Trying To Remember What Causes Brain Damage: This is Your Brain on Poverty

By Jarrett Chapin

If we suppose that intellectual aptitude is the result of teaching, we may be partially wrong. Likewise, if we believe that economics and socio-environmental conditions are responsible for quantifiable intellect then we might only be somewhat right. Such are the implications of various bits of research that, over the past decade, has suggested that socioeconomic status (SES) bears considerably on the ability of children growing up in low (LSES) or high (HSES) socioeconomic environments. Although cognitive science and public fiscal policy may seem to have only tentative links at times, in education they dovetail splendidly.

You may prefer, for instance, that former account of aptitude, “aptitude is the result of teaching,” you might understand “reform” to be a function of staffing or certification. It would follow then that the dysfunction in our American schools can be solved by either reforming the process of certification for teachers, devising ways of ferreting out poor educators in our school system with a system of performance-based pay incentives and deprivations, or the reallocation of good teachers who tend to migrate to suburban middleclass schools.

On the other hand, if you, a social critic, prefer the idea “that economics and socio- environmental conditions are responsible for quantifiable intellect” then one favors reforms like after school programs, Head Start, welfare assistance, and environmental improvement i.e., school construction and community rehabilitation.

On both accounts, according to a growing population of researchers in various fields, we may be wrong. Not completely wrong, but wrong nonetheless. And being incorrect in our assumptions, in this case, would mean spending on facets of education reform which bear little effect on the real education problems of LSES children. Correct spending suggests targeting areas in the sphere of education that are usually understood to be outliers or ancillary elements.

Researchers from Cornell University, Gary Evans and Michelle Shamberg, and in another study Mark Kishiyama and associates from the University of California, Berkeley, have suggested that it is not simply poverty but the consequent stress indicative of poverty that diminishes the aptitude of LSES children. They suggest that the education problems of LSES children are neurological as much as they are environmental. Both the Berkeley and Cornell studies build upon the work of Martha J. Farah and associates at the University of Pennsylvania who seem to have been the first to transpose SES onto neurophysiology to make this argument. In 2005, Dr. Farah, joined by Kimberly G. Noble and Hallam Hurt produced a study, one of several authored by Farah, that seeks to change the way we think about the performance of students and the solutions on which we are willing to spend money.

The first stage of the University of Pennsylvania study was an attempt to describe and account for lower neurocognitive performance between 30 LSES and 30 middle SES (MSES) public school kindergarteners. Designed to “assess the functioning of five key neurocognitive systems” such as spatial and visual cognition, cognitive control, language, and memory, the Berkeley researchers found that MSES children scored a standard deviation higher than LSES children in language tests and two-thirds of a standard deviation higher for “executive function” mentioned before as “cognitive control.” In most of the tests, spatial and visual results became irrelevant.

It is interesting that test scores are interpreted neurophysiologically, i.e., a test of “Language” becomes a test of the “left perisylvian cortical region” just as a test of working memory becomes a test of the “prefrontal cortex.”[1] On this note, the original test of memory was a test of the medial temporal cortical region, a region where the hippocampus is located, the hippocampus being responsible for the creation of long term memories.

To support the assumption about the socioenvironmental effect of stress hormones on brain anatomy and function, the researchers cite a laboratory experiment in which rat pups are separated from their rat mothers. The separation, it seems, adds rodent sized emphasis to the theory that stress can cause brain damage. The separation of rat pups from their mothers, according to researchers Meany, Diorno, and Francis, was found to have altered both anatomy and function in their rat brains and had a particular effect on “the medial temporal area needed for memory, although prefrontal systems involved in the regulation of the stress response are also impacted.”[2] Farah, Noble, and Hurt suppose that, similarly, living in low socioeconomic environments produces the kind of stress which, as did the stress of separation experienced by rats in the experiment above mentioned, damages human brains. Such stressful stimuli, according to the researchers, might include: “concern about providing for basic family needs, dangerous neighborhoods, and little control over one’s work life.”[3]

The research leads us to adopt a theory that high stress in LSES environments produces greater amounts of stress hormones like cortisol, a secretion from the adrenal gland that affects blood pressure and insulin levels, and catecholamine, also an adrenal secretion responsible for preparing the body for the “fight-or-flight” response. Stress hormones in large amounts are thought to cause a deregulation of blood pressure and the immune system but also to the prefrontal cortex and the medial cortex which contains hippocampus, the organ responsible for the creation of long term memories. The University of Pennsylvania researchers suppose that the cognitive disparities they found between LSES and MSES children in all three experiments can be accounted for in this way.

