Let me provide you with one more summary of a brownbag lunch talk at Penn’s Center for Cognitive Neuroscience. Last December, I attended the talk and I found myself once more building my own sets of questions on top of those the presenter asked and tried to answer. The speaker, Allyson Mackey from the Gabrieli Lab at MIT, is one more scientist who is employing their academic work to tackle social problems: her research on plasticity in the developing brain explores the intersection between income inequality and education. More specifically, she investigates into individual differences in academic outcome as a response to targeted cognitive training, and the associated brain structures and functions.
Using imaging methods such as event-related and resting-state fMRI as well as DTI, Mackey investigates at the cellular level into plasticity that is induced by learning in the highly composite environments that we call schools. Her two major questions are: How does the brain shape learning? (Or: At what age are cognitive skills most malleable?) And: How does learning shape the brain? (Or: How does the brain change as students learn new cognitive skills?)
Earlier studies provided evidence that cortical thickness and the pace of cortical development correlate with cognitive abilities, and the same is assumed for the size and connectivity of the hippocampus. These research findings, however, are not easily translatable to “academic success” for several reasons. Firstly, research findings so far have focused on controlled laboratory situations and not on real-life classroom settings, and, secondly, it is not clear to what extent isolated cognitive abilities predict overall academic achievement.
Over the course of the last decades, pioneering studies have suggested that lower SES correlates with less cortical gray matter. One of the experiments Mackey introduced (Neuroanatomical Correlates of the Income Achievement Gap) aims to add to the literature on income achievement gaps. These inequalities seem to be highly stable in the US regardless of what distributions are used. Reasons for the imbalance between high-SES high-achievers and low-SES low-achievers could be differences in school quality, language exposure, expectations, and stress levels. Mackey was able to confirm that cortices are thicker in higher-income students; even if controlled for income group, greater cortical thickness is still positively correlated with test scores. However, against her prediction, this finding did not occur in a circumscribed prefrontal area. Interestingly, only the thickness in temporal, inferior parietal, and occipital regions was positively correlated with performance, not in the PFC.
Since adolescence is a time of cortical thinning, it is unclear whether low-income test subjects had less cortical gray matter from birth on, or if their cortices are thinning more rapidly. Nonetheless, from this experiment and earlier studies it is clear that brain development matters in the context of SES and school performance and that the relationship between cortical thickness and cognitive development is complex. The question of whether thicker or thinner cortices are associated with better performance may even depend on the environment individuals find themselves in. But since both SES and cortical thickness are variable, the neuroscientific approach shows promise to find biomarkers for effective educational interventions.
It is crucial, however, to carry out longitudinal studies, as Mackey emphasized by introducing another experiment (Neural Basis of Individual Differences in Achievement Gains). Students who attended a charter school in Boston improved significantly in math exams during middle school. These improvements were associated with bigger hippocampi and a higher connectivity between the hippocampi and the rest of the brain. Resting exclusively on brain data from after the achievement gain, however, it is impossible to determine whether bigger and highly connected hippocampi allowed for the improvement or if these characteristics occurred due to the targeted tutoring over the course of the three years.
Thus, Mackey and her colleagues conclude, more brain data should be acquired to trace the development of individuals’ performance and brain characteristics over a longer period of time. In addition, data on stress levels and other environmental factors that are known to influence the brain as well as cognitive abilities could be collected and tested for their correlation with brain structure, cognitive function, and behavioral data.
A historian, philosopher, or sociologist of science, however, would ask different kinds of questions. Only a tiny subset of those is provided in the following:
The historian: What is the pre-history of the claim that lower SES children have thinner cortices? Can we relate those measurements to Stephen Jay Gould’s work? When did we start conceiving adolescence as a distinct brain entity? Has the gathering of more data always contributed to “more” scientific knowledge that could be used in favor of the target population?
The philosopher: What is cognition? What is learning? Can we quantify cognitive abilities and/or learning success? To what extent does academic success relate to those?
The sociologist: Why are there certain schools that seem so prone to the study of low-SES children? How do we classify low SES? Does the structure of our education system and working environments allow low-SES students to become high achievers?
Neuroscience is neither philosophy, nor history, nor sociology. I won’t accuse Allyson Mackey of the reification of social inequalities or any sort of neo-phrenology. However, I will ask: How can she and her team, if at all, integrate the abovementioned questions into their work in order to make it more beneficial to the researched population, instead of solely providing a scientific specification of the biological effects of existing inequalities?