Brain scans, modularity, and the interpretation of subtractive images

Seeing a brain tumor and seeing a thought are not the same kind of claim. The opening grants that a brain scan can be medically useful: like an X-ray, it lets us look inside the body. The author’s concern is what happens when we treat scans as pictures of mental activity. Keep that distinction in place. The criticism does not begin by declaring the technology useless.

The psychological interpretation depends on a theory. The modular theory of mind divides mental activity into distinct components and places those components in particular brain regions. On this picture, identifying activity in a location could help us identify the mental process happening there. The issue is whether the mind really separates into those localized units.

Uttal offers an alternative: particular mental processes may be properties of a more general activity distributed throughout the brain. The author illustrates the problem with a clean division between emotion and reason. When someone gets angry, there is generally a reason for the anger. Reason-giving and emotion can be intertwined rather than two neatly separate parts of experience.

That example challenges the simple separation, but it leaves an obvious question. If brain functions are distributed, why do scans show particular areas lighting up during particular tasks? Those images seem to support the very localization the author questions. The third paragraph raises this apparent objection rather than abandoning the criticism.

To understand the response, follow how the image is made. Oxygen use serves as a measure of metabolic activity. Researchers first measure it under a control condition, then measure it while the subject performs a cognitive task. They subtract the baseline from the on-task measurement. What remains is the difference between the two measurements, not a picture of all the activity occurring during the task.

That distinction matters. A brain region could be active in both conditions yet contribute little or nothing to the difference if its measured activity stays the same. Subtracting its baseline activity would remove it from the resulting display. An area that shows an increase would stand out, even though it is not the only area doing anything.

The tempting interpretation is that whatever remains after subtraction is the activity belonging solely to the task. The author objects that this can create a false impression of neatly localized functions. The entire brain is active in both conditions, but the display removes distributed activity and draws our attention to the differences. An accurate difference between measurements can still invite a misleading conclusion about where a mental process resides.

That answers the apparent objection: localized bright areas in a difference image do not by themselves defeat the distributed-activity account. We need to distinguish what the method displays from the psychological conclusion someone draws from it.

The final question adds a reason for the modular theory’s appeal. The striking images make it easy to picture separate mental functions in separate places. But that visual appeal is part of what the author wants us to examine critically. Our main point needs both concerns: the psychological premise may be questionable, and the subtraction method can make the premise look better supported than it is.