Saturday, December 17, 2011

More Neanderthal Data

One of the better sources for up-to-date information regarding Neanderthal admixture has been John Hawks' weblog. Hawks is a paleoanthropologist at the University of Wisconsin, and in addition to providing links to and commentary upon some of the latest research, Hawks and his students have been working at replicating and extending some of the more interesting findings regarding the Neanderthal genome and its presence in modern humans.

An excellent overview of the process is contained in this post from last week, and in it you will find a thorough but easily accessible explanation for how scientists arrive at the 1-4% admixture estimate of Neanderthal genetic material into non-African humans. In the original paper describing Neanderthal admixture (Green 2010), only five modern human genomes—two African and three non-African—were compared to the Neanderthal draft sequence. Hawks and his students have been making much the same comparison against a much larger database of modern genomes, with some clarifying results. Their work is of course still preliminary and not reviewed (and thus should be taken with a grain of salt), but if it proves to be accurate, it leads to some interesting conclusions:

  • The admixture finding from (Green 2010) is being convincingly replicated and confirmed as the genomes from more and more present-day humans are being compared to the draft sequence of the Neanderthal genome. Either the draft sequence has some very bad data, or the assumptions behind the genome comparison techniques are totally without merit, or Neanderthal admixture into non-African humans is a confirmed reality—there doesn't seem to be any other way around it.
  • The mean amount of Neanderthal admixture into non-African populations (over and above any admixture into African populations) looks to be around 3%.
  • Although 3% is the mean, the variability remains significantly wide: any given non-African individual might easily fall within the range of 2-4% admixture, and there are indications some individuals will fall well outside that range. Even within families, there might be significant variability.
  • All indications are that the Neanderthal admixture is shuffled throughout much of the modern human genome. That is, person A might have 3% admixture and person B might have 3% admixture, but the two may share relatively little admixture in common. It's not specific parts of the modern genome that are Neanderthal derived; instead, it appears that much of the modern non-African genome has been impacted in a shuffled-up sort of way.
  • The mean Neanderthal admixture difference between Europeans and Asians is extremely small. This result goes against the idea that an out-of-Africa migrating population interbred primarily with Western European Neanderthals around 35 to 40 thousand years ago. Instead, this result points more plausibly to the idea that admixture happened earlier in time and/or at another locale (such as the Middle East).
  • All these findings are based solely upon SNP analysis. The other sources of genetic differentiation—such as duplication, deletion, insertion, inversion—still await less costly and more accurate analysis. But in these areas too, Neanderthal admixture might be expected to have significant impact.

None of these findings provide any direct evidence for my idea that autism is a species-differentiation event (in particular, a species dis-recognition event), but neither do any of these findings contradict the idea. As I've said elsewhere, what we can look for now are studies comparing the Neanderthal draft sequence to modern humans diagnosed with such things as autism, schizophrenia and bipolar. If my idea is going to have any legs, then what we might expect to find are Neanderthal-distinguishing genetic signatures among these diagnosed populations, either admixture amounts outside the norm (most likely higher than the typical range) or a specific pattern of relationship against the Neanderthal genome unique to the diagnosed population. Given the rapid rate of technology advancement in this area, we might not have to wait all that much longer.

One other thing: as I read through these Neanderthal admixture findings, I can't help but be reminded of the genetic research conducted so far in the areas of autism, schizophrenia and bipolar. In each instance, researchers have been uncovering extremely lengthy lists of candidate genetic markers for the condition, markers that show up in only a tiny percentage of the affected population and markers that are scattered almost at random throughout the genome. This of course also does not provide any direct evidence that Neanderthal admixture is playing a role in such conditions, but the genetic similarities are difficult not to notice. And contrast this with the dubious conclusion being drawn by the autism research community, a community that insists on casting all these genetic variations as genetic defects, each leading (by miraculous coincidence, it would seem) to the same neurological and phenotypic outcome. There is a difference between the words “speculative” and “implausible,” and I see that difference on display here.

(Green 2010): Green, Richard E. and others. 2010. “A Draft Sequence of the Neandertal Genome.” Science 328:710–22

Wednesday, December 14, 2011

Ami Klin's Good Science

Can someone explain to me how Ami Klin's research team in (Shultz 2011) could have possibly overlooked the idea of running a second version of their experiment, one employing a visual scene more naturally appealing to autistic perception? It would have been so easy: remove the humans from the scene and through automated means have the toy wagon's door open and shut on a regular basis. Bring in another group of controls and autistic children, and make the same measurements as were made for the first scenario. The comparisons across populations and across scenarios might have provided a wealth of information for both autistic and non-autistic perception, probably far more than was provided by the study as conducted.

Any decent research team might have recognized the potential in that second scenario, but it was criminally stupid for the Klin research team not to have recognized it, because the team had already been through that experience!

The whole beauty of (Klin 2009) was that instead of the nothing-new experiment the authors had originally designed, and prompted by the “serendipitous” observations of a fifteen-month-old autistic girl, the study team was able to re-configure the experiment to test for both non-autistic appealing and autistic appealing scenarios, thereby setting up a wealth of cross-population, cross-scenario information for comparison and contrast. That serendipitous enhancement literally made the study.

So you would think the Klin team would know by now the value of taking that broader approach with every experiment going forward. You would think. But apparently the fifteen-month-old autistic girl didn't hit them upside the head hard enough.

I can't describe how dumbfounded I am by the scientific blindness I see on display almost everywhere in the autism research community. People go on and on to me about the need to weed out bad science and to expose all the charlatans and to have better standards, and on and on and on they go. For what, I ask? For this? For more of Ami Klin's inability to see beyond the end of his own nose? Is this the good science everyone's striving for? Hell, I'd rather have the charlatans. At least the charlatans know what they're doing.

(Shultz 2011): Shultz, Sarah; Klin, Ami; Jones, Warren. 2011. “Inhibition of eye blinking reveals subjective perceptions of stimulus salience.” PNAS (in press).

(Klin 2009): Klin, Ami; Lin, David J.; Gorrindo, Phillip; Ramsay, Gordon; Jones, Warren. 2009. “Two-year-olds with Autism Orient to Non-Social Contingencies Rather than Biological Motion.” Nature 459: 257–61.

Saturday, December 10, 2011

Competing Speculative Hypotheses

About a year ago, I wrote a rather lengthy essay (Griswold 2011) inspired by the paleoanthropologist Richard Klein and his ideas regarding the behavioral and cultural changes known as the human great leap forward. Here ( you can watch a recent 45-minute lecture by Dr. Klein on the topic, and although I think his written works (for instance, (Klein 2002) and (Klein 2008)) provide better detail, the video lecture does give a reasonable introduction and overview to Klein's approach and is well worth the short investment of time.

Near the very end of the lecture, after having outlined arguments and evidence for the notion that the human great leap forward was a sudden event—occurring around fifty thousand years ago—Klein states his hypothesis that the sudden event must have been launched by a genetic mutation, one producing significant cognitive effect. Klein then briefly mentions the recent and preliminary work on the mapping of the Neanderthal genome (Green 2010), work that has led to two major (and still preliminary) results: 1. there is now a list (a fairly short list) of modern human/non-Neanderthal gene sequences that would serve as obvious candidates for recent genetic mutation; and 2. nearly all non-African modern humans carry a small influence (estimated in the neighborhood of 2.5%) of Neanderthal-originated genetic material.

As Klein mentions in his lecture, the first result opens the door to a means for testing his genetic mutation hypothesis. The idea would be to obtain a list of modern human genetic sequences that differ significantly from those of ancient humans (including Neanderthals) and see if a subset of these produce the kind of neurological impact consistent with human behavioral and cognitive change. There are of course some technical challenges that stand in the way of this approach: for one, the genome mappings are still in need of greater clarification and accuracy, and furthermore, scientists have not yet been all that successful in connecting genetic material to phenotypic effect, be it cognitive or otherwise. Nonetheless, these are challenges that might be expected to be overcome through technological advances, and so indeed, Klein's hypothesis might one day soon be put to a thorough scientific test.

As I outlined in (Griswold 2011), I'm fairly convinced Klein is going to be disappointed in the results of that test. Although I concur with Klein's assertion that the human great leap forward was a sudden anthropological event, I see an incongruity undermining Klein's explanation for that event. Although most scientists, Klein included, would take for granted that human cognitive advancement must have been driven by a genetic/neurological/evolutionary change, nearly everything scientists can actually demonstrate about genetics, neurology and evolution runs counter to the type of sudden, population-wide, large-scale event Klein is describing. It's as though the animal world had gone sightless for billions of years and then overnight one of the species popped up a pair of excellent eyes and immediately conquered the rest of the planet through this new-found vision. It makes for a dramatic story, but biology doesn't seem to work that way.

Genetically-driven evolutionary change does of course happen, but not on the time or impact scale Klein is proposing (and not even on the time or impact scale that cultural evolutionists would propose). On the one hand, Klein's description of sudden human behavioral and cognitive change looks accurate enough based upon the archaeological evidence, but on the other hand, his explanation for that sudden change looks utterly implausible based upon the logic of biology.

That said, it would still be prudent to wait for the science.

