Thursday, March 26, 2015

The Flynn Effect's Unseen Hand, Revised


[Three years ago I wrote an essay summarizing my ideas regarding human intelligence and the Flynn effect. I would describe that effort as less than successful. There are several reasons for this—laziness, too much excitement, too little time, trying to fit to an academic form. And unfortunately I have had little opportunity in recent years to make the necessary revisions. Recently however I have managed to revisit the work and insert some improvements—mostly changes in terminology, expansions of explanations, and a simpler tone. I won't suggest that the new version will prove any more influential than its predecessor, but it does set my mind at ease to know that when I had something I thought worthy to say, I at least made the attempt to say it well.]



The Flynn Effect's Unseen Hand


Introduction. The Flynn effect is a well known but insufficiently explained phenomenon. Many different causes have been suggested for the population-wide generational increases in raw intelligence scores—including heterosis, better nutrition, more abundant education, environmental complexity and various combinations of the above—but no explanation offered so far has proven to be scientifically or logically compelling. This lack of progress might be indicative of a misunderstanding of human intelligence itself, which is depicted these days almost entirely in terms of brain-based functioning alone. That brain-based focus however has often been the bedevilment in the many offered explanations of the Flynn effect, for it has been difficult to reconcile purported neural mechanisms producing individual intelligence differences with purported neural mechanisms producing widespread intelligence gains.

Accordingly, this essay will propose an alternative model of human intelligence, one decidedly not centered on the human brain. This model will highlight two complementary aspects of human intelligence: 1. environmental intelligence, defined as the total amount of non-biological pattern, structure and form tangibly contained within the human environment, and 2. neuronal intelligence, defined as an individual's capacity to absorb and respond to environmental intelligence. It can be shown that it is environmental intelligence that serves as the sole driver of the Flynn effect, independently of neuronal intelligence. It will also be demonstrated how environmental intelligence is similar to but far more comprehensive than the concept known as environmental complexity. Finally, it will be shown that this dual-aspect model of human intelligence can effectively answer many of the Flynn effect paradoxes enumerated by James Flynn himself.


Background. Generational gains in raw intelligence scores were first noticed by several individuals—including Reed Tuddenham and Richard Lynn—but it was James Flynn in the 1980s who most clearly demonstrated the ubiquitous nature of what has come to be known as the Flynn effect. In the thirty years since, the Flynn effect has attracted a good deal of study and ink—in part because the phenomenon has been regarded as surprising, and in part because the phenomenon has continued to defy adequate explanation.

This situation stands in contrast to many other areas of intelligence research, including investigations into the source and impact of individual and group intelligence differences. Using factor analysis, identical twin studies and many other tools of modern cognitive science, researchers have been able to demonstrate consistently that individual intelligence differences produce significant impact in such areas as academics and career, and that these individual differences are driven mostly by genetics and are almost certainly neuronally based. These discoveries and achievements have led to a nearly unanimous consensus that intelligence is to be regarded exclusively as a brain-produced activity—in short, greater intelligence is spawned by a more effective brain.

The Flynn effect, however, throws something of a monkey wrench into this widely held view. To accept the conclusion that intelligence is exclusively a brain-produced activity—an activity determined primarily by genetics—one must anticipate that overall human intelligence will remain relatively stable across time, in accordance with all standard biological and evolutionary principles. That is why the Flynn effect has been regarded as so surprising: the sizable and widespread raw intelligence gains recorded across the entire twentieth century far outstrip any brain-based improvements that might be anticipated under a biological/neuronal/evolutionary framework.

One response to this dilemma has been to search for an orthogonal influence underlying the Flynn effect, and James Flynn himself (1999) has uttered that very reaction in almost those exact same terms ("it is as if some unseen hand is propelling scores upward"). Richard Lewontin (1976) has already provided a convincing description for how such an orthogonal influence would work [Flynn's description of Lewontin's idea: "(Lewontin) distinguished the role of genes within groups from the role of genes between groups. He imagined a sack of seedcorn with plenty of genetic variation randomly divided into two batches, each of which would therefore be equal for overall genetic quality. Batch A is grown in a uniform and optimal environment, so within that group all height differences at maturity are due to genetic variation; batch B is grown in a uniform environment which lacks enough nitrates, so within that group all height differences are also genetic. However, the difference in average height between the two groups will, of course, be due entirely to the unequal quality of their two environments....genes (could) explain 100 percent of IQ differences within generations, and yet, environment might explain 100 percent of the average IQ difference between generations."]. But no one has ever pursued this line of reasoning to its ultimate conclusion, in part because what paralyzes the pursuit is the widespread certainty—a dogma really—that intelligence is strictly a brain-produced phenomenon. Any offered explanation for the Flynn effect—be it heterosis, better nutrition, improved education, environmental complexity, or any combination or alternative to the above—any explanation it seems has to be brought back ultimately to human neurology, has to induce a material impact upon the human brain. Vigor, nutrients, schooling, video games—whatever is driving intelligence gains, it must somehow change the human brain, must make it more effective, make it more intelligent. Unfortunately, this circling back to neurology serves only to heighten the original tension of the problem: if there are neural mechanisms explaining individual intelligence differences, and there are different neural mechanisms driving population-wide intelligence gains, how are these mechanisms supposed to co-exist within the same human brain and not interfere with the intelligence-producing impact of the other. If Lewontin's suggestion has been offered as the pathway to a more straightforward explanation of human intelligence, its application to human neurology has proven to be anything but.

A decisive alternative would be to drop the dogma altogether. Nothing actually compels acceptance of the idea that intelligence is strictly a brain-produced phenomenon. Despite the widespread consensus, no one has yet to demonstrate an actual neural mechanism producing an actual intelligence effect. Neural activity certainly accompanies intelligence behavior—there is plenty of evidence for that—but the correlation does not go so far as to prove causation. Furthermore, with the Flynn effect still a puzzle and a mystery, bumping against many fundamental assumptions regarding biology, evolution and intelligence, it would seem there is reasonable motivation for casting the cognitive net a little wider.

This essay describes a model of human intelligence that removes the location of intelligence away from the human brain and places it more squarely within the human environment, a concept that will be dubbed environmental intelligence. Thus freed from the constraints of biology, neurology and evolution (that is, freed from the constraints of the human brain), human intelligence can be seen as able to change and accumulate at a significant pace, which indeed it must if it is going to produce the phenomenon known as the Flynn effect. The human brain still gets to play an important role within this new model—under a concept defined as neuronal intelligence—but this role will be seen as necessarily secondary. Instead of producing human intelligence, the human brain will be depicted as responding to the intelligence contained within the surrounding environment, an idea not as radical as it might at first appear, since responsiveness of course has always been the activity traditionally reserved for neural systems.


Environmental intelligence and neuronal intelligence. A fresh perspective can be gained on human intelligence by considering it as the product of two orthogonal components—environmental intelligence and neuronal intelligence.

Environmental intelligence is defined as the total amount of non-biological pattern, structure and form tangibly contained within the human environment. Every artifact a human encounters, every synthesized product that crosses his path, every constructed invention helping to mark his way—all the way from the simplest spoken hello to the intricacies of the latest and greatest microchip—each formulated element enveloping a modern man's existence, an envelopment now so thorough it practically eclipses the natural world from view, all of this, every last patterned piece of it, forms the sum total of environmental intelligence. Even the human body, still the most biological, non-artificial entity to be found within a modern human's sensory world, even the human body comes these days invariably clothed, manicured, bespectacled, bejeweled and perfumed, which is to say the human body comes these days abundantly encased in many of the diverse varieties of environmental intelligence. Everywhere a man looks, every moment he listens, every texture he brushes against, he finds himself inundated with a constructed cornucopia built up out of order, shape, number, rule; and this cornucopia in turn incessantly broadcasts back into his neural system the elements of its underlying characteristics—symmetry, repetition, pattern, structure, form. Environmental intelligence is the influence so easily overlooked because it is the influence so invariably right there, right before one's very eyes. These days environmental intelligence is utterly ubiquitous, composing the very fabric of modern human existence, thoroughly embodied in the furniture, the transportation, the words, the games, the weapons, the gifts, the gardens, the laboratories, the music—thoroughly embodied in quite literally, or at least quite literally once the few remaining biological elements have been removed, quite literally the everything.

