Acupuncture Takedown

At the end of last week a study was (surprisingly) published in Nature Neuroscience claiming to justify acupuncture. I say surprisingly, because many aspects of the paper are dubious. The fundamental biochemical findings were interesting and quite promising (which is presumably how the study made it into Nature Neuroscience), but the connections to acupuncture were overly belaboured and should have been harshly criticized during review. There have been three excellent blog reviews published since the article came out, and I highly recommend giving them a read:

Ed Yong's at Not Exactly Rocket Science

Orac's at Respectful Insolence

Steve Novella's at NeuroLogica Blog

The main take-home point (and why much of the language in this article should have raised flags for reviewers) is the fact that this study demonstrated at most a plausible mechanism for the localized pain relief claims of acupuncture. The actual efficacy of acupuncture as a legitimate pain treatment modality, like any other medical treatment, still needs to be demonstrated clinically (something which it has largely failed to do despite years of research), and this study has no bearing on the non-pain treatment claims of acupuncture. Unfortunately, the article fails to acknowledge the lack of clinical support for acupuncture as a treatment modality, as well as failing to acknowledge the many aspects of acupuncture which are in no way validated by these results (non-pain treatment claims, body meridians, and all the rest of the unsupported magic an acupuncturist spends years learning), all while claiming validation for acupuncture.

What angers me the most about situations like this is that negative result studies for alternative 'medicine' modalities never receive the same sort of coverage. The prestige and respect of the journal of Nature Neuroscience will now be co-opted by the alternative medicine community to justify far more than the only somewhat plausible technique of poking people with needles to provide temporary pain relief - all, of course, for a 'reasonable' price.

I swear to think the truth, the whole truth, and nothing but the truth

Lie detection and eye-witness testimony is a messy business. There are a myriad of issues that can arise, from outright lying to false memories created by after-the-fact suggestions and rationalizations. It is the latter effect that makes the proposed use of fMRI in a civil case so troublesome, because even if fMRI were a perfectly accurate technology for identifying lies and truths (which it is not) there would still be the problem of knowing whether the person telling the truth was telling an objective truth or a falsehood he believed to be true. Although the argument could be made that fMRI would function simply as a method for screening outright liars from sincere witnesses (at which point other evidence must be relied upon to determine the veracity of the sincere witnesses' statements), there remain two major problems:

1.) Outright liars can turn into sincere but misled witnesses given enough time and repetition. Since access to MRI scans is not always available in a timely fashion, this would likely be a major issue (for example, the scan in this case comes four years after the incident).

2.) fMRI scans look so damn impressive that any jury (and even most judges) are likely to give them far more weight than they deserve. Considering the degree to which neuroscience peer reviewers can be dazzled by the colourful pictures (and these are people who spend much of their careers dealing with the nuances and limitations of fMRI) it is hard to expect a judge and jury, no matter how intelligent, to give fMRI evidence the appropriate level of ambivalence.

While I am all for the the use of science and technology to improve the justice system, new techniques must be introduced very carefully. The ease with which people can be dazzled by fancy technical words and graphics must be acknowledged along with just what exactly an fMRI 'truth scan' is showing.

Edit: According to alexismadrigal on Twitter, the use of fMRI has been rejected by the court. I found that out about five minutes after posting this... oh the magic of Twitter.

New Post at Computing Intelligence

I put up a new post over at Computing Intelligence on some sloppy language in what are usually excellent science news briefs from ScienceNOW. The subject is similar to the Animal Intelligence post and its continuation I wrote a while ago, but is a little more targeted and better referenced.

Free Will Revisited

Just over a year ago, at the request of Cornucrapia, I made a post discussing my outlook on the concept of free will. Since free will seems to be one of those topics that refuses to keep its ugly head down, Robert sent me an email mentioning that it had come up at the new atheist group at the University of North Carolina at Greensboro. While I directed him to have a look at my original free will post, I thought it might be worth rattling off some more musings on the matter. The subject of free will, after all, is ostensibly a question about the function of the brain, so one would think that I might have something sage-like to say on the matter.

