Socioeconomic Status and Reading Development
Taylor, Abdurokhmonova & Romeo (2023) published “Socioeconomic Status and Reading Development: Moving from ‘deficit’ to ‘adaptation’ in neurobiological models of experience-dependent learning” in Mind, Brain & Education. This is a long post but I think it’s an important one. I have edited the article extensively and put some information in bold:
Despite numerous policy reforms, income-related achievement gaps in reading scores (based on eligibility for free/reduced-price lunch) have remained remarkably consistent over the last 20+ years (National Assessment of Educational Progress, 1998-2019). In essence, models largely propose that cognitive stimulation influences the development of perisylvian association cortex, and ultimately language and literacy development, whereas increased family stress affects prefrontal and limbic development, and ultimately executive function and emotional regulation. However, there is increasing evidence that these are not fully distinct pathways and that the myriad social factors associated with socioeconomic disadvantage interact in complex ways to influence development.
Children from lower SES backgrounds are, on average, exposed to less speech from caregivers, and this language experience often differs across certain qualitative metrics. Similarly, there is an abundant literature demonstrating that stress experienced by both caregivers and children influences EF development. The human stress response is partially coordinated by the interaction between the hypothalamus, pituitary gland, and adrenal glands, or HPA axis. However, prolonged exposure to stress without protective buffers—termed “toxic stress”—can chronically elevate cortisol levels (National Scientific Council on the Developing Child, 2005). Hypercortisolism can then negatively affect several neurobiological systems associated with memory (hippocampus), emotion regulation (amygdala), and EF (prefrontal cortex) due to higher concentrations of glucocorticoid receptors. Indeed, studies have shown that higher cortisol levels are associated with reduced activation in prefrontal areas during EF tasks and reduced prefrontal cortical thickness that is in turn associated with reduced EF performance.
Recent work suggests that SES differences in children's EF skills may be partially explained by earlier differences in their language skills. This is not entirely surprising, given that both language and EF are supported by overlapping/adjacent frontotemporal brain regions that undergo rapid maturation during the preschool years. Furthermore, there is recent evidence that increased parental stress, and specifically stress associated with financial scarcity, reduces parents' child-directed speech. Also, emerging evidence suggests that poverty reduction causes changes to patterns of childhood brain functioning in ways that have been previously linked to higher language and cognitive skills.
In sum, there is robust evidence supporting multiple pathways through which SES influences cognitive development. This includes affecting the frequency and quality of children's cognitive stimulation as well as stress experienced by both caregivers and children. However, while these are hypothesized to be distinct pathways, increasing evidence suggests that they are intertwined, such that there is no single explanation for SES differences in language and EF development. One thing that is consistent across all studies reviewed thus far is that SES is positively correlated with neurocognitive functioning, such that higher SES is associated with better outcomes, while lower SES is associated with worse outcomes. However, as the field of SES neuroscience grows, there is growing evidence of a much more complex relationship between children's early experiences and neurocognitive development that necessitates evolving theories and models.
When SES is found to correlate with performance, it is thus assumed that higher SES participants exhibit better cognitive skills because they recruit more neural processing resources than lower SES participants. This is inherently a deficit perspective, suggesting that there is an optimal pattern of neurocognitive functioning (that was likely identified in higher SES convenience samples), and children from lower SES backgrounds exhibit deficient versions of these neural patterns. However, as Farah writes, “it is possible that higher and lower SES participants are performing the task in different ways, and, therefore, that high-SES participant's success is correlated with one pattern of activity and low-SES participants' success with a different one” (Farah, 2017, p. 58). Indeed, there is a growing body of literature finding that SES systematically moderates brain–behavior relationships in the domains of EF, language, and literacy.
Within the EF domain, SES has been found to moderate associations between brain activation and multiple EF components. For example, in middle schoolers (age 14) completing a working memory task, higher SES children performed better when they exhibited greater frontotemporal recruitment, while lower SES children performed better when they recruited these regions less to perform the task. Similarly, in a study of adolescent males (age 16–17) completing a response inhibition task (go/no-go), SES was correlated with activation differences in the classic inhibitory network (right inferior frontal gyrus + subthalamic nucleus + globus pallidus), despite no effect of SES on task performance. Analogous relationships are seen in studies of brain structure. In a large sample of children ranging from 3 to 21 years old, cognitive flexibility was associated with white matter microstructure and macrostructure in children from lower SES backgrounds, but no such relationship was seen in children from higher SES backgrounds. Other studies find similar patterns of SES moderating relationships between brain structure/function and performance on EF more broadly, reasoning, and attention-based rule learning.
SES also moderates brain–behavior relationships for language and reading skills. In a study of 5-year-olds completing an auditory rhyme judgment task, higher SES children exhibited positive relationships between phonological processing and activation in right superior temporal regions, while lower SES children exhibited similar relationships but with left-lateralized activation. Similarly, in a diffusion imaging study of 7–13-year-olds, higher SES children exhibited positive relationships between reading scores and integrity of multiple left hemisphere frontotemporal white matter tracts, while lower SES children exhibited relationships with reading skill in right hemisphere homologues. Additional studies find that lower SES children exhibit stronger brain–behavior correlations for language/reading skills than their higher SES peers, suggesting that higher SES environments may protect against the potential negative outcomes associated with lower baseline language skills, while the lack of resources associated with lower SES may exacerbate negative outcomes. However, a recent study of SES-diverse children with and without reading disorders found that different neurocognitive mechanisms were predictive of reading status depending on SES. Specifically, for higher SES children, reading disorders were more strongly predicted by differences in neural responses to phonological processing—often considered the “core deficit” of the reading disorder dyslexia. However, for lower SES children, reading disorders were more strongly predicted by neural responses to orthographic processing. This suggests that differences in children's early environments affect not only the neurocognitive systems called upon for typical learning and cognitive functioning but also the systems that break down in the context of learning disabilities. This has important implications for screening, diagnosis, and treatment approaches and may further contribute to SES disparities in academic skills.