One advantage of this definition of environmental intelligence is that it directly and observably connects human intelligence to the sudden advancement of the human species. Prior to the human great leap forward, there would have been essentially no environmental intelligence to be found anywhere within the human surroundings. No written words. No constructed buildings. No artifacts of even the simplest kind. Prior to the human great leap forward, man would have been surrounded by only the most natural of settings, all the way from his skin to the farthest horizon, exactly as would have been the case for all the other animals; and not coincidentally it would have been perfectly correct to assess man's overall, absolute level of intelligence at that time as essentially zero. (Ancient homo sapiens certainly were not capable of taking an IQ exam, let alone answering its questions correctly, let alone constructing such an exam in the first place.) But beginning with ostrich shell beads, bone awls, rudimentary clothing and cave paintings, and proceeding straight through to domesticated crops, mud-plastered abodes and towering pyramids, and crescendoing today in highways, skyscrapers, televisions and rockets to the moon, the one indisputable observation that can be made throughout that entire course of recent human progress is that no matter in what environs man suddenly found himself, he found himself always surrounded by an ever growing totality of non-biological pattern, structure, symmetry, repetition and form. For the last fifty thousand years, man has been increasingly enveloping himself in the many and diverse material artifacts that ultimately compose the sum total of environmental intelligence, and not coincidentally, man has been displaying throughout that entire fifty thousand year interval the ever more abundant signs of intelligence.

To actually measure environmental intelligence would be admittedly a pragmatic nightmare. The sheer enormity of pattern, structure and form contained within the modern world would alone overwhelm any genuine effort to size it, and furthermore, there could be no easy agreement on how best to quantify the structure contained within for instance an automobile or a library book. But these practical difficulties do not nullify the material certainty of measurement—environmental intelligence tangibly exists, one can touch it, hear it, talk about it, it is there right before one's very eyes. Plus there is no need to actually measure environmental intelligence in order to attest to its ever increasing presence and influence. Think of the North American continent alone. Only a few hundred years ago, man dwelled there in but a handful of places, and outside of a few isolated civilizations the amount of non-biological pattern, structure and form to be found within the Western Hemisphere would have been extremely modest. But by one hundred years ago, man had taken up residence from nearly coast to coast and had augmented his New World surroundings with an entire patchwork of fields, houses, roads, signs and machines. Today just one glance at the skylines of such cities as Chicago and Toronto would be more than sufficient to convince even the most dire skeptic that by almost any reasonable means of measurement, it would have to be calculated that the total amount of human environmental intelligence has been persistently, indeed rapidly, on the rise.


The second component of human intelligence, neuronal intelligence, is in nearly every respect nothing at all like the first. Neuronal intelligence is defined as an individual's capacity to absorb and respond to environmental intelligence, making it clear that neuronal intelligence is considered here to be a secondary (a responding) construct. This of course runs counter to the prevailing wisdom. The prevailing wisdom would claim that all the many material artifacts forming the sum total of environmental intelligence are not so much the embodiment of intelligence as they are the results of intelligence, the results of the wondrous if still somewhat mysterious mechanisms of the human brain (where indeed all the intelligence must reside). This presumed equivalence between intelligence and human neurology has arisen in large measure—and quite understandably enough—from the many successful results and findings of modern intelligence research. With the employment of IQ exams now widespread, and with the correlation of their results against twin and other family studies, against career and academic outcomes, against neuroimaging and other laboratory techniques, intelligence researchers have been able to formulate a great deal of predictive insight into what drives individual and group intelligence differences and have been able to demonstrate with a high degree of confidence that such differences are for the most part genetically derived and are almost entirely neuronally based. Neuronal intelligence has become the component of intelligence with which everyone is most familiar, because it is the component of intelligence that has been the most accurately and thoroughly measured, and the most successfully understood.

The one pitfall in these many informative findings of modern intelligence research is that they have been so successful in tying individual and group intelligence differences to genetics and neurology that they have managed to convince researchers—to the point of near unanimity and to the point of dogma—that all intelligence differences must be tied to genetics and/or neurology, including intelligence differences that manifest across time (the Flynn effect). Invariably these days, when an explanation for the Flynn effect is offered—whatever that explanation may be—it is offered first and foremost as a temporal and population-wide influence on the human brain. But in point of fact, all the evidence backing the neuronal, genetic basis for individual and group intelligence differences is evidence both gathered at and applicable for only a particular moment in time; the evidence remains utterly silent when applied across time. All the illuminating findings of statistical analysis, including the resultant concept of a general intelligence (Spearman's g), arise strictly from comparisons made against one's contemporaries, and not against one's ancestors or descendants. Indeed most intelligence researchers recognize this distinction well enough to realize that any neural mechanisms that might explain individual intelligence differences would likely be very poor candidates as neural mechanisms underlying the Flynn effect; and yet no researcher is able to carry this distinction to its most logical conclusion, namely that there might not be any neural mechanisms to be associated with the Flynn effect, not in any way whatsoever. Having been witness to so much present-moment evidence for the neural/genetic causation of individual intelligence differences—causation that is perfectly plausible applied across the range of biological diversity within the human population—researchers then cannot let the idea go, even when considering intelligence differences that span an entirely separate domain. And thus nearly every explanation for the Flynn effect continues to be offered with its seemingly mandatory tie back to human neurology, and thus nearly every explanation for the Flynn effect continues to fail, and fail for nearly the same reason—the seemingly mandatory tie back to human neurology becomes downright implausible applied across just a handful of generations.

The way past this predicament begins first with a more thorough examination of that preeminent tool for measuring neuronal intelligence—the IQ test. It is the comparative, normed results of IQ tests that provide nearly all the basis for the present understanding regarding individual and group intelligence differences, and so naturally it is the results that get most of the attention. But an IQ test is more than just its normed results; an IQ test has content—and not just any content. An IQ test does not assess for instance an individual's capacity to scavenge food, ward off predators or procreate, and an IQ test does not measure one's ability to run, leap or throw. The challenges that one faces on an IQ test are challenges composed almost entirely out of a particular set of material artifacts—language, arithmetic, geometrical puzzles, and so on—artifacts which are in turn built up out of a basic set of underlying characteristics—symmetry, pattern, structure, repetition, form. These underlying characteristics are of course the very same characteristics already encountered under the description of environmental intelligence. When examined carefully, an IQ test reveals its content as made up out of miniaturized, formalized versions of the types of structural material artifacts one encounters nearly everywhere in the everyday world; which is to say, the content of an IQ test stands as a proxy for environmental intelligence. When an individual takes an IQ test, what he demonstrates is his relative capacity for absorbing and responding to these proxies for environmental intelligence, which in turn points to his relative capacity for absorbing and responding to the environmental intelligence he will encounter in his everyday world. Therefore it is not in the least bit surprising that those individuals who demonstrate greater ability in mastering the complexities of an IQ test are also the individuals who tend to demonstrate greater ability when navigating the complexities of the real world. This analysis of the content and challenge of an IQ test leads directly back to the stated definition of neuronal intelligence: neuronal intelligence is an individual's capacity to absorb and respond to environmental intelligence, with a strong emphasis to be placed on both a. capacity, and b. response to environmental intelligence. By itself, neuronal intelligence cannot explain human intelligence, because by itself, neuronal intelligence is merely a capacity in need of a target. That target is environmental intelligence—the total amount of non-biological pattern, structure and form tangibly contained within the human environment, the other essential component in any comprehensive description of human intelligence.


It is important to emphasize one more time the orthogonal relationship of environmental intelligence and neuronal intelligence. Neuronal intelligence is a biological capacity, a human behavioral ability, and thus there is no objection to associating neuronal intelligence with neural and genetic causes. But environmental intelligence is not biological at all; it is instead a collection of characteristics from physical, mostly man-made artifacts, quantifiable, changeable and accumulative within the material world, and thus environmental intelligence stands completely independent of any neurological or evolutionary constraint.

Environmental intelligence and neuronal intelligence are each an essential component of human intelligence, but each delivers its influence in an entirely separate domain.


