Modern brain mapping calls on a number of non-invasive techniques, such as functional MRI or EEG, to investigate which parts of the brain participate in which neurological and cognitive functions. But, in the unfortunate event that you would suffer from a brain disorder requiring surgery, your doctor would not design her operative strategy on an fMRI scan–at least, not on fMRI alone: she would most likely resort to direct electrical stimulation of your cerebral cortex to map out the areas that are absolutely necessary for you to speak or move your hand. In a study recently published in Brain, Tate and colleagues report the results of just such a procedure in a large cohort of patients at the University Hospital of Montpellier, France. Their collected experience will prove extremely useful for clinicians and scientists involved in clinical brain mapping. In addition, their findings challenge the role of Broca’s area, traditionally thought to be crucial for speech output.
How the brain was mapped
In the study, the authors performed brain mapping during surgery in 165 patients suffering from low-grade gliomas, slow-growing brain tumors. The procedure is called awake craniotomy because direct cortical stimulation mapping requires the patient to speak and move; thus, the surgery is performed under local as opposed to general anesthesia. The neurosurgeon applied an electrode to points on the patient’s cerebral cortex and intermittently delivered stimulation while the patient was asked to count or name pictures. Electrical stimulation of a patch of cortex at frequencies of around 50 Hz for a couple of seconds transiently impaired the function of that area, offering a glimpse of what would befall the patient if that area were removed or damaged. A neuropsychologist or a speech therapist present in the operating room graded the patient’s spoken output, differentiating anomia (the inability to name objects despite being able to speak), dysarthria (improperly articulated speech), paraphasias (the replacement of a word or part of a word by another one, related either semantically or phonologically), or speech arrest (the incapacity to speak at all). Sites where speech was altered by stimulation (“hits” in the jargon) were then mapped onto a standard atlas of the brain’s surface to allow comparisons between patients. Sites that induced motor or somatosensory responses were also mapped, which confirmed the method’s accuracy in that the vast majority of those hits were found in the precentral and postcentral gyri, respectively.
The authors used cluster analysis to summarize the anatomical locations where each type of language disturbance was most commonly observed. The detailed results of the study, beautifully presented on maps of the brain surface with colored dots indicating the effect of cortical stimulation on each area, will likely be referred to often by professionals who perform similar procedures as well as neuroscientists interested in the anatomical correlates of cognitive functions.
Tate and colleagues found that dysarthria most often resulted from stimulation of the lateral precentral and postcentral gyri of both hemispheres, implicating the somatosensory cortex in the network necessary for proper articulation. Crucially, they observed that speech arrest, often considered a good marker of Broca’s area, in fact most often resulted from perturbation of the ventral premotor cortex, irrespective of whether the left or right hemisphere was stimulated. By contrast, stimulation of Broca’s area itself (the opercularis and triangularis parts of the left inferior frontal gyrus, according to the anatomical definition) very rarely caused speech arrest; patients more often produced semantic and phonological paraphasias. These findings attribute a role to Broca’s area in higher-order aspects of speech production rather than as a motor or premotor output region. Anomia and paraphasias were generally found in the superior temporal and inferior frontal regions of the left hemisphere.
Many strengths and a few weaknesses
This study is exceptional in several aspects: the high number of patients (single-center functional MRI studies of language very rarely enroll so many participants!), the rigorous protocol for testing speech production, and the clarity of the graphic representation of the results. There are a couple of shortcomings to the study as well: the anatomical location of sites that failed to disrupt speech was not recorded in detail, so that the probabilistic aspect of the cortical maps presented here only applies at the level of entire gyri. Also, by mapping stimulation sites to a template rather than to the patients’ own cortical anatomy, the authors “discarded” information that would have been helpful to investigate the variability of anatomical-functional relationships in the human cortex. Nevertheless, these minor drawbacks do not diminish the major significance of the present study in any respect.
Is there a chance that functional brain mapping with non-invasive methods will one day allow physicians to dispense with direct cortical stimulation? Not according to corresponding author Professor Hugues Duffau: “Non-invasive brain mapping using functional neuroimaging will never be able to localize language functions as accurately as invasive stimulation, because fMRI or DTI [diffusion tensor imaging, a technique that uses MRI to image fiber tracts in the brain’s white matter] are only able to provide indirect information on brain processes.”
Another frequently highlighted difference between fMRI and direct stimulation studies of cognitive functions is that the former reveals all the cerebral areas that participate in a given task, whereas the latter focuses on those that are absolutely necessary in order to accomplish it, and is thus the preferred approach in the specific situation of neurosurgery.
Direct cortical stimulation is not limited to the exploration of speech production: Prof. Duffau and his team have also probed the anatomical correlates of semantic and syntactic processing, visual-verbal congruence judgments, spatial awareness, mentalistic inferences (the attribution of mental states to others), and the anatomical correlates of consciousness. More generally, direct electrical stimulation of the brain offers unique insights into the localization of neurological and cognitive functions in humans, and future studies will likely continue to enrich our understanding of the human brain’s anatomical and functional organization.
Tate, M., Herbet, G., Moritz-Gasser, S., Tate, J., & Duffau, H. (2014). Probabilistic map of critical functional regions of the human cerebral cortex: Broca’s area revisited Brain, 137 (10), 2773-2782 DOI: 10.1093/brain/awu168