Because they preferred this particular interpretive model, one which posits that stress hormones cause brain damage, researchers find evidence of both medial temporal as well as prefrontal damage in their LSES subjects. Having found such evidence, it would have followed—or it would have at least been very suggestive—that something in a LSES subject’s background causes brain damage. However, at the experiment’s conclusion researchers did not find the evidence they anticipated. Instead, they found that LSES subjects differed little from MSES subjects in all but two categories, language (left perisylvian cortical region), and executive function or cognitive control, (prefrontal cortex).

Having found no memory deficiency with regard to the medial temporal region, researchers seem to have decided instead to mine memory from the prefrontal cortex whereon LSES subjects scored one-third lower than MSES subjects—not quite as low as LSES scored in language. Prefrontal or executive function was thus split into 3 different categories, cognitive control, reward processing, and most important, working memory. Not surprising, researchers also tried to induce a memory response in the medial temporal lobe (memory) by placing a delay between the stimulus and the response portion of the experiment with the hope that such a delay would give LSES subjects ample time to forget.

Hoping to replicate their findings in the prefrontal and left perisylvian region, researchers conducted two more experiments, one on 150 first graders “of varying ethnicities” and SES backgrounds, and one on of 60 middle school students divided, as were the subjects of the first study, into equal parts LSES and MSES. In both of the latter studies, LSES subject’s scores were similar to those in the first round. Researchers thus concluded that it is that stress affects working memory and that, perhaps, is the cause of lowered ability in larger numbers of LSES subjects.

In their account of the correlation of LSES and low achievement, Evans and Schamberg lean heavily on a model developed by researcher Bruce S. McEwen who supposes that stress hormones like cortisol, which may help the body respond and adapt to various environmental stimuli, may also build up under prolonged strain causing long term physical damage in conditions of perpetual stress, just like the rats in the study described by the University of Pennsylvania study.[4] McEwen calls the effects of perpetual or “chronic stress,” “allostatic load” which can be described as a weakening of health marked by a prolonged secretion of survival hormones. Hormones or “physiological mediators,” like emergency switches or rocket packs for movie heroes, allow humans to adapt to new conditions and rise to challenges that are taxing to our bodies and minds. Cortisol, for instance, can provide a burst of energy in a dangerous moment or make you intellectually sharper when you are confronted with an emergent problem. In high doses, cortisol can overtax the body and cause effects like cognitive impairment and obesity. Wrote McEwan in 2000, “Both cortisol and catecholamines are mediators of the adaptation of many systems of the body to acute challenges, while, at the same time, these mediators also participate in pathological changes over long periods of time ranging from immunosuppression to obesity, hypertension, and atherosclerosis.”

Based upon McEwan’s account, Evans and Schamberg suppose that the frequency of accumulated stress varies by the duration of childhood poverty which in turn can be correlated positively with lowered working memory capacity in adult life. This thesis, combined with an examination of longitudinal data derived from a study of 195 white young adults, led the researchers to suggest that the working memory of an adult from a LSES background is quantifiably affected by stress. Evans and Schamberg conclude that, “on average, poor adults raised in middle-income families could hold in working memory a sequence of 9.44 items, whereas poor adults who grew up in poverty had a working memory capacity of 8.50 items.”[5]

The researchers do entertain the possibility that their findings might be indicative of a reversed situation wherein decreased memory ability is what causes the prolonged stress or allostatic load “poverty–>working memory–>allostatic load,” rather than poverty–>allostatic load–>working memory. They entertain this idea, but not for long as they insist that the inferential links between allostatic load and working memory are very suggestive of the former causal chain, that the stress of poverty causes lowered working memory. Further, the researchers add, “the relationship between duration of childhood poverty and allostatic load was not attenuated when working memory was partialed from the equation.” Said simply, if lowered working memory, rather than poverty, was indeed the cause of stress observed in the data, then the elimination of working memory, in particular places of the sample, should have removed instances of allostatic load. It did not.

