Another group used part of the ENIGMA data set and data from the European Union IMAGEN consortium to investigate gene associations with cortical thickness assessed with MRI and intelligence in 1,583 adolescents aged 14 years old (Desrivieres et al., 2015). Measures of cortical thickness (CT) have some technical advantages over VBM measures of gray matter volume and CT has been associated with intelligence in large representative samples of children and adolescents (Karama et al., 2009, 2011). This IMAGEN report started by examining the relationship between CT and nearly 55,000 SNPs. One variant (rs7171755) was associated with thinner cortex in the left frontal and temporal lobes, and in a small subsample, with WAIS scores. They further found that this variant affected expression of the NPTN gene, implicated in the production of a glycoprotein required for synaptic health. This finding illustrates how synaptic events may influence brain maturation of CT, which in turn may influence intelligence. The cascade of intervening steps surely is complex, but it also is a finite problem to solve. The methodology and complexity of analyses that led to these findings require an advanced technical background. But even this summary, along with the studies summarized in Section 2.6, illustrates how sophisticated the hunt for specific genes has become.
All these studies demonstrate that identifying the many genes with small effects thought to be involved in intelligence is not an insurmountable challenge. Once there is more progress, epigenetic effects can be investigated for individual genes, but as of yet, there are not enough data to test specific epigenetic hypotheses. At minimum, these very large samples showing DNA associations with measures of intelligence should put to rest any doubts about a genetic role for intelligence. Recall our three laws once again: no story about the brain is simple; no one study is definitive; and it takes many years to sort out conflicting and inconsistent findings and establish a weight of evidence. The studies summarized in this section illustrate the rapid pace of progress for understanding the complexities surrounding the genetic aspects of intelligence.
We opened Chapter 3 referring to our 1988 PET report with a question, “Where in the brain is intelligence?” Nearly 30 years later, neuroimaging is providing informative data on this question, and the question itself is becoming more refined. Genetic studies in combination with neuroimaging are beginning to suggest specific brain mechanisms involved in individual differences in intelligence. Neuroscience research aimed at understanding intelligence has a firm basis and is progressing at a fast pace based on evolving neuroimaging and genetic technologies and methods. There is now a context for thinking about how brain parameters might be used to predict or even define intelligence. There is also a developing empirical context for thinking about how brain mechanisms might be manipulated to enhance intelligence and these are the subjects of the next chapter.
Chapter 4 Summary
New neuroimaging methods, especially graph analysis, have revealed structural and functional brain networks related to intelligence test scores.
Overall, the brain networks identified in many intelligence studies are consistent with the PFIT framework with possible modifications for consideration.
Although many studies find correlations between brain measures and IQ scores, predicting intelligence from brain images is not yet practical for several reasons, but there is exciting progress.
In general, neuroimaging studies of reasoning report results that are consistent with studies of intelligence, although many studies of reasoning avoid discussing any overlap. More collaboration between reasoning researchers, with cognitive expertise, and intelligence researchers, with psychometric expertise, is the best way to integrate these two rich empirical traditions.
The combination of quantitative genetics and neuroimaging reveals that individual differences in brain measures and intelligence have genes in common.
The combination of molecular genetics and neuroimaging has identified specific genes and related brain mechanisms that may influence individual differences in intelligence.
Review Questions
1. What is the strongest evidence that intelligence depends on multiple areas distributed throughout the brain?
2. What are different ways that brain efficiency can be measured?
3. Which brain measure shows the strongest correlation to IQ scores and why is this not sufficient for predicting IQ from brain images?
4. Why are studies of analogical reasoning related to studies of intelligence?
5. Describe how the combination of quantitative genetics and neuroimaging has advanced intelligence research.
6. How has the combination of molecular genetics and neuroimaging advanced the search for intelligence genes?
Further Reading
“Where smart brains are different: A quantitative meta-analysis of functional and structural brain imaging studies on intelligence” (Basten et al., 2015). This is the most recent comprehensive, technical review of neuroimaging studies of intelligence.
“What Does a Smart Brain Look Like?” (Haier, 2009b). Written for a lay audience, this is an overview of the PFIT of intelligence and what imaging studies may mean for education.
“Genetics and intelligence differences: five special findings” (Plomin & Deary, 2015). Latest review of key genetic findings related to intelligence and what they mean.
“Rich-club organization of the human connectome” (van den Heuvel & Sporns, 2011). This is a technical article for readers wanting more detail on rich clubs and brain connections revealed by graph analysis.
Chapter Five
The Holy Grail: Can Neuroscience Boost Intelligence?
We who have worked on this project at Beekman University have the satisfaction of knowing we have taken one of nature’s mistakes and by our new techniques created a superior human being.
(Professor Nemur’s fictional character delivers this remark during his presentation at a psychology conference describing how he increased the IQ of a mentally retarded man to super genius level. Flowers for Algernon, Keyes, 1966)
I know Kung Fu.