As I mentioned in my original post, I am actually not a fan of the topic. My dislike is not based on finding the topic itself dull, but rather because it is such an old topic weighed down by the nonsensical baggage of eminent names that have come before it. For some reason the combination of trundling theological dogma overly concerned with the divine judgement and punishment of immortal souls, the psychologically pressing intimacy of the question, and our current dearth of information about many aspects of mental life makes the subject of free will (as well as 'consciousness') burdened by a disproportionate number of eminent thinkers from completely unrelated fields all deciding that it is a perfect problem to which they should devote their retirement treatises. The opinions of these thinkers are then bandied back and forth, all with a great deal of undeserved weight given the phenomenal intellectual prestige of the thinkers' earlier works.

While the explanation of my discontent turned a little more vitriolic than I had originally planned, it is nice to have gotten it out of the way. Now I can press on with my own meandering thoughts on the matter. One of the difficulties that plague many discussions of free will is a lack of definition. With such a vague (though intuitive) definition as "do we control our own actions", it is difficult to engage the topic in a meaningful manner. To start with, I think the anyone who brings up the debate must also seriously consider the rejoinder, "Does it matter either way?" To a great extent, the consternation gripping many people over the topic of free will rests with the theological roots I was railing against in my previous paragraph. After all, if we live in a deterministic universe (which itself is not a settled matter, but most people treat it as such), how can we be divinely judged on actions we had no choice but to perform?

Treating the matter outside of the theological realm in the domain of empirical philosophy, I admit the question of free will can still carry some weight when it comes to the issue of earthly justice (such as our criminal justice system). The justice system is a complex entity, however, and, though some people view it as such, does not exist solely for the purpose of delivering retribution. People, including criminals, are remarkably complex dynamical systems. As I mentioned in my previous post, such systems are virtually impossible to fully model, and sometimes impossible to even remotely predict, and thus we have no recourse but to act as if free will exists even if there is no mystical soul or tiny homunculus making choices. In my mind imprisonment and fines therefore remain ethical and necessary institutions. I tried to more fully elucidate my feelings on the matter, but it threatened to take over the rest of my discussion, and I had one more area that I wanted to address. If people take issue with my brief remarks on crime and punishment, let me know and I will try to more completely discuss the matter in another post.

As an atheistic scientist, I strongly doubt the existence of the aforementioned immaterial soul or decision-making homunculus. Of course, there is always the possibility of discovering some previously unsuspected aspect of our mental lives (after all, we only recently uncovered the quantum nature of photosynthesis) which makes our brains fundamentally different than other computing devices, but at the same time that does not mean we will not be able to reproduce our cognitive abilities following such a leap in knowledge. While the strong AI hypothesis (basically, that the brain is a computing device akin to any other computational model) is by no means proven, it is an open question with I think very little current evidence against it. As I said, even if our brain operates in a fundamentally different and as-yet unknown manner from a Turing Machine, every piece of evidence we currently have still points to it being a physical device beholden to physical laws. Damage the brain and you damage your mental faculties. Accepting this physical nature, however, does not equate to relegating our mental lives to that of deterministic automatons. As I have said before, we are still simply too complex to fully predict.

There is one final argument that I would like to address along the lines of neurophysiology. I do not know if I have the argument entirely correct, as I am getting the report of the argument second-hand, but it is a supposed proof against free will. Rather than further mangle the argument by summarizing it again in my own words, I will reproduce it here as it was sent to me:

When I think about moving my finger I am already moving it, and therefore the decision to move my finger must have been made before I thought to do it. Free will would, in this case, be an illusion. Because, the argument goes, there is a slight delay in the signal being sent from my brain to move my finger. Therefore, if I were to have conscious control (and actually making decisions about such things) then I would think about moving my finger, and, half or a quarter of a second later, my finger would move. Instead, at the same time I think about moving my finger, my finger moves, implying to those advancing the argument, that there must be something beyond our control in our heads making us do stuff. So, we do not have free will.