Mark Kishiyama and his research associates from the University of California, Berkeley, have produced a behavioral experiment for which 26 children from LSES and HSES backgrounds were fitted with electrodes and asked to perform various cognitive feats. Taking also the University of Pennsylvania study as a research benchmark, the Berkeley researchers used Electroencephalography (EEG) equipment in an attempt to find behavioral data to support the poverty-neurocognition correlation and to confine the problem, as did the prior pair of researchers, to the prefrontal cortex.[6] Though researchers admit the limitations of a largely behavioral test, the study is an advance over prior studies because it uses brain imaging to show more concretely what other experimenters only supposed by projecting onto human brains, the theories of other researchers derived from the brains of rats. That is, imaging actually showed on video monitors, the degree of prefrontal activity of each subject.

On the 5 tests issued, HSES subjects scored significantly higher than LSES subjects who, for the most part, scored within the predetermined mean. LSES subjects, in other words, performed many of the tasks with average or near average aptitude. For instance, on a test which required subjects to count forward and then in reverse as means of testing their working memory, “digit span,” HSES children scored higher than 1 standard deviation over the predetermined mean. LSES subjects achieved mean scores on the low end. Another test of working memory included a verbal component, “semantic fluency,” a test in which subjects were required produce all the words they could conjure in order to satisfy a cue from proctors, i.e., the names of animals, food, or words that begin with “sh.” HSES subjects excelled in the semantic fluency test, but seem to glow brightest in the area of language overall. In a general “language” test, subjects were asked simply to define words. HSES children scored as high as 2 standard deviations above the mean; LSES children scored 1 standard deviation below.

The tests may have shown HSES children to be semi-savants when it comes to counting backwards and forwards, though in regard to the poverty-neurocognition correlation, the experiment failed to prove that LSES subjects had their prefrontal cortices wounded by their SES. Their scores were average but not indicative of brain damage and the only thing that seems to wound them was the comparison to HSES children.

Of course the researchers insist that their experiment on 26 children drawn only from the Bay Area confirms that SES bears neurophysiologically on the prefrontal function in the brains of children. However, these researchers do not seem to have transcended a theory built upon observations of rat pups. They note themselves that behavioral tests can only prove so much about the neurological operations as “they provide only indirect measures of brain function.” And, as the authors also note, they have not completely isolated the observed effects to “prefrontal dysfunction.” In a sense, they are no closer to proving the poverty-neurocognition correlation than the University of Pennsylvania researchers who preceded them.

If the Berkeley study demonstrated anything convincingly it was that HSES children excel in language, an ability attributed to the left perisylvian cortical region which Farah, Noble, and Hurt in their earlier work found to be predicted by “cognitive stimulation,” otherwise known as conversation. The latter note that their research suggests cognitive stimulation “was the sole factor identified as predicting language ability […] along with the child’s gender and the mother’s I.Q..” Not really so different from the contrasting account of medial temporal, or memory ability which was attributed to “average social/emotional nurturance,” otherwise known as a lack of stress. However, we must doubt that a lack of conversation causes neurophysiological brain damage in the way that researchers assume stress hormones cause a deformation of the prefrontal cortex and the medial temporal lobe. Evidence that HSES children excel in language ability is not surprising. What is surprising is that the Berkeley researchers would use their test of “semantic fluency” which pairs working memory with language in order to demonstrate something about working memory.