(Keanu Reeves as Neo in The Matrix sci-fi movie several seconds after a learning program on fighting is uploaded directly into his brain, 1999)
A tablet a day and I was limitless … I wasn’t high. I wasn’t wired. Just clear. I knew what I needed to do and how to do it … I read the Elegant Universe by Brian Greene in 45 minutes, and I understood all of it!
(Two characters in the movie Limitless after taking an IQ pill, 2011)
Learning Objectives
Why is the “weight of evidence” concept especially important for claims about increasing intelligence?
What are three examples of research findings that claimed sizeable increases in IQ which proved incorrect?
Explain the concept of “transfer” in the context of studies purporting to show IQ increases.
What are five methods of brain stimulation that may influence cognition whether or not they increase IQ?
What are six ethical issues concerning the use of drugs for cognitive enhancement?
Introduction
Why would the Governor of Georgia ask the state legislature to buy a classical music CD for every newborn infant in the state? Why learn to memorize longer and longer strings of random numbers? Why do school systems with limited funding purchase expensive computer games for children to play during class? What exactly are the best 5 or 7 or 10 tips for increasing your IQ by 17–40 points?
This chapter is about sense and nonsense regarding the possibility of increasing intelligence. The good news is that neuroscience may someday offer the possibility of increasing intelligence based on an understanding of the brain mechanisms involved, including mechanisms that can be influenced by a variety of means. The bad news is that the claims that we already know how to do this are naïve, wrong, or misrepresentations. These claims are not just on the Internet or in books written by authors with no particular scientific expertise. Some are found in research reports published after peer review in highly respected scie
ntific journals. How could this happen?
Higher intelligence is better than lower intelligence; no one seriously disagrees. All intelligence research speaks to the goal of enhancement, either directly or indirectly. This is a worthy goal; just ask the parents of a child with low IQ or a cognitive disability. It is also a primary goal of all parents for their children whether articulated so bluntly or not. There may be some people who do not care to be smarter, but I do not know any of them. Achieving the goal of increasing intelligence will require an understanding of what intelligence factors are, how best to measure them, how they develop, how they relate to specific brain mechanisms, and how those mechanisms may be malleable. There is a long history of trying to increase intelligence. I cannot document it, but I suspect this was a subject of interest to the alchemists, ancient builders, and even earlier mystics. So far as modern scientific efforts, there is no appreciable success when success is defined by independent replication of empirical research results that last over time based on sophisticated assessments of intelligence in well-designed studies.
In Chapter 1, we noted the critical measurement problem that IQ scores are not a ratio scale, making change scores before and after an intervention nearly impossible to interpret. To repeat, IQ points are not measures like inches or pounds. This key problem, at the heart of claims about increasing intelligence, is all but ignored in the studies we will review in this chapter. In Chapter 2 we reviewed the failures of earnest compensatory and early childhood education programs to boost IQ. These programs have other positive results, but the weight of evidence does not support any claims concerning increased intelligence as assessed with IQ or other psychometric tests. One hypothesis is that this failure may be due in large part to the genetic influences on intelligence demonstrated in the many studies we have discussed in Chapters 2 and 4. Nonetheless, apparently undaunted by past failures and inherent measurement problems, or ignorant of them, there are newer reports in the scientific literature that claim to raise IQ scores dramatically in children and adults. We will examine three of these specific claims under the implied heading, “Don’t let this happen to you.” These claims are based on the use of classical music, memory training, and computer games to raise IQ. By showing how such claims should be evaluated skeptically I hope to inoculate you against future declarations of alleged breakthrough or landmark results. Following these cautionary case studies, we then examine equally dubious claims about drugs that increase IQ. After that, we move on to the exciting possibility of increasing intelligence by neuroscience means that test the boundary between science and science fiction.
5.1 Case 1: Mozart and the Brain
Mozart died in 1791, but 202 years later, Mozart became the focus of a craze. It started with a brief letter published in the respected scientific journal Nature. The letter claimed that listening to a particular Mozart sonata for 10 minutes temporarily increased IQ by 8 points (Rauscher et al., 1993). Eight points is about half a standard deviation, quite a large effect for a mere 10-minute intervention. Intelligence researchers recognized immediately that this sounded too good to be true. It was not true but, amazingly, it took six years to dampen popular enthusiasm after a critical review (Chabris, 1999). Another 11 years were needed to finally put this claim to rest with the publication of a comprehensive review article titled, “Mozart effect – Shmozart effect: A meta-analysis” (Pietschnig et al., 2010). The title speaks for itself. Over the 17 years this popular myth endured, an uncountable number of Mozart and other classical music CDs were purchased with the expectation of increasing IQ by just listening. School music programs gained new support and music lessons gained a new rationale. Enumerable high school science fair projects investigated various aspects of the “Mozart Effect,” tested mostly on friends and family. In fairness, these were hardly terrible consequences of a wrong idea. Possibly with the exception of a few accordion lessons, no one was harmed, but no one’s IQ increased either.