The Model. With the path now prepared by these definitions and descriptions of environmental intelligence and neuronal intelligence, an example can be developed illustrating how these two components, working simultaneously and yet independently, combine to explain the known and observable characteristics of intelligence as a whole, including the characteristic known as the Flynn effect. All that is required further are two straightforward assumptions: 1. the practical difficulties in measuring environmental intelligence can be theoretically overcome; and 2. consistent with observations from human history, environmental intelligence can be assumed to increase over any significant interval of time.

In the example to be developed, intelligence characteristics will be assessed at two different moments in time, call them Time 1 and Time 2, with an interval of several generations passing between these moments. The intelligence characteristics of the individuals living at Time 1 and Time 2 will be described in the usual way, via results on intelligence exams, and at these two moments the intelligence characteristics of the environment will also need to be detailed. Drawing upon the assumption that the practical difficulties in measuring environmental intelligence can be theoretically overcome, a system of measurement will be assumed that is able to accurately assess the total amount of non-biological pattern, structure and form tangibly contained within the human environment, quantifying this amount in something called environmental intelligence units (EIU). At Time 1, the total amount of pattern, structure and form within the human environment will be assumed to be measured at 200 EIU. Then several generations later, at Time 2, the total amount of pattern, structure and form within the human environment will be measured at double the previous amount, at 400 EIU. Such a sizable increase across several generations might seem too large at first but is actually quite reasonable by recent human standards (consider for instance the enormous amount of environmental change from the late 1800s to the late 1900s). And at any rate, the hypothesized numbers are not critical in and of themselves: any significant increase in environmental intelligence across the interval of time being considered will be sufficient to demonstrate the principles pertinent to the example.

At Time 1, with the amount of environmental intelligence having been measured at 200 EIU, a standard battery of intelligence tests is administered to a broad sampling from the population, and as is done with real world intelligence exams, the raw scores are then normed and delineated into ranked categories. The essential outcome of this process can be summarized through the exam results of just three individuals—call them A1, B1 and C1—individuals who represent respectively results consistent with high intelligence, medium intelligence, and low intelligence. Their raw scores might be stated in a variety of ways: a) as the actual number of questions answered correctly, or b) as the percentage of questions answered correctly, or c) as the percentage of environmental intelligence successfully absorbed and mastered. This last approach is derived from the discussion above, where the content of an IQ exam has been described as a proxy for environmental intelligence. If the battery of tests administered to A1, B1 and C1 is in fact a perfect proxy for environmental intelligence, then the percentage of questions answered correctly can stand as a percentage measure of the amount of environmental intelligence successfully mastered. For instance, when it is discovered that A1 can correctly answer 80% of the test questions, the result could be stated as follows: A1 has demonstrated the capacity to master roughly 80% of the environmental intelligence contained in the IQ exam, which indicates a capacity to master roughly 80% of the environmental intelligence to be found in his everyday world. In a similar vein, when B1 and C1 respectively answer 70% and 60% of the test questions correctly, it can be said they are demonstrating the capacity to master corresponding percentages of environmental intelligence.

The results of both the environmental and individual intelligence measures at Time 1 are summarized in the following chart:

Time 1 (Environmental Intelligence: 200 EIU)


Test Scores
Population Rank

A1
80%
High Intelligence

B1
70%
Medium Intelligence

C1
60%
Low Intelligence


At this point, all the standard types of analysis regarding individual intelligence differences can be performed quite adequately, leading to the type of informative findings that fall under the heading of neuronal intelligence. Using relative intelligence rankings, and employing factor analysis and incorporating an assortment of statistical and biological information gathered from the population at large, scientists will be able to show with considerable confidence that, all other things being equal, A1 can expect greater success than his B1 and C1 peers in such areas as academics and career, and that the individual intelligence differences between A1, B1 and C1 can be attributed in large degree to biological and genetic causes. The comparative, normed intelligence scores at Time 1 (or at any given time) are sufficient to provide a wealth of information regarding the characteristics of neuronal intelligence.

An absolute measure of intelligence for A1, B1 and C1 has not yet been determined, but it would be a simple matter to do so. With a measurement of 200 EIU having been assigned to Time 1's environmental intelligence, and with the raw test results able to be stated as a percentage of environmental intelligence effectively mastered, a quick calculation reveals that A1's absolute level of intelligence is 160 EIU (200 EIU x 80%), B1's is 140 EIU, and C1's is 120 EIU. The chart can be updated to reflect these figures:

Time 1 (Environmental Intelligence: 200 EIU)


Test Scores
Population Rank
Absolute Intelligence
A1
80%
High Intelligence
160 EIU
B1
70%
Medium Intelligence
140 EIU
C1
60%
Low Intelligence
120 EIU

It should be noted that this additional calculation of an absolute intelligence score does not aid at all in the understanding of neuronal intelligence. As far as individual and group intelligence differences are concerned, the inclusion of an absolute intelligence score is nothing but a superfluous addendum—the normed, relative intelligence rankings are more than sufficient by themselves to make present-moment findings regarding neuronal intelligence. However, the inclusion of an absolute intelligence score will nonetheless prove to be invaluable, for it will turn out to be an essential feature in the comparison of intelligence characteristics between Time 1 and Time 2.


As a reminder, environmental intelligence is assumed to increase over time, and at Time 2 the total amount of pattern, structure and form contained within the human environment is assessed to have increased to 400 EIU. Since Time 2 occurs several generations after Time 1, A1, B1 and C1 are no longer alive. But since A1, B1 and C1 were only representative individuals culled from the overall Time 1 test results, it is perfectly reasonable at Time 2 to call upon their equivalent descendants—call them A2, B2 and C2—all of whom can be taken as biologically and genetically similar to their Time 1 ancestors. Indeed, when the standard battery of intelligence tests is administered to the Time 2 population, A2, B2 and C2 score in a familiar pattern: A2 answers 80% of the test questions correctly, which is interpreted as reflecting an 80% mastery of Time 2's environmental intelligence, and B2 and C2, to no surprise, score 70% and 60% respectively. Once again the population results are normed and delineated into ranked categories, and just as was the case with their ancestors, A2 falls within the range of high intelligence, B2 falls within the range of medium intelligence, and C2 falls within the range of low intelligence. These Time 2 results can be summarized as follows:

Time 2 (Environmental Intelligence: 400 EIU)


Test Scores
Population Rank

A2
80%
High Intelligence

B2
70%
Medium Intelligence

C2
60%
Low Intelligence


Once again, all the standard types of analysis regarding individual and group intelligence differences can now be performed quite adequately, and at Time 2 the findings regarding neuronal intelligence will look almost identical to the findings from Time 1. Using factor analysis and population statistics, scientists will once again be able to state that A2 can anticipate greater success than his B2 and C2 peers, and that the individual intelligence differences between A2, B2 and C2 are to be attributed in large degree to biological and genetic influences. If the scientists were to look at just the pattern of individual and group intelligence differences from Time 1 to Time 2, they would be led to believe that the overall intelligence characteristics are quite stable within this population, just as might be anticipated for a capacity strongly under the influence of biological/evolutionary forces.

And yet at Time 2, the scientists will decidedly not be talking about the stability of intelligence. Instead they will be talking about a significant anomaly that has taken place.

There are several ways to characterize this anomaly. The first is to begin by examining what has taken place as the IQ tests have been administered to the general population. The first intelligence tests offered to the Time 2 population were the exact same tests given to the Time 1 population, but as it turns out, those tests are now laughably easy, to the point that nearly everyone scores in the uppermost ranges. This prevents a meaningful comparison of results, since no one is being challenged anymore and nearly everyone is scoring the same. In order to restore the tests to their former condition of being able to provide meaningful comparisons, the test producers find they must beef up the exams, make the questions more difficult, after which the relative rankings reemerge. It is only after such modifications have been made that the tests can be effectively administered to the population, with the resulting scores as shown.