Also in the video lecture, almost as an aside, Klein dismisses the other major finding from Neanderthal genome mapping—the admixture of Neanderthal genetic material into modern humans—suggesting that as details of the respective genomes become more complete and accurate, this finding will prove to be false. Logically speaking, however, Klein doesn't need to make that dismissal: it's perfectly possible that Neanderthal admixture will continue to hold upon further analysis, and yet its impact on human behavioral and cognitive change will prove nonetheless to be benign. I think what's driving Klein's desire to dismiss the admixture finding is that he wants to emphasize how modern Homo sapiens—post genetic mutation—were so cognitively and behaviorally advanced over their Neanderthal contemporaries that all they could do was swamp the Neanderthals into extinction, not interact with or incorporate them. That's a reasonable conclusion to draw given what we already know about European replacement of Neanderthals (evidence for which Klein has intimate working knowledge), but in point of fact there's nothing about the Neanderthal admixture finding that implies it had to be a post out-of-Africa event—the evidence for the timing of that admixture remains inconclusive, and it's quite possible any such admixture could have taken place near the beginning or even prior to the out-of-Africa migration.

Here too, it would be prudent to wait for the science.

My own interest in the Neanderthal admixture finding is that it serves as possible evidence for an alternative explanation to the human great leap forward, an explanation I find logically more plausible—albeit perhaps just as speculative—as Richard Klein's. In (Griswold 2011) I outline how the introduction of autism into the human population—autistic perception in particular—could have served as the catalyst driving human cognitive and behavioral change. I won't repeat the details here, but the concepts at work are a description of autism as a lack of species recognition; autistic perception as a compensatory foregrounding of non-biological pattern, structure, symmetry and form; human cognitive and behavioral advancement as the environmental accumulation of these very same elements of non-biological pattern, structure, symmetry and form; and Neanderthal admixture as the conceivable biological cause for autism-related species dis-recognition. Under my scenario, all these concepts would have come together in a kind of circumstantial stew that began cooking around fifty thousand years ago.

With the ongoing advancements in human genome sequencing, parts of my autism hypothesis might become just as amenable to scientific testing as Klein's genetic mutation hypothesis. The key evidence to look for is whether large or distinctive presences of Neanderthal-derived genetic material within individuals correlates significantly to diagnoses of autism (and perhaps to similar conditions of schizophrenia and bipolar). A high and distinctive correlation would be supportive for describing autism as a condition of species dis-recognition, and thereby indirectly supportive of an autism-related explanation for mankind's great leap forward.

Of course it's also possible that neither speculation—mine nor Richard Klein's—will prove to be helpful, and it will be some other explanation, perhaps one not yet thought of, that manages to untangle the mysteries from fifty thousand years ago. But one thing is for certain: these recent advancements in human genome mapping are opening an intriguing window onto our anthropological past. It's an excellent time to be alive if one is prone to asking such questions as, what caused human beings to become so distinctively human?

(Green 2010): Green, Richard E. and others. 2010. “A Draft Sequence of the Neandertal Genome.” Science 328:710–22

(Griswold 2011): Griswold, Alan. 2011. Autistic Songs. Bloomington, IN: iUniverse.

(Klein 2002): Klein, Richard G. 2002. The Dawn of Human Culture. New York: Wiley.

(Klein 2008): Klein, Richard G. 2008. "Out of Africa and the Evolution of Human Behavior." Evolutionary Anthropology 17:267–81.

Wednesday, December 7, 2011

The Joy of Meta

Here is an abstract from the very latest in autism research:

Recent studies have implicated physiological and metabolic abnormalities in autism spectrum disorders (ASD) and other psychiatric disorders, particularly immune dysregulation or inflammation, oxidative stress, mitochondrial dysfunction and environmental toxicant exposures (‘four major areas’). The aim of this study was to determine trends in the literature on these topics with respect to ASD. A comprehensive literature search from 1971 to 2010 was performed in these four major areas in ASD with three objectives. First, publications were divided by several criteria, including whether or not they implicated an association between the physiological abnormality and ASD. A large percentage of publications implicated an association between ASD and immune dysregulation/inflammation (416 out of 437 publications, 95%), oxidative stress (all 115), mitochondrial dysfunction (145 of 153, 95%) and toxicant exposures (170 of 190, 89%). Second, the strength of evidence for publications in each area was computed using a validated scale. The strongest evidence was for immune dysregulation/inflammation and oxidative stress, followed by toxicant exposures and mitochondrial dysfunction. In all areas, at least 45% of the publications were rated as providing strong evidence for an association between the physiological abnormalities and ASD. Third, the time trends in the four major areas were compared with trends in neuroimaging, neuropathology, theory of mind and genetics (‘four comparison areas’). The number of publications per 5-year block in all eight areas was calculated in order to identify significant changes in trends. Prior to 1986, only 12 publications were identified in the four major areas and 51 in the four comparison areas (42 for genetics). For each 5-year period, the total number of publications in the eight combined areas increased progressively. Most publications (552 of 895, 62%) in the four major areas were published in the last 5 years (2006–2010). Evaluation of trends between the four major areas and the four comparison areas demonstrated that the largest relative growth was in immune dysregulation/inflammation, oxidative stress, toxicant exposures, genetics and neuroimaging. Research on mitochondrial dysfunction started growing in the last 5 years. Theory of mind and neuropathology research has declined in recent years. Although most publications implicated an association between the four major areas and ASD, publication bias may have led to an overestimation of this association. Further research into these physiological areas may provide insight into general or subset-specific processes that could contribute to the development of ASD and other psychiatric disorders.

I'm not sure how badly you'll want to delve into all that verbiage, but in short it's an assessment of various "abnormalities" associated with autism through the means of looking at research publication trends. For instance, the association of mitochondrial dysfunction with autism is assessed by looking at the rising number of mitochondria-autism research articles published over the calendar years from 1971 to 2010. It's the type of work that can be done with a good search engine and a spreadsheet program (3D color graphics would be a bonus).

In other words, it's a shitty piece of meta-analysis.

Of course, there have been better examples of meta-analysis applied to autism research, but this raises an intriguing question. How does one differentiate good meta-analysis from bad meta-analysis (think of how important this will be to Ben Goldacre as he's writing his next book, Bad Meta Science)? Well obviously, what we need to do is institute some standards, protocols and ethics for the general practice of meta-analysis, and once these are in place we can begin performing meta-analyses of all the meta-analyses. And if there happen to be any lingering problems, we can just ask Janet Stemwedel to remind the tribe of meta scientists of all their meta social duties.

No really. I think this would be a fruitful line of endeavor. As far as I can tell, the possibilities for employment are unlimited.

Monday, December 5, 2011

Autism Etiology

Genetics, neurology, environment—permutations and combinations aplenty, plausible mechanisms afew.

Saturday, December 3, 2011

Dispelling the Fog

Why try to deny the Flynn effect or pass it off as a short-term anomaly? If we embrace the Flynn effect for exactly what it says, the faulty explanations disappear.

Wednesday, November 30, 2011

Read All About It

From Rose Eveleth comes an absolutely gorgeous article on autistic potential, spurred by the work of Mottron, Soulières, Dawson and colleagues. The article makes a nice blend of realism and affirmation, and might be the best bit of autism-related journalism I've yet to see.

Tuesday, November 29, 2011

On the Prowl

Scientists are perfectly free to keep looking for the source of intelligence inside the human brain. But they might as well keep looking for the luminiferous ether while they're at it.

Monday, November 21, 2011

A Balanced Approach

I will say this much in defense of Ami Klin. His description of autistic perceptual characteristics comes with a corresponding (and fairly accurate) description of non-autistic perceptual characteristics, along with an appreciation for the merit and deeply ingrained nature of those characteristics (see for instance the answer to the next-to-last question here). This is an approach I wish the Mottron research team would seriously consider. Indeed, if we could combine the best of Ami Klin's observations and Laurent Mottron's research and ideas, we might arrive at an informative and mutually clarifying description for these two fundamental forms of human perception. But alas, blind spots continue to be the norm within autism research.

Sunday, November 20, 2011

Calibrating the Eye-Tracker

Ami Klin is making a fundamental mistake in emphasizing early autistic attention towards objects. When autistic toddlers line up toys, they're not intrigued by the toys—they're intrigued by the line.

Friday, November 11, 2011


The one consistent finding throughout autism research is that whatever autistics do, it's wrong.

Wednesday, November 9, 2011

Liars and Hypocrites

If I'd known that curing autism would be so easy, I might not have raised such a big objection. But according to recent reports, all we need to do is train autistic individuals to be better liars and hypocrites, and they'll become virtually indistinguishable from their neurotypical peers.

Tuesday, November 8, 2011


Many people attack science because they don't want to think. I attack science because scientists don't want to think.

Saturday, November 5, 2011

How Do I List Thee? Let Me Count the Ways

Oh, I just relish authorship discussions like this one. They always leave me wondering how Darwin or Tolstoy would have handled such dilemmas.

