intelligence

IQ,SES and heritability

ResearchBlogging.org
A reader of this blog wrote to me recently regarding a series of posts I have written regarding IQ,SES and heritability, and I thought it would be good to share the comments with the rest of the mouse trap community and to delineate my position on the matter (and what I believe the studies show the relation is). First I’ll like to quote extensively from the comment (private mail):

Let me start with my belief that I think we share the same idea about IQ and its origins.
That is, genetics endows each and every person with a maximum IQ that can be achieved if and only if the environment is perfect for the development of this IQ.

Agreed!

Consequently, in identical twins, environment is the only cause of differences in IQ; IQ differences between persons that are not identical twins must be related to both environment and genes.

Sibling-correlation-422There are subtle nuances here. (no I’m not being pedantic, the importance of these will become clear in the course of this post). First if MZ twins are raised in the same family, they share the same genotype (A), they share the same ‘shared environment’ (C), so the difference is due to the non-shared environmental factor (E) only. In case of DZ twins raised in the same family, they share half the genotype (A), they share the same shared environment (C) and the difference in say IQ, is due to both genotype (A) and non-shared environmental factors (E). Twin studies with MZ and DZ twins (and in some cases siblings, half-siblings etc ) raised in the same family are used to tease apart the contribution of shared environmental factor, as opposed to genetics and non-shared environmental factors and can be used to find at a broad level if the trait is highly ‘genetically’ heritable (correlation between MZ>>DZ>>siblings), has high ‘shared environmental’ factors operating (MZ~DZ~sibling , but correlation still high) or is largely controlled by random non-shared environmental influences (low correlation in DZ/MZ/siblings). Please see accompanying Wikipedia figure.

Adoption studies are another method that is used to tease apart the shared environmental factors from genetic factors in calculating the heritability of a trait. Thus, even if correlation in MZ twin IQ is high, the effect could be due to shared environment factors (if say both MZ and DZ twin show similar correlation) , or it may be largely genetic (if MZ>>DZ when it comes to correlation between the trait in affected twins.

So it is necessary to qualify your statement: In identical twins, raised in the same family, , non-shared environment is the only cause of differences in IQ. IQ differences between persons that are not identical twins must be related to both shared environment , non-shared environment and genes.

I fully back up this citation:
“suffice it to say that I believe (and think that I have evidence on my side) that shows that in low SES conditions, a Low SES does not lead to full flowering of genetic Intelligence potential and is thus a leading cause of low IQ amongst low SES populations.”
In this case, I think that the problem starts only with your following comments; while you write “a leading cause”, I think that, by yourself, you mean “way more important than genetics”. If this is the case, would you explain to me why you think that?

Here goes. Consider a large sample of children in say low SES populations. IQ may be represented by a formula IQ= aA+cC+eE; where A reflects genotype, C shared environment and E non-shared environment. Here we are assuming no interaction of IQ with SES, so this equation (given values of a, c, e ) should hold for all SES data (both high as well as low SES cohort) . Unfortunately, life is not that simple, and one can not fit the same equation to low SES as well as high SES data set without changing the slopes of variables involved. Thus, Turkheimer and colleagues in two sets of studies have shown that there is an interaction of IQ and SES and there is direct affcet (SES mediated by s) and indirect interaction effects mediated via effects on A(a’), C(c’) and E(e’). Thus our equation becomes
IQ= sSES+(a+a’SES)A+(c+c’SES)C+(e+e’SES)E.
This equation, with suitable values of s, a, a’, c, c’, e and e’ now holds for all values of SES and IQ and the data fits nicely and can be interpreted. Remember that a+a’SES is sort of indicative of contribution of genetic factor to IQ and the proportion of variance due to genetic factor(at any given SES) can be found by squaring this and dividing this by sum of all other variances .
Var(A) =SQR(a+a?SES)
Similarly heritability or proportion of variance due to A :SQR(h)=SQR(a+a?SES)/(SQR(a+a?SES)*SQR(c+c?SES)*SQR(e+e?SES))

Turkheimer have plotted nice plot of their data which sows clearly that in LOW SES situations, the proportion of variance in IQ is largely due to shared environmental factors (C) , while in HIGH SES situations, the proportion of variance in IQ is largely due to genetic factors (A). the figures (in the free PDF available at Turkheimer’s side) is a must see to grasp the significance of this. I am quoting a bit from the paper:

Figure 3 shows the FSIQ variance accounted for by the three components, with 95% confidence intervals. In the most impoverished families, the modeled heritability of FSIQ is essentially 0, and C accounts for almost 60% of the variability; in the most affluent families, virtually all of the modeled variability in IQ is attributable to A.

Let us pause here and reflect on what this means. This means that in low SES families, IQ is independent of genotype and is mostly dependent on the SES status. Let us take some concrete examples. Say the mean IQ of low SES sample (income as a proxy for SES ranging from 1000-5000 rs p.a.; mean 2500 and variance 250)) is 80 with mean variance of 20. Thus, in this sample a typical child has IQ in range 80 +-20 or between 60 and 100. Suppose further that there are 5 alleles that confer differential advantage for IQ on a locus thus representing 5 genotypes, then having either of the genotype will essentially give us no predictive power to say whether the IQ of a particular sample is 80 or 60 or 100. Also, let us assume that there are 5 classes of C and they are highly correlated with SES. First class of C (is 1000-1500 SES range) and so on and so forth. Then knowing whjat kind of family (C) the child grew up in we could easily predict his IQ (if he is of class C where SES ranges from 1000-1500), his IQ is most probably 60. This is what I mean when I say that in low SES environments IQ is largely determined by environment and not by genetics. Now , I have taken a jump here and equated C with SES, but that is a justified leap in my opinion (more about that later).

What this also means is that given the right type of environment (say class C with SES in upper range of 3500-5000 rs p.a.) , all children (irrespective of their genotype (any 5 variants of genotype) can still achieve an IQ in the upper range , say 100 as the environment is the only predominant factor operating at this level and the impact of genetics is still not felt. Thus, if we do increase SES and provide the right C, then every child in this group can have mean IQ of 100.

Contrast this with the case at the upper end of the strata (SES). Here most of the variation in IQ is predominantly due to genetics (A) and shared environment C does not seem to play a big factor. Thus, knowing a genotype of a child has greater predictive power in this sample, than knowing his C (or family income or SES). Thus, evenif we provide a very enriched environment to all children (increase their C to the highest percentile), it would have no effect on increasing the mean IQ of the sample as now the IQ is mostly under genetic control.

This in a nutshell, is what I mean when I claim that low SES is the leading cause of low IQ in low SES families.