In one sense, the reason that the IQ tests have to be modified at Time 2 is clear from the parameters of the example itself. Since the content of an IQ test stands as a proxy for environmental intelligence, and since environmental intelligence has significantly increased from Time 1 to Time 2, the tests must be reconstituted in order to reflect this fact; that is, the additional amount of pattern, structure and form to be found within the Time 2 environment must be incorporated into the Time 2 exams in order to assess the population's relative dexterity with this new structural material. But in an entirely different sense, another reason emerges for explaining why the Time 2 exams have to be modified—namely, that this is precisely what has been taking place in the real world throughout the entire last century. Ever since IQ exams were first administered, each successive generation has been scoring progressively better on the existing exams, to the point that test makers find they must modify the exams in order to keep them challenging, in order to maintain their usefulness for comparative purposes. These modifications generally take the shape of more difficult questions, questions that incorporate a greater amount of pattern, structure and form. Thus by virtue of the parameters of the example itself and by virtue of the evidence from the real world, it can be seen that intelligence tests must be strengthened in order to counteract the persistent influence of the increasing amount of complexity within the human environment.

And it is not just the tests that need to be reconsidered. The intelligence characteristics of A2, B2 and C2 must also be reexamined, because they are now evincing two seemingly contradictory facts:
  1. The neuronal intelligence characteristics of A2, B2 and C2 are essentially identical to the neuronal intelligence characteristics of their A1, B1 and C1 ancestors.
  2. The overall amount of intelligence being displayed by A2, B2 and C2 is essentially double the amount of intelligence that was displayed by their A1, B1 and C1 ancestors.
The second fact arises from recognizing that A2, B2 and C2 are correctly answering the same percentage of questions as did their Time 1 ancestors but are doing so while taking a far more difficult test. This comes out also through the calculation of absolute intelligence scores for A2, B2 and C2. With Time 2 environmental intelligence assessed at 400 EIU, A2's test results reflect an absolute intelligence score of 320 EIU (400 EIU x 80%). B2 scores at 280 EIU, and C2 scores at 240 EIU—in each case a doubling over A1, B1 and C1:

Time 2 (Environmental Intelligence: 400 EIU)


Test Scores
Population Rank
Absolute Intelligence
A2
80%
High Intelligence
320 EIU
B2
70%
Medium Intelligence
280 EIU
C2
60%
Low Intelligence
240 EIU

As contradictory as these results might at first appear, this example reflects exactly what has been happening in the real world. The only difference is that the example also includes an assessment of environmental intelligence, as well as an analysis of the impact of environmental intelligence on individual intelligence differences, differences that in this instance manifest over an interval of time. And what arises from this example is a clear indication of exactly what produces the population-wide generational increases in raw intelligence scores. The sole driver of raw intelligence gains is the increasing amount of environmental intelligence, the increasing amount of non-biological pattern, structure and form tangibly contained within the human environment.

And by corollary, neuronal intelligence—including any mention at all of neurology or genetics—has absolutely nothing to do with intelligence gains over time. Neuronal intelligence, the biological capacity to absorb and respond to environmental intelligence, that capacity will remain nearly constant over time, but that capacity will encounter an ever expanding target.


Flynn's Paradoxes. In his book What is Intelligence?, Flynn (2007) describes four paradoxes he associates with the Flynn effect. To someone not obsessed with the brain's monopoly on human intelligence, however, these paradoxes are not paradoxical at all—each can be answered simply and directly using this essay's dual-component model of human intelligence.

Two of the paradoxes, labeled as the intelligence paradox and the mental retardation paradox, state the apparent incongruity that if the Flynn effect were literally true, then humans from one generation would be too implausibly dumb or too implausibly smart compared to humans from a different generation. In Flynn's words:
"If huge IQ gains are intelligence gains, why are we not struck by the extraordinary subtlety of our children's conversation? Why do we not have to make allowances for the limitations of our parents? A difference of some 18 points in Full Scale IQ over two generations ought to be highly visible.

"If we project IQ gains back to 1900, the average IQ scored against current norms was somewhere between 50 and 70. If IQ gains are in any sense real, we are driven to the absurd conclusion that a majority of our ancestors were mentally retarded."
The resolution to these two paradoxes is to recognize that Flynn is confusing the two different aspects of intelligence; he is confusing environmental intelligence with neuronal intelligence. In particular, he is using the changed levels in one aspect (environmental intelligence) to infer a corresponding change in the other aspect (neuronal intelligence). That inference is entirely unwarranted.

Consider the individual named A1 in the model. At Time 1, A1 is assessed to be highly intelligent. He demonstrates an above-average ability to absorb and respond to environmental intelligence by correctly answering 80% of the test questions presented to him, and as A1 navigates through his Time 1 world, it can be anticipated he will experience relatively greater achievement in such areas as academics and career compared for instance to his B1 and C1 peers. But when A1's absolute (raw) intelligence score of 160 EIU is compared to the population of Time 2, A1 suddenly appears much less smart. 160 EIU scores far below the 240 EIU of C2, a person assessed to be of low intelligence at Time 2. If 240 EIU is considered to be of low intelligence at Time 2, then A1's score of 160 EIU seems to mark him as a borderline imbecile.

So which is it? Is A1 highly intelligent or is he an imbecile? This paradox is resolved by recognizing that A1's neuronal intelligence is not subject to change. A1's absolute intelligence score of 160 EIU has as much to do with the time period during which it was registered as it has to do with A1's biological capacity. If A1 could be magically transported forward in time and raised in the Time 2 world, he would absorb and respond to about 80% of the Time 2 environmental intelligence and would score correspondingly on a Time 2 intelligence exam, making it clear once again that he is a highly intelligent individual. A1's apparently low score of 160 EIU has nothing to do with A1's intelligence abilities; it has everything to do with the change in environmental intelligence from Time 1 to Time 2.

This works exactly the same way going backwards in time. Consider C2, who is assessed at Time 2 to be of low intelligence. But when C2's absolute (raw) intelligence score of 240 EIU is compared to the Time 1 population, where a score of 160 EIU is considered to be highly intelligent, C2 suddenly comes across as a Mensa candidate, and one wonders if C2 simply had the misfortune of being born too late.

So which is it? Is C2 of low intelligence or a Mensa candidate? Once again, the resolution is to recognize that C2's neuronal intelligence is not subject to change. If C2 could be magically transported back in time and raised in the Time 1 world, he would absorb only about 60% of the Time 1 environmental intelligence and would score relatively poorly on the Time 1 intelligence exam. The timing of one's birth does not alter one's personal intellectual ability.

In addition to these hypothetical examples from the model, Flynn provides a real world scenario that brings out both the paradox and its resolution in the most enlightening of ways. After noting that the average raw intelligence score from around the year 1900 would translate to an IQ of about 50 to 70 on today's scale, Flynn raises the specter of the following tableau:
"Jensen relates an interview with a young man with a Wechsler IQ of 75. Despite the fact that he attended baseball games frequently, he was vague about the rules, did not know how many players were on a team, could not name the teams his home team played, and could not name any of the most famous players.

"When Americans attended baseball games a century ago, were almost half of them too dull to follow the game or use a scorecard? My father who was born in 1885 taught me to keep score and spoke as if this was something virtually everyone did when he was a boy. How did Englishmen play cricket in 1900? Taking their mean IQ at face value, most of them would need a minder to position them in the field, tell them when to bat, and tell them when the innings was over."
This is a quintessential example of mistaking a change in raw intelligence scores as evidence for a change in neuronal intelligence, when in fact it is evidence for a change in environmental intelligence. Think about incorporating questions dealing with baseball rules into an intelligence test. If such questions had appeared on an exam in say the year 1800, no one at all, including the smartest people who then lived, would have been able to answer such questions correctly (other than by random luck). By contrast, if such questions were to appear on today's intelligence exams, many individuals, including those of low-to-average intelligence, would be able to answer the questions correctly—baseball and its rules have become an established part of the human environment, their widespread presence and influence are now thoroughly absorbed by a large percentage of the human population. As Flynn indicates, it would be only those with an IQ of around 75 or under who would have limited potential to answer such questions correctly.

So does this imply that the smartest people from the year 1800 must have had the same intellectual capacity as Jensen's young man? It of course does not imply that at all.

