Large scale brain activations predict reasoning profiles
For many centuries, philosophers and scientists have been fascinated by a form of reasoning that allows us to draw necessary conclusions from some premises. For example, if we hear that All men are mortals, and that Socrates is a man, we know that Socrates is mortal, by simply reasoning from these premises. This way to draw novel knowledge by means of necessary inferences is so ubiquitous in human reasoning that we often do not even realize that we are making logical inferences. However its neural bases are poorly understood. Furthermore, notoriously, different individuals follow different strategies to find solutions. In our study, we propose a new approach to understand the neural basis of deductive reasoning and explore in detail the inter-individual variability.
In general, when studying the neural basis of behavior, researchers start from a certain behavior and try to identify the underlying patterns of brain activity. We reverse this strategy and show that specific patterns of brain activity can predict what strategy an individual will use when reasoning about elementary deductive problems. We also show that the predictive power of the activity profiles is distributed in a non-uniform way across the areas activated during the corresponding cognitive operation. Thus, the activation of left ventro-lateral and occipital cortex (BA47 and BA9) predicts if participants will draw logically valid solutions. By contrast, the activations of the left lateral and superior frontal cortex ((BA44/45 and BA6/8) predict if participants will be consistent in their responses, even when they make mistakes.
We conclude that deductive reasoning can best be described as a cascade of cognitive processes that require the concerted operation of several, functionally different brain areas. In general, the activation of some areas of the left brain are essential to reason logically. Potentially, this discovery may allow future research to better characterize some cognitive deficits that attain patients with brain injuries, and set up better rehabilitation therapies. This research may also help creating novel and more efficient educational tools inspired by how the brain actually processes problems.
The work, realized in collaboration with Carlo Reverberi, Paolo Cherubini and Eraldo Paulesu (University of Milano-Bicocca), Richard Frackowiak (Centre Hospitalier Universitaire Vaudois, Lausanne) and Emiliano Macaluso (Fondazione Santa Lucia, Rome), won the Editor's Choice Award at the 18th World Congress of the Human Brain Mappin