for national differences in ability.

Assuming that the survey is representative of expert opinions, genetic factors should receive more attention in future research and public debates. To fairly consider different hypotheses, future research should incorporate procedures (e.g., rules for methods of argumentation) that reduce zeitgeist or political pressures that may bias responses on controversial issues (e.g., Segerstråle, 2000; Jussim et al., 2015).

Education was measured with two items, environmental factors with 11 items, and genetics with one item. To better estimate the importance of nature and nurture, a single binary question could be added to future surveys (e.g., “Which is more influential, genetic or environmental factors?”).

A single empirical study can contradict expert opinion, and the results of the current survey must be validated in future empirical research.

This last observation is crucial. Expert opinion is no more than an indication of the current state of the argument among informed persons. A new discovery can change the argument. More genetic research, particularly on racial differences on intelligence (which as far as I know has never been funded) would be very likely to strengthen the genetic interpretation, but a negative result would be very interesting. Solid proof of environmental manipulations and educational techniques to overcome national and racial differences in intelligence would swing the argument in favour of environmental explanations.

Perhaps we should predict what such a survey will find by 31 December 2020. I assume there will be no funding for such research, and that results will have to be obtained by making assumptions about censored data. I am not a super-forecaster, but I think that genetic explanations for national and racial differences in intelligence, as shown in Table 1, will rise from 20% to 30% but the standard deviation will remain just as large.  

What are your precise predictions?

Sunday, 20 March 2016

What makes problems difficult?


Psychologists have been better at measuring intelligence than explaining how they do so. “The indifference of the indicator” is all very well, but this dictum has been met with public indifference and incomprehension. This is because psychometricians keep saying that intelligence matters, but then put their foot in it by saying “but how you test it doesn’t matter”. Technically, this is correct: it does not matter precisely what the test is, so long as it has sufficient difficulty to stretch minds and grade them. In that sense the actual indicator of intelligence is a matter of indifference, but only so long as it has the necessary psychometric properties.

I try to get round this problem of understanding by giving the example of digit span: remembering digits forward is easy (and only weakly predictive of general ability) but remembering digits backwards is harder (and more strongly predictive of general ability). In that difference lies the essence of difficulty.

It then gets rather technical. Some tests are good indicators at the lower end of ability, others at the higher end. They all have characteristics and quirks. Hence the reification of intelligence test results into g which satisfies most researchers but bemuses the general public.

Compare this with the forced expiratory volume test.

Forced expiratory volume (FEV) measures how much air a person can exhale during a forced breath. The amount of air exhaled may be measured during the first (FEV1), second (FEV2), and/or third seconds (FEV3) of the forced breath. Forced vital capacity (FVC) is the total amount of air exhaled during the FEV test.

Neat, isn’t it? (You can then study whether 30 mins of aerobic exercise over 8 weeks raises the volumes. It does, a bit.) Can psychometrics define an intelligence measure in as simple a way?

Diego Blum, Heinz Holling, Maria Silvia Galibert, Boris Forthmann.  Task difficulty prediction of figural analogies. doi:10.1016/j.intell.2016.03.001

The purpose of this psychometric study is to explain performance on cognitive tasks pertaining Analogical Reasoning that were taken into consideration during the construction of a Test of Figural Analogies. For this purpose, a general Linear Logistic Test Model (LLTM) was mainly used for data analysis. A 30-itemed Test of Figural Analogies was administered to a sample of 422 students from Argentina, and eight of these items were administered along with a Matrices Test to 84 participants mostly from Germany. Women represented 77% and 76% of each respective sample. Indicators of validity and reliability show acceptable results. Item difficulties can be predicted by a set of nine Cognitive Operations to a satisfactory extent, as the Pearson correlation between the Rasch model and the LLTM item difficulty parameters r = .89, the mean prediction error is slightly different between the two models, and there is an overall effect of the number of combined rules on item difficulty (F(3,23) = 15.16, p < .001) with an effect sizeη2 = .66 (large effect). Results suggest that almost all rotation rules are highly influential on item difficulty. (my emphasis).


Rules for figure test

Figural matrices are a good test of intelligence. Raven dreamed his up from logical principles, using patterns he had seen on pottery in the British Museum. His test works very well, even though one difficult item among the 60 is placed a little too early in the B sequence. Incidentally, to my mind this placing error is one of the proofs that the test is reasonably culture fair, in that all racial groups find it difficult, without having to confer across continents about it.