The critical moment in time would have been around the year 1900. If intelligence questions regarding baseball rules had appeared on intelligence exams at that time, the results would have been decidedly mixed. Some people would have been able to answer such questions correctly, but many others would not, including those of otherwise average-to-high intelligence, and this only because baseball had not yet become widely entrenched within the human environment (it was just then catching on). But after the exam was finished, if one of those baseball-ignorant, question-misanswering persons of average-to-high intelligence had been taken to the ballpark, bought a ticket, sat with in the grandstands, explained the rules, given a scorecard and pencil, a perfectly capable set of behaviors would have swiftly emerged. After all, this is a person of average-to-high intelligence, he can absorb and respond to baseball rules just fine, they will give him not the slightest bit of trouble. And around the year 1900, this scene would have actually been taking place, again and again and again, not just a hypothetical example but instead a real world, fully surveyable experience—an experience of human intelligence observably on the rise.

The increase in raw intelligence scores from 1900 to 2000 has everything to do with the increasing amount of environmental intelligence (including the addition of baseball rules). It has nothing to do with individual intellectual abilities. It has nothing to do with neuronal intelligence.


Another Flynn paradox is called the identical twins paradox. Flynn's words again:
"There is no doubt that twins separated at birth, and raised apart, have very similar IQs, presumably because of their identical genes. Indeed a wide range of studies show that genes dominate individual differences in IQ and that environment is feeble. And yet, IQ gains are so great as to signal the existence of environmental factors of enormous potency. How can environment be both so feeble and so potent?"
The short answer to this paradox is to say that environment, despite Flynn's doubts, is indeed both feeble and potent. It is feeble when considering individual and group intelligence differences that manifest at a particular moment in time—the domain in which neurology and genetics hold full sway. And environment is potent when considering intelligence differences that manifest across time—the domain in which neurology and genetics remain utterly silent. But although the short answer resolves the paradox precisely, it does not address what is actually the problem here, namely why does Flynn think this is a paradox.

There could be several ways one might analogize this essay's model of human intelligence. For instance, Lewontin's example of the seed corn would do fine. Also, one might consider the height of ships floating in a harbor, which differ from one another because of each ship's inherent characteristics (individual differences at a moment in time) and yet might change overall because of the rising and falling tide (environmental influence across time). Flynn would not find either Lewontin's seed corn or the rising and falling ships to be paradoxical, and yet the exact same mechanism applied to human intelligence seems to leave him utterly baffled. The question is why.

Flynn's bafflement arises from the ingrained assumption common to all intelligence researchers: each has become completely convinced that all intelligence differences and characteristics must ultimately be described as neural differences and characteristics. In other words, if an influence has no direct or indirect impact upon the human brain, then it cannot be an influence related to human intelligence. And so when Flynn considers environmental forces, which he can see have the perfect potential for explaining the Flynn effect, he stops short when he cannot find a plausible, straightforward way to tie those forces back to the supposedly requisite change in human genetics and/or neurology. This would be equivalent to not seeing how the nitrates in the soil can impact the seed corn's genetic structure, or not seeing how the water in the harbor can alter the ships' physical characteristics. But Flynn does not fall for this false dilemma with the seed corn or with the ships because he understands that the actions of the nitrates are orthogonal to the seed corn's genetics, and he understands that the water level in the harbor is independent of each ship's physical characteristics. It is only in the field of human intelligence that he finds himself unable to countenance this orthogonality and independence.

And yet that is all it takes. When one accepts the orthogonal relationship between neuronal intelligence and environmental intelligence, when one finally drops the the unnecessary and unsupported requirement that all intelligence characteristics are essentially neural characteristics, then the bafflement of the identical twins paradox swiftly disappears.


The remaining Flynn paradox is called the factor analysis paradox:
"How can intelligence be both one and many at the same time or how can IQ gains be so contemptuous of g loadings? How can people get more intelligent and have no larger vocabularies, no larger stores of general information, no greater ability to solve arithmetical problems?"
The first part of Flynn's statement is handled with ease: IQ gains across time can be so contemptuous of g loadings because IQ gains across time have absolutely nothing to do with neuronal intelligence, and therefore have absolutely nothing to do with Spearman's g. In fact, contemptuous is not the right word; utter indifference would more precisely capture the relationship.

The second part of the statement—why are intelligence gains differential across the many aspects of intelligence—that question is more intriguing and brings out additional features of environmental intelligence. In this essay so far, environmental intelligence has always been taken as a whole, with an emphasis on the principle that as a whole, environmental intelligence will be inexorably increasing. But when environmental intelligence is broken down into its component pieces and aspects, differing rates of increase can emerge. This will be seen for instance geographically, where around this planet's surface the rate of increase in environmental intelligence can vary from place to place. In the late 1900s through today, for example, the largest gains in environmental intelligence, and therefore the largest gains in IQ scores, most likely occurred in locales such as India and China, where there was a sudden and tangible surge of the overall amount of pattern, structure and form being added into the surroundings. Flynn's subcategories of vocabulary, arithmetic and general knowledge too, although more stable now, must have passed through epics where rapid increases undoubtedly occurred. Words, both spoken and written, obviously infiltrated the human environment at some point, as did numbers and their practical uses, and although no one was recording the surge in corresponding intelligence at that time, the surge clearly had to have taken place. That fewer common words and numerical techniques are being added into the environment today is compensated for by the palpable expansions in such areas as electronic logic, transportation networks, and so on. Plus none of these variable rates within the components of environmental intelligence should cause one to lose sight of the bigger picture, which is that the total amount of environmental intelligence will tend to increase persistently, and will do so without any influence upon, or any influence from, neuronal intelligence. These several features of environmental intelligence can be summarized as follows:
  1. Over time, the total amount of environmental intelligence will increase.
  2. Different aspects of environmental intelligence will increase at different rates at different times.
  3. Increases in environmental intelligence are independent of neuronal intelligence.
These principles of environmental intelligence (in particular, principle 2) are adequate to address the questions raised by the factor analysis paradox, and can do so without any unnecessary reliance upon the characteristics of the human brain.


Environmental Complexity. Of the many offered explanations for the Flynn effect, the one most similar to the model proposed here is the notion of environmental complexity. Schooler (1998) and Greenfield (1998) offer introductions to the idea, and it is not uncommon in general discussion to hear someone suggest that the modern usage of such things as puzzles, graphics and games might have something to do with the increasing levels of tested intelligence. Such suggestions are certainly on the right track, but when they are examined carefully and thoroughly, it can be seen that in many crucial respects their overall ability to explain the Flynn effect falls a good deal short.

The first problem with the notion of environmental complexity is that its proponents focus on certain things within the human environment, and ignore the impact of the environment as a whole. For instance, two commonly cited examples of the type of environmental complexity that can increase intelligence are the widespread use of video games, and the growing complexity and multivariate plot lines in television shows and movies. Others might highlight the expanded presence of visual imagery and puzzles within everyday life. But no matter what thing or set of things is being considered, it becomes immediately clear that by itself it cannot account for the ubiquitous and relentless nature of the Flynn effect. The Flynn effect was working its magic long before there even were video games and television sets, and the Flynn effect remains prominent in locations where video games and sophisticated dramas have yet to take much hold. Alternative candidates for environment complexity might be offered instead, but inevitably all must fall victim to the same problem of limited temporal and spatial impact. The Flynn effect is a population-wide, time-persistent phenomenon, and so any explanation for the Flynn effect has to have population-wide, time-persistent effect. Specific instances of environmental complexity are almost guaranteed to never fit the bill.

The second problem with the notion of environmental complexity is that its proponents—like nearly everyone else—insist on tying their explanation back to human neurology. Playing video games, for instance, is seen as expanding the capacity of working memory. Modern movie plots are described as forming a larger number of simultaneous connections within the logical neural circuitry. It would seem that environmental complexity by itself is rather useless, that its only real purpose is to prompt a major restructuring within the neurons, a massive rewiring between the ears. Such ideas now run amok within modern science, but they lack biological parsimoniousness and plausibility, they beg plasticity miracles within the human head. In truth, the human brain does not need to be changed by instances of environmental complexity, the human brain needs merely to respond to the stimulus of environmental complexity, a mechanism conforming quite nicely—and quite plausibly—to the traditional description of a neural system.

