Why Ogi Ogas and Sai Gaddam Are Phrenologists
Why Ogi Ogas and Sai Gaddam Are Phrenologists
Although I believe it would be edifying for our two favorite 'cognitive neuroscientists' to read this essay, they are not my audience. Rather, I write to fandom, to hopefully shed some light on why these two wankers think they have scientific validity, and why they in fact do not.
sabrina_il wrote a fantastic essay that is about psychology and what it can and cannot tell you. I am going to try and talk more about the history of neuroscience and why I keep referring to the two Dr. Fails as phrenologists (which is a pretty hefty insult among neuroscientists).
Simply put, phrenology was the 'science' and 'studies' surrounding an old theory (most prominent in the 1800s and early 1900s) that the personality and abilities of a person could be determined by examining their skull, on the assumption that the skull reflected the qualities of the brain underneath it.
This is not in fact true.
The shape of the skull has no relation to the properties of the brain underneath it. None. There was no scientific validity to phrenology then, and there is no scientific validity to phrenology today.
While it was still in vogue, phrenology was used to promote and confirm the racism and sexism present in that time period. What better way to confirm that white people were superior to everyone else than to measure their brains? What better way to confirm that women and inferior to men than to note that their heads are smaller?
Eventually scientists took the one idea that phrenology kind of got right (that the brain was somewhat modular; there are different areas devoted to different tasks) and rightfully shoved phrenology into a corner and tried to pretend they were never that stupid.
This brings us to Ogi Ogas, Sai Gaddam, and modern day cognitive neuroscience.
Cognitive neuroscience is the science of linking cognitive processes to the underlying neural structures. This is a fairly new field, and the primary reason for this is that until recently there simply weren't very many ways to look inside a live brain, and there's only so much a dead one can tell you.
Common imaging technologies include EEG, PET, and the current gold standard, fMRI. fMRI is a beautiful thing. It is also incredibly complicated, and our best machines still only have a resolution of about 1-2 mm. This is fantastic compared to any imaging techniques that came before, but is still much larger than a neuron and therefore does not let us take in everything that the brain is doing. In addition, the temporal resolution on fMRI is good but not great, so we still perceive actions at a lag rather than seeing what happens in 'real time'. It is also very difficult to image the subcortical structures (which are deep inside the brain) or certain areas of the brain near sinus and other physiological features in the head.
Possibly the only truly correct aspect of imaging studies Ogi Ogas and Sai Gaddam have mentioned is that often people extrapolate too far from imaging studies. Imaging studies are notorious for revealing differential activation in one area of the brain and then having the researchers (or the PR people, which is worse) stating that they have found 'the area of the brain that does x'. I promise you, if you ever read something that says 'this area of the brain does x' and 'x' is ANY kind of complicated behavior, you aren't getting the whole story.
...oh, and I forgot my favorite problem with imaging studies. We don't actually have a very good idea of which areas of the brain are where in different people. There's a lot of variation. So when doing imaging, researchers average all of the scans together and do a hell of a lot of processing on the images. It's not at all like taking picture or a video. Which just makes 'this area of the brain does x' even more problematic.
So what are Ogi and Sai trying to do? They are, apparently, modelers. Which means they are attempting to build a model which will mimic the human brain well enough to have predictive power. They used as comparison models of the visual system, which are paragons of simplicity compared to social behavior *or* the subcortical structures. Their examples used the responses of single neurons in the model (whereas now, they propose to use whole STRUCTURES). At one point they actually said that the human visual system was *more* complicated than subcortical structures because they developed later in human evolution. The older = simpler idea is a fallacy so great any evolutionary biologist would find it worthy of beating their heads in, but what can you say. They aren't biologists anymore than they are social scientists.
Needless to say current research is nowhere *near* ready to start putting together a *neural* computer model of human behavior. Cutting edge research today has people finally being able to predict from imaging studies *what* a person is looking at rather than just the fact that they are looking at something. This is a pretty big step forward. It's three worlds removed from being able to link neural structures to social behavior.
Essentially Ogi and Sai looked at current imaging research and said 'I don't think their data justify their conclusions. I know! Let's START with the conclusions and make our data up as we go! We can totally make a model to predict what we already know!' This is why I called them phrenologists. Because they are starting from bad data, and confirming themselves as they go.
It's a testament to how screwed up their ideas are that I can't even link in how they thought their gigantic survey would help them. When people model the brain right now they tend to look at simpler responses; perception of sensation, simple movement. Erotic thought? Social Interaction? No.
There is an area of research right now looking at moral decisions, and finding that they are emotional rather than rational (and hey! arbitrary!). The imaging studies that support these conclusions also show that the areas of the brain implicated in moral decision making encompass most of the cortex and subcortical structures as well. Essentially showing that for higher-order (the highest order) interactions such as those defined by CULTURAL VALUES we use many, many areas of our fascinating and complicated brain. Not individual subcortical structures, but a detailed and right now ineffable interaction between subcortical structures and our cortex.
Ogi Ogas and Sai Gaddam are attemping to profit off of bad science by cloaking it in complicated terminology and cutting edge technology. We shouldn't let them.