Thursday, November 3, 2011


It's nice to see the media attention being given to Laurent Mottron, Michelle Dawson, and their immediate colleagues in conjunction with the appearance of Dr. Mottron's commentary Changing perceptions: The power of autism (which sadly is behind a paywall). I've never been one to be hesitant about expressing my dissent from some of these researchers' positions and views, but this has always been done in the context of great admiration and respect for their overall effort. These are scientists who have consistently led the way in providing autism research that is affirmative, encouraging, accurate and productive for autistic individuals, a stance which has demonstrated both courage and insight. Any positive attention they get is well deserved and well earned.

[Update: It appears that the paywall restriction has been removed. More applause.]

Tuesday, November 1, 2011

Kierkegaard's Lemma

When being a Christian was unpopular, unrewarding, even dangerous, then we had true Christianity. Science is no different.

Saturday, October 29, 2011

Lab Rats

The best science is descriptive. If you need an experiment to make your point, you've already missed the point.

Wednesday, October 26, 2011

Lateral Glance

What makes a brain non-autistic? That would be an important question too.

Monday, October 24, 2011


You know, to be a scientist these days is to engage in a form of “mandatory hierarchies of processing.” I wonder why people can't choose to be an independent thinker instead.

Saturday, October 15, 2011

A Novel Approach to Ad Nauseam

Here is the abstract from the latest autism-genetics breakthrough paper (Casey 2011):

Autism spectrum disorder (ASD) is a highly heritable disorder of complex and heterogeneous aetiology. It is primarily characterized by altered cognitive ability including impaired language and communication skills and fundamental deficits in social reciprocity. Despite some notable successes in neuropsychiatric genetics, overall, the high heritability of ASD (~90%) remains poorly explained by common genetic risk variants. However, recent studies suggest that rare genomic variation, in particular copy number variation, may account for a significant proportion of the genetic basis of ASD. We present a large scale analysis to identify candidate genes which may contain low-frequency recessive variation contributing to ASD while taking into account the potential contribution of population differences to the genetic heterogeneity of ASD. Our strategy, homozygous haplotype (HH) mapping, aims to detect homozygous segments of identical haplotype structure that are shared at a higher frequency amongst ASD patients compared to parental controls. The analysis was performed on 1,402 Autism Genome Project trios genotyped for 1 million single nucleotide polymorphisms (SNPs). We identified 25 known and 1,218 novel ASD candidate genes in the discovery analysis including CADM2, ABHD14A, CHRFAM7A, GRIK2, GRM3, EPHA3, FGF10, KCND2, PDZK1, IMMP2L and FOXP2. Furthermore, 10 of the previously reported ASD genes and 300 of the novel candidates identified in the discovery analysis were replicated in an independent sample of 1,182 trios. Our results demonstrate that regions of HH are significantly enriched for previously reported ASD candidate genes and the observed association is independent of gene size (odds ratio 2.10). Our findings highlight the applicability of HH mapping in complex disorders such as ASD and offer an alternative approach to the analysis of genome-wide association data.

Allow me to interpret that abstract without the spin: Here are a hundred-some authors admitting that all their previous data-mining techniques have failed to find what they were looking for, but fear not, because they've discovered yet another data-mining technique that alas, also fails to find what they're looking for. But there are some secondary benefits—namely getting one's name attached to yet another massive-author publication—and from that perspective, all autism-related data-mining techniques seem to be equally effective.

(Casey 2011): Casey JP, Magalhaes T, Conroy JM, Regan R, Shah N, Anney R, Shields DC, Abrahams BS, Almeida J, Bacchelli E, Bailey AJ, Baird G, Battaglia A, Berney T, Bolshakova N, Bolton PF, Bourgeron T, Brennan S, Cali P, Correia C, Corsello C, Coutanche M, Dawson G, de Jonge M, Delorme R, Duketis E, Duque F, Estes A, Farrar P, Fernandez BA, Folstein SE, Foley S, Fombonne E, Freitag CM, Gilbert J, Gillberg C, Glessner JT, Green J, Guter SJ, Hakonarson H, Holt R, Hughes G, Hus V, Igliozzi R, Kim C, Klauck SM, Kolevzon A, Lamb JA, Leboyer M, Le Couteur A, Leventhal BL, Lord C, Lund SC, Maestrini E, Mantoulan C, Marshall CR, McConachie H, McDougle CJ, McGrath J, McMahon WM, Merikangas A, Miller J, Minopoli F, Mirza GK, Munson J, Nelson SF, Nygren G, Oliveira G, Pagnamenta AT, Papanikolaou K, Parr JR, Parrini B, Pickles A, Pinto D, Piven J, Posey DJ, Poustka A, Poustka F, Ragoussis J, Roge B, Rutter ML, Sequeira AF, Soorya L, Sousa I, Sykes N, Stoppioni V, Tancredi R, Tauber M, Thompson AP, Thomson S, Tsiantis J, Van Engeland H, Vincent JB, Volkmar F, Vorstman JA, Wallace S, Wang K, Wassink TH, White K, Wing K, Wittemeyer K, Yaspan BL, Zwaigenbaum L, Betancur C, Buxbaum JD, Cantor RM, Cook EH, Coon H, Cuccaro ML, Geschwind DH, Haines JL, Hallmayer J, Monaco AP, Nurnberger JI Jr, Pericak-Vance MA, Schellenberg GD, Scherer SW, Sutcliffe JS, Szatmari P, Vieland VJ, Wijsman EM, Green A, Gill M, Gallagher L, Vicente A, Ennis S. 2011. “A novel approach of homozygous haplotype sharing identifies candidate genes in autism spectrum disorder.” Human Genetics (in press).

Tuesday, October 11, 2011

Evidence Christ Was Autistic?

Here is the abstract from a recently published paper (Izuma 2011):

People act more prosocially when they know they are watched by others, an everyday observation borne out by studies from behavioral economics, social psychology, and cognitive neuroscience. This effect is thought to be mediated by the incentive to improve one's social reputation, a specific and possibly uniquely human motivation that depends on our ability to represent what other people think of us. Here we tested the hypothesis that social reputation effects are selectively impaired in autism, a developmental disorder characterized in part by impairments in reciprocal social interactions but whose underlying cognitive causes remain elusive. When asked to make real charitable donations in the presence or absence of an observer, matched healthy controls donated significantly more in the observer's presence than absence, replicating prior work. By contrast, people with high-functioning autism were not influenced by the presence of an observer at all in this task. However, both groups performed significantly better on a continuous performance task in the presence of an observer, suggesting intact general social facilitation in autism. The results argue that people with autism lack the ability to take into consideration what others think of them and provide further support for specialized neural systems mediating the effects of social reputation.

It's difficult to read that passage without being reminded of Matthew 6:1–4, from the Sermon on the Mount:

Take heed that ye do not your alms before men, to be seen of them: otherwise ye have no reward of your Father which is in heaven. Therefore when thou doest thine alms, do not sound a trumpet before thee, as the hypocrites do in the synagogues and in the streets, that they may have glory of men. Verily I say unto you, They have their reward. But when thou doest alms, let not thy left hand know what thy right hand doeth: That thine alms may be in secret: and thy Father which seeth in secret Himself shall reward thee openly.

What would Jesus do? Apparently not what non-autistics would do.

(Izuma 2011): Keise Izuma, Kenji Matsumoto, Colin F. Camerer, and Ralph Adolphs. 2011. “Insensitivity to social reputation in autism.” PNAS 2011: 1107038108v1-201107038.

Monday, October 10, 2011

Autistics Think Differently, Part 3

[Edit 02/11/2017: The final version of this essay can be found here.]

In “The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome?” (Soulières 2011), the following sentence appears:

“Autistics can maintain more veridical representations (e.g. representations closer to the actual information present in the environment) when performing high level, complex tasks.”

That's an intriguing statement, and I think I have a rough idea of what the authors are driving at (and I would generally agree with that rough idea). But the sentence as stated strikes me as slightly off key and a bit misleading. Let me see if I can explain.

I'm going to lay out a hypothetical visual scene (something similar could be done in the auditory domain if so desired), and once I've described the scene, I'm going to have three different entities survey it, including one that represents autistic perception and another that represents non-autistic perception. But when it comes to veridical representation, it's going to be the third entity that emerges as the clear winner in “capturing the actual information present in the environment.”

Here's the scene: It's a fairly open and sparse field, maybe in a large park or reserve. Near the front section of the field is a bench, where a woman and a girl are seated and talking. Rising behind them are four tall light poles, evenly spaced and situated so that they form a diagonal across the visual plane. The sky is mostly blue with a few nondescript clouds, and there is nothing else worthy of note.

Here are three entities viewing this scene from the same perspective, along with a rough description of what each perceives:

  • Entity 1 surveys this scene entirely as light and color stimulus. You can think of it as a pixelated view, where the perception of this scene is best described as a series of points, each point determined by its relative position and its light qualities, such as brightness and hue.
  • Entity 2 surveys this scene and is immediately drawn to the woman and girl on the bench. If asked about this perception, Entity 2 might say something like, “Yes, I can see the mother and daughter on the bench over there. See, the daughter is extremely upset—she's crying.” If asked about the light poles, Entity 2 might offer the observation that it's a good thing to have them here, because people can come at night without being afraid.
  • Entity 3 surveys this scene and is struck by the particular arrangement of the light poles. Entity 3 might note that there are four of them, or might point out that they are evenly spaced, or might remark on the angle they form in the visual plane. If asked about the people on the bench, Entity 3 might say they were noticed and there were two of them and they looked small beneath the light poles towering above them.