Tests of this sort are known as the A:B::C:D analogies (A is to B as C is to D). When a problem is based on finding the missing element D of the analogy (I.e., A:B::C:?), then C:D becomes the target analog and A:B becomes the source analog. What needs to be extrapolated from one domain to the other is the compound of structural relations that binds these two entities, and not just superficial data (Gentner, 1983). The basic problem A:B::C:? can be applied to different types of contents, namely: verbal, pictorial and figural (Wolf Nelson & Gillespie, 1991).

How does one describe the difficulty level of each item? Mulholland, Pellegrino, and Glaser (1980) studied the causes of item difficulty in geometric analogy problems, and concluded that the number of item elements, as well as the number of transformations, had a significant effect on error rates.

These authors decided to build a test with designed levels of item difficulty, and chose to keep the same standard figures in all items, so as to reduce surface complexity and concentrate on underlying operational differences between items. They used 9 main rules to build the items, rotating the figures by 45, 90 and 180 degrees, using X and Y axis reflections, line subtractions and dot movements. You can call this: “How to build your own IQ test” and the supplementary material shows you how to do this. Note that certain rule combinations lead to some imprecisions and, therefore, the process of rule-based item generation should not be considered a pure-mechanical procedure. As a consequence, the authors have further explanations about their design guidelines which need to be understood


Rules and parameters Table 2


Based on the data provided in Table 2, specific rule-based contributions to item difficulty can be interpreted. The short clockwise main shape rotation, the subtraction and the dot movement rules make some contributions in this regard. Most interestingly, the best predictors of item difficulty are all the other rotation rules (I.e., both counter clockwise rotations, both long rotations, and the short clockwise trapezium rotation), followed by the reflection rule. Special mention must be given to the long clockwise trapezium rotation, which has the biggest influence on item difficulty. In other words, people found it most difficult to manipulate rotations during task resolution. In fact, the two easiest items according to the Rasch model (items 2 and 4) do not comprise rotation rules, nor does item 25 which is the 7th easiest item. Also, combining rules within a single item has an impact on item difficulty by itself, since both the ANOVA results and the Box Plot show that the higher the number of combined rules, the greater the item difficulties.

Rules and difficulty

I am aware that some of this has been done before, if only because I attended conferences years ago showing that an intelligence test could be constructed out of general principles of learning, and that it had good predictive value.

I think that this is a good paper which should be mentioned whenever critics assume that test material is arbitrary and unrepresentative in some way. This work establishes that rules of design complexity are strongly associated with the ease or difficulty human subjects experience when they solve problems.

One fly in the ointment: it seems that psychology is now 76% a girly subject and women are less good at mental rotation of shapes, so it might be good to check this with boys studying something other than psychology.

The authors found that this test works well at low as well as high levels of ability, which is particularly useful.

A high positive correlation (r = .89) reveals that item difficulties are strongly associated with the predicted difficulties of each rule, and these item difficulties remain practically unchanged in a further study.

By wary of comparison only, the test-retest correlation of the Wechsler after 6  months is 0.93, so the above correlation of 0.89 is a very strong endorsement of the design principles of the test created by the authors.

Perhaps we have taken a step towards finding out what makes problems difficult.

Take a closer look at the paper here:

Tuesday, 15 March 2016

Admixture in the Americas: European intelligence


John Fuerst and Emil Kirkegaard have been looking at some interesting data sets, and coming to interesting and succinct conclusion, with all their workings openly available for further testing.

John Fuerst and Emil Kirkegaard. Admixture in the Americas: Regional and National Differences. MANKIND QUARTERLY 2016 56:3 256

There are also 6 critical comment for readers to look at.

We conducted novel analyses regarding the association between continental racial ancestry, cognitive ability and socioeconomic outcomes across 6 datasets: states of Mexico, States of the United States, states of Brazil, departments of Colombia, sovereign nations and all units together. We find that European ancestry is consistently and usually strongly positively correlated with cognitive ability and socioeconomic outcomes (mean r for cognitive ability = .708; for socioeconomic well-being = .643) (Sections 3-8). In most cases, including another ancestry component, in addition to European ancestry, did not increase predictive power (Section 9). At the national level, the association between European ancestry and outcomes was robust to controls for natural-environmental factors (Section 10). This was not always the case at the regional level (Section 18). It was found that genetic distance did not have predictive power independent of European ancestry (Section 10). Automatic modelling using best subset selection and lasso regression agreed in most cases that European ancestry was a non-redundant predictor (Section 11). Results were robust across 4 different ways of weighting the analyses (Section 12). It was found that the effect of European ancestry on socioeconomic outcomes was mostly mediated by cognitive ability (Section 13). We failed to find evidence of international colorism or culturalism (I.e., neither skin reflectance nor self-reported race/ethnicity showed incremental predictive ability once genomic ancestry had been taken into account) (Section 14). The association between European ancestry and cognitive outcomes was robust across a number of alternative measures of cognitive ability (Section 15). It was found that the general socioeconomic factor was not structurally different in the American sample as compared to the worldwide sample, thus justifying the use of that measure. Using Jensen's method of correlated vectors, it was found that the association between European ancestry and socioeconomic outcomes was stronger on more S factor loaded outcomes, r = .75 (Section 16). There was some evidence that tourist expenditure helped explain the relatively high socioeconomic performance of Caribbean states (Section 17).

