Much of neuroscience rests on certain critical unproven assumptions that if wrong, could collapse much of the field as we know it today.
There are certain fundamental pillars on which huge parts of neuroscience have been constructed. Yet, at their base lie assumptions that are not proven and could crumble. And if they did, it would collapse much of the field as we know it today.
Assumption #1: All mammalian brains behave sort of the same
All brains are made of similar kinds of cells, with electrical signaling properties, and all mammalian brains have a roughly similar anatomical architecture. Consequently, since brain tissue from species like rats and mice are far easier to experiment with, it is convenient to assume that a human brain is mostly just a larger rat brain. The majority of neuroscience literature makes this implicit assumption, drawing conclusions about ‘the brain’ and how it works based on studies in rodents (and fruit flies and monkeys..). Rarely does a paper qualify that its findings refer to the rodent brain but are not yet proven in the human brain. Such qualification is simply not even expected. Yet increasingly we see that the differences between species are likely to be significant – from different types of neurons, to distinct gene expression and fundamentally different capabilities of glial cells (see related post From Mouse Brain to Human Brain). Of course, some things are the same. Yet even a single gene can fundamentally change the way neurons signal and all sorts of behaviors. And when it comes down to it, it is really the differences that matter. Consider:
If you were to study one kind of atom can you assume what you find is true of all atoms because they were all made of protons and electrons?
If you studied one organic molecule can you implicitly assume the findings extend to all organic molecules because they are all composed of similar kinds of atoms and bond types?
If you study one type of mammal can you assume their behavior can extend to all mammals because they all have common organs?
Yet, this is exactly what is done in Neuroscience. There is no qualification about where to draw the line in this extrapolation, no real need seen to do so.
Assumption #2: The brain is all about neuronal signaling
Neurons are unique and magnificent looking cells with their branching dendrites and long axons. They are also exciting in their fast electrical signaling. So much so that the field itself has been named for this cell as ‘Neuro’science and all models and theories are built on networks of neurons. Yet, neurons are not even the most common cell type in the brain. This distinction belongs to glia – not nearly as magnificent to look at nor electrically as zingy. Yet, increasingly there is an emerging picture that suggests that not only are glia not simply ‘support cells’ but may in fact be in charge. It is the glia that control the blood flow and therefore resources, and the glia that probably modulate and synchronize neurons, shaping their behavior. Yet there is hardly a model in neuroscience that considers them (see related post The Crisis of Computational Neuroscience), hardly a theory of the brain that includes them and a wild host of assumptions built on an implicit neuron-only kind of framework. Consider if aliens were only to study our cell phones and telecom networks but ignore all humans as support structures (maybe we are?). It’s not a perfect analogy of course, but it can give you a sense of how the framing completely changes the way you study something and the conclusions you draw. This could be a very big reason that real theory and insight has been elusive.
see related post Einstein, Astrocytes and EEG
Assumption #3: Bigger is more important
Another big centerpiece of neuroscience is the idea of synchronization as the functional readout that matters. Synchronization is considered in many ways in the cortex from spike measurements to field potentials at various scales, and even extrapolated at some level to blood flow measures. The essential implicit assumption here is that big events matter more. This is a nice assumption to make because big things are easier to measure and spot with statistics. Small things can be obscured in the noise. Yet, small things can sometimes be the most important. In human society for instance (consider that it is a network of brains or a meta-neuronal (?) network so may produce real parallels), the conversations that matter are not always the ones yelled in unison. Sometimes they are – but often they are not. Furthermore, it may not be the synchrony per se that matters but the underlying message of the synchrony. It could even be that the important events are in fact not the most synchronous at all and synchrony is just an aftermath or possible fallout. Consider say a conversation that led to an important action that was then tweeted about and retweeted. The tweeting and retweeting may be a big synchronous event but one that is secondary to the original conversation, and may not even occur at times. Indeed, while synchrony has yielded correlates to behavior, so far these correlations are not terribly strong and far from predictive under most circumstances. It could turn out that we have built all our theory on what can be easily measured rather than what is actually important.
The decade of the brain came and went without much to report on the human brain that was of consequence. Yet science as a collective enterprise hates questioning its assumptions, because once you are well down a path there is increasingly more at stake when you are wrong – big programs have been built, big money has been spent, expensive equipment purchased, tools created, papers published, prizes given and entire careers built on their backs. The consequences are too horrifying to consider. So, we carry on, turning these assumptions to dogma, protected by ever growing edifices of weak correlations and differences with p-values <0.05. Yet still, it’s possible that it’s all wrong and the nature of humans to question is unrelenting. Slowly but surely, the cracks will be revealed and perhaps then, the real decade of the brain will finally arrive.