Although these descriptions are meant to highlight only general tendencies, I think most people would agree that the perceptions of Entity 1 closely match those of a camera, the perceptions of Entity 2 are fairly typical of a non-autistic person, and the perceptions of Entity 3 are more indicative of someone who might be on the autism spectrum. Each entity sees the exact same visual stimulus, but each extracts from that stimulus an entirely different set of information. That is the essence of what we mean when we talk about the concept perception.

In my way of thinking, Entity 1 has by far the most veridical representation here—it comes the closest to perceiving this visual scene as it truly is. The key to a camera creating an accurate visual representation is, ironically enough, not to do much of anything at all with it; in particular, not to impose any form upon the visual scene. No foregrounding. No backgrounding. No extracting of signal from noise. Just reproduce the visual scene as it visually is—that's all a camera is required to do.

By contrast, both Entity 2 and Entity 3 come to their particular perceptions by imposing some kind of structure on the raw visual stimulus, which is to say some elements in the visual scene form perceptual foreground while other elements fade mostly unnoticed into the background. The perceptual process is quite similar for both Entity 2 and Entity 3 (and quite different from the perceptual process of Entity 1). What distinguishes the perceptions of Entity 2 and Entity 3 is the material of the signal itself; that is, there is a categorical difference in what tends to foreground within the perceptions of Entity 2 and Entity 3.

It's not obvious yet that Entity 3's representations are more veridical than those of Entity 2, but let's keep exploring all these perceptions in more detail.

It's actually quite fortunate that Entity 1 is not a biological or responsive agent. If it were, having the most veridical representation of the visual scene would manifest as a huge liability. To be responsive to an environmental stimulus requires that information be extracted from it, exactly what Entity 1 cannot do. This has actually been a problem in the world of robotics, where despite having extremely accurate cameras, it has been nonetheless difficult to get machines to respond flexibly and constructively to various visual stimuli, precisely because it is difficult to get machines to recognize what constitutes the necessary foreground and what needs to be dismissed as inconsequential background.

Possessing a perfectly veridical representation of an environmental stimulus is tantamount to experiencing sensory chaos. Everything comes across as noise, nothing appears as signal. And with no signal, there is no information. And with no information, there is no ability to respond with purpose. We must keep in mind these thoughts about perfectly veridical representations and their corresponding sensory chaos, because when we later consider the perceptions of Entity 3 (autistic perception), we'll discover this very same concept comes into play.

First, however, let's talk in more detail about the perceptions of Entity 2 (non-autistic perception). It's my contention that what signifies and distinguishes non-autistic perception is its strong tendency to focus upon human-related events in the sensory environment; that is, it is the human-specific features that most commonly foreground within non-autistic perception, and it is this human-specific focus that provides the necessary structure for extracting signal from the sensory noise. You can see this at work in my depiction of Entity 2's perceptions, where the attention is drawn primarily to the people in the scene—and in a very detailed way—and even the elements which are not so apparently human-related are often tied back to humanity by some indirect means (the light poles, for instance, are comprehended as helping people see at night and not be afraid).

I wouldn't use the terms “global processing” or “hierarchies of processing” to describe this phenomenon, but I would consider such a term as “common thread” to suggest how these people-specific perceptions tie non-autistic cognition together into a cohesive package. And this works at more than just the individual level. Since nearly all humans share the characteristic of these people-specific perceptions, these perceptions serve to coalesce not just individual thoughts and behaviors but also the conventions and actions of the species as a whole. Humans formulate their shared species-specific perceptions into a series of cohesive thoughts, behaviors, conventions and environments, exactly as we might expect from a species-driven, biologically essential phenomenon.

Indeed, it is important to note that it's not just in humans that we observe this common form of species-specific perception. All across the animal kingdom, we can observe abundant evidence that creatures attend most strongly to the other members of their own species and to the species-related elements in their surroundings while most everything else in the sensory environment is ignored as inconsequential background. It is in this sense that I think the word “mandatory” comes into play. The common thread of a species-specific focus is extremely powerful, it has been forged through the long-burning furnace of evolutionary time. I think most organisms inherently rely upon this common thread, and would find it extremely difficult to step outside it.

In my previous post (Autistics Think Differently, Part 2) I complained that the authors of (Soulières 2011) have not provided an affirmative, distinguishing description for non-autistic perception and cognition. If it were up to me to provide that description, I would do it in much the same way I have here, highlighting the species-specific perceptions non-autistics share and that help cement the common cognitions and behaviors across that particular class of the population. I am of course open to criticisms of this idea and am willing to consider alternative suggestions, but so far I've not see much of anything forthcoming, anything beyond that is just a settling for describing non-autistic individuals as constituting the norm.

It is also my contention that what most fundamentally distinguishes autistic perception is that it lacks the species-specific focus that is characteristic of non-autistic perception. For reasons not yet clearly identified, autistic individuals do not tend to naturally foreground human-related elements within the sensory environment, and as a baseline, this would leave autistic individuals in much the same situation as that of Entity 1: autistics would perceive the raw environmental stimulus almost exactly as it is and would have no natural means of obtaining signal from the various aspects of that stimulus.

It's primarily in this sense that I think it's fair to say autistic individuals experience more veridical representations—representations that don't come with as many pre-imposed filters, such as the human-focused filters that get routinely and naturally applied within non-autistic perception. But this also means that the natural (beginning) state of autistic perception is one that comes dangerously close to sensory chaos, and I believe this goes a long ways towards explaining why autistic individuals tend to experience sensory difficulties, difficulties that vary in domain and range and seem to have no discernible physical cause. It also goes a long ways towards explaining the developmental difficulties autistic individuals experience in their early years, because both as an individual and as a member of a species that has built its environmental surroundings out of a shared perceptual experience, an autistic individual would find himself closed off from those species-shared experiences and all their coalescing and foregrounding effects. As was stated in the discussion for Entity 1, the possession of a perfectly veridical representation is actually a huge liability when it comes to acting as a biological or responsive agent.

The saving grace for autistics is that there are features within the sensory environment that seem to inherently foreground in the absence of any stronger means of perceptual organization. It's not entirely clear to me (logically or biologically) what causes these particular elements to form signal against an otherwise chaotic background, but we recognize these mostly non-biological features through such names as symmetry, repetition, pattern, mapping, structure, and form. The authors of (Soulières 2011) routinely invoke said features in describing the distinguishing characteristics of autistic perception and cognition [see for instance (Mottron 2009) and Principles 6 and 7 in (Mottron 2006)], and in my depiction of Entity 3's perception, you'll notice the emphasis being placed on such things as number, repetition, pattern, geometry and so on. And just as non-autistics will often apprehend non-biological features in their sensory environment through a referential connection to humanity, autistics will often reverse this process, apprehending humans through such things as number, categorization and measure. It seems to me that there is a good deal of evidence backing the idea that autistic perception is drawn primarily to those environmental features consisting of non-biological pattern, structure and form, and it is out of such features that autistics gain the majority of their perceptual foregrounding. Apprehension of non-biological environmental structure forms the backbone of an autistic individual's atypical means of overcoming sensory chaos, allowing that individual to respond productively as a biological agent.

One of the fascinating aspects of autistic perceptual foregrounding is that it can bring forth incredible variety and novelty. An autistic individual is apt to pull almost any kind of information from a sensory environment (it wouldn't have been all that surprising, for instance, if Entity 3 had ignored both the light poles and the people on the bench and had fixated instead on the harmonious colors in the clouds and sky). Because of the variety, novelty and non-biological (objective) nature of the information autistic individuals tend to gather, this too might be considered a valid reason for classifying autistic representations as more veridical (and I think the authors of (Soulières 2011) actually have something of this consideration in mind). But here I think we should be a bit more cautious. Once an autistic individual has actually foregrounded some aspect of the sensory environment, he has already moved far away from the realm of true veridical representation, is no longer perceiving reality anything at all like a camera. And no matter what structure is being applied to the environmental stimulus—be it the biological, species-driven form common to non-autistic perception, or the more pattern-based variety familiar to autistic perception—it can have equally valid potential to be a good or poor reflector of environmental reality. When we consider the entirety of human history, as well as the entirety of modern human society, I think it's fair to say that all types of human perceptual foregrounding are potentially informative and valuable, we wouldn't want to be deprived of hardly any of these perspectives.

In summary, I think there is merit and truth in the (Soulières 2011) claim that autistic individuals have a tendency to experience more veridical representations under many circumstances, but I wouldn't want to make that statement as obvious, simple fact. To come to that conclusion requires a deep understanding of both autistic perception and non-autistic perception, and in particular a deep understanding of what fundamentally distinguishes them. And this is just one more example of why I think it's inadequate to describe either autistic or non-autistic cognition as merely a deficit or norm. What's needed here are clear, affirmative, distinguishing descriptions—descriptions that enlighten us about both autism and non-autism, descriptions that bring out the essential value in each of these points of view.