Various studies, including the first study conducted by the University of Pennsylvania researchers and that just mentioned, bear out the idea that HSES children have greater language ability than LSES. The second study conducted by Evans and Schamberg did not test subjects for language aptitude but did seem to suggest that, as did the Berkeley researchers, working memory/the prefrontal cortex is somehow morphologically bound up with language/the left perisylvian cortical region: “working memory is essential to language comprehension, reading, and problem solving, and it is a critical prerequisite for long-term storage of information.” In all of the studies beginning with the first conducted by researchers at the University of Pennsylvania, an attempt has been made to prove the existence of both prefrontal and medial temporal impairment in a sample of LSES subjects. The existence of both would justify the use of the stress hormone interpretive model discussed in the first study and called allostatic load in the second. Rather than justifying the use of the stress hormone model, however, each study has tried to attach working memory in some way to language aptitude which was found in two of the three studies to be consistently high in HSES and comparatively low in LSES subjects. Contrary to the hopeful conclusions of each study, it would seem that language aptitude is the only thing damaged by poverty.

In the past couple of years, the Educational Testing Service has released a report called “The Family: America’s Smallest School.” In the 2007 report, authors note that by the age of four, “the average child in a professional family hears about 20 million more words than children in working-class families hear, and about 35 million more than the children in welfare families hear.”[7] Also, similar to the findings of the researchers mentioned above, ETS reports that “at the highest SES quartile, 62 percent of parents reported reading to their children every day, compared to only 36 percent of parents at the lowest SES quartile.” ETS reports the findings of Child Trends, an organization that, according to ETS, gathered information from 7 research papers, reports, and books spanning the work of 19 researchers to conclude that “by the age of two, children who are read to regularly display greater language comprehension, larger vocabularies and higher cognitive skills than their peers […] In addition, being read to aids in the socioemotional development of young children.”

The famous 17th century political thinker Thomas Hobbes once supposed, unscientifically, that the function of language was the conversion of our mental discourse into verbal discourse. Having no language, he inferred, we would have no way of indexing our thoughts and therefore no capacity for memory. We can say this in another way: memory is a function of verbal ability. Perhaps there’s something to this 400-year-old conjecture. In order to raise language ability or, in the idiom of the neurophysiologists, to decrease brain damage to the left perisylvian cortical region, ETS suggests that parents equip homes with reading material and a quiet place to study such as a desk. They also recommend reading to children which, as it is said to aid the “socioemotional development” of children, may also reduce some of the damage to the prefrontal cortex that was found to be typical of the LSES subjects in all three studies.

The social critics were right after all. If a LSES environment/family is poor in substance then we must give these children substitute environments and caretakers. In the absence of parents who are not willing or able to read to their own children, Head Start type intervention and after school counselor surrogates may do a lot of good for LSES children. Some other solutions such as diverting pay and benefits from teachers and administrators toward the improvement of urban essential infrastructure such as convenience stores, libraries, police departments, and transportation, raising minimum wage, and creating more jobs for LSES families may also decrease some of the neurophysiological strain of poverty. That, perhaps, is a no brainer.


[1] Farah, M. J., Noble, K. G. & Hurt, H. Poverty, privilege and brain development: Empirical findings and

ethical implications. In J. Illes (Ed.),. (2005). Neuroethics in the 21st Century. New York: Oxford University Press.

[2] Meaney, M. J., Diorio, J., Francis, D., & al, e. Early environmental regulation of forebrain glucocoricoid receptor gene expression: implications for adrenocortical responses to stress. Developmental Neuroscience. (1996). 18, 49-72: in Farah, Martha, Poverty, Privilege, And Brain…

[3] iBid of 1

[4] McEwen BS. Allostasis and allostatic load: Implications for neuropsychopharmacology. Neuropsychopharmacology. (2000). 22: 108–124: “altered states of brain chemistry and function make the afflicted individual more susceptible to the physiological impact of life events and, in turn, more vulnerable to the impact of the stress hormones themselves. Furthermore, these considerations of stress and health are becoming useful in understanding gradients of health across the full range of education and income, referred to as “socioeconomic status” or SES” (2).

[5] Gary W. Evans, G. W. & Schamberg, M. A. Childhood poverty, chronic stress, and adult working memory. PNAS (2009) 106: 6545-6549.

[6] Kishiyama, M. M. et al. Socioeconomic disparities affect prefrontal function in children. J. Cogn. Neurosci. ( 2009). 21, 6: 1106-111.

[7] Barton, P., & Coley, R. America’s smallest school: The family. Policy Information Report, Policy Information Center, Educational Testing Services. (2007).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: