Philosophy of Cognitive Neuroscience

Cognitive neuroscience emerged as a field in the late twentieth century and brought dramatic changes to our ability to systematically investigate the relationship between the mind and brain. The philosophy of cognitive neuroscience seeks to clarify conceptual issues in the methods and inferences employed by cognitive neuroscientists.

One central point of debate in the field involves notions of functional specialisation and localisation in the brain – that is, whether and to what extent the brain can be broken down into specialised components that perform discrete functions related to cognition and behaviour. A foundational method in cognitive neuroscience involves inferring the specialised functions of brain regions from patients who have suffered localised brain damage, similar to how one might reverse engineer a machine by systematically removing its components and seeing how its behaviour changes. In cognitive neuroscience, these inferences are not as straightforward as they first seem because, in complex systems like the brain, damage to a single component may produce behaviours that are only indirectly related to the function of that component.

Neuroimaging studies aim to make similar inferences about the functions of brain regions by observing the brain activity of healthy subjects during a variety of task and control conditions. Neuroimaging is also used to draw conclusions about the psychological states of subjects from measurements of their brain activity. Both forms of inference are controversial in part because individual regions of the brain are often active during a diverse array of tasks. This many-to-one relation between tasks and the activity of neural areas has cast doubt on the idea that specific areas perform specialised functions. It has also generated concerns that the correspondence between brain activity and psychological states is too loose to determine anything substantive about subjects’ psychological states from brain activity alone.

These debates about functional inferences in both neuroimaging studies and studies of patients with brain damage have generated reflection on whether and how neuroscientific evidence should impel revisions to our taxonomy of psychological states. Such revisions may enable a tighter correspondence between psychological constructs and neuroscientific constructs. One problem with this proposal is that it is unclear what the appropriate unit of analysis within the brain is for such purposes. Without such a unit, it is unclear how neuroscientific evidence can anchor revisions to our psychological taxonomy.