Before I rest, some objections might be readily apparent to a keen observer. First is the assumption inherent that SES and C are the same. I would like to propose here a new shared and universal sub-threshold environmental factor and would like to elaborate with a couple of examples. Let us say that those below poverty level do not have access to iodized salt and are thus prone to goiter and also mysteriously to low IQ as there is a module of brain (5 diff alleles at a particular locus leading to differences in abilities using this module) that needs iodine for its flowering and in absence of iodine, none of the alleles have any effect whatsoever- the module itself does not develop, so there are no questions of differences in ability or IQ due to differences in genotype etc. Now, given this state of affairs and also the fact that low SES families do not have access to iodine, when IQ is measured (then because of absence of this factor X), all children in this proband will have an IQ that does not measure abilities of this module (say this module adds 20 points of IQ) and thus all of them will have an IQ less by 20 points than was actually possible.Say the mean IQ measured is 80. Given the fact that some of the higher SES within this low SES group may have partial and sporadic access to iodine , the variance will be entirely environmental and no genetic variance would be found with some people having IQ close to 100 , who are in relatively upper start and have decent access to iodine. Contrast this the higher SES proband all of whom have access to iodized salt and thus can use their additional 20 points advantage on IQ tests. It would not be surprising if most of the variation here was genetic based on factor X allele) rather than due to income level or SES.

Another example to ruminate on is another universal and shared sub-threshold factor like having a golf course in the house. Let us assume that within higher SES group, this environmental enrichment factor plays a role, with some lower strata of higher SES (the middle class) not able to afford a golf course, while the higher higher SES strata (the upper class) abvle to afford a golf course and expose their children to them . Further, suppose that there is a module in the brains and genes switched on only if exposure to golf course takes place. Then within this higher SES group, what we will observe is that though the genetics plays a good role (due to factor X-iodine: remember, which is available to all in this group) ; still there would also be variation due to environment (golf course exposure) and that a full 20 points more can be added to all people of this group (with mean IQ 100 raising their IQ to 120), if all were exposed to a golf course and a intelligence-module-dependent-on-golf-course-exposure was allowed to develop. And on the higher end of IQ (and SES) what we would find is that most of the variance now is genetic (due to this golf-course module coming into play), while at the lower end, most of the variance is still environmental within this ‘high’ SES group.

If the above seems far fetched this is exactly what Turkheiemrs et al found in their follow up study focusing on mid to high SES children. I quote from it (again the pdf has beautiful figures and you should see them) :

Figure 2 illustrates the relations between income and genetic and shared environmental proportions of variance, as implied by the parameters estimated in Model 3. Genetic influences accounted for about 55% of the variance in adolescents’ cognitive aptitude and shared environmental influences about 35% among higher income families. Among lower income families, the proportions were in the reverse direction, 39% genetic and 45% shared environment. Although the shared environmental proportion of variance decreased with income, shared environmental variance per se did not decrease. The interactive effect was driven entirely by the increase in genetic variance. Genetic variance in cognitive aptitude nearly doubled from 4.41 in families earning less than $5000 annually to 8.29 in families earning more than $25,000 annually.

Our investigation supports our hypothesis that the magnitude of genetic influences on cognitive aptitude varies with socioeconomic status. This partially replicates the results presented by Turkheimer et al. (2003); however, no shared environmental interaction effects were demonstrable in the current study. Genetic influences accounted for about 55% of the variance in adolescents’ cognitive aptitude and shared environmental influences about 35% among higher income families. Among lower income families, the proportions were in the reverse direction, 39% genetic and 45% shared environment. This pattern is similar to the pattern seen in Turkheimer et al. (2003), although less marked.

So, I want you to pause here and grasp the significance of this- at every level of IQ-SES, there may be threshold factor that giverns whether IQ modules flower to full potential and this is the putative mechanism that leads to SES causing low or high IQ directional and causal relation. At each level, as the threshold factors become available,. more and more IQ starts coming under genetic control, but , and this is important, for jumps in IQ to take place , increasing SES (removing the sub-threshold conditions) is VERY important.

I mean “not following up on the ‘a leading cause'”, because in a later post, you write:
“Now, I have shown elsewhere that low SES causes low IQ”
Here, there is no mention of any other possible cause besides the environment anymore.

Yes, because as shown very strongly by Turkeihems and team , at low SES, shared environment/SES is the putative mechanism and genetics has no/negligible role to play. So for low SES, low SES causes low IQ. period.

in another post, you write
“A series of studies that I have discussed earlier, clearly indicate that in the absence of good socioeconomic conditions, IQ can be stunted by as large as 20 IQ points. ”
This same post also contained this citation “Children of well-off biological parents reared by poor/well -off adopted parents have Average IQ about 16 point higher than children of poor biological parents”
In my opinion, the latter would indicate the approximate range of genetic IQ differences for the samples in this study, while the former would indicate the approximate maximal environmental gain that can be hoped for in the environments that were encountered in these studies.

No they don’t. They talk about different SES groups, so as shown findings from one cannot be extrapolated to the other. In the low SES group, there is no genetic variation. We can thus not conclude that that (16 points diff.) is the ‘average’ genetic component taken the entire sample together. what one can say is that if mean IQ of high SES children was 100, the mean IQ of low SES children was 84 . Period. The difference is likely due to the fact, that the module X has not developed in low SES people (more later) .

Regarding the former, yes I agree that that is the maximum gain that one can hope for if all children of low SES were given the right environment (raised to high SES). Put another way, if mean IQ of poor/low SES children is 84 , then given the right conditions the mean of the low SES children can be raised to 104 (greater than high SES children’s mean :-).

As both of them do cover the same IQ range (10-20), the logical consequence for a broad statement on IQ and genetics seems therefore to be, that these studies may say that overall, IQ changes can be expected to be determined to approximately equal parts by genetics and environment, with environment being responsible for a typically larger part in low SES families, and genetics playing a relatively larger part in high SES families.

Agreed partially, but that glosses over the fact of sub-threshold universal shared environments and the fact that the role of genetic and environmental component varies with SES, an therefore an ideal statement would be IQ is under gentic controltolarge extent, but that gentics needs threshold environments to flower and thus the importance of environment component- not in explaining variance , but by its direct effect on IQ enabling/flowering.

This same post also contained this citation “Children of well-off biological parents reared by poor/well -off adopted parents have Average IQ about 16 point higher than children of poor biological parents”
In my opinion, the latter would indicate the approximate range of genetic IQ differences for the samples in this study, while the former would indicate the approximate maximal environmental gain that can be hoped for in the environments that were encountered in these studies.
As both of them do cover the same IQ range (10-20), the logical consequence for a broad statement on IQ and genetics seems therefore to be, that these studies may say that overall, IQ changes can be expected to be determined to approximately equal parts by genetics and environment, with environment being responsible for a typically larger part in low SES families, and genetics playing a relatively larger part in high SES families.
There also is this citation:
“The normal observation that identical twins belonging to well-off/middle class families have IQ rates similar as compared to fraternal twins, thus indicates that for children from well-off background (biological/adopted), the IQ (observed phenotype) is mostly due to genetic factors (underlying genotype) and environmental factors are not a big determinant.

The paradoxical observation that identical twins belonging to poor families have IQ rates as varying as compared to fraternal twins, should indicate that for children from poor background (biological/adopted), the IQ (observed phenotype) is mostly due to environmental factors and genetic factors (the underlying genotype ) are not a big determinant.”

