A brain area that’s specialized to recognize faces has a unique structure in each of our brains – and mapping that area’s connectivity patterns can tell us how each of our brains use it, says a new study.
The fusiform gyrus in the temporal lobe plays a part in our recognition of words, numbers, faces, colors, and other visual specifics – but it’s becoming increasingly clear that no two people’s fusiform gyrus structure is identical. By studying this region in a larger connectomic framework, though, researchers can now predict which parts of a certain person’s fusiform gyrus are specialized for face recognition.
Since the early days of neurophysiology – way back into the 1800s – scientists have been working to pinpoint certain types of brain activity to certain structures within the brain. Simple experiments and lesion studies – many of them pretty crude by today’s standards – demonstrated that, for instance, the cerebellum is necessary for coordinating bodily movement; and that the inferior frontal gyrus (IFG) is involved in speech production.
Things get trickier, though, when we try to study more abstract mental tasks. For example, debates over the possible existence of “grandmother cells” – groups of neurons whose activity might represent complex concepts like “my grandmother” – have raged for decades, with no clear resolution in sight. The story’s similar for “mirror neurons” – networks of cells that some scientists think are responsible for our ability to understand and predict the intent of another person’s action.
All these debates reflect a far more fundamental gap in our understanding – one that many scientists seem reluctant to acknowledge: To this day, no one’s been able to demonstrate exactly what a “concept” is in neurological terms – or even if it’s a single type of “thing” at all.
This is why you’ll sometimes hear theoretical psychologists talk about “engrams” – hypothetical means by which neural networks might store memories – a bit like computer files in the brain. But the fact is, no one’s sure if the brain organizes information in a way that’s at all analogous to the way a computer does. In fact, a growing body of research points toward the idea that our memories are highly interdependent and dynamic - more like ripples in a pond than files in a computer.
This is where connectomics comes in. As researchers become increasingly aware that no two brains are quite alike, they’ve begun to focus on mapping the neural networks that connect various processing hubs to one another. As an analogy, you might say they’ve begun to study traffic patterns by mapping a country’s entire highway system, rather than just focusing on the stoplights in individual cities.
They accomplished this through a technique called diffusion imaging, which is based on a brain-scanning technology known as diffusion MRI (dMRI). Diffusion imaging applies a magnetic field to the brain, causing water to flow along axons – the long “tails” of neurons that connect them to other areas – allowing the MRI scan to detect which areas are sending out lots of signals to others during certain mental activities. As you can imagine, this technique has been revealing all sorts of surprising new facts about the brain’s functionality.
In this particular study, the researchers found that during face-recognition tasks, certain parts of the fusiform gyrus lit up with active connections to areas like the superior and inferior temporal cortices, which are also known to be involved in face recognition. Intriguingly, they also detected connectivity with parts of the cerebellum – an ancient brain structure involved in bodily balance and movement, which no one expected to be part of any visual recognition pathway. Sounds like a Science Mystery to me!
The team even discovered that they could use the connectivity patterns they found to predict which faces a person would recognize:
By using only structural connectivity, as measured through diffusion-weighted imaging, we were able to predict functional activation to faces in the fusiform gyrus … The structure-function relationship discovered from the initial participants was highly robust in predicting activation in a second group of participants, despite differences in acquisition parameters and stimuli.
In short, they’ve discovered patterns of functional connectivity that directly correspond to our ability to recognize a particular face.
It’s still a far cry from an engram – after all, we still don’t know exactly what information these connections encode, or how the brain encodes that data, or what other conditions might need to be met for an “a-ha!” recognition to take place – but still, network mapping appears to be a very promising starting point for investigating questions like these.
The researchers plan to use this approach to study connectivity patterns in the brains of children with severe autism, and other patients who have trouble recognizing faces. They hope it’ll also be useful for understanding how we recognize scenes and other objects – eventually, a network-oriented approach may even offer clues about how we recognize familiar ideas.
In other words, for the first time in history, we’re using our brains’ natural love of connections to understand just how our brains form those connections in the first place. For those of us who love a good mystery, it’s an exciting time to be studying our own minds!