Technology is often times overwhelmingly amazing, and the advances afforded us by science-minded engineers in particular. They give us scientists on the experiment-side of the equation the ability to push the frontier of understanding in entirely new directions. When this happens it can range from a simple shove to a complete paradigm shift. It just so happens that quite recently an article was published in Nature that might accomplish the latter, and as it is at an interface I have recently become very involved in, I figured I would take a stab at sharing the findings with you.

At the interface of biomedical engineering and neuroscience, researchers Yevgeniy Sirotin and Aniruddha Das of Columbia University have developed a “dual wavelength optical imaging technique” that allows neuroscientists to image and monitor in real-time the cortical blood volume and its oxygenation simultaneously by switching between two wavelengths (Hence the “dual wavelength” in the name). While this may seem overly boring to most of you, its implications as demonstrated just within this relatively short article alone are far-reaching.

Using two macaques as their subjects, the authors set up visual tasks to occupy the monkeys while using their newfangled gadget. What they found quite quickly, and to their amazement, was a stimulus-independent haemodynamic signal. To translate, this means that they found a change in the dynamics of the blood flow in the brain that was seemingly independent of a stimulus, yet consistent. They then realized that the signal itself appeared to entrain the the timing of the visual task, thus anticipating, rather than happening simultaneously to or after, the stimulus.

In order to observe this more closely, they set up a controlled experiment in which they could minimize the visual input to the subjects, but maintain its timing. To accomplish this the researchers placed the subjects in a dark room where the animal was taught to fix its gaze periodically on a fixation point for a reward (Juice). The fixation point itself was maintained constantly, but was switched between two colours that cued the animal to either fix their gaze on the point or to relax. The visual input, however, was so minimal that it was “askin to seeing nothing apart from one single twinkling star in an otherwise black night sky.”

Once the animals were able to learn this task the haemodynamic signal was manifest at the trial frequency, despite the environment of total darkness and the fixation point lying outside of the imaging area. The next obvious step was to test the relationship between the signal and neuronal activity, to challenge the long-held assumption that haemodynamic signals are caused by neuronal responses to a stimulus.

Through quite complex analytical methods Sirotin and Das found very enticing evidence that while visually evoked haemodynamic signals are predicted by local neuronal activity, as has been already established, this particular model completely fails under trial-related haemodynamics. Altering the trial timing unexpectedly after a subject had established a rythm of trial timing, the haemodynamic signal continued to oscillate at the earlier established period for a few trials before adapting to the new timing. Amazingly, this occured despite the animal itself adapting to the new trial pace immediately.

This finding, as conceptually simple as it appears to be, challenges the assumption that there is a linear productive relationship between neuronal and haemodynamic signals, which has implications for fMRI studies. It also raises the question of whether or not there are other exceptions to the accepted paradigm that neurosciensts have simply been failing to challenge. Secondly, and more obviously, the findings suggest that there is an anticipatory brain mechanism in place that may play a role in preparing the cortex for anticipated tasks by bringin in the proper amount of arterial blood.

(References)

Sirotin, Y, Das, A, 2009, 'Anticipatory haemodynamic signals in sensory cortex not predicted by local neuronal activity', Nature, vol. 457, pp. 475-479. 10.1038/nature07664