These are extremely nice observations. I would be interested in the conclusions one might be tempted to draw from them. Reading the latter part of this sentence, one might come to the following conclusion (conclusion 1): “if in low-SES families the variations in IQ are largely determined by environmental factors, then providing a positive environment for the development of IQ would increase the IQ levels in these families impressively (up to 20 points; but, this is an up to value, means would be more interesting).”
While I completely agree with this thinking, one might also be tempted to draw the conclusion that (conclusion 2) “As IQ variations in low-SES families are largely due to environment, providing an IQ-stimulating environment in low-SES families might completely eliminate the IQ differences between low-SES families and high-SES families”
At the least, a non-cautious reader might understand your words as such. I am not sure whether you think that way or not. I would like to hear your opinion on that. I think that this citation “Children of well-off biological parents reared by poor/well -off adopted parents have Average IQ about 16 point higher than children of poor biological parents” provides an argument that precludes conclusion 2. It would rather say that (conclusion 3), ” providing a perfect IQ-stimulating environment for low-SES families as encountered in these studies, one should think that their offspring would achieve an IQ level that is 16 points lower than that of the offspring of high-SES families.”
I would like to hear your opinion on my conclusion 3.

I agree with conclusion 1. I also agree with conclusion 2 (not based on political correctness, but hard data). The paper on which these figures are based can be found here. The mean IQ of high SES persons is 113.5 and the mean IQ of low SES children is 98.00, thus a difference of ~16 points. The variation in IQ of high SES children raised in high SES families is 12.25; as shown this variance is likely due to genetics (say hundred percent is due to genetics); then changing the SES within the given range should have no effect on average IQ and it would remain 119 (for this high +/high+ group). On the other hand, the variance in low SES, reared by low SES families is 15.41 and mean is 92.40;thus if all were given enriched environment, their mean IQ would become 92.4+ 15.4 = ~ 108 . We still have a 10 point difference which can be accounted for by the fact that genetics had not come into play for low SES , low SES group yet and as genetics enters and increases the variance due to genetic flowering,, their IQ would be in the same league as high IQ/High IQ children.

So definitely the conclusion 3 is flawed- the difference would not be close to 16 points, but negligible, as the 16 points nowhere measures gentic difference in abilities, but reflects the genetic factor not yet active in low SES, due to improper environmental exposure.

I think that this is a rather important conclusion, as it tells us something about the differences in IQ that can be expected to exist between distinct population stratums (don’t know whether this is an appropriate word for what I try to say; I hope you understand what I mean).
If this is the ballpark of figures that we can expect between low-and high SES IQ differences, this would have important effects on future IQ-distributions. Population-wide stability of IQ-performance, if measured in a saturated environment (maximum stimulation of all members of society), can then only be achieved if all stratums of society have the same number of offspring per individuum. If low-SES families have more children, we have to expect that the 16-point lower IQ will decrease the whole-population IQ.
Here, the 16 points only apply to the sample as measured in your example; the true value of the saturated stratum-dependent-IQ together with stratum-dependent birthrates will determine the shift of the saturated IQ-distribution for the generations to come.

Do you agree with this point of view?

To use a very strong and negative connotation word, the above smacks of eugenics. And I wont comment further on this. Each according to his own philosophy, but beware that science does not support your conclusions. Instead of population controlling the poor, please try to elevate their vicious loop of undeserved poverty, low IQ and harmful stigma.

Turkheimer, E., Haley, A., Waldron, M., D’Onofrio, B., & Gottesman, I. (2003). Socioeconomic status modifies heritability of iq in young children Psychological Science, 14 (6), 623-628 DOI: 10.1046/j.0956-7976.2003.psci_1475.x

Harden, K., Turkheimer, E., & Loehlin, J. (2006). Genotype by Environment Interaction in Adolescents’ Cognitive Aptitude Behavior Genetics, 37 (2), 273-283 DOI: 10.1007/s10519-006-9113-4

Capron, C., & Duyme, M. (1989). Assessment of effects of socio-economic status on IQ in a full cross-fostering study Nature, 340 (6234), 552-554 DOI: 10.1038/340552a0

Reblog this post [with Zemanta]

Living on the edge of chaos; implications for autism and psychosis

SMI32-stained pyramidal neurons in cerebral co...Image via Wikipedia

I serendipitously came cross this article today about how our brains are self-organized criticality or systems living on the edge of chaos. There are many interesting ideas and gold nuggets in that article, and I’ll briefly quote from it.

In reality, your brain operates on the edge of chaos. Though much of the time it runs in an orderly and stable way, every now and again it suddenly and unpredictably lurches into a blizzard of noise.

Neuroscientists have long suspected as much. Only recently, however, have they come up with proof that brains work this way. Now they are trying to work out why. Some believe that near-chaotic states may be crucial to memory, and could explain why some people are smarter than others.

In technical terms, systems on the edge of chaos are said to be in a state of “self-organised criticality”. These systems are right on the boundary between stable, orderly behaviour – such as a swinging pendulum – and the unpredictable world of chaos, as exemplified by turbulence.

The quintessential example of self-organised criticality is a growing sand pile. As grains build up, the pile grows in a predictable way until, suddenly and without warning, it hits a critical point and collapses. These “sand avalanches” occur spontaneously and are almost impossible to predict, so the system is said to be both critical and self-organising. Earthquakes, avalanches and wildfires are also thought to behave like this, with periods of stability followed by catastrophic periods of instability that rearrange the system into a new, temporarily stable state.

Self-organised criticality has another defining feature: even though individual sand avalanches are impossible to predict, their overall distribution is regular. The avalanches are “scale invariant”, which means that avalanches of all possible sizes occur. They also follow a “power law” distribution, which means bigger avalanches happen less often than smaller avalanches, according to a strict mathematical ratio. Earthquakes offer the best real-world example. Quakes of magnitude 5.0 on the Richter scale happen 10 times as often as quakes of magnitude 6.0, and 100 times as often as quakes of magnitude 7.0.

These are purely physical systems, but the brain has much in common with them. Networks of brain cells alternate between periods of calm and periods of instability – “avalanches” of electrical activity that cascade through the neurons. Like real avalanches, exactly how these cascades occur and the resulting state of the brain are unpredictable.

Two of the power laws that are found in human brains relate to the phase shift and phase lock periods of EEG/fMRI or human brain systems etc. As per this PLOS comp biology paper:

Self-organized criticality is an attractive model for human brain dynamics, but there has been little direct evidence for its existence in large-scale systems measured by neuroimaging. In general, critical systems are associated with fractal or power law scaling, long-range correlations in space and time, and rapid reconfiguration in response to external inputs. Here, we consider two measures of phase synchronization: the phase-lock interval, or duration of coupling between a pair of (neurophysiological) processes, and the lability of global synchronization of a (brain functional) network. Using computational simulations of two mechanistically distinct systems displaying complex dynamics, the Ising model and the Kuramoto model, we show that both synchronization metrics have power law probability distributions specifically when these systems are in a critical state. We then demonstrate power law scaling of both pairwise and global synchronization metrics in functional MRI and magnetoencephalographic data recorded from normal volunteers under resting conditions. These results strongly suggest that human brain functional systems exist in an endogenous state of dynamical criticality, characterized by a greater than random probability of both prolonged periods of phase-locking and occurrence of large rapid changes in the state of global synchronization, analogous to the neuronal “avalanches” previously described in cellular systems. Moreover, evidence for critical dynamics was identified consistently in neurophysiological systems operating at frequency intervals ranging from 0.05–0.11 to 62.5–125 Hz, confirming that criticality is a property of human brain functional network organization at all frequency intervals in the brain’s physiological bandwidth.