This paper is long, very detailed and content rich, and important. It follows the new and noble practice of showing all its workings, so the diligent reader can test the findings, and in that same spirit of openness includes critical opinions from other scholars: research as it ought to be.

12 zero-order correlational analyses found a substantial positive relationship of European ancestry with both cognitive ability and general socioeconomic well- being. Multiple regression results generally found that European ancestry remained a non-redundant positive predictor when including natural- environmental predictors in the models. Socioeconomic (S factor) scores in the United States were the sole exception. More research is needed on the relationship between socioeconomic outcomes and racial ancestry in that country.

Our path analysis and semi-partial analyses indicated that cognitive ability scores can largely statistically explain the association between ancestry and socioeconomic outcomes.

The association between racial ancestry and outcomes could be mediated by genetic, cultural or other factors (Rindermann, 2015). As it has been demonstrated that indices of genetic ancestry track an array of inter- generationally transmitted cultural traits (Spolaore & Wacziarg, 2015), the results are consistent with a cultural mediation model. While the association between racial ancestry and outcomes is also consistent with an evolutionary genetic model, to obtain decisive evidence in support of such a model, one would need to identify specific alleles that vary between ancestral groups which are directly (e.g., Piffer, 2015b) or (plausibly) indirectly (e.g., Fedderke et al., 2014) associated with cognitive and/or socioeconomic outcomes at the individual level (Rindermann, 2015).

Based on a review of dozens of studies, Fuerst and Kirkegaard (2015) found that racial ancestry was associated with inter-individual socioeconomic outcome differences within admixed populations (e.g., Black Trinidadian and Toboggans) throughout the Americas (e.g., in Brazil, Chile, Colombia, Mexico, Peru, Puerto Rico, Trinidad and Tobago and the United States). Across studies, African and Amerindian ancestry was negatively and European ancestry was positively associated with socioeconomic outcomes. It is an open question, however, as to whether, on this same level of analysis, cognitive ability is robustly associated with racial ancestry and as to whether cognitive ability mediates the biogeographic ancestry-socioeconomic outcome association. There are, at present, a number of datasets which allow for the testing of these hypothesis, such as the US based Pediatric Imaging, Neurocognition and Genetics (PING) survey and The National Longitudinal Study of Adolescent to Adult Health (Add Health) survey.

As for limitations, we wish to emphasize that our national level cognitive measures were suboptimal. Our intra-national level indexes of ancestry were often likewise. Also, measurement error can give problems with multiple regression type approaches resulting in false positives (Westfall & Yarkoni, under review), and it is unknown how measurement error and the SAC measure and control methods interact. We suspect that better measures will not substantively alter the results, as they generally were robust across different analyses and different measures. Nonetheless, replicating the analyses using better and more fine- grained measures would be worthwhile.

The conclusion the authors come to is that to understand the intelligence and social achievements of people in the United States of America, Mexico, Brazil and Colombia you need to know only one thing: how much European ancestry they have. This is pretty much a consistent finding in their samples, but they look through many other possible explanations, such as the contributions of other genetic groups, the special contribution of tourism to economies, and the depredations of other factors like parasite load, all covered in detail in their paper.

Here are the pictures for each of the states or provinces, and then the composite picture.

European ancestry and Mexican states

European ancestry and US states

European ancestry and Brazilian state

European ancestry and Colombian departments

European ancestry and nations

Above is the picture for sovereign nations.


European ancestry, states and nations combined


So, that’s the Americas sorted out. With the partial exception of US states, to the extent that they are European, they are bright and successful.