This essay's model of environmental intelligence, while similar to the notion of environmental complexity, avoids the shortcomings of environmental complexity by incorporating two significant improvements. One, environmental intelligence embraces a far more comprehensive context than does the notion of environmental complexity—comprehensive enough to have population-wide, time-persistent impact. Environmental intelligence achieves this comprehensiveness by eschewing the focus on particular things within the human environment and incorporating instead nothing short of the total amount of non-biological pattern, structure and form tangibly contained within the human environment. Two, environmental intelligence, unlike the notion of environmental complexity, severs the unnecessary tie back to human neurology, allowing environmental intelligence to accumulate and change without biological restriction and without resort to any biological miracle. Environmental intelligence takes the seed offered by the notion of environmental complexity and expands it to its full logical limit, expands it into a fully functioning component of human intelligence, one capable of serving as the embodiment of human intelligence, and one capable of serving as the orthogonal partner to the workings of the human brain.


Conclusion. A consequence that becomes readily apparent from this essay's dual-component model of human intelligence is that the Flynn effect cannot be regarded—as it too often is—as merely a twentieth-century anomaly. Tracking the historical increase in environmental intelligence, the Flynn effect must have begun near the time of the human great leap forward and will have been shadowing human existence ever since. And there is no reason to expect the Flynn effect will end anytime soon.

The unseen hand propelling intelligence scores upward is environmental intelligence, the total amount of non-biological pattern, structure and form tangibly contained within the human environment. It has remained unseen for so long because it has become so inextricably right there, right before one's very eyes, the very fabric of modern human existence, the map by which humans now navigate their world. If there is something worthy of being called a miracle in human intelligence, it would have to be this, environmental intelligence, for no other species on this planet has built its own version of environmental intelligence, and humans did not build theirs for a very long time.

The human brain—or at least researchers' obsessive focus on the human brain—has been given a thorough chastening within this essay, but that does not nullify the importance of the brain to human intelligence. The human neural system is still a necessary component of human intelligence; all that has been demonstrated here is that the human neural system is not a sufficient component of human intelligence. Any comprehensive description of intelligence, one capable of explaining individual and group intelligence differences as well as explaining the Flynn effect, will incorporate both neuronal intelligence and environmental intelligence, two components working simultaneously and orthogonally, producing an overall human intelligence that varies throughout the population and that increases year after year after year.



References
Flynn, J. R. (1999). Searching for justice: The discovery of IQ gains over time. American Psychologist, 54, 5–20.
Flynn, J. R. (2007). What is intelligence? Beyond the Flynn effect. New York: Cambridge University Press.
Greenfield, P. (1998). The cultural evolution of IQ. In U. Neisser (Ed.), The rising curve. Washington, DC: American Psychological Association.
Lewontin, R. C. (1976). Further remarks on race and the genetics of intelligence; Race and intelligence. In N. J. Block & G. Dworkin (eds.), The IQ controversy. New York: Pantheon Books.
Schooler, C. (1998). Environmental complexity and the Flynn effect. In U. Neisser (Ed.), The rising curve. Washington, DC: American Psychological Association.

Saturday, March 21, 2015

There Are Cesspools, and Then There Are Cesspools


Some of you may be following Dorothy Bishop's series of hubbub regarding Johnny Matson's reign of self-serving dishonor while in the role of editor for two Elsevier journals, Research in Autism Spectrum Disorders (RASD) and Research in Developmental Disabilities (RIDD). Really, the whole episode is kind of humorous and sad, in the way that most human activities are. But I would be remiss if I were not to point out here some observations that I am certain Bishop will overlook.

First of all, Matson's activities, while certainly extreme, are not at all unfamiliar. His reliance upon self citation, reciprocating citation and collegial nepotism differ from the remainder of the research community only in degree, not in kind. Everyone knows (wink, wink) that such activities have become the bedrock of survival for those scrambling to stay afloat in today's competitive academic industry. I am certain some might try to deny it or try to justify it, but if I were to rule out protests from those who have ever cited themselves, have ever cited a mentor or colleague, or have ever curried favor for position or assignment, then I am going to anticipate the push-back will be rather meager.

Second of all, Matson's activities are extremely old news (if they can even be called news, given how out in the open they were). I myself had noted all the way back in 2009 that RASD should be renamed The Matson Mouthpiece, and Michelle Dawson has been tweeting about Matson's self-referencing cesspool on practically a weekly basis since she opened her Twitter account. Bishop—who was acquainted with Matson, was aware of his work, and was listed as an editor on RASD—Bishop arrived extraordinarily late and ingĂ©nue to the party. But perhaps because she has been so successful in getting grant funding and publishing papers, when she raised a ruckus the research community finally took some notice.



Bishop's latest contribution to this episode is an attempt to shame Elsevier into apologizing for the Matson incident. The reason Bishop believes that an apology is in order is about as informative and revealing as any reason can be. Her words:
It matters because RIDD and RASD are presented to the world as peer-reviewed journals, backed up by the 'distinguished brand' of Elsevier. We live in times when there is competition for jobs and prizes, and these will go to those who have plenty of publications in peer-reviewed journals, preferably with high citations. If an editor bypasses peer review and encourages self-citation, then the quality of the work in the journal is misrepresented and some people gain unfair advantages from this. The main victims here are those who published in RASD and RIDD in good faith, thinking that acceptance in the journal was a marker of quality. They will be feeling pretty bitter about the 'added value' of Elsevier right now, as the value of their own work will be degraded by association with these journals.
Think about that statement for a moment. Think about it good and hard. Any of you who have ever put forth science as the preeminent means for acquiring truth and understanding about our world, think good and hard about what that statement must imply about what is actually valued in today's science. I certainly cannot fault the statement for being inaccurate, but I would think that for many of today's scientists, the exposure is just too embarrassing.

Or let me put it this way. If in today's research environment, Einstein were to publish his special relativity essay in a journal such as Science, the essay would of course be held in the highest regard by everyone—and this before anyone had even bothered to read it or understand it. On the other hand, if Einstein were to publish his essay in something like RASD as it was run under Johnny Matson's leadership, then the essay would now be regarded as forever tainted. And if Einstein were to publish his essay on the back of a set of cocktail napkins, it would be universally and instantaneously panned as totally worthless, not worth a glance. The contents of the essay? Well, what do they have to do with anything? (By the way, for anyone interested, the journal in which Einstein's essay first appeared was run with many remarkable similarities to Matson's RASD. Just saying.)

Listen, I am going to be blunt and crude about what I think is going on here: today's scientists do not give a shit about science. What they give a shit about is publication, reputation, grants and jobs. End of story.

Saturday, November 1, 2014

Mired

It was almost exactly ten years ago I was introduced into the world of autism. Much has changed for me since then, nearly all of it good. I can't say the same however for the world of autism -- if anything it has taken a few steps backwards.

Friday, April 4, 2014

The Flynn Effect In Its Entirety

Here are two recent research articles highlighting the ongoing relentlessness of the Flynn effect:
The appearance of these articles is in no way surprising: reports such as these have been popping up on a regular basis for nearly thirty years now. Of course, it was all the rage around ten years ago to proclaim the Flynn effect had been just a twentieth-century anomaly and now was coming to an end. I guess someone must have forgotten to tell the Saudi Arabian deaf children. I guess someone must have forgotten to tell the Chinese. I guess someone must have forgotten to tell just about everyone.

So let's make the current status of the Flynn effect a little more explicit, shall we: the Flynn effect is an ongoing phenomenon within the human population and all its subpopulations. That is, the Flynn effect is both active and ubiquitous. But I can go much further than that, because of course every piece of evidence from human history insists we must go further. Eschewing the narrowness of modern scientific vision, I insist we add a clause: the Flynn effect is an ongoing phenomenon within the human population and all its subpopulations, and has been so for at least the last ten thousand years, probably much longer than that. That is, the Flynn effect has been active, ubiquitous and relentless ever since man first displayed signs of intelligence and began scattering off the savannas. Far from being just a twentieth-century anomaly, the Flynn effect has become a deeply ingrained aspect of the human condition, and holds the key to intelligence itself and its significant non-neuronal component.

At the present time, no intelligence researcher recognizes the depth and breadth of the Flynn effect. Even James Flynn, who probably comes the nearest to understanding we are dealing with an ideal measure of human modernity, insists nonetheless on limiting the temporal range of that insight to mostly the last century alone, and certainly no further back than the industrial revolution. Such limited perspective is an unnecessary mistake. Limited perspectives engender limited explanations, and limited explanations are a priori inadequate, because the Flynn effect shows no evidence of being a limited phenomenon.