Further, as per research by Thatcher et al, the EEG phase shift is larger in people with high IQ, while phase lock is smaller in the people with high IQ.

Phase shift duration (40–90 ms) was positively related to intelligence (P < .00001) and the phase lock duration (100–800 ms) was negatively related to intelligence (P < .00001). Phase reset in short interelectrode distances (6 cm) was more highly correlated to I.Q. (P < .0001) than in long distances (> 12 cm).

Further, in this paper , thatcher eta look at autistics and conclude that the people with autism show some deficits in phase shift and phase lock.

Results: In both short (6 cm) and long (21 – 24 cm) inter-electrode distances phase shift duration in ASD subjects was significantly shorter in all frequency bands but especially in the alpha-1 frequency band (8 – 10 Hz) (P < .0001). Phase lock duration was significantly longer in the alpha-2 frequencyband (10 – 12 Hz) in ASD subjects (P < .0001). An anatomical gradient was present with the occipitalparietal regions the most significant.
Conclusions: The findings in this study support the hypothesis that neural resource recruitment occurs in the lower frequency bands and especially the alpha-1 frequency band while neural resource allocation occurs in the alpha-2 frequency band. The results are consistent with a general GABA inhibitory neurotransmitter deficiency resulting in reduced number and/or strength of thalamo-cortical connections in autistic subjects 

It is interesting that in the original new scientist article , thatcher speculates that the pattern in schizophrenia may be reverse of what is seen in autism (exactly my thoughts, though the confounding of low IQ with autism may explain his autism results to an extent):

He found that the length of time the children’s brains spent in both the stable phase-locked states and the unstable phase-shifting states correlated with their IQ scores. For example, phase shifts typically last 55 milliseconds, but an additional 1 millisecond seemed to add as many as 20 points to the child’s IQ. A shorter time in the stable phase-locked state also corresponded with greater intelligence – with a difference of 1 millisecond adding 4.6 IQ points to a child’s score (NeuroImage, vol 42, p 1639). Thatcher says this is because a longer phase shift allows the brain to recruit many more neurons for the problem at hand. “It’s like casting a net and capturing as many neurons as possible at any one time,” he says. The result is a greater overall processing power that contributes to higher intelligence. Hovering on the edge of chaos provides brains with their amazing capacity to process information and rapidly adapt to our ever-changing environment, but what happens if we stray either side of the boundary? The most obvious assumption would be that all of us are a short step away from mental illness. Meyer-Lindenberg suggests that schizophrenia may be caused by parts of the brain straying away from the critical point. However, for now that is purely speculative. Thatcher, meanwhile, has found that certain regions in the brains of people with autism spend less time than average in the unstable, phase-shifting states. These abnormalities reduce the capacity to process information and, suggestively, are found only in the regions associated with social behaviour. “These regions have shifted from chaos to more stable activity,” he says. The work might also help us understand epilepsy better: in an epileptic fit, the brain has a tendency to suddenly fire synchronously, and deviation from the critical point could explain this. “They say it’s a fine line between genius and madness,” says Liley. “Maybe we’re finally beginning to understand the wisdom of this statement.”

Thus, it seems Autism and Psychosis are just two ways in which self-organized criticality can cease to do what it was designed to do- live on the edge , without falling on either side of order or chaos.

Reblog this post [with Zemanta]

Major conscious and unconcoscious processes in the brain: part 3: Robot minds

This article continues my series on major conscious and unconscious processes in the brain. In my last two posts I have talked about 8 major unconscious processes in the brain viz sensory, motor, learning , affective, cognitive (deliberative), modelling, communications and attentive systems. Today, I will not talk about brain in particular, but will approach the problem from a slightly different problem domain- that of modelling/implementing an artificial brain/ mind.

I am a computer scientist, so am vaguely aware of the varied approaches used to model/implement the brain. Many of these use computers , though not every approach assumes that the brain is a computer.

Before continuing I would briefly like to digress and link to one of my earlier posts regarding the different  traditions of psychological research in personality and how I think they fit an evolutionary stage model . That may serve as a background to the type of sweeping analysis and genralisation that I am going to do. To be fair it is also important to recall an Indian parable of how when asked to describe an elephant by a few blind man each described what he could lay his hands on and thus provided a partial and incorrect picture of the elephant. Some one who grabbed the tail, described it as snake-like and so forth.

With that in mind let us look at the major approaches to modelling/mplementing the brain/intelligence/mind. Also remember that I am most interested in unconscious brain processes till now and sincerely believe that all the unconscious processes can, and will be successfully implemented in machines.   I do not believe machines will become sentient (at least any time soon), but that question is for another day.

So, with due thanks to @wildcat2030, I came across this book today and could immediately see how the different major approaches to artificial robot brains are heavily influenced (and follow) the evolutionary first five stages and the first five unconscious processes in the brain.
The book in question is ‘Robot Brains: Circuits and Systems for Conscious Machines’ by Pentti O. Haikonen and although he is most interested in conscious machines I will restrict myself to intelligent but unconscious machines/robots.

The first chapter of the book (which has made to my reading list) is available at Wiley site in its entirety and I quote extensively from there:

Presently there are five main approaches to the modelling of cognition that could be used for the development of cognitive machines: the computational approach (artificial intelligence, AI), the artificial neural networks approach, the dynamical systems approach, the quantum approach and the cognitive approach. Neurobiological approaches exist, but these may be better suited for the eventual explanation of the workings of the biological brain.

The computational approach (also known as artificial intelligence, AI) towards thinking machines was initially worded by Turing (1950). A machine would be thinking if the results of the computation were indistinguishable from the results of human thinking. Later on Newell and Simon (1976) presented their Physical Symbol System Hypothesis, which maintained that general intelligent action can be achieved by a physical symbol system and that this system has all the necessary and sufficient means for this purpose. A physical symbol system was here the computer that operates with symbols (binary words) and attached rules that stipulate which symbols are to follow others. Newell and Simon believed that the computer would be able to reproduce human-like general intelligence, a feat that still remains to be seen. However, they realized that this hypothesis was only an empirical generalization and not a theorem that could be formally proven. Very little in the way of empirical proof for this hypothesis exists even today and in the 1970s the situation was not better. Therefore Newell and Simon pretended to see other kinds of proof that were in those days readily available. They proposed that the principal body of evidence for the symbol system hypothesis was negative evidence, namely the absence of specific competing hypotheses; how else could intelligent activity be accomplished by man or machine? However, the absence of evidence is by no means any evidence of absence. This kind of ‘proof by ignorance’ is too often available in large quantities, yet it is not a logically valid argument. Nevertheless, this issue has not yet been formally settled in one way or another. Today’s positive evidence is that it is possible to create world-class chess-playing programs and these can be called ‘artificial intelligence’. The negative evidence is that it appears to be next to impossible to create real general intelligence via preprogrammed commands and computations.

The original computational approach can be criticized for the lack of a cognitive foundation. Some recent approaches have tried to remedy this and consider systems that integrate the processes of perception, reaction, deliberation and reasoning (Franklin, 1995, 2003; Sloman, 2000). There is another argument against the computational view of the brain. It is known that the human brain is slow, yet it is possible to learn to play tennis and other activities that require instant responses. Computations take time. Tennis playing and the like would call for the fastest computers in existence. How could the slow brain manage this if it were to execute computations?