(Soulières 2011): Soulières I, Dawson M, Gernsbacher MA, Mottron L, 2011 The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome? PLoS ONE 6(9): e25372. doi:10.1371/journal.pone.0025372

(Mottron 2009): Mottron L, Dawson M, Soulières I (2009) Enhanced perception in savant syndrome: patterns, structure and creativity. Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1385–1391.

(Mottron 2006): Mottron L, Dawson M, Soulières I, Hubert B, Burack J (2006) Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. Journal of Autism and Developmental Disorders 36: 27–43.

Sunday, October 9, 2011

Autistics Think Differently, Part 2

In the discussion section of “The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome?” (Soulières 2011), you will find the following paragraph:

“We have proposed that autistics' cognitive processes function in an atypically independent way, leading to “parallel, non-strategic integration of patterns across multiple levels and scales” and to versatility in cognitive processing. Such “independent thinking” suggests ways in which apparently specific or isolated abilities can co-exist with atypical but flexible, creative, and complex achievements. Across a wide range of tasks, including or perhaps especially in complex tasks, autistics do not experience to the same extent the typical loss or distortion of information that characterizes non-autistics' mandatory hierarchies of processing. Therefore, autistics can maintain more veridical representations (e.g. representations closer to the actual information present in the environment) when performing high level, complex tasks. The current results suggest that such a mechanism is also present in Asperger syndrome and therefore represents a commonality across the autistic spectrum. Given the opportunity, different subgroups of autistics may advantageously apply more independent thinking to different available aspects of information: verbal information, by persons whose specific diagnosis is Asperger's, and perceptual information, by persons whose specific diagnosis is autism.”

That's a lovely paragraph in many respects, highlighting several of the strengths in these authors' approach to describing (affirmatively) the characteristics of autistic perception and cognition. It's practically all there: atypical information processing, independent thinking, patterns, versatility, creativity, veridical representations of complex stimuli, specfic/isolated abilities, and so on. With nary a reference to a neural construct, this paragraph manages nonetheless to capture a broad, descriptive range of autistic cognitive concepts, all translating well into observable and distinctive autistic behaviors. While the rest of the autism research community continues to focus on describing autistic perception and cognition as deficient relative to their non-autistic counterparts, the authors of (Soulières 2011) make an informative rebuttal simply by providing consistent evidence and detailed descriptions outlining what autistic perception and cognition actually are, as opposed to obsessing on what they are not.

On the other hand, the above paragraph also highlights the glaring weakness in these authors' approach. Here we have yet one more time from these authors an instance of an extremely detailed, extremely descriptive, extremely compelling description of autistic perception and cognition that goes entirely unaccompanied by a corresponding description for non-autistic perception and cognition.

In the above paragraph, the description of non-autistic perception and cognition is reduced to a single phrase: “non-autistics' mandatory hierarchies of processing.” It's bad enough that this description is so sparse, but what makes it worse is that there is no clear indication of what that particular phrase is supposed to mean—it actually sounds empty to me. When I read it, the questions that pop into my head are: What is the nature of these hierarchies? Are there examples? What's at their root? What's at their leaves? What is it about these mysterious hierarchies that make them mandatory? And above all else, how does any of this shed light on non-autistic perception, cognition, and behavior? Listen, I may be a very poor reader, but I honestly can't find an adequate answer to those questions in any of these authors' work.

If you follow the reference that is attached to the phrase “non-autistics' mandatory hierarchies of processing,” you'll be taken to (Soulières 2009). Unfortunately that doesn't help very much, because (Soulières 2009) simply repeats the terminology, not explaining it in any greater detail than does (Soulières 2011). I suppose a subtle interpretation that could be made from this reference to (Soulières 2009) is that the authors are intending the phrase to be taken neuronally: that is, “mandatory hierarchies of processing” is meant to invoke a particular form of non-autistic brain structure and organization. I would note, however, that even if this interpretation is valid, it would only be a theory, not an observation or a description. I would also note it wouldn't let the authors off the hook: these authors have provided similar theories about autistic brain structure and organization, but that has never stopped them from augmenting such theories with a broad array of non-neuronal depictions of autistic perception and cognition—precisely the type of supplementary explanation that is conspicuously absent on the non-autistic side of the equation.

Perhaps a more helpful reference for the phrase would have been (Mottron 2006), which actually does discuss global hierarchical processing in some detail (albeit a bit haphazardly). There you'll find an assortment of loose explanations for the term, including some of which are more neuronal in nature (as was perhaps being suggested through the reference to (Soulières 2009)). But more commonly in (Mottron 2006), global hierarchical processing is discussed in terms of relative autistic/non-autistic performances on various laboratory tasks, with the conclusion from this analysis being summed up succinctly in the statement of Principle 5: higher-order processing is optional in autism and mandatory in non-autistics. Which is to say, the authors are suggesting that autistics can process all the information that non-autistics can, and on top of that autistics can process more. Which is to say, the authors are suggesting non-autistics can process less. Which is to say, the authors are suggesting non-autistic cognition is deficient relative to autistic cognitive processing. Which is to say, the apparent purpose of using the phrase “non-autistics' mandatory hierarchies of processing” is to lead the reader back to what must undoubtedly be these authors' greatest autism research discovery so far, namely the deficit-based model of non-autism! (Maybe they'll win a Nobel Prize for that.)

Well really, what am I supposed to think? Look at the above quoted paragraph and consider what it must imply, in the absence of other information, about non-autistic perception and cognition. Autistic thinking is independent. So non-autistic thinking is not? Autistic thinking is flexible. So non-autistic thinking is not? Autistic thinking is creative. So non-autistic thinking is not? Autistic thinking is veridical. So non-autistic thinking is not? After having convincingly rebutted the research community's nonstop reliance on deficit-based models of autism through their use of informative, detailed, affirmative descriptions of autistic perception and cognition, these authors then make the tragic mistake of turning right around and presenting non-autistic perception and cognition as nothing more than a series of unflattering comparisons to autistic perception and cognition. Heck, if I were to rely on what these authors have implied so far about non-autistic cognitive abilities, I would have to conclude early intervention wouldn't even be worth the bother for this population, we might as well just dispatch the poor non-autistic souls straight off to the institutions.

Or on the other hand, we could just go back to assuming I'm a very poor reader. Here is what I would propose to set me straight:

Start with a clean sheet of paper. Head one column with “Autistic Perception and Cognition.” Head the other with “Non-autistic Perception and Cognition.” Fill in the columns with the descriptions and observations from the authors of (Soulières 2011). I could do the autistic column myself—the material for it is impressively abundant. The above quoted paragraph would be a good start and (Mottron 2009) is practically an entire tone poem in description of autistic perception and cognition. Throw in a couple dubious sentences about local processing and related neuronal theories, and the column is done. But for me, the non-autistic column continues to look mostly blank. Outside some hypothetical musings about non-autistic brain structure and organization, all I can think of to add would be the phrases “global processing” and “non-autistics' mandatory hierarchies of processing,” and of course I would have to put an asterisk next to those terms, because I really don't have a clue as to what they're supposed to mean. Now if someone—the authors, anyone—wants to come along and fill in the rest of the column for me, show me what I've been missing, I would be happy to give them thanks, apologize for the trouble, and go on about my sheepish way. But if on the other hand the non-autistic column can't be adequately filled in—not in the same way that the autistic column can—then my complaint remains legitimate and unaddressed, and I will continue to be vocal about it.

As much as I like and generally agree with the authors' descriptions of autistic perception and cognition, I also believe those descriptions fall flat when there is nothing to contrast them against. And on top of that, I'm also of the firm belief that non-autistic perception and cognition has an affirmative, distinctive richness all its own, a richness that possesses deep inner logic, a richness that has been biologically essential, and a richness that provides ongoing and immense value to the entire human population. It's long past time for that richness to be acknowledged and described.

(Soulières 2011): Soulières I, Dawson M, Gernsbacher MA, Mottron L, 2011 The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome? PLoS ONE 6(9): e25372. doi:10.1371/journal.pone.0025372

(Soulières 2009): Soulières I, Dawson M, Samson F, Barbeau EB, Sahyoun CP, et al. (2009) Enhanced visual processing contributes to matrix reasoning in autism. Human Brain Mapping 30: 4082–4107.

(Mottron 2009): Mottron L, Dawson M, Soulières I (2009) Enhanced perception in savant syndrome: patterns, structure and creativity. Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1385–1391.

(Mottron 2006): Mottron L, Dawson M, Soulières I, Hubert B, Burack J (2006) Enhanced perceptual functioning in autism: an update, and eight principles of autistic perception. Journal of Autism and Developmental Disorders 36: 27–43.

Friday, October 7, 2011

Autistics Think Differently, Part 1

I would encourage everyone to read the recently published paper “The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome?” (Soulières 2011). It appears in the online journal PLoS One (which happily is open access) and although there appears to have been a glitch with the original publication of the paper (wording changes that were made without the permission of the authors), all seems to be rectified now.