Monday, 14 March 2016

An urchin writes


Even in the cobbled streets streets of London one is not free from the impudence of urchins. The under-butler, rightly embarrassed, told me that when he answered the front door the bedraggled youth thrust the missive into his hand, muttered “Anapestic trimeter” and ran away into the fading foggy lamp-light, not even demanding a penny. The front door, I ask you! I blame the 1870 Education Act.

I show it exactly as found, including a telling reference to ISIR (the International Society for Intelligence Research) and offer a prize to anyone who can track the youth down, and turn him in to the proper authorities, for transportation.

Rushton and Jensen and Me


I took a young lady to dinner

In hopes I might take her to bed:

She looked at me over the table

And “What do you study?” she said.

I chewed on my nail and pondered

Just how in the world to reply;

But I could not contrive an invention,

So I told her the truth, with a sigh.

“You really––” she ventured (half-trembling),

“Ah, waiter, please bring us our checks.”

It’s been a full month and a few days,

And she’s not returned one of my texts.

            O, in this dark era a g-man

            Is a terribly hard thing to be:

            Suspicion and hate are our doom and our fate––

            Just ask Rushton, or Jensen, or me.


‘Twas a dull August morning in psych class,

When a-sudden the lecturer cited

A gap with which we’re all familiar…

Well, didn’t that leave me delighted!

Hurrah and three cheers! A good fellow!––

Then he said: “It’s environment, you see,

For the diff’rence I mentioned is absent

In children whose age is but three.”

I could not contain my annoyance,

And (beshrew it!) I shot up my hand….

The tale of the Racist Young Freshman

Still rings ‘cross the length of the land.

            O, in this dark era a g-man

            Is a terribly hard thing to be;

            Our data and figures are “aggressions” and “triggers”––

            Just ask Nyborg, or Watson, or Herrnstein or Murray,

            Or Rushton, or Jensen, or me.


In lecture-halls, seminars, meetings

Where we students have gathered to chat,

From the lesbian slam poets’ dorm room

To the den of the seediest frat––

Any mention of my avocation,

Without fail, will draw gasps and hisses;

The fellows will groan; and I’ll notice

A blanch on the cheeks of the misses.

So I hole up, alone, and I forego

The parties and clubs and such stuff;

If only they’d love and accept me…

My professors are assholes enough.

            O, in this dark era a g-man

            Is a terribly hard thing to be:

            Is there no one on Earth who perceives our works’ worth?

            Just ask Summers, Satoshi, or Sarich or Scarr,

            Or Nyborg, or Watson, or Herrnstein and Murray,

            Or Rushton, or Jensen, or me.


Will we tolerate these constant aggressions?

Will we just let The Man keep us down?

Or will we rise, and shake off our shackles,

And take to the streets of the town?

Let’s march on the quad and the library

And strike up an infernal clatter––

Our banner emblazoned with Matrices,


We will throw out the meanies and bigots…

And then we will finally be free,

In a world where no one is enragèd

At an I or a Q or a g.

         O, in this dark era a g-man

         Is a terribly hard thing to be:

         But stand up and fight, for you know we are right!

         (Gabriel there.) Soon we shall be free….

         Sing the names of our martyrs: of Richwine or Scott,

         Or Rindermann, Charlton, Shockley, or Lahn,

         Or Harpending, Levin, Bouchard, or Cattell,

         Or Summers, Satoshi, or Sarich or Scarr,

         Or Nyborg or Watson or Herrnstein or Murray

         Or Rushton, or Jensen, or me.

Thursday, 10 March 2016

Threatened stereotype threat


the dark road

I’m in a dark place. I feel like the ground is moving from underneath me and I no longer know what is real and what is not.

Thus laments Professor Michael Inzlicht of Toronto, a man with honest doubts. Steve Hsu has described Inzlicht’s predicament here:

Inzlicht is very open about the personal advantages his work on ego depletion has brought him, and now his ego is depleted. He was also a strong believer in stereotype threat, and that is under threat. What has caused the ground to move beneath his feet?

Inzlicht has found that his frequently replicated effect has not survived the more intense scrutiny of a large sample, pre-registered study. You know very well that I keep repeating the complaint that so much of psychology uses unrepresentative, small samples that it produces weak results. Social psychology seems to be the most florid example of this lamentable practice. Now a strong method has superseded oft-repeated weak methods, as it should, and we are closer to the truth.