The artificial neural networks approach, also known as connectionism, had its beginnings in the early 1940s when McCulloch and Pitts (1943) proposed that the brain cells, neurons, could be modelled by a simple electronic circuit. This circuit would receive a number of signals, multiply their intensities by the so-called synaptic weight values and sum these modified values together. The circuit would give an output signal if the sum value exceeded a given threshold. It was realized that these artificial neurons could learn and execute basic logic operations if their synaptic weight values were adjusted properly. If these artificial neurons were realized as hardware circuits then no programs would be necessary and biologically plausible artificial replicas of the brain might be possible. Also, neural networks operate in parallel, doing many things simultaneously. Thus the overall operational speed could be fast even if the individual neurons were slow. However, problems with artificial neural learning led to complicated statistical learning algorithms, ones that could best be implemented as computer programs. Many of today’s artificial neural networks are statistical pattern recognition and classification circuits. Therefore they are rather removed from their original biologically inspired idea. Cognition is not mere classification and the human brain is hardly a computer that executes complicated synaptic weight-adjusting algorithms.

The human brain has some 10 to the power of 11 neurons and each neuron may have tens of thousands of synaptic inputs and input weights. Many artificial neural networks learn by tweaking the synaptic weight values against each other when thousands of training examples are presented. Where in the brain would reside the computing process that would execute synaptic weight adjusting algorithms? Where would these algorithms have come from? The evolutionary feasibility of these kinds of algorithms can be seriously doubted. Complicated algorithms do not evolve via trial and error either. Moreover, humans are able to learn with a few examples only, instead of having training sessions with thousands or hundreds of thousands of examples. It is obvious that the mainstream neural networks approach is not a very plausible candidate for machine cognition although the human brain is a neural network.

Dynamical systems were proposed as a model for cognition by Ashby (1952) already in the 1950s and have been developed further by contemporary researchers (for example Thelen and Smith, 1994; Gelder, 1998, 1999; Port, 2000; Wallace, 2005). According to this approach the brain is considered as a complex system with dynamical interactions with its environment. Gelder and Port (1995) define a dynamical system as a set of quantitative variables, which change simultaneously and interdependently over quantitative time in accordance with some set of equations. Obviously the brain is indeed a large system of neuron activity variables that change over time. Accordingly the brain can be modelled as a dynamical system if the neuron activity can be quantified and if a suitable set of, say, differential equations can be formulated. The dynamical hypothesis sees the brain as comparable to analog feedback control systems with continuous parameter values. No inner representations are assumed or even accepted. However, the dynamical systems approach seems to have problems in explaining phenomena like ‘inner speech’. A would-be designer of an artificial brain would find it difficult to see what kind of system dynamics would be necessary for a specific linguistically expressed thought. The dynamical systems approach has been criticized, for instance by Eliasmith (1996, 1997), who argues that the low dimensional systems of differential equations, which must rely on collective parameters, do not model cognition easily and the dynamicists have a difficult time keeping arbitrariness from permeating their models. Eliasmith laments that there seems to be no clear ways of justifying parameter settings, choosing equations, interpreting data or creating system boundaries. Furthermore, the collective parameter models make the interpretation of the dynamic system’s behaviour difficult, as it is not easy to see or determine the meaning of any particular parameter in the model. Obviously these issues would translate into engineering problems for a designer of dynamical systems.

The quantum approach maintains that the brain is ultimately governed by quantum processes, which execute nonalgorithmic computations or act as a mediator between the brain and an assumed more-or-less immaterial ‘self’ or even ‘conscious energy field’ (for example Herbert, 1993; Hameroff, 1994; Penrose, 1989; Eccles, 1994). The quantum approach is supposed to solve problems like the apparently nonalgorithmic nature of thought, free will, the coherence of conscious experience, telepathy, telekinesis, the immortality of the soul and others. From an engineering point of view even the most practical propositions of the quantum approach are presently highly impractical in terms of actual implementation. Then there are some proposals that are hardly distinguishable from wishful fabrications of fairy tales. Here the quantum approach is not pursued.

The cognitive approach maintains that conscious machines can be built because one example already exists, namely the human brain. Therefore a cognitive machine should emulate the cognitive processes of the brain and mind, instead of merely trying to reproduce the results of the thinking processes. Accordingly the results of neurosciences and cognitive psychology should be evaluated and implemented in the design if deemed essential. However, this approach does not necessarily involve the simulation or emulation of the biological neuron as such, instead, what is to be produced is the abstracted information processing function of the neuron.

A cognitive machine would be an embodied physical entity that would interact with the environment. Cognitive robots would be obvious applications of machine cognition and there have been some early attempts towards that direction. Holland seeks to provide robots with some kind of consciousness via internal models (Holland and Goodman, 2003; Holland, 2004). Kawamura has been developing a cognitive robot with a sense of self (Kawamura, 2005; Kawamura et al., 2005). There are also others. Grand presents an experimentalist’s approach towards cognitive robots in his book (Grand, 2003).

A cognitive machine would be a complete system with processes like perception, attention, inner speech, imagination, emotions as well as pain and pleasure. Various technical approaches can be envisioned, namely indirect ones with programs, hybrid systems that combine programs and neural networks, and direct ones that are based on dedicated neural cognitive architectures. The operation of these dedicated neural cognitive architectures would combine neural, symbolic and dynamic elements.

However, the neural elements here would not be those of the traditional neural networks; no statistical learning with thousands of examples would be implied, no backpropagation or other weight-adjusting algorithms are used. Instead the networks would be associative in a way that allows the symbolic use of the neural signal arrays (vectors). The ‘symbolic’ here does not refer to the meaning-free symbol manipulation system of AI; instead it refers to the human way of using symbols with meanings. It is assumed that these cognitive machines would eventually be conscious, or at least they would reproduce most of the folk psychology hallmarks of consciousness (Haikonen, 2003a, 2005a). The engineering aspects of the direct cognitive approach are pursued in this book.