I would also encourage a quick glance at this paper's predecessors, (Dawson 2007) and (Bolte 2009). Although there are some subtle variations among these studies, the general theme remains quite consistent, namely that individuals across the autism spectrum evince a markedly different cognitive profile compared to their non-autistic peers, a profile that consistently reveals many affirmative signs of intellectual capacity while also displaying a significant atypicality in the presentation of that capacity.

This is an important finding not just in the world of autism research but also in the world of intelligence research. Intelligence researchers routinely compare individuals by their intelligence performance (as measured by the standard battery of intelligence tests), and these comparisons have been shown to have predictive value in the real world. But until now, such differences have been characterized almost entirely by reference to the level of individual intelligence. With (Soulières 2011), (Dawson 2007), and (Bolte 2009), researchers have been given a clear-cut instance in which individual intelligence differences are more meaningfully characterized by reference to the type of intelligence being displayed. In other words, in the titles of (Dawson 2007) and (Soulières 2011) it is the word “nature” that needs to be emphasized—autistic individuals are displaying an entirely different kind of perception and cognition. That the level of autistic intelligence is also being underestimated is due primarily to the fact researchers routinely assume autistic cognition is little more than a damaged version of its non-autistic counterpart. (Soulières 2011) is helping put another nail in the coffin of that assumption.

(Soulières 2011): Soulières I, Dawson M, Gernsbacher MA, Mottron L, 2011 The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome? PLoS ONE 6(9): e25372. doi:10.1371/journal.pone.0025372

(Dawson 2007): Dawson M, Soulières I, Gernsbacher MA, Mottron L (2007) The level and nature of autistic intelligence. Psychological Science 18: 657–662.

(Bolte 2009): Bolte S, Dziobek I, Poustka F (2009) Brief report: The level and nature of autistic intelligence revisited. Journal of Autism and Developmental Disorders 39: 678–682.

Wednesday, September 28, 2011

The Mathematics of Robison's Males

I'm not sure if anyone has remarked on this yet (someone should have by now—it's pretty obvious), but John Elder Robison's observation that most families with an autistic child have more boys than girls is almost certainly true. It's not a mystery, it's simple mathematics.

Start with the assumption that on average every newborn has a 50% chance of being male and a 50% chance of being female. Then add the assumption that on average every autistic newborn has an 80% chance of being male and a 20% chance of being female. In the absence of other information, the following two scenarios will emerge as mathematical results:

  • Given that a family has at least one autistic child, the family will likely have more males than females.
  • Given that a family has no autistic child, the family will likely have more females than males.

The effect is more pronounced (and thus more readily observable) in the first scenario. But this gets balanced by the fact that the second scenario is far more common.

Note that when we ask questions such as what is the male/female ratio in families given that at least one child has autism, we are in the land of conditional probabilities. A 50/50 male/female ratio is fine if there are no other conditions, but by saying that the family has (or does not have) a child with autism, you are creating conditional probabilities, and in this case those conditional probabilities do not come out to be 50/50.

It's easiest to see the phenomenon in families that have at least one child with autism. Some of those families will have only one child. In that case, we know that the one child has autism and so 80% of those families will have more males than females and 20% of those families will have more females than males.

With two children, we have two cases to consider. In one case, one of the children has autism (80% male/20% female) and one does not (roughly 49.7% male/50.3% female). Put those odds together and you will find that about 39.8% of such families have more males than females, 50.1% have equal sex distribution, and 10.1% have more females than males. The other case is that both children have autism, and in this case it comes out as 64% have more males than females, 32% have equal sex distribution, and 4% have more females than males.

For families with more than two children with at least one of them having autism, the calculations will be similar with similar results: the odds will always be greater that the family has more male children than female children. Again, there is nothing mysterious about this outcome, it's just simple mathematics.

In contrast to all these scenarios, we will have the situations where the family does not have a child with autism. In all those instances, the results will skew slightly towards having more females in the family than males. The numbers aren't as dramatic in these scenarios, but of course we have to keep in mind that these families are far more common than the families who have at least one autistic child.

This example is a good reminder that science is based on a foundation of logic and mathematics. I see an awful lot of scientific work that appears to be fine as far as the scientific technique goes but is utterly abysmal when it comes to the underlying logic and mathematics. In other words, even if you do RCTs out the ass, if the work is still based on lousy logic and lousy mathematics, then the science is going to be lousy too.

Sunday, September 25, 2011

Isomorphic Redundancy

If the material form of intelligence exists inside our neurons, what would be its structure? Well, take a good look around you. How would the neuronal structure of intelligence be any different than the structure you experience each and every day?

Saturday, September 24, 2011

The Popular History of Intelligence Gains

The Flynn effect began just before humans started administering intelligence tests, and ended right after the Flynn effect's discovery—surely any hunter-gatherer would have been able to tell you that.

Sunday, September 18, 2011

Neural Damage

The correct distinction is between that of intelligent behavior, which is a biological response, and intelligence itself, which is a material artifact. Cognitive scientists confuse the two concepts constantly, and in particular insist each must be generated from inside the human brain. That misfiring has spawned a tremendous paralysis.

Thursday, September 15, 2011

Flynn Effect 104

A paper recently published in the journal Intelligence, “The Flynn effect puzzle: A 30-year examination from the right tail of the ability distribution provides some missing pieces” (Wai and Putallaz 2011), serves a valuable purpose—it makes it more difficult for scientists to dismiss the Flynn effect.

I think most scientists would agree that the Flynn effect—the persistent, well-documented rise in raw intelligence scores across nearly every measured population—is a phenomenon that has yet to be explained very well; the words “mystery” and “puzzle” frequently get bandied about in association with the Flynn effect. Perhaps uncomfortable with the idea of a mystery or a puzzle sitting right in the middle of all the psychometric progress they’ve been making in recent years, some cognitive scientists seem content to mostly dismiss the Flynn effect, describing it as little more than a twentieth-century anomaly, driven perhaps by some underprivileged populations catching up to the norm and raising the mean, a process that may have already run its course, thereby allowing the Flynn effect to calm down and politely go away. The slightly understated suggestion is, why bother to explain a blip?

But (Wai and Putallaz 2011) shouts a great big No! to all that. (Wai and Putallaz 2011) provides evidence that the Flynn effect holds just as strongly among those of high intelligence and advanced intellectual ability as it does across the remainder of the population, and (Wai and Putallaz 2011) also provides evidence that, at least for the population being studied, the Flynn effect continues to hum along at a significant pace. The unmistakable conclusion from this new study is that the Flynn effect appears to be both ubiquitous and relentless; it shows no evidence of being anything like a blip.

I can't say I'm surprised by that conclusion. By my reckoning, the Flynn effect has been both ubiquitous and relentless for at least the last fifty thousand years.

In considering explanations for the Flynn effect that would possess the necessary characteristics of being both widespread and ongoing, some discussions surrounding (Wai and Putallaz 2011) highlight the impact of an increasing level of environmental complexity. This notion is generally described as a set of fast-paced, highly structured items the population (children in particular) gets routinely exposed to and that helps individuals acquire the kinds of skills that translate well to problem-solving abilities on intelligence tests. The two most commonly cited examples are the increased use of video games and the increasingly complex and multivariate nature of movie and television show plots.

Highlighting increased environmental complexity is certainly a step in the right direction, but it is also clear from the context of these discussions that what’s being made is a small and tentative step, one not quite sure of where it's going. Therefore let me take this opportunity to convince the tentative proponents of environmental complexity that they can take a much larger step, and take it with a good deal more confidence.

The first problem I see with discussions surrounding environmental complexity is that they tend to focus on things in the environment instead of considering the impact of the environment as a whole. No matter what thing or set of things is being contemplated—including video games and entertainment plots—one immediately realizes huge swaths of the population never get exposed to that particular thing or set of things, and yet they too fall under the full sway of the Flynn effect. Indeed, the Flynn effect was working its magic long before there even were video games and television sets, and the Flynn effect continues to be prominent in locales where video games and sophisticated dramas have yet to take much hold. Suggesting alternative candidates for environmental complexity would fail to solve the problem as well: anything (any thing) we might mention is going to betray the same weakness, is going to be not universal or timeless enough to explain the Flynn effect's widespread and nonstop power.

But the repair to this problem is actually quite simple: just put all the things together.

From the beginning of man's great leap forward and across nearly the entirety of this planet's surface, the types of environmental complexity have varied greatly from time to time and from place to place, but the one consistent observation that can be made about environmental complexity is that no matter what time and no matter what place is under consideration, the total amount of environmental complexity is nearly always on the rise. Man has been accumulating an ever-increasing supply of non-biological pattern, structure and form into his surroundings, has been absorbing these surroundings and responding constructively to them, and through these means has exhibited an increasingly sophisticated set of behaviors. As the hypothetical example in Flynn Effect 101 demonstrates, the total quantification of environmental intelligence is all that is needed to drive population-wide raw intelligence gains. No particular piece of environmental complexity need ever be mentioned; it is the accumulative impact of environmental complexity that produces the Flynn effect.

The second problem I see with discussions surrounding environmental complexity is that everyone feels compelled to translate environmental complexity into a lasting physical impression upon the human brain. Playing video games, for instance, is seen as expanding the capacity of working memory; modern movie plots form a larger number of connections within our logical neural circuitry. It's as though environmental complexity is useless as it is; its only purpose is to prompt restructuring inside our neurons, spawn massive rewiring between our ears. This notion is certainly scientifically popular, but it also betrays an extremely poor conception of biology.

