Funded Grants

Relational reasoning: Neural mechanisms, development, and plasticity

Every good teacher knows that there is no better way to help someone understand a new idea than by drawing comparisons to something she or he already knows. And so, when students ask me about a career in research, I use an analogy to describe my own work: my colleagues, students, and I have been collecting pieces of a large jigsaw puzzle, and fitting them together bit by bit. We have found some of these pieces close to home in the fields of cognitive neuroscience and developmental psychology, and other pieces farther afield in areas such as comparative neuroanatomy, education, and public health. To understand the analogy above, a student would have to glean the abstract properties that are shared across two distinct content areas, including a part-to-whole relationship, and the idea of a large project that can be tackled bit by bit. This ability to jointly consider disparate sets of mental representations, known as relational reasoning, is central to human cognition, both because it supports learning and enables abstract thought.

Indeed, it is relational reasoning itself - often measured with such tests as propositional analogies, transitive inference, and Raven's-like matrices tests - that I seek to understand:

Is this capacity unique to humans, and if so, why? Which neural mechanisms support relational reasoning and its development? Why is reasoning ability, as measured by tests of Jluid intelligence', such a good predictor of academic achievement? Can it imprave with intensive training, and if so, through what mechanism(s)?

Although reasoning ability may seem too complex and multifaceted to gain much traction on at the level of brain mechanisms, it can be broken down into a set of component cognitive processes. Over the last decade, my colleagues and I have identified the neural underpinnings of several cognitive processes that support reasoning.

Building on this research, we have examined the structural and functional brain changes that support improvements in relational reasoning over childhood and adolescence. We have also begun to identify the neural changes associated with reasoning training in young adults. Finally, working with children from low socioeconomic backgrounds, we have shown an average lO-point increase in non-verbal IQ after reasoning training.

In our future work, we seek to (1) examine the effects on reasoning ability and academic achievement of focal brain lesions early in development, with a view to developing effective cognitive interventions, (2) use transcranial magnetic stimulation in adults to test whether the regions identified by our brain imaging research are necessary for relational reasoning, and (3) use new eye movement technology to better understand individual differences in reasoning ability, and to monitor improvements in reasoning over the course of training. Our research - at the intersection of cognitive neuroscience, developmental psychology, and education - provides important theoretical insights, and also promises to yield practical benefits for society.