Now to me these computational approaches are all unidimensional-

  1. The computational approach is suited for symbol-manipulation and information-represntation and might give good results when used in systems that have mostly ‘sensory’ features like forming a mental represntation of external world, a chess game etc. Here something (stimuli from world) is represented as something else (an internal symbolic represntation).
  2. The Dynamical Systems approach is guided by interactions with the environment and the principles of feedback control systems and also is prone to ‘arbitrariness’ or ‘randomness’. It is perfectly suited to implement the ‘motor system‘ of brain as one of the common features is apparent unpredictability (volition) despite being deterministic (chaos theory) .
  3. The Neural networks or connectionsim is well suited for implementing the ‘learning system’ of the brain and we can very well see that the best neural network based systems are those that can categorize and classify things just like ‘the learning system’ of the brain does.
  4. The quantum approach to brain, I haven’t studied enough to comment on, but the action-tendencies of ‘affective system’ seem all too similar to the superimposed,simultaneous states that exits in a wave function before it is collapsed. Being in an affective state just means having a set of many possible related and relevant actions simultaneously activated and then perhaps one of that decided upon somehow and actualized. I’m sure that if we could ever model emotion in machine sit would have to use quantum principles of wave functions, entanglemnets etc.
  5. The cognitive approach, again I haven’t go a hang of yet, but it seems that the proposal is to build some design into the machine that is based on actual brain and mind implemntations. Embodiment seems important and so does emulating the information processing functions of neurons. I would stick my neck out and predict that whatever this cognitive approach is it should be best able to model the reasoning and evaluative and decision-making functions of the brain. I am reminded of the computational modelling methods, used to functionally decompose a cognitive process, and are used in cognitive science (whether symbolic or subsymbolic modelling) which again aid in decision making / reasoning (see wikipedia entry)

Overall, I would say there is room for further improvement in the way we build more intelligent machines. They could be made such that they have two models of world – one deterministic , another chaotic and use the two models simulatenously (sixth stage of modelling); then they could communicate with other machines and thus learn language (some simulation methods for language abilities do involve agents communicating with each other using arbitrary tokens and later a language developing) (seventh stage) and then they could be implemented such that they have a spotlight of attention (eighth stage) whereby some coherent systems are amplified and others suppressed. Of course all this is easier said than done, we will need at least three more major approaches to modelling and implementing brain/intelligence before we can model every major unconscious process in the brain. To model consciousness and program sentience is an uphill task from there and would definitely require a leap in our understandings/ capabilities.

Do tell me if you find the above reasonable and do believe that these major approaches to artificial brain implementation are guided and constrained by the major unconscious processes in the brain and that we can learn much about brain from the study of these artificial approaches and vice versa.

SES and the developing brain

I have written about poverty/SES and its effects on brain development/IQ earlier too,and this new review article by Farah and Hackman in TICS is a very good introduction to anyone interested in the issue.

The article reviews the behavioral studies that show that SES is correlated with at least the two brain systems of executive function and language abilities.It also review physiological data that shows that even when behavioral outcomes do not differ ERP can show differential activation in the brains of people with low and middle SES , thus suggesting that differences that may not be detected on behavioral measures may still exist. They also review (f)MRI data that shows no structural differences in the brains of low and middle SES children, but definite functional differences.they also review experimental manipulation of social status in labarotaories, and show how those studies also indicate that SES and executive function are correlated.

They then turn to the million dollar question of the direction of causality and for this infer indirectly based on the SES-IQ causal linkages.

What is the cause of SES differences in brain function? Is it contextual priming? Is it social causation, reflecting the influence of SES on brain development? Alternatively, is it social selection, in which abilities inherited from parents lead to lower SES? Current research on SES and brain development is not designed to answer this question. However, research on SES and IQ is relevant and supports a substantial role of SES and its correlated experience as causal factors.

Slightly less than half of the SES-related IQ variability in adopted children is attributable to the SES of the adoptive family rather than the biological. This might underestimate environmental influences because the effects of prenatal and early postnatal environment are included in the estimates of genetic influence. Additional evidence comes from studies of when poverty was experienced in a child’s life. Early poverty is a better predictor of later cognitive achievement than poverty in middle- or late-childhood, an effect that is difficult to explain by genetics. SES modifies the heritability of IQ, such that in the highest SES families, genes account for most of the variance in IQ because environmental influences are in effect at ceiling in this group, whereas in the lowest SES families, variance in IQ is overwhelmingly dominated by environmental influences because these are in effect the limiting factor in this group. In addition, a growing body of research indicates that cognitive performance is modified by epigenetic mechanisms, indicating that experience has a strong influence on gene expression and resultant phenotypic cognitive traits . Lastly, considerable evidence of brain plasticity in response to experience throughout development indicates that SES influences on brain development are plausible.

Differences in the quality and quantity of schooling is one plausible mechanism that has been proposed. However, many of the SES differences summarized in this article are present in young children with little or no experience of school , so differences in formal education cannot, on their own, account for all of the variance in cognition and brain development attributable to SES. The situation is analogous to that of SES disparities in health, which are only partly explained by differential access to medical services and for which other psychosocial mechanisms are important causal factors .

The last point is really important and can be extended. Access to health services for low SES people may be a reason why , for eg, more schizophrenia incidence is found in low SES neighbourhoods. which brings us to the same chicken-and-egg question of the drift theory of schizophrenia- whether people with schizophrenia drift into low SES or low SES is a risk factor in itself. Exactly this point was brought to my attention when I was interacting with a few budding psychiatrists recently, this Martha Farah theory about the SES leading to lower IQ/ cognitive abilities. It is important to acknowledge that low SES not only leads to left hypo-frontality (another symptom of schizophrenia), schizophrenia is supposed to be due to lessened mylienation and again nutritional factors may have a role to play; also access to health care, exposure to chronic stress and lesser subjective feelings of control may all be mediating afctors that lead low SS to lead to schizophrenia/ psychosis.Also remember that schizophrenia is sort of a devlopmenetal disorder.

Well, I digressed a bit, but the idea is that not only does low SES affect ‘normal’ cognitive abilities, it may even increase the risk for ‘abnormal’ cognitive abilities that may lead to psychosis, and his effect of SES on IQ/cognitive abilities/ risk of mental diseases is mediated by the effect of SES on the developing brain. I have already covered the putative mechanisms by which SES may affect brain development, but just to recap, here I quote from the paper:

Candidate causal pathways from environmental differences to differences in brain development include lead exposure, cognitive stimulation, nutrition, parenting styles and transient or chronic hierarchy effects. One particularly promising area for investigation is the effect of chronic stress. Lower-SES is associated with higher levels of stress in addition to changes in the function of physiological stress response systems in children and adults. Changes in such systems are likely candidates to mediate SES effects as they impact both cognitive performance and brain regions, such as the prefrontal cortex and hippocampus, in which there are SES differences.

We can only hope that the evil of low SES is recognized as soon as possible and if for nothing else, than for advancing science, some intervention studies are done that manipulate the SES variables in the right direction and thus ensure that the full cognitive potential of the children flowers.

ResearchBlogging.org

HACKMAN, D., & FARAH, M. (2009). Socioeconomic status and the developing brain Trends in Cognitive Sciences, 13 (2), 65-73 DOI: 10.1016/j.tics.2008.11.003

Neurological correlates of Poverty

While people generally do not squirm on reading a headline claiming neural correlates of religion, god, trust, consciousness, political/ sexual orientation etc, I am sure the title neural correlates of Poverty would have lead to some uneasy shuffling around. How can poverty that is clearly a result of economic opportunities/ capabilities be reduced to brain? Are we claiming that low inherent IQ and the neural correlates thereof define and lead to poverty? Or is the claim instead that poverty leads to definite changes in the brain, which may lead to manifestation of low IQ and the sustenance of the vicious circle of poverty? The regular readers of the blog will know which side of the fence I am sitting on!

The blogosphere is normally abuzz with controversial topics like atheism, meaninglessness of evolution and race and gender differences(for eg.  in IQ) and people defend these sacred dictum doggedly, claiming that ‘is’ and ‘ought’ need not be confused, especially in a cold, logical science which deals with all facts and should not be guided by values. Yet, the same blogosphere generally silently ignores, or does not take a stand , when the ‘is’ and ‘ought’ are in sync and something morally significant is also found to be scientifically valid. Rather the apology for such facts is made very cautiously, with the spirit of not offending the people who have a different, and in my view, an inferior moral system.