At bottom, the human neural system is simply the biological means of stimulus and response, just as it is for all the other animals, just as it was for early Homo sapiens. The human brain can produce behavior, including intelligent behavior, but it cannot store quantities of content, including the content of intelligence. There are three compelling reasons why the material form of human intelligence will not be found inside the human skull:

  1. There is no concrete evidence for it. Although neuroscience has certainly produced a prodigious amount of data, statistics and pictures, it has produced not even the first step towards describing how connected sets of neurons produce anything from a simple hello to the theory of relativity. That description is yet only a distant hope, certainly not an accomplishment.
  2. It would require an evolutionary miracle. Nearly everything man counts as intelligent behavior (nearly everything man measures on intelligence tests) had its human origin within only the last several thousand years. Early Homo sapiens—biological equivalents to ourselves—displayed almost nothing of what we currently describe as intelligent behavior. Thus if the human brain is to be conceived of as both the physical source and the physical location of human intelligence, then the current structure of the human brain must have sprung up all at once population wide, in violation of everything we know about biological evolution.
  3. It would be utterly superfluous. Every aspect of human intelligence can be described by appealing to something material within the human environment. Language, mathematics, logic—every feature of human intelligence has a tangible and lasting form within the physical surroundings, form that thoroughly defines the feature. To repeat that tangible and lasting form within the structures of the human brain would be, to put it mildly, excessively redundant.

The desire to translate the effects of environmental complexity into physical impacts upon the human brain is nothing but the consequence of a scientific prejudice—a prejudice without evidence, likelihood, or need.

In summary, those who are proponents of increasing environmental complexity as an explanation for the Flynn effect are certainly on the right track, but to arrive at their destination, they must learn to be more bold. In particular:

  • They must stop emphasizing things or sets of things when characterizing environmental complexity. It is the total amount of environmental complexity—that is to say, the landscape-wide accumulation of environmental intelligence—that has been the continuous force driving the Flynn effect.
  • They must accept that the combined elements of environmental complexity are quite literally the material form of human intelligence itself. They must quit looking for intelligence within our neurons, they must not stop expecting to find intelligence inside our heads.

(Wai and Putallaz 2011): Wai, Jonathan; Putallaz, Martha. 2011. “The Flynn effect puzzle: A 30-year examination from the right tail of the ability distribution provides some missing pieces.” Intelligence (in press).

Wednesday, September 14, 2011

Flynn Effect 103

James Flynn, in his book What is Intelligence (Flynn 2007), lists several paradoxes he associates with intelligence and the Flynn effect. One of his paradoxes (labeled as the identical twins paradox) is described by Flynn as follows:

“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?”

In a previous post, Flynn Effect 101, I provide a description of human intelligence that portrays it as consisting of two separate aspects. The first aspect is labeled as environmental intelligence, and is defined to be the amount of non-biological pattern, structure and form tangibly contained within a given physical environment. The second aspect is labeled as individual intelligence, comparable to what gets measured on intelligence tests and defined as an individual's ability to absorb environmental intelligence and respond constructively to it.

One of the merits of this description is that it resolves Flynn's identical twins paradox, by showing that it is not paradoxical at all.

Here is a short summary of the hypothetical example created in Flynn Effect 101 to demonstrate the concepts of environmental intelligence and individual intelligence:

At time 1, the environment contains a certain amount of non-biological pattern, structure and form (environmental intelligence) measured as 200 ei. [Both the number and unit are arbitrary. For the example, all that's required is that the quantified amount be less than it will be at time 2.] Also at time 1, there are three individuals named A1, B1, and C1, who based upon their population-normed performances on intelligence tests display the following set of intelligence results:

Time 1 Environmental Intelligence: 200 ei.

Time 1 Test Performance:

  • A1: 80% of environmental intelligence mastered (high intelligence)
  • B1: 70% of environmental intelligence mastered (medium intelligence)
  • C1: 60% of environmental intelligence mastered (low intelligence)

Time 1 Absolute (Raw) Intelligence Score:

  • A1: 160 ei (80% times 200 ei)
  • B1: 140 ei (70% times 200 ei)
  • C1: 120 ei (60% times 200 ei)

At time 2—assumed to be several generations later—the amount of non-biological pattern, structure and form located in the human environment has increased to the point where it can now be measured at 400 ei. A2, B2 and C2, identified as biologically equivalent descendents of A1, B1, and C1, perform as follows on the time 2 intelligence tests:

Time 2 Environmental Intelligence: 400 ei.

Time 2 Test Performance:

  • A2: 80% of environmental intelligence mastered (high intelligence)
  • B2: 70% of environmental intelligence mastered (medium intelligence)
  • C2: 60% of environmental intelligence mastered (low intelligence)

Time 2 Absolute (Raw) Intelligence Score:

  • A2: 320 ei (80% times 400 ei)
  • B2: 280 ei (70% times 400 ei)
  • C2: 240 ei (60% times 400 ei)

At any given point in time—whether it be time 1, time 2, or any other we might consider—it is individual intelligence, in the form of relative intelligence scores, that dominates the psychometric landscape. Using point-in-time scores alone, cognitive scientists can gather a wealth of information about relative human intelligence, including such statistics as g factor analysis and correlations to academic and career achievement. And as Flynn has rightly noted, in the real world these statistics lead to the inevitable conclusion that individual intelligence differences are produced primarily by genetically-driven biological forces. At any given point in time, environment scarcely gets to play a role at all; it is, as Flynn suggests, utterly feeble.

Furthermore, individual intelligence differences, as seen in both the hypothetical example and in real-world numbers, remain extremely constant over time, precisely as we might expect for a phenomenon being produced from a biological source. The human biological form—genetics, neurons, and all—does not rapidly transform over a stretch of time. And indeed, we have every reason to expect that the human biological factors determining individual intelligence would have been working in much the same way as they do now from as early as tens of thousands of years ago. When it comes to individual intelligence, biology is more than just potent—it is constant and dominant.

And yet … over time, while human biology has remained constant, the absolute level of human intelligence has relentlessly and ubiquitously continued to grow.

Cognitive scientists immediately go off track by trying to attribute this phenomenon known as the Flynn effect to those same biological components underlying individual intelligence, when in fact there is no logically compelling reason to do so. It is only a prejudice that convinces scientists to believe that the material form of human intelligence must be generated inside the human brain. In the hypothetical example above, it is clearly not a biological force that is generating the across-time raw intelligence advances. The increase in absolute levels of intelligence from time 1 to time 2 are driven solely by the increase in environmental intelligence, by the increase in non-biological pattern, structure and form tangibly contained within the physical environment.

The phenomenon of environmental intelligence is not widely recognized, but one can hardly say this is because the phenomenon is unobservable. Just a cursory glance at human history from the time of man's great leap forward, as well as a cursory glance at the rapidly transforming human landscape, should be more than adequate to convince even the greatest skeptic that there has been an ever-increasing amount of non-biological pattern, structure and form being continuously introduced into the human environment. Neural biology is constant—the constant ability to absorb pattern, structure and form—but that constant ability encounters an ever-expanding target. Over time, it is the environment—the entire environment—that becomes the dominant factor driving massive, population-wide intelligence gains. Biology, including genetics, scarcely gets to play a role at all; it is, in this domain, utterly feeble.

There is a simple analogy for the relationship between individual intelligence and environmental intelligence. Think of some ships in a harbor, including lightweight ships such as pleasure craft and heavy duty ships such as battle cruisers. We can take a measurement (relative to a fixed vertical point on land) of the bottommost part of each ship and we'll discover that there are some significant differences. Some ships will sit higher, some will sit lower. Through further analysis of such characteristics as geometric structure and material density (equivalent to psychometric analysis), we can identify ship-based factors that determine relative vertical positions in the water, exactly as we presently determine factors such as genetics that seem to drive intelligence differences among individuals.

If we then move forwards in time, we might discover that a comprehensive measurement reveals all the same relative differences in bottommost positions of the ships (with the same factors driving these relative differences), and yet we also discover that the absolute position of each ship has risen by a significantly large amount. If we make the mistake of trying to explain this increase by appealing to the characteristics of the ships themselves, we will get nowhere; the characteristics of the ships can help explain their relative vertical positions, but they do nothing for explaining their change in absolute vertical position. For that, we must turn to the context (the environment) of these ships, and in this case it is of course the water that is serving (literally) as the rising tide that raises all ships.

Environmental intelligence, the amount of non-biological pattern, structure and form tangibly contained within the physical environment, serves as the universal context of human intelligence—the relentless, ubiquitous increase in environmental intelligence has been the dominant factor driving the Flynn effect. Individual intelligence, defined as the ability to absorb environmental intelligence and respond constructively to it, remains stable in the human population over time, as is to be expected from a genetically-determined, biologically-driven skill. Both components are potent, each in its own domain; and viewed in this light, Flynn's identical twins paradox swiftly disappears.

(Flynn 2007): Flynn, James R. 2007. What Is Intelligence?: Beyond the Flynn Effect. Cambridge: Cambridge University Press.