The original 1993 report was based on 36 college students who were tested on measures of abstract reasoning after a 10-minute exposure to three different conditions. The reasoning measures were three different spatial reasoning tests from the Stanford–Binet intelligence test battery. The three experimental conditions were listening to Mozart’s Sonata for Two Pianos in D major, listening to a relaxation tape, and listening to silence (I know you cannot listen to silence, but the sentence requires parallel construction). After each condition, one of the three tests was given. For the Mozart condition, the standard test score was 57.56. This was statistically higher than 54.61 for the relaxation condition and 54.00 for the silence condition. These standard test scores were “translated” to spatial IQ scores of 119, 111, and 110, respectively. These findings are shown in Figure 5.1. The authors stated that, “Thus, the IQs of subjects participating in the music condition were 8–9 points above their IQ scores in the other conditions.” They also noted the enhancing effect only lasted for the 10–15 minutes of testing. They encouraged additional research on the duration of the time between listening and testing, on variations in the time of listening, on effects on other measures of intelligence and memory, on other compositions and styles of music, and on possible differences between musicians and non-musicians. And so the Mozart Effect for increasing IQ was born, along with countless science fair projects.
Figure 5.1 The bar graphs that launched the Mozart Effect. Spatial intelligence test scores and IQ equivalents (y-axis) are shown after listening for 10 minutes to Mozart, a relaxation tape, or silence. A different test and different participants were used for each condition.
Reprinted with permission, Rauscher et al. (1993).
Whoever the peer reviewers were for Nature, they apparently were unaware that treating the three different reasoning tests as equal measures of abstract reasoning, based on the fact they were correlated to each other, was a terrible psychometric procedure. They also failed to require information about the participants with respect to IQ or musical experience and ability. And most distressing, the translation of individual test scores to spatial IQ scores and the claim of an eight-point increase was psychometrically naïve, as we have discussed in Chapter 1. A case could be made that this report of a single experiment in a small sample with an extraordinary finding was not based on extraordinary evidence and therefore was not ready for publication, especially in Nature.
Although the authors focused on spatial IQ, the resulting media coverage was not so specific and the Mozart Effect was widely understood to enhance IQ in general. In addition to the media coverage, some intelligence researchers at the time seized on any findings that suggested IQ was highly malleable as evidence against a strong role for genetic influences on intelligence. The Nature report also spoke to a desire for an easy way to acquire greater intelligence at a relatively small cost and no risk. According to an article in the New York Times (January 15, 1998), Governor Zell Miller of Georgia proposed spending $105,000 a year from the state budget to purchase classical music tapes or CDs for the approximately 100,000 children born in Georgia each year (iPods were not yet a reality). During his budget address to lawmakers, the Governor played part of Beethoven’s Ode to Joy and asked, “Now don’t you feel smarter already?” He also remarked that from his experience growing up, “Musicians were folks that not only could play a fiddle but they also were good mechanics.” The article also quoted a skeptic: ‘‘I’m familiar with those findings and, at the moment,’’ said Sandra Trehaub, a professor of psychology at the University of Toronto who studies infants’ perception of music, ‘‘I don’t think we have the evidence to make that statement unambiguously. If we really think you can swallow a pill, buy a record or a particular book or have any one experience and that that’s going to be the thing that gets you into Harvard or Princeton, then that’s an illusion.’’ Notably, she did not mention Yale.
Independent replication of findings is a cornerstone of the scientific method. Additional research was catching up on Mozart. Five years after the o
riginal publication, two critical letters and a response from the original paper’s primary author appeared in Nature (Chabris, 1999). The first critical letter from Dr. Christopher Chabris was a meta-analysis of 16 studies of the Mozart Effect that included 714 participants and several different reasoning tests. This kind of analysis combines all the results across studies. The overall Mozart Effect was tiny for general intelligence and a bit larger for spatial–temporal tests. He too converted individual test scores to IQ equivalents and found about 1.4 points for general intelligence and about 2.1 points for spatial–temporal reasoning. These conversions are always suspect, as we have discussed, but they serve a narrow purpose of showing how small the effect looks because such small fluctuations are likely due to standard measurement errors found in all mental tests. Chabris concluded that these small effects were likely due to the effect of positive mood invoked by an enjoyable experience like listening to Mozart. In his formulation, enjoyable experience increases arousal, especially in the right hemisphere, which processes spatial–temporal information. A second letter in this citation from Dr. Kenneth Steele and colleagues reported a complete failure to replicate the findings from the original 1993 experiment. Some of their results indicated poorer test performance after listening to Mozart. In response, Dr. Rauscher noted that the original report did not claim an increase in intelligence. She contended that the claim was limited to spatial–temporal tasks involving mental imagery and temporal ordering. She pointed out that the smaller number of studies in Chabris’ meta-analysis that used only tests of spatial–temporal reasoning did show an increase after Mozart listening, and she critiqued his enjoyment arousal hypothesis. She also critiqued the studies from the Steele group because they were not exact replications. She acknowledged that there were inconsistent results from independent studies and concluded, “Because some people cannot get bread to rise does not negate the existence of a ‘yeast effect’.”
The Neuroscience of Intelligence Page 19