I believe whenever people discuss poverty/SES, they have either of the two moral systems: first, the world is unfair and poor people are poor because of some external factors/ circumstances; addressing them may solve/ eliminate the problem of poverty;  and second: the world is fair (like an idealized free market) and if someone is poor they are due to either inherent internal flaws (bad genes) or maybe bad choices (they want to be poor/ are lazy and unindustrious etc); so the problem of poverty cannot/ should not be solved.  I subscribe to the first moral system and believe in interventions to solve the problem of poverty. I am glad to have scientific facts to my side and have been addressing these issues in a series of posts .

The latest impetus to write on the topic comes form reading Lehrer’s post titled Poverty and the brain at the Frontal cortex and I am glad to have found a fellow blogger who doesn’t mind speaking on such controversial topics and take a stand for ‘is’ that is in sync with ‘ought’. It is an excellent post regarding how early interventions can help alleviate poverty and how a poor person suffers from the viscous circle of poverty by the mediating influence of brain and IQ.

Lehrer also mentions the work of Martha Farah (of Visual Agnosia fame whose earlier work was on vision) on the same and I recommend reading at least this article by Martha and colleagues, although many other invaluable gems are present on her site.

The article begins with an anecdotal reference to how Martha first became aware of the gravity of the issue, when she saw her babysitters / maids steeped in poverty and the low IQ and SES viscous circle. this resonates with me and I can easily relate to this as my child enjoys a lot of toys while our maid’s children are faced with lack.

I would now quote extensively from the aforementioned article:

It seemed to me that children’s experience of the world is very different in low and middle SES environments. Most middle SES children have abundant opportunities to explore the world, literally, in terms of people met and places seen, and figuratively, in terms of the world of ideas. In contrast, low SES children generally have fewer interactions with the wider world and much of what they do experience is stressful. Basic research with animals has established the powerful effects of both environmental impoverishment and stress on the developing brain.

She then goes on to make out the case for NCC of poverty:

For the sake of exploring the cognitive neuroscience perspective on transgenerational poverty, and discovering what, if anything, it can contribute to correcting socioeconomic inequality, the first order of business is to ask whether socioeconomic status bears any straightforward relation to brain development. On the face of things it might seem unlikely that characteristics such as income, education and job status, which are typically used to estimate SES, would bear any systematic relationship to physiological processes such as those involved in brain development. It is, however, well established that SES affects physical health through a number of different causal pathways (Adler et al. 1994), many of which could play a role in brain development. It is also clear that poverty is associated with differences in brain function on the basis of the differences in standardized test performance cited earlier, as cognitive tests reflect the function of the brain. However, for a cognitive neuroscience approach to be helpful, the relations between socioeconomic status and the brain must be relatively straightforward and generalizable. The first question that my collaborators and I addressed was therefore: Can we generalize about the neurocognitive correlates of socioeconomic status? Once we have established the neurocognitive profile of childhood poverty, we can begin to test more specific hypotheses about causal mechanisms.

I will now digress a little from the main topic and introduce the five neurocognitive systems that Martha and colleagues have identified and how they tested some children from low and middle SES for finding their capabilities in these systems.

The children were tested on a battery of tasks adapted from the cognitive neuroscience literature, designed to assess the functioning of five key neurocognitive systems. These systems are described briefly here.

The Prefrontal/Executive system enables flexible responding in situations where the appropriate response may not be the most routine or attractive one, or where it requires maintenance or updating of information concerning recent events. It is dependent on prefrontal cortex, a late-maturing brain region that is disproportionately developed in humans.

The Left perisylvian/Language system is a complex, distributed system encompassing semantic, syntactic and phonological aspects of language and dependent predominantly on the temporal and frontal areas of the left hemisphere that surround the Sylvian fissure.

The Medial temporal/Memory system is responsible for one-trial learning, the ability to retain a representation of a stimulus after a single exposure to it (which contrasts with the ability to gradually strengthen a representation through conditioning-like mechanisms), and is dependent on the hippocampus and related structures of the medial temporal lobe.

The Parietal/Spatial cognition system underlies our ability to mentally represent and manipulate the spatial relations among objects, and is primarily dependent upon posterior parietal cortex.

The Occipitotemporal/Visual cognition system is responsible for pattern recognition and visual mental imagery, translating image format visual representations into more abstract representations of object shape and identity, and reciprocally translating visual memory knowledge into image format representations (mental images).

Not surprisingly, in view of the literature on SES and standardized cognitive tests, the middle SES children performed better than the low SES children on the battery of tasks as a whole. For some systems, most notably the Left perisylvian/Language system and the Prefrontal/Executive system, the disparity between low and middle SES kindergarteners was both large and statistically significant.

Thus, they found, in a small group of children , that Language and Executive systems’ performance differed in low and middle SES children and they were able to replicate this finding with a larger group of children too. This time they broke executive function further into components and found a finer granularity of how SES affects the brain:

As before, the language system showed a highly significant relationship to SES, as did executive functions including Lateral prefrontal/Working memory and Anterior cingulate/Cognitive control components and the Parietal/Spatial cognition system. With a more demanding delay between exposure and test in the memory tasks, we also found a difference in the Medial temporal/Memory system. Performance on the Parietal/spatial system tests also differed as a function of SES.

They also did some studies with older children and to summarize the results of all these studies in their own words:

In sum, although the outcome of each study was different, there were also commonalities among them despite different tasks and different children tested at different ages. The most robust neurocognitive correlates of SES appear to involve the Left perisylvian/Language system, the Medial temporal/Memory system (insofar as SES effects were found in both studies that tested memory with an adequate delay) and the Prefrontal/Executive system, in particular its Lateral prefrontal/Working memory and Anterior cingulate/Cognitive control components. Children growing up in low SES environments perform less well on tests that tax the functioning of these specific systems.

Next they look at the causal versus correlational nature of findings and if causal, then the directions of causality. It is this paragraph , that amazed me, for they seem to be apologetic for the fact that their findings are also ethically good ones.

Do these associations reflect the effects of SES on brain development, or the opposite direction of causality? Perhaps families with higher innate language, executive and memory abilities tend to acquire and maintain a higher SES. Such a mechanism seems likely, a priori, as it would be surprising if genetic influences on cognitive ability did not, in the aggregate, contribute to individual and family SES. However, it seems also seems likely that causality operates in the opposite direction as well, with SES influencing cognitive ability through childhood environment. Note that the direction of causality is an empirical issue, not an ethical one. The issue of whether and to what extent SES differences cause neurocognitive differences or visa versa should not be confused with the issue of whether we have an obligation to help children of any background become educated, productive citizens.

Then, quite important from this blog’s point of view, they review the literature that supports SES to IQ direction of causality.