Tuesday, September 13, 2011

Flynn Effect 102

James Flynn, in his book What is Intelligence (Flynn 2007), lists several paradoxes he associates with intelligence and the Flynn effect. Two of these paradoxes (labeled as the intelligence paradox and the mental retardation paradox) deal with the problem that people from one time period appear to be too implausibly dumb or too implausibly smart compared to people from a different time period. Flynn describes these paradoxes as follows:

“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 each of these paradoxes is to recognize that Flynn is mixing up two different aspects of intelligence; he is mixing up environmental intelligence and individual intelligence. In particular, he is using changed rankings in one aspect (environmental intelligence) to infer that there must be a corresponding change in rankings for the other aspect (individual intelligence). It simply does not work that way.

In my previous post, Flynn Effect 101, I created a hypothetical example to demonstrate the concepts of environmental intelligence and individual intelligence, and to show how these concepts can produce the kinds of intelligence differences we routinely observe in individuals while at the same time producing the Flynn effect. Here is a short summary of that example:

At time 1, the environment contains a certain amount of non-biological pattern, structure and form (environmental intelligence) measured as 200 ei. [Both the number and unit are arbitrary. For the example, all that's required is that the quantified amount be less than it will be at time 2.] Also at time 1, there are three individuals named A1, B1, and C1 who based upon their population-normed performances on intelligence tests display the following set of intelligence results:

Time 1 Environmental Intelligence: 200 ei.

Time 1 Test Performance:

  • A1: 80% of environmental intelligence mastered (high intelligence)
  • B1: 70% of environmental intelligence mastered (medium intelligence)
  • C1: 60% of environmental intelligence mastered (low intelligence)

Time 1 Absolute (Raw) Intelligence Score:

  • A1: 160 ei (80% times 200 ei)
  • B1: 140 ei (70% times 200 ei)
  • C1: 120 ei (60% times 200 ei)

At time 2—assumed to be several generations later—the amount of non-biological pattern, structure and form located in the human environment has increased to the point where it can now be measured at 400 ei. A2, B2 and C2, identified as biologically equivalent descendents of A1, B1, and C1, perform as follows on the time 2 intelligence tests:

Time 2 Environmental Intelligence: 400 ei.

Time 2 Test Performance:

  • A2: 80% of environmental intelligence mastered (high intelligence)
  • B2: 70% of environmental intelligence mastered (medium intelligence)
  • C2: 60% of environmental intelligence mastered (low intelligence)

Time 2 Absolute (Raw) Intelligence Score:

  • A2: 320 ei (80% times 400 ei)
  • B2: 280 ei (70% times 400 ei)
  • C2: 240 ei (60% times 400 ei)

Let's begin by concentrating on the individual named A1. At time 1, A1 is assessed to be a smart person. He is able to absorb and constructively respond to about 80% of the environmental intelligence to be found around him, an ability he demonstrates through his performance on intelligence tests—tests which are composed out of representative examples from that same environmental intelligence. And as A1 navigates through his time 1 world, we can expect relatively high achievement compared to that of, for instance, B1 and C1.

But if we take A1's absolute (raw) intelligence score of 160 ei and compare it to those who live in the time 2 world, A1 suddenly doesn't look so smart. 160 ei is well below the 240 ei score of C2, a person assessed to be of low intelligence at time 2. If 240 ei is considered to be of low intelligence at time 2, then A1's score of 160 ei must make him out to be a borderline imbecile.

So which is it, is A1 a smart person or an imbecile? The answer is that A1 is a smart person, no matter what time period is being considered.

It is a logical mistake to think that A1's relatively low score of 160 ei at time 2 has anything to do with A1's intellectual ability, has anything to do with A1's individual intelligence. A1 has the capacity to absorb around 80% of the environmental intelligence around him, no matter what that environment might happen to be, and so if we could magically transport A1 forward in time and raise him in a time 2 world, he would absorb and constructively respond to about 80% of the time 2 environment. That means he would score at an absolute level of 320 ei, which at time 2 is pretty darn good. A1's score of 160 ei at time 1 is produced not just by his individual ability, it is produced also in concert with the amount of environmental intelligence existing at that time. His raw score of 160 ei looks low at time 2 solely due to the fact that the level of environmental intelligence has so dramatically increased from time 1 to time 2.

If we concentrate on the individual named C2 and contemplate going backwards in time, we'll witness the same phenomenon at work in reverse. At time 2, C2's absolute (raw) score of 240 ei puts him in the category of low intelligence, suggesting life circumstances that can be expected to be more difficult than for his cohorts A2 and B2. But if we take C2's score of 240 ei and place it in the time 1 context, C2 suddenly comes across as a Mensa candidate, and we might be wondering if C2 could have been a high achiever, if only he had had the good fortune to be born at time 1.

So which is it, is C2 a person of low intelligence or a Mensa candidate? The answer is that C2 is a person of low intelligence, no matter what time period is being considered.

Just as in the analysis of A1's circumstances, the relative positioning of C2's raw intelligence score at times 2 and 1 has nothing to do with C2's intellectual ability (individual intelligence). It is solely a consequence of the difference in environmental intelligence between the two time periods. If C2 could be magically sent back and raised at time 1, he would score at an absolute level of 120 ei and be judged once again to have low intelligence. The timing of his birth cannot alter his personal intellectual ability.

Allow me to tackle a non-hypothetical example that Flynn mentions in his book: the learning of baseball rules. After noting that the average raw intelligence score from about 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 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 perfect example of taking a change in raw intelligence scores as evidence for a change in individual intelligence, when in fact it is evidence for a change in environmental intelligence.

Think about putting some questions dealing with baseball rules on an intelligence test. If you had put those questions on a test administered in 1800, absolutely no one would have answered the questions correctly, including the smartest people who then lived. On the other hand, if you had put such questions on a test administered in 2000, lots of folks, including those of average intelligence, would have been able to answer the questions correctly—by then baseball and its rules had become an established part of the human environment. Following Flynn's anecdote, you would have had to drop down to those with an IQ of about 75 before you might have consistently expected to get wrong answers on basic baseball questions in the year 2000.

So does this imply that the smartest people in 1800 had about the same intellectual capacity as Jensen's young man? Flynn would like us to ponder that question as though it were a genuine paradox, but in fact the question itself is based upon nothing but a logical fallacy.

If you had placed problems about baseball rules on an intelligence test administered in 1900, you would have gotten a mixed set of results. Some people would have been able to solve them correctly, but others would not, including folks of otherwise average to high intelligence, since baseball had not yet become widely entrenched in the human environment (in 1900, the game was just then catching on). But let's take one of those folks of average to high intelligence who had just failed to solve the problems correctly (because of having never been exposed to baseball), and let's buy that person a ticket and take him to the ballpark. Let's sit with him in the grandstands, go over some of the basic rules, offer him a scorecard and pencil. How do we expect him to perform? Keep in mind that this is a person of average to high intelligence, he has the capacity to absorb quite a bit from what the environment has to offer, there is really nothing amiss in his individual intellectual ability. He will pick up the rules of the game quite easily, he will hardly give them a second thought. And around 1900, this is a scene that would have been repeating itself again and again and again.

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) and it has nothing to do with individual intellectual abilities.

Allow me just one more example, this one dear to both Professor Flynn and myself. Allow me to consider the intelligence of an ancient Greek named Aristotle.

If an intelligence test appropriate for the classical Athenian environment had been administered to Aristotle, I'm certain he would have scored quite well on it, likely among the top tier of his fellow citizens; Aristotle left behind ample evidence he could absorb a very large portion of the classical Athenian surroundings. But if on the other hand the equivalent of a modern intelligence test had been administered to Aristotle, he would have found nearly all its questions to be utterly incomprehensible. Aristotle had no notion of the number 0. His arithmetic and algebraic skills were crude at best. He had been speaking a language with a more limited vocabulary and simpler grammar than our own. (Not to mention, his baseball knowledge was completely nil.) Almost the entirety of a modern intelligence test would have looked unfamiliar to Aristotle, and his raw score by necessity would have registered exceptionally low.

Does this mean Aristotle was intellectually inferior to us? As I hope the reader must realize by now, it does not mean that at all.

Aristotle's low raw intelligence score on a modern intelligence exam would be driven entirely by the difference in amounts of environmental intelligence existing between classical Athens and the modern world. There would be nothing defective about Aristotle's individual intelligence, and indeed if we could magically transport Aristotle forwards in time and raise him in modern Athens, his ample intellectual capacity would no doubt allow him to absorb as much of the current intellectual environment as do the smartest among us. Dare I say, he might even blossom into a modern-day genius.

As all these examples demonstrate, the Flynn effect is being driven entirely by the ever-increasing amount of non-biological pattern, structure and form existing within the human environment. The Flynn effect has everything to do with environmental intelligence and has nothing to do with individual intelligence. In particular, the Flynn effect has nothing to do with our biological intellectual capacity. It has nothing to do with the neurons inside our head.

(Flynn 2007): Flynn, James R. 2007. What Is Intelligence?: Beyond the Flynn Effect. Cambridge: Cambridge University Press.