Archive | Cognitive ability RSS feed for this section

Big Think’s list of world’s smartest people

13 Sep

Ten bests lists, and the like, should not be taken too seriously. Generally, a question such as “who was the greatest musician,” is really just a chance to talk about talent and accomplishment. The rankings themselves should be taken with a grain of salt. That is the way I feel about this list of the world’s smartest people. The list is inconsistent applying different criteria for different people, sometimes it gives IQ scores and other times it ignores them. Some of the IQs reported are really just guesses, and some, such as the claim that Cleopatra had an IQ, are based on very scanty evidence.

There is some reason to believe that creativity and cognitive ability (IQ) are different abilities. Thus, the inclusion of Shakespeare, despite his great artistic accomplishments may not be warranted. Also, why would Shakespeare be rated more highly than some non-anglophone literary figure, such as Tagore or Tolstoy?

If I were to draw up such a list, I certainly would include Godel and Cantor.

The Guardian’s terrible article on giftedness

28 Jul

Let me stipulate, before I begin, that I believe that most students could substantially improve their academic performance. But I cannot accept the assertion implied in a recent Guardian headline: “Why there’s no such thing as a gifted child.”

What evidence does the Guardian supply for its assertion? It’s almost entirely anecdotal. For example, we are told that:

Most Nobel laureates were unexceptional in childhood. Einstein was slow to talk and was dubbed the dopey one by the family maid. He failed the general part of the entry test to Zurich Polytechnic – though they let him in because of high physics and maths scores. He struggled at work initially, failing to get academic post and being passed over for promotion at the Swiss Patent Office because he wasn’t good enough at machine technology. But he kept plugging away and eventually rewrote the laws of Newtonian mechanics with his theory of relativity.

There is a lot wrong with this account. The story of a family maid years after Einstein was a grown and famous man is hardly strong evidence. Just how slow was he to talk? Doesn’t “high physics and maths scores” count as evidence of giftedness? As a child Einstein taught himself calculus and his employment problems were largely due to Antisemitism.

Te article also says:

Lewis Terman, a pioneering American educational psychologist, set up a study in 1921 following 1,470 Californians, who excelled in the newly available IQ tests, throughout their lives. None ended up as the great thinkers of their age that Terman expected they would. But he did miss two future Nobel prize winners – Luis Alvarez and William Shockley, both physicists – whom he dismissed from the study as their test scores were not high enough.

I didn’t know about Luis Alvarez, but I did know the story about Shockley. Terman study was a longitudinal study of gifted children who had an IQ above a defined cut score. Shockley’s IQ was high but just not high enough to be in Terman group. The students identified by Terman did well in life, having above average academic and professional achievement. It might be worth pointing out that Nobel prizes are not that common and while IQ does a good job of capturing highly valued cognitive skills it may not be a good measure of creativity.

Later the article makes this astonishing admission:

While the jury is out on giftedness being innate and other factors potentially making the difference, what is certain is that the behaviours associated with high levels of performance are replicable and most can be taught – even traits such as curiosity.

Wait a second, the article is supposed to be arguing that giftedness doesn’t exist, but the author not only concedes that it exists, but also that some aspects could be innate. There certainly are “behaviours associated with high levels of performance” but we do not know to what extent these behaviors can be taught.



Growth mindset: A failure to replicate

10 Jul

Carole Dweck’s work has received a great deal of attention. Essentially she argues that an individual’s beliefs about intelligence is a powerful predictor of scholastic attainment. In a Scientific American article she wrote:

 Many people assume that superior intelligence or ability is a key to success. But more than three decades of research shows that an overemphasis on intellect or talent—and the implication that such traits are innate and fixed—leaves people vulnerable to failure, fearful of challenges and unmotivated to learn.
Teaching people to have a “growth mind-set,” which encourages a focus on “process” rather than on intelligence or talent, produces high achievers in school and in life.

Now a study, published in Personality and Individual Differences, fails to replicate these findings. Here is the abstract:

Implicit theories of intelligence have been proposed to predict a large number of different outcomes in education. The belief that intelligence is malleable (growth mindset) is supposed to lead to better academic achievement and students’ mindset is therefore a potential target for interventions. The present study used a large sample of university applicants (N = 5653) taking a scholastic aptitude test to further examine the relationship between mindset and achievement in the academic domain. We found that results in the test were slightly negatively associated with growth mindset (r = − 0.03). Mindset showed no relationship with the number of test administrations participants signed up for and it did not predict change in the test results. The results show that the strength of the association between academic achievement and mindset might be weaker than previously thought.