The latest Flynn effect explanatory fad has the Flynn effect being produced by greater guessing on standardized tests, a perfectly suitable hypothesis I would say, as long as the theory's authors are willing to boldly and courageously step forward and insist the entire human population is currently engaged in greater guessing on standardized tests and has been doing so for at least the last ten thousand years. However, if the theory's authors are for some reason hesitant to make such a claim, then I am going to insist on dismissing their puny hypothesis as entirely inadequate to the task, a dismissal they need not feel all that bad about, since of course they will have plenty of company.



Liu, J. & Lynn, R. (2013). An increase of intelligence in China 1986–2012. Intelligence, 40, 139–144. http://dx.doi.org/10.1016/j.intell.2013.06.017

Bakhiet, S., Barakat, S. & Lynn, R. (2014). A Flynn effect among deaf boys in Saudi Arabia. Intelligence, 44, 75–77. http://dx.doi.org/10.1016/j.intell.2014.03.003

Armstrong, E. & Woodley, M. A. (2013). The rule-dependence model explains the commonalities between the Flynn effect and IQ gains via retesting. Learning and Individual Differences, 29, 41–49. http://dx.doi.org/10.1016/j.lindif.2013.10.009

Tuesday, December 3, 2013

Connectivity Explained

There seems to be a bit of a hullabaloo developing over the state of neural connectivity in autism. But to be honest, I fail to see what all the fuss is about — a little autism science logic can clear up the matter instantly. To wit:
  • Some studies have shown that various regions of autistic brains are under-connected relative to typical controls. This is a problem because everything associated with autism is bad.
  • Some studies have demonstrated that various regions of the autistic brain are over-connected (hyper-connected) relative to typical controls. This is a problem because everything associated with autism is bad.
  • Some studies have suggested that various regions of the autistic brain are indistinguishable from typical controls. This is a problem because everything associated with autism is bad.
There, see how easy it is to clear up these matters when you apply a little autism science logic.

Monday, September 23, 2013

Every Disney Drama Has a Cruel Villain

I'll tell you what Autism Speaks understands: the big money it garners from corporations like Disney.

And I'll tell you what Autism Speaks doesn't understand: autistic kids.

Sunday, August 11, 2013

Cohorts, Time and the Flynn Effect

Agbayani (2013) is a recently published paper on the topic of the Flynn effect and age-related IQ decline, and is a continuation of the ideas presented in Agbayani (2011) and Dickinson (2010). (Of these, only Agbayani (2011) is not paywalled, but each paper comes to roughly the same conclusion from essentially the same set of data, so reading Agbayani (2011) is enough to get the gist of the idea.) What these papers do is take the results from Wechsler IQ exams given at different time periods and for a variety of age categories and combine these with Flynn effect norming information to produce what is in essence a directly comparable grid of intelligence scores along the two dimensions of calendar year and age. It is the same technique I have employed (in idealized form) in previous posts, for instance in Intelligence as Field. As such, the empirical results of Agbayani (2013), although admittedly limited in scope, provide some corroboration for my idealized charts.

There are three temporal patterns that emerge from the Agbayani (2013) data:
  • For any given calendar year, the age-related pattern of raw intelligence scores shows a peak for test takers in early adulthood, followed by a gradual yet significant decline for test takers at increasingly older ages.
  • For any given age category, raw intelligence scores show a consistent increase across calendar years, with the rate of increase being similar for all age categories.
  • The intelligence profile for any given birth cohort (which would be obtained by reading diagonally across the grid) shows a relatively flat level of scores from early adulthood until around age 70 or so. This is consistent with most longitudinal intelligence studies applied to individuals.
An idealized chart of comparable intelligence scores roughly matching the empirical results of Dickinson (2010), Agbayani (2011) and Agbayani (2013) would look something like Chart A:

AgeChart A
Comparable Intelligence Scores by Calendar Year and Age
60 80 88 97
40 90 99 109
20 100 110 121
1960 1980 2000
Calendar Year

Note that this chart is consistent with the three temporal patterns listed, and in particular note that for the cohort born in 1940 (age 20 in 1960, age 40 in 1980, age 60 in 2000), the raw intelligence profile is relatively flat across adulthood, just as Agbayani (2013) highlights.

In Dickinson (2010), Agbayani (2011) and Agbayani (2013), these results are interpreted in the following way (I will call this the Agbayani interpretation):
  • The Flynn effect is a function of cohorts.
  • In the absence of a Flynn effect, the cross-sectional age-related declines in intelligence scores will mostly disappear.
In essence, these authors look at the flat intelligence profile for a given cohort and combine this with the assumption that the Flynn effect remains constant over the lifetime of that cohort, and conclude that the flat age-related intelligence profile that emerges is therefore the natural age-related intelligence profile. Or in other words, these authors are saying that the cross-sectional age-related declines in intelligence scores evident for any given calendar year will disappear once the cohort-induced Flynn effects are factored out.

That interpretation is almost certainly wrong.

An alternative interpretation to the Agbayani interpretation, one that I find far more natural and plausible, can be outlined as follows (I will call this the Griswold interpretation):
  • The Flynn effect is a continuous, universal function of time (not of cohorts).
  • In the absence of a Flynn effect, the age-related decline in intelligence scores will emerge in each cohort, demonstrating that such decline is in fact the natural age-related intelligence profile.
Under the Griswold interpretation, the cross-sectional age-related declines in intelligence scores at each calendar year are indicative of the actual age-related intelligence profile and are unaffected by the presence of any Flynn effect. The flat intelligence profile that is today evident in each birth cohort arises from the combination of two independent and mutually offsetting influences: one, the natural decline in intelligence that occurs as one ages, and two, the increase in intelligence that results from the continuous influence of the Flynn effect throughout one's lifetime. (It is something of a coincidence that these two opposite-direction influences are nearly equal in magnitude.)

To compare the Agbayani and Griswold interpretations and assess their consequences and resulting plausibility, we need to examine each interpretation under a scenario in which there is no Flynn effect, then follow this with a description of how each interpretation can be transitioned into a scenario in which the Flynn effect has full impact. Such an investigation will highlight the fundamental contrast between the two interpretations and will demonstrate how each necessitates a dramatically different understanding of how the Flynn effect must work.

We can begin with the Agbayani interpretation. Assuming a scenario in which there is no Flynn effect and given the Agbayani assumption that there will be little age-related decline in intelligence scores in the absence of a Flynn effect, the resulting two-dimensional grid of raw intelligence scores would now have to look something like Chart B:

AgeChart B
Agbayani interpretation, no Flynn effect
60 97 97 97
40 99 99 99
20 100 100 100
1960 1980 2000
Calendar Year

Chart B meets the requirements of the Agbayani interpretation. Note that there is practically no age-related decline either at any calendar year or for any cohort, and of course there is also no telltale sign of a Flynn effect. From this base, the trick now is to figure out how to introduce a Flynn effect — that is, transition from Chart B to Chart A — and do it in a way that is consistent with the Agbayani interpretation.

When I first contemplated the Agbayani interpretation, I could see no reasonable way to make that transition work, but upon further reflection I recognize I was being too hasty. In truth, there is a way to make the math work out. What needs to be done is to add something to each birth cohort — call it a Flynn effect boost (FEB) — and in order to account for the fact that the Flynn effect is evident across all age groups (including child age groups), it is necessary to add this boost right from the very beginning (at birth, if you will) and then have its influence remain constant over the cohort's lifetime. Mathematically, the technique looks something like Chart C:

AgeChart C
Agbayani interpretation, Flynn effect transition
60 97 x (1 - 2 x FEB) 97 x (1 - 1 x FEB) 97 x (1 + 0 x FEB)
40 99 x (1 - 1 x FEB) 99 x (1 + 0 x FEB) 99 x (1 + 1 x FEB)
20 100 x (1 + 0 x FEB) 100 x (1 + 1 x FEB) 100 x (1 + 2 x FEB)
1960 1980 2000
Calendar Year

Chart C starts from the no Flynn effect base of Chart B, then each successive cohort is given a Flynn effect boost that is larger than the boost provided to the previous cohort. Furthermore, each cohort is given the entirety of its Flynn effect boost right from the beginning, with the influence of the boost remaining constant thereafter over the cohort's existence. Reading diagonally across Chart C and observing how the Flynn effect is introduced and maintained in each cohort, we see that this technique is exactly what is required in order to remain consistent with the Agbayani assumption that the Flynn effect is purely a function of cohorts.