Cross-fostering studies of within- and between -SES adoption suggest that roughly half the IQ disparity in children is experiential (Capron & Duyme, 1989; Schiff & Lewontin, 1986). If anything, these studies are likely to err in the direction of underestimating the influence of environment because the effects of prenatal and early postnatal environment are included in the estimates of genetic influences in adoption studies. A recent twin study by Turkheimer and colleagues (2003) showed that, within low SES families, IQ variation is far less genetic than environmental in origin. Additional evidence comes from studies of when, in a child’s life, poverty was experienced. Within a given family that experiences a period of poverty, the effects are greater on siblings who were young during that period (Duncan et al. 1994), an effect that cannot be explained by genetics. In sum, multiple sources of evidence indicate that SES does indeed have an effect on cognitive development, although its role in the specific types of neurocognitive system development investigated here is not yet known.

Next they tried to tease out what specific SES related factors can affect the different neurocognitive systems. They list both physical and psychological factors that have been hypothesized and researched on in relation to SES and IQ.

Potential causes, physical and psychological

What aspects of the environment might be responsible for the differences in neurocognitive development between low and middle SES children? A large set of possibilities exist, some affecting brain development by their direct effects on the body and some by less direct psychological mechanisms. Three somatic factors have been identified as significant risk factors for low cognitive achievement by the Center for Children and Poverty (1997): inadequate nutrition, substance abuse (particularly prenatal exposure), and lead exposure.

As with potential physical causes, the set of potential psychological causes for the SES gap in cognitive achievement is large, and the causes are likely to exert their effects synergistically. Here we will review research on differences in cognitive stimulation and stress.

They then discuss the psychological factors, which they then investigated, in more detail.

One difference between low and middle SES families that seems predictable, even in the absence of any other information, is that low SES children are likely to receive less cognitive stimulation than middle SES children. Their economic status alone predicts that they will have fewer toys and books and less exposure to zoos, museums and other cultural institutions because of the expense of such items and activities. This is indeed the case (Bradley et al. 2001a) and has been identified as a mediator between SES and measures of cognitive achievement (Bradley and Corwyn 1999; Brooks-Gunn and Duncan 1997; McLoyd 1998). Such a mediating role is consistent with the results of neuroscience research with animals. Starting many decades ago (e.g., Volkmar & Greenough, 1972) researchers began to observe the powerful effects of environmental stimulation on brain development. Animals reared in barren laboratory cages showed less well developed brains by a number of different anatomical and physiological measures, compared with those reared in more complex environments with opportunities to climb, burrow and socialize (see van Praag et al 2000 for a review).

The lives of low SES individuals tend to be more stressful for a variety of reasons, some of which are obvious: concern about providing for basic family needs, dangerous neighborhoods, and little control over one’s work life. Again, research bears out this intuition: Turner and Avison (2003) confirmed that lower SES is associated with more stressful life events by a number of different measures. The same appears to be true for children as well as adults, and is apparent in salivary levels of the stress hormone cortisol (Lupien et al. 2001).

Why is stress an important consideration for neurocognitive development? Psychological stress causes the secretion of cortisol and other stress hormones, which affect the brain in numerous ways (McEwen 2000). The immature brain is particularly sensitive to these effects. In basic research studies of rat brain development, rat pups are subjected to the severe stress of prolonged separation from the mother and stress hormone levels predictably climb. The later anatomy and function of the brain is altered by this early neuroendocrine phenomenon. The brain area most affected is the medial temporal area needed for memory, although prefrontal systems involved in the regulation of the stress response are also impacted (Meaney et al. 1996).

They then go on to discuss how this information can be used to formulate mechanisms that mediate the effect of low SES on diffrent neurocognitive systems.

The latest phase of our research is an attempt to make use of the description of the SES disparities in neurocognitive development in testing hypotheses about the causal pathways. Drawing on our previous research that identified three neurocognitive systems as having the most robust differences as a function of SES (Perisylvian/Language, Medial temporal/Memory, and Prefrontal/Executive), we are now testing hypotheses concerning the determinants of individual differences in the development of these systems in children of low SES. Specifically, we are investigating the role of childhood cognitive stimulation and social/emotional nurturance (Farah et al. 2005; Childhood experience and neurocognitive development: Dissociation of cognitive and emotional influences).

They then describe an observational study of interaction between children and parents and how they assess the cognitive simulation an social/emotional nurturance using HOME assessment battery. What they found follows:

Children’s performance on the tests of Left perisylvian/Language was predicted by average cognitive stimulation. This was the sole factor identified as predicting language ability by forward stepwise regression, and one of three factors identified by backwards stepwise regression, along with the child’s gender and the mother’s IQ. In contrast, performance on tests of Medial temporal/Memory ability was predicted by average social/emotional nurturance. This was the sole factor identified as predicting memory ability by forward stepwise regression and one of three factors identified by backwards stepwise regression, along with the child’s age and cognitive stimulation. The relation between memory and early emotional experience is consistent with the animal research cited earlier, showing a deleterious effect of stress hormones on hippocampal development. Our analyses did not reveal any systematic relation of the predictor variables considered here to Lateral prefrontal/Working memory or Anterior cingulate/Cognitive control function. In conclusion, different aspects of early experience affect different systems of the developing brain. Cognitive stimulation influences the development of language, whereas social/emotional nurturance affects the development of memory but not language.

Here is what they conclude:

What are the implications for society of a more mechanistic understanding of the effects of childhood poverty on brain development? To different degrees, and in different ways, we regard children as the responsibility of both parents and society. Parents’ responsibility begins before birth and encompasses virtually every aspect of the child’s life. Society’s responsibility is more circumscribed. In the United States, for example, society’s contribution to the cognitive development of children begins at age 5 or 6, depending on whether public kindergarten is offered. The physical health and safety of all infants and children is a social imperative, however, well before school age. Laws requiring lead abatement in homes occupied by children exemplify our societal commitment to protect them from the neurological damage caused by this neurotoxin. Research on the effects of early life stress and limited cognitive stimulation has begun to show that these concomitants of poverty have negative effects on neurological development too, by mechanisms no less concrete and real. Thus, neuroscience may recast the disadvantages of childhood poverty as a bioethical issue rather than merely one of economic opportunity.

In my view the societal implications are far reaching, if low SES leads to lowered cognitive functioning, it becomes our duty to provide more cognitive stimulation and ensure that all children get sufficient social/ emotional nurturance so that their IQ can flower to its full potential.

I would have liked to end on this note, but cant help pointing that the five neurocognitive systems Martha has identified, to me seems to follow in stages, with the later systems maturing later :

1) Occipital/ visual : describe/ perceive the world/ self
2) Parietal/ spatial:explain the world/self (may be involved in consciousness)
3) Temporal/ Memory: predict the world/self
4) Frontal/ executive: control the world/ self
5) Sylvian/ Language: improve the world/ self

We all know that language abilities develop the oldest and vision is more or less developed at birth; also the fact that SES should affect the latter stages of neurocognitive systems also gels in. the fact that cognitive stimulation affects language and emotional/social nurturance affects memory to me also fits in.

Anyway whatever the implication sof this research for stage theories, they have far reaching and imprortanat implications for social policy and education.
ResearchBlogging.org
Farah, M.J.,Noble, K.G. and Hurt, H. (2005). Poverty, privilege and brain development: Emprical findings and ethical implications. In J. Illes (Ed.) Neuroethics in the 21st Century. New York: Oxford University Press.

Go to Top