What Do Math Skills Look Like in the Brain?

School’s out for summer – at least in most places in the Northern Hemisphere. That means most kids will not be giving a lot of thought to math for a couple months, but for some cognitive neuroscientists, math will continue to be an important area of study. Cognitive scientists have long been interested in the neural underpinnings of individual differences in math abilities. And a new review paper suggests that people harness multiple brain regions to support complex math skills and that these neural patterns can change with effective math training.

“The inspiration for this review arose from the pressing need to enhance mathematical abilities in American children and adults,” says Xueying Ren, a doctoral candidate at the University of Pittsburgh and lead author of the review paper in the Journal of Cognitive Neuroscience. “Disturbing statistics from the 2022 National Assessment of Educational Progress reveal that only 36% of 4th graders and 26% of 8th graders achieve math proficiency or surpass it.”

Yet despite the importance of math skills to the global economy, Ren says, the neural underpinnings of the individual differences in math abilities remain largely unclear. “We hope that this review will contribute to a better understanding of the neural origins of these differences and provide insights for further research that can guide the development of effective interventions, particularly for individuals with math learning deficits,” she says. The paper specifically looked at math competence beyond basic number processing and comprehension, including the capacity to apply mathematical concepts in various contexts to solve novel problems. 

I spoke with Ren about the new paper, including what motivated her personal interest in the topic and next steps for the research.

CNS: How did you become personally interested in this research topic?

Ren: I developed a personal interest in the neural bases of mathematical cognition due to my fascination with the human brain and its intricate workings. I have always been intrigued by the variations in cognitive abilities among individuals despite having similar brain structures. This curiosity was further sparked by my own experiences, as my husband, who is a physicist, excelled in mathematics while I found myself less inclined towards it. This led me to question the underlying reasons for such differences and motivated me to explore the neural mechanisms responsible for individual differences in learning and intelligence. 

CNS: What new insight were you seeking with this review paper?

Ren: The main goal of this paper was to synthesize existing evidence on the neural basis of math abilities in both children and adults. We aimed to understand the neural markers associated with math competence beyond just focusing on one specific brain region or a limited set. For example, the intraparietal sulcus has been widely recognized for its role in math cognition. However, we propose a new perspective that considers the brain as a complex network working to minimize the energy required for information processing. We introduce the concept of the fronto-parietal network efficiency (FP-NET) as an alternative explanation of individual differences in math abilities.

CNS: Can you explain what the FP-NET is?

Ren: FP-NET suggests that the fronto-parietal network acts as a hub that efficiently integrates information from different brain areas. It forms optimal routes for rapid information transfer between various brain regions based on the specific task demands. Think of the brain networks like a map with streets connecting different locations. For simple and familiar tasks, the brain takes a direct route from one area to another (e.g., fronto-parietal), ensuring fast information transfer. However, for complex and unfamiliar tasks, multiple paths are engaged, and information is gathered from different areas and sources (like taking different roads that lead to a central hub). In some cases, the direct route may be efficient, but integrating information from different paths might be challenging, increasing the brain’s workload. So, instead of solely relying on the direct route, the brain sacrifices a bit of efficiency there and increases efficiency in other pathways to the central hub. This helps the brain process information more effectively overall. By adapting and utilizing different routes for information processing, the brain optimizes its overall efficiency, making it more efficient in handling complex tasks.

CNS: What were you most excited to find? 

Ren: I was particularly excited to discover that effective training can induce changes in brain activity, neural representations, and enhance neural efficiency. These findings suggest that tutoring-induced functional changes, observed through altered brain activation and neural representations, have the potential to be an effective intervention strategy for individuals with math learning deficits. This is especially significant for educators seeking to improve the math abilities of children facing such difficulties. 

CNS: Were there any novel papers or techniques included in your review that you would like to highlight?

Ren: A notable study by Chen et al. (2021) employed a novel network-level analysis, revealing that children with lower math abilities exhibited less differentiated neural representation patterns between different math problems across various brain regions, including the fusiform gyrus and intraparietal sulcus, compared to children with stronger math abilities. These findings suggest that a lack of distinct neural representations could serve as a neural marker for lower math abilities. More importantly, this study emphasizes the significance of examining neural representations at the network level, rather than relying solely on activation patterns within a specific set of brain regions, to gain a comprehensive understanding of the neural foundations of math competence.

CNS: What do you most want people to understand about this body of work?

Ren: The crucial message we want the general public to understand from this work is that individual differences in math abilities are not solely due to the functioning of one specific brain region but more about how different brain regions work together. By considering the whole brain network, we can develop targeted interventions that support individuals with different math abilities. 

For cognitive neuroscientists, this work challenges the conventional view that math competence resides solely in fixed cortical regions. This shift in thinking at the network level can advance our understanding of the complex relationship between cognitive skills (i.e., math abilities in our case) and the human brain, opening up new avenues for future research.

CNS: What’s next for this line of work?

Ren: An important question that warrants further investigation is whether the observed neural markers are specific to math or if they reflect more general cognitive properties applicable to a wide range of skills. To address this, future studies should thoroughly examine the influence of other cognitive factors, such as executive control and working memory, on math competence at the neural level. This comprehensive approach will shed light on the specific neural correlates of math abilities and their relations to broader cognitive processes.

-Lisa M.P. Munoz