ETA: I now have a second post talking more modeling and what I think Ogi Ogas and Sai Gaddam intend to do with the 'data'.
Although I believe it would be edifying for our two favorite 'cognitive neuroscientists' to read this essay, they are not my audience. Rather, I write to fandom, to hopefully shed some light on why these two wankers think they have scientific validity, and why they in fact do not.
sabrina_il wrote a fantastic essay that is about psychology and what it can and cannot tell you. I am going to try and talk more about the history of neuroscience and why I keep referring to the two Dr. Fails as phrenologists (which is a pretty hefty insult among neuroscientists).
Simply put, phrenology was the 'science' and 'studies' surrounding an old theory (most prominent in the 1800s and early 1900s) that the personality and abilities of a person could be determined by examining their skull, on the assumption that the skull reflected the qualities of the brain underneath it.
This is not in fact true.
The shape of the skull has no relation to the properties of the brain underneath it. None. There was no scientific validity to phrenology then, and there is no scientific validity to phrenology today.
While it was still in vogue, phrenology was used to promote and confirm the racism and sexism present in that time period. What better way to confirm that white people were superior to everyone else than to measure their brains? What better way to confirm that women and inferior to men than to note that their heads are smaller?
Eventually scientists took the one idea that phrenology kind of got right (that the brain was somewhat modular; there are different areas devoted to different tasks) and rightfully shoved phrenology into a corner and tried to pretend they were never that stupid.
This brings us to Ogi Ogas, Sai Gaddam, and modern day cognitive neuroscience.
Cognitive neuroscience is the science of linking cognitive processes to the underlying neural structures. This is a fairly new field, and the primary reason for this is that until recently there simply weren't very many ways to look inside a live brain, and there's only so much a dead one can tell you.
Common imaging technologies include EEG, PET, and the current gold standard, fMRI. fMRI is a beautiful thing. It is also incredibly complicated, and our best machines still only have a resolution of about 1-2 mm. This is fantastic compared to any imaging techniques that came before, but is still much larger than a neuron and therefore does not let us take in everything that the brain is doing. In addition, the temporal resolution on fMRI is good but not great, so we still perceive actions at a lag rather than seeing what happens in 'real time'. It is also very difficult to image the subcortical structures (which are deep inside the brain) or certain areas of the brain near sinus and other physiological features in the head.
Possibly the only truly correct aspect of imaging studies Ogi Ogas and Sai Gaddam have mentioned is that often people extrapolate too far from imaging studies. Imaging studies are notorious for revealing differential activation in one area of the brain and then having the researchers (or the PR people, which is worse) stating that they have found 'the area of the brain that does x'. I promise you, if you ever read something that says 'this area of the brain does x' and 'x' is ANY kind of complicated behavior, you aren't getting the whole story.
...oh, and I forgot my favorite problem with imaging studies. We don't actually have a very good idea of which areas of the brain are where in different people. There's a lot of variation. So when doing imaging, researchers average all of the scans together and do a hell of a lot of processing on the images. It's not at all like taking picture or a video. Which just makes 'this area of the brain does x' even more problematic.
So what are Ogi and Sai trying to do? They are, apparently, modelers. Which means they are attempting to build a model which will mimic the human brain well enough to have predictive power. They used as comparison models of the visual system, which are paragons of simplicity compared to social behavior *or* the subcortical structures. Their examples used the responses of single neurons in the model (whereas now, they propose to use whole STRUCTURES). At one point they actually said that the human visual system was *more* complicated than subcortical structures because they developed later in human evolution. The older = simpler idea is a fallacy so great any evolutionary biologist would find it worthy of beating their heads in, but what can you say. They aren't biologists anymore than they are social scientists.
Needless to say current research is nowhere *near* ready to start putting together a *neural* computer model of human behavior. Cutting edge research today has people finally being able to predict from imaging studies *what* a person is looking at rather than just the fact that they are looking at something. This is a pretty big step forward. It's three worlds removed from being able to link neural structures to social behavior.
Essentially Ogi and Sai looked at current imaging research and said 'I don't think their data justify their conclusions. I know! Let's START with the conclusions and make our data up as we go! We can totally make a model to predict what we already know!' This is why I called them phrenologists. Because they are starting from bad data, and confirming themselves as they go.
It's a testament to how screwed up their ideas are that I can't even link in how they thought their gigantic survey would help them. When people model the brain right now they tend to look at simpler responses; perception of sensation, simple movement. Erotic thought? Social Interaction? No.
There is an area of research right now looking at moral decisions, and finding that they are emotional rather than rational (and hey! arbitrary!). The imaging studies that support these conclusions also show that the areas of the brain implicated in moral decision making encompass most of the cortex and subcortical structures as well. Essentially showing that for higher-order (the highest order) interactions such as those defined by CULTURAL VALUES we use many, many areas of our fascinating and complicated brain. Not individual subcortical structures, but a detailed and right now ineffable interaction between subcortical structures and our cortex.
Ogi Ogas and Sai Gaddam are attemping to profit off of bad science by cloaking it in complicated terminology and cutting edge technology. We shouldn't let them.
ETA: I now have a second post talking more modeling and what I think Ogi Ogas and Sai Gaddam intend to do with the 'data'.