If we plug in an FEB value of approximately 0.1, then Chart B transitions quite smoothly into Chart A, thereby demonstrating that the suggested technique provides a plausible mechanism for how the Flynn effect must work under the Agbayani interpretation. I leave open the possibility that I have overlooked something, but as far as I can tell, this technique is the only one mathematically plausible under the conditions of the Agbayani interpretation.



By contrast, the scenarios and transitions look quite different under the Griswold interpretation. We begin once again by considering the scenario in which there is no Flynn effect, along with the Griswold assumption that the age-related intelligence decline will continue to be evident under such a scenario. This means that the two-dimensional grid of raw intelligence scores must now look something like Chart D:

AgeChart D
Griswold interpretation, no Flynn effect
60 80 80 80
40 90 90 90
20 100 100 100
1960 1980 2000
Calendar Year

Chart D meets the requirements of the Griswold interpretation. Note that the age-related decline is still evident for all calendar years, and furthermore the age-related decline is now evident also for each cohort. Plus there is no hint of a Flynn effect anywhere in these numbers. As before, the next step is to introduce a Flynn effect — that is, transition from Chart D to Chart A — and do it in a way that is consistent with the Griswold interpretation.

Here, the math is fairly straightforward and practically suggests itself. Consistent with the Griswold assumption that the Flynn effect produces a continuous and universal impact over time, the transition is produced simply by introducing a Flynn effect boost (FEB) at each calendar year and across all age groups. Mathematically, the technique looks something like Chart E:

AgeChart E
Griswold interpretation, Flynn effect transition
60 80 x (1 + 0 x FEB) 80 x (1 + 1 x FEB) 80 x (1 + 2 x FEB)
40 90 x (1 + 0 x FEB) 90 x (1 + 1 x FEB) 90 x (1 + 2 x FEB)
20 100 x (1 + 0 x FEB) 100 x (1 + 1 x FEB) 100 x (1 + 2 x FEB)
1960 1980 2000
Calendar Year

Chart E starts from the no Flynn effect base of Chart D, then a Flynn effect boost is introduced continuously and universally over time, impacting all age groups and all cohorts the same, exactly as required under the assumptions of the Griswold interpretation. If we plug in an FEB value of approximately 0.1, then Chart D transitions quite smoothly into Chart A, demonstrating that the suggested technique provides a plausible mechanism for how the Flynn effect must work under the Griswold interpretation.



Since each interpretation appears to be mathematically plausible, we need to search further for evidence to rule out either (or both). The ideal approach to this task would be to examine an adult population for which there is no Flynn effect: if the cross-sectional age-related intelligence scores are flat over such a population, the Griswold interpretation could be ruled out, and if the cross-sectional age-related intelligence scores are in decline over that population, the Agbayani interpretation could be ruled out. Unfortunately, it is unclear whether any extant human population can be characterized as untouched by the Flynn effect (the Flynn effect's relentless ubiquitousness being one of its more tantalizing features), and thus the ideal approach appears to be unavailable.

If the Flynn effect were to come to a halt, then that would also provide a means for assessing the two interpretations, because the halting would produce distinctly different signatures under each interpretation. Under the Griswold interpretation, a stop in the Flynn effect would immediately impact all age groups and all cohorts, something like Chart F:

AgeChart F
Griswold interpretation, halt of the Flynn effect at year 2000
60 80 88 97 97 97 97
40 90 99 109 109 109 109
20 100 110 121 121 121 121
1960 1980 2000 2020 2040 2060
Calendar Year

Under the Agbayani interpretation, a halt to the Flynn effect would produce a more complex pattern, because the halt would impact only future cohorts and not any existing cohorts. This would produce something like Chart G:

AgeChart G
Agbayani interpretation, halt of the Flynn effect at year 2000
60 80 88 97 107 118 128
40 90 99 109 120 130 130
20 100 110 121 131 131 131
1960 1980 2000 2020 2040 2060
Calendar Year

Under the Agbayani interpretation, for a period of time after the halt of the Flynn effect, intelligence scores would continue to increase in the older age categories but would begin to level off at the younger age categories (including childhood ages). Such a signature would be strong evidence against the Griswold interpretation and in favor of the Agbayani interpretation. Unfortunately, it is again unclear whether in the real world a halt to the Flynn effect is imminent or whether it could be quickly and easily recognized; and thus this technique, while theoretically interesting, would appear to be pragmatically out of reach.

Despite these obstacles, I do believe some strong arguments can be made against the Agbayani interpretation. The first difficulty is with the assumption that raw intelligence abilities would remain relatively level across the adult years in the absence of a Flynn effect. Although this is certainly possible, it runs counter to many other known biological abilities, such as athleticism and sexual vitality, and it would seem that a pattern of peak during early adulthood followed by gradual decline in later years would be the preferred assumption. To assume otherwise should require some evidentiary explanation, and of course Agbayani (2013) does not provide that explanation — it is only the Agbayani interpretation that supports the assumption of a flat level of intelligence across adulthood, and we have seen that there is at least one alternative interpretation that runs exactly counter to such an assumption.

More problematic still is the need for a Flynn effect boost to be provided to each cohort quite early in its existence, a need driven by the assumption that the Flynn effect is a function of cohorts. Although such a mechanism is theoretically possible, it runs counter to almost everything that is commonly understood about intelligence and the Flynn effect. For instance, you cannot say something like better education might produce the Flynn effect, because under the Agbayani interpretation a cohort's Flynn effect boost has to be fully in place before even the first day of school. Many similar intelligence explanations are ruled out for the exact same reason. By making the Flynn effect a function of cohorts, one removes the element of time, and much of what we understand about the acquisition of intelligence depends upon the passage of time, not upon the introduction of cohorts. Under the Agbayani interpretation, it would appear we must look only for Flynn effect causal candidates that are materially different across cohorts, present in nearly every member of each cohort, and present essentially right from birth. The list of plausible such candidates would seem to be conspicuously small. Although it is extremely common and popular to assume that the Flynn effect must be a cohort-driven phenomenon (see for instance the first sentence here), I think the proponents of that assumption fail to appreciate the difficult-to-explain consequences that must inevitably arise.

Admitting freely to my bias, it seems to me that the Griswold interpretation suffers from far less strain. In the first place it adopts the more physically natural assumption that unaided by a Flynn effect, human intelligence would peak in early adulthood then gradually decline towards old age — mirroring similar abilities in athleticism, health maintenance and sexual vitality. More importantly, by describing the Flynn effect as a universal, continuous, incremental function of time (not of cohorts), the Griswold interpretation opens the door to a straightforward, environmental explanation for the Flynn effect. Any consistently changing phenomenon that is essentially present for all people in all places at all times becomes a viable candidate as a causal explanation for the Flynn effect, and such candidates are quite conceivable. I have described elsewhere that I believe it is the increasing amount of non-biological pattern, structure and form tangibly contained within the human environment that serves as the most likely driver for the Flynn effect, and although here is not the place to argue the merits of that explanation, it is permissible for me to note it is entirely consistent with the Griswold interpretation. I am unaware of any competing Flynn effect explanation that is entirely consistent with the Agbayani interpretation.




Agbayani, K.A. & Hiscock, M. (2013). Age-related change in Wechsler IQ norms after adjustment for the Flynn effect: Estimates from three computational models. Journal of Clinical and Experimental Neuropsychology, 35(6), 642-654.

Agbayani, K. A. (2011). Patterns of age-related IQ changes from the WAIS to WAIS-III after adjusting for the Flynn effect. Retrieved online from http://repositories.tdl.org/uh-ir/handle/10657/236.

Dickinson, M. D. & Hiscock, M. (2010). Age-related IQ decline is reduced markedly after adjustment for the Flynn effect. Journal of Clinical and Experimental Neuropsychology, 32(8), 865-870.