He adds, with commendable frankness: I edited an entire book on stereotype threat, I have signed my name to an amicus brief to the Supreme Court of the United States citing stereotype threat, yet now I am not as certain as I once was about the  robustness of the effect. I feel like a traitor for having just written that; like, I’ve disrespected my parents, a no no according to Commandment number 5. But, a meta-analysis published just last year suggests that stereotype threat, at least for some populations and under some conditions, might not be so robust after all. P-curving some of the original papers is also not comforting. Now, stereotype threat is a politically charged topic and I really really want it to be real. I think a lot more pain-staking work needs to be done before I stop believing (and rumour has it that another RRR of stereotype threat is in the works), but I would be lying if I said that doubts have not crept in. ...

I digress, but I had always thought that stereotype threat was monumentally silly, particularly when it was advanced to account for racial differences in ability. What this body of work asserted was that African Americans, far from being lower in ability than Europeans, were in fact so sensitive about being thought to be of lower ability that they performed as if they were lower in ability. Anything which triggered a sense of threat (black people not as bright as white people) immediately made them go into a tizzy of incompetence. Frankly, this seemed to be an excellent excuse not to employ them in any occupation which required a modicum of thought. It was a case of the remedy being worse than the illness, and of treating every African American as a delicate flower to be shielded from so much as a scintilla of intellectual challenge. This was nonsense, from which all should be spared, particularly African Americans, but also women, who apparently suffered from their own stereotype threat when asked to do sums.

By the way, I do not doubt that you can temporarily affect a person’s motivation by demoralising them in some way. Decades ago Seligman reported on the bad effects on problem-solving of inducing a sense of helplessness. It is a temporary manipulation, and part of normal fluctuation in the extent to which people will persist at tasks, though those most affected by such discouragement need to avoid challenging work, if they can.

But, my pleasure in seeing this research go down the plughole is tempered by two considerations:

1 How did psychology make such a damn fool mistake and repeat it so often?

2 What findings that I cherish are about to go down that same plughole?

Here, I go so far as to quote myself:

We must judge our cherished ideas by the same harsh standards we apply to the ideas we find most repellent.

Why are we so resistant to disconfirmation, so wedded to our opinions? Many fanciful ideas have been proposed: deep personal failings, deep political biases, a shallow love of peer approval and social status, and shallow, sanctimonious, posturing mendacity. I have a more modest proposal: effort justification. I think everyone is prone to being proud that they have actually read something, particularly a long book. At the end of that, having been in the company of one mind, they tend to see the author as their friend, a trusted companion. Having read the damn thing, the last thing they want to hear is that their efforts have been wasted. The situation is compounded if they have taken some notes and repeated the main arguments to friends. Any objections from critics can be countered by saying: Read the book yourself. The identification with the thesis is considerably strengthened if you give a talk about the work, showing slides covering the main points, and receive the approbation of an audience, particularly when audience members provide examples which substantiate the thesis. I call this the PowerPoint Effect: if it looks pretty and gets approbation, it is probably true. Lecturers are particularly prone to it. Once again, effort justification come into play.

Multiply all this tenfold if you had the original research idea, ran a pilot which gave positive results on a small sample, applied for a grant, appointed a research assistant, struggled to get experimental subjects, and finally got a result which seemed good, just like the pilot study did, and then wrote a book about the further studies which seemed to provide confirmations of your thesis.

The problem of trying to save a cherished idea from going down flames is not restricted to the unhappy Prof Inzlicht, nor to social psychologists with political views, nor to psychologists generally. It has general application, a potency of territorial proportions, and the seductive quality of a warm bed. Of course, intelligence research is different in that it occupies a lofty sanctuary of the mind, well protected by the teachings of the wise, but who would lose, full of pain, the Collects that give comfort and sustain faith, and suffer all of that travail just because of an awkward fact, a spoke in the spinning wheel of belief?

As the Alexandrian poet says in his magisterial The God Abandons Antony

Above all, don’t fool yourself, don’t say
it was a dream, your ears deceived you:
don’t degrade yourself with empty hopes like these.

As one long prepared, and graced with courage,
as is right for you who proved worthy of this kind of city,
go firmly to the window
and listen with deep emotion, but not
with the whining, the pleas of a coward;
listen—your final delectation—to the voices,
to the exquisite music of that strange procession,
and say goodbye to her, to the Alexandria you are losing.

There are times when we must kiss ideas goodbye.

Postscript: By the way, I don’t know if Cognitive Dissonance, of which Effort Justification is a variant, still holds up. Has it been subjected to a large sample, pre-registered replication? I wonder. I have lectured about cognitive dissonance often. Too often, perhaps. Another psychological fact going down in flames?  Say it isn’t true.