Moreover, these findings point to the importance of diverse SES representation in developmental neuroscience research studies, and human subjects research more generally. Historically, both developmental psychology and neuroscience research have disproportionately relied on convenience samples of participants, which in turn disproportionately represent WEIRD (western, educated, industrialized, rich, and democratic) populations. As aptly noted by Roberts, “how diversity is dealt with in psychology both reflects and affects the ideologies of psychologists” (Roberts, 2022, p. 10). If we had a broader characterization of cognitive development and a fuller picture of heterogeneity in patterns of developmental brain and behavior changes, we might be in a much better position to understand how and why two brains might differ from one another, rather than automatically assuming that one is a deficient version of the other. Beyond informing theory and reducing stigmatization, this could have significant real-world implications; for example, in informing clinical decisions of true developmental disabilities/impairments as opposed to natural variation within the broad, normal range of development in context.
Importantly, many scholars actively advocate for greater representation in neuroscience research and to stop automatically labeling experiential differences as deficits, especially when outcomes are still within a functionally typical range. Such practices will allow for a more thorough characterization of the heterogeneity in human brain functioning and how early experiences shape brain development processes. They may also result in better translation of findings to policy and clinical/educational practice and, ultimately, to better strategies for reducing inequitable outcomes for students from diverse backgrounds. Thus, a more representative neuroscience field is critical for advancing both basic science and translational outcomes.
Just as our human ancestors adapted to varying environments to ensure their survival, present-day humans adapt to their unique environments over the course of development. Thus, we are likely to invest in attributes and skills that support our survival—in our specific context—at the expense of other skills. Adaptive models of adversity recognize this tradeoff and help to explain why certain neurocognitive disparities along the SES continuum exist in the first place. One prominent theory, the stress-acceleration hypothesis, posits that exposure to significant early-life adversity is associated with hastened maturation of brain regions associated with threat and emotion processing, such as the amygdala, hippocampus, and prefrontal cortex. Although this may advantage survival in the short term (e.g., by prioritizing associative learning and memory), it can have long-term consequences on health and later neuroplasticity underlying certain skills, such as academic knowledge. Tooley et al. expand on how SES specifically affects the pace of brain development, such that high SES provides opportunities for rare and positive events (e.g., enriching family vacations) that trigger surprise, delay brain maturation, and ultimately enhance plasticity. Meanwhile, chronic negative experiences (e.g., illness, financial hardship) increase allostatic load, encourage faster maturation, and potentially restrict plasticity. This restricted plasticity may ultimately affect language and EF skills that underlie reading achievement.
Importantly, though, such stress-adapted skills are only brought forth in environmental contexts and conditions involving a lack of predictability, which contrast with highly controlled laboratory environments. This is consistent with recent findings that lab-based measures of executive function are more strongly related to real-world academic performance measures for children from higher SES backgrounds, yet these relationships are weaker for children of lower SES backgrounds (Ellwood-Lowe, Irving, & Bunge, 2022).
To fully understand the effects of context on development, one must consider both proximal micro-contexts that are nested in macro-contexts, as well as potentially confounding relationships between contextual factors (e.g., race and socioeconomic status). In addition to developing more inclusive imaging methodologies, researchers may consider incorporating community-engaged research methods and broader community-partnership efforts.
However, better measures and methods are only the start. An evaluation of the underlying assumptions behind mainstream measures and tools is also critical to advancing our knowledge of experience-dependent brain development and maximizing translational its impact on education. When considering a body of evidence largely normed to WEIRD brains, the very notion of differences as disparities maintains oppression by upholding the privileged context as the aspirational norm. This perspective leads to intervention programs largely designed to bolster the specific neurocognitive mechanisms that have been shown to be advantageous for children growing up in higher SES environments. However, as reviewed above, the “optimal” neurobiological path toward cognitive and academic success is context dependent. What if, instead of continuously trying to usher lower SES students on the neurocognitive path taken by their higher SES peers, we aim to harness their stress-adapted relative strengths to support their own unique learning?
As argued by Lindsey, Karns, and Myatt (2010), “While we may not be able to solve the socioeconomic disparities of class within our country, we do have the moral responsibility to believe our students, make certain they understand our belief in their capacity to learn, and militate within our schools and districts for an equitable distribution of resources” (p. 51), which we believe includes the resource of neurobiologically supported learning opportunities. In sum, it is critical to further explore and seek to honor diversity in brain development, not only to expand scientific validity but also to push toward greater educational equity by viewing young learners from an adaptive, asset-based perspective as opposed to a deficit-based one.
I love this work because (1) it highlights the bias in science that favors white privilege and (2) it reconsiders brain development in ways that reject deficit models and, instead, argue for huge changes in how psychological science is conducted and what we infer from research.