This is, to me, an almost entirely nonsensical argument. As far as I can understand it, the claim is that because one's actions appear to happen in the same instant that one thinks about doing the action, there must be some sort of unconscious automatic decision making device controlling both the thought that the move should be made and the move itself. What the argument is actually doing is basing a conclusion off of the acknowledgement of the latency of some neurological processes but not others. We do not entirely understand at what point one becomes aware of a conscious desire for action, but even assuming that the command is sent to the motor cortex at the same instant it is consciously acknowledged, there is still the latency of the visual and proprioceptive systems in checking that the command was executed. So perhaps there is a two hundred and fifty millisecond delay before one's finger starts to waggle, but one could reasonably expect an equally large or larger delay in the visual and somatosensory cortices as they decide whether or not the waggle is going on, and then report that knowledge back to the administrative cortical regions. What is actually an amazing property of our brain is that it gives the impression of a complete, simultaneous, and coherent picture of the world.

Avoiding Neuronal Tangle

Every so often, I run across a study that elegantly elucidates a solution to a problem I failed to think of, but which in retrospect appears to be a very serious problem I really ought to have wondered about. When this happens, it is both exciting (I am learning something really neat, after all) and disheartening (why didn't I even wonder about that?) at the same time. A recent study pointed out and summarized by Neuroskeptic resulted in such an experience for me.

Since Neuroskeptic has already summarized the study quite nicely and readably, I will only briefly explain what was going on here to convince you it is worth following the link. Basically, the problem that I failed to wonder about was how branching neuronal processes managed to avoid entangling themselves and mainly forming self-connections. I had wondered about the problems of axonal guidance, particularly in relation to long distance connections, but I also should have wondered about the lack of tangling in the dendritic trees and local axonal processes as well. I also knew chemical markers would be involved, as previous evidence for chemical markers guiding axon growth has been found. However, somehow developing a unique chemical marker for each neuron to keep it from entangling itself seems like something that would be rather difficult to do. Fascinatingly, though, researchers from the Department of Biological Chemistry at the Howard Hughes Medical Institute and the David Geffen School of Medicine at UCLA have found evidence for a remarkably elegant solution. I highly recommend reading Neuroskeptic's summary and checking out the paper itself.

Holographic Stimulation

There hasn't been a lot of activity in the past few days for some reason, which is odd because I think I have actually been writing both more regularly and substantively. Of course, it is summer time and apparently the rest of the northern hemisphere is actually enjoying some warm weather (people may complain about British weather, but it seems Germans don't have it much better), so perhaps people are just off having real life fun instead of sitting inside reading my ramblings. Also, I haven't actually written a lot about science lately, so it is entirely possible that what I think have been substantive posts have simply been amateur attempts at besting the triviality that so readily consumes a blogger's body of work. Ah well, I guess what I am really trying to say is I am intellectually vain and enjoy it when people at least appear to be reading what I write, so you should all tell your friends about this site.

In the meantime, here is a quick return to science. We had a symposium at the Institute today with two rather interesting talks, so I will give a brief summary of each of them (the first talk tonight, the second talk gets a summary tomorrow).

The first talk was by Dr. Christoph Lutz from the Université Paris Descartes. He was describing a new technique his group has developed for more effectively stimulating neurons optically. This requires a bit of background, though. One rather interesting experimental technique for analyzing neuronal properties is optical stimulation (technically called photolysis excitation or inhibition depending, naturally, on whether you excite or inhibit the neurons). I believe it is a fairly recent technique, but I might be mistaken. The basic idea is that you bind a neurotransmitter (in the case of the experiment Dr. Lutz described, they chose the most common excitatory transmitter in the brain: glutamate) to a specific molecule which essentially prevents normal interactions with the transmitter (this is called 'caging' the neurotransmitter). You then bathe the neurons (in this case, a slice of tissue from a rat hippocampus) with the