Note, however, the limitations reported by the authors:

In sum, we found that mindset had virtually no association with results in a scholastic aptitude test used for university admissions. While the association between mindset and goal achievement was previously shown to be weak (Burnette et al., 2013), our study presents a large amount of new data suggesting that the association may be even weaker than previously thought. Given that recent large scale experiments suggest that learning growth mindset improves academic achievement (Paunesku et al., 2015; Yeager et al., 2016a, b), our study does not invalidate the notion that implicit theories of intelligence might be a promising target for educational interventions. However, it suggests that mindset might not be as useful for predicting future success or that its predictive abilities are at least limited to specific circumstances. Yet, we note that our study has several limitations including possible self selection and range-restriction effects, a short measure of mindset, and a short duration between subsequent administrations of the test. We also did not include measures of hypothesized mediating variables, such as the amount of practice, and the mindset measure was not directly tailored to assess beliefs about the possibility of improvement in the GAP test. Future studies may overcome these limitations and thus better explain differences between results of the present and past studies.


Higher childhood IQ related to a lower risk mortality

3 Jul

The Lothian birth-cohort studies continue to contribute to our understanding of cognitive aging. Here is the latest paper, published in the journal Intelligence. And here are the paper’s highlights;

• 94% of the participants of the Scottish Mental Survey 1947 were traced.
• Higher childhood IQ was related to a lower risk of all-cause mortality by age 79.
• The effect was slightly stronger in women than in men.
• The decline in risk across categories of IQ scores was graded across the full range.
•This is the only study of IQ and mortality in an entire year-of-birth cohort.

Here is Ian Deary’s brief description of the Lothian birth-cohort studies

Do your siblings lower your IQ?

7 Jun

As the number of siblings increases, the fraction of parental resources received by each one declines. Thus, it has been suggested that larger sib size should be associated with lower IQ and lower school performance. A recent paper in the journal Intelligence looked at the evidence. Here is the abstract:

We examine the effects of child sibship size on intelligence, school performance and adult income for a sample of Swedish school children (n = 1326). These children were measured in grade three in 1965 (age 10) and in grades six (age 13) and nine (age 16), and the women and men were later followed up in adulthood at ages 43 and 47, respectively. Using Bayesian varying-intercept modeling we account for differences between school classes in each of our three response variables: IQ-scores, school grades and adult income, and control for background variables such as gender, socioeconomic status, and maternal- and paternal age. Consistent with previous research, we find patterns of decreasing IQ scores for increasing sibship sizes, specifically for an increasing number of older siblings. No relationships between sibship size and children’s school grades are found. We find, however, patterns of decreasing adult income for an increasing number of younger siblings. In addition, considerable amounts of variations in intelligence scores as well as school grades are found between school classes. Some implications of the findings and suggestions for future research are provided.

James Flynn on the Flynn Effect

24 Oct

I showed this video to my doctoral students last week. It’s a nice summary of Flynn’s work. I regard the Flynn Effect as one of the most important discoveries, a greatest mysteries, of scientific psychology.

Special brag point: I am the author of the study he mentions on Ohio school examinations.

Brain training increases grey matter volume

15 Aug

On Friday, I reported on a meta-analysis that presented evidence that working memory brain training does not transfer to other cognitive skills. The most recent issue of Personality and Individual Differences carries a paper titled:  “Gray matter volumetric changes with a challenging adaptive cognitive training program based on the dual n-back task.” The n-back task is the most widely used procedure for working memory training in academic research.

Surprisingly, these results do not, necessarily, contradict each other. As noted in the abstract:

“Changes in the gray matter volume of these clusters were correlated with a) behavioral changes across the training program and b) changes in four psychological factors assessed before and after training (fluid and crystallized intelligence, working memory capacity, and attention control). None of these correlations were statistically significant, and therefore, psychological and biological changes were seen as independent.”

Since there working memory training does improve performance on the trained task, we would expect there to be some kind of measurable physical change in the brain. But this does not mean that the training effects are transferable to other cognitive domains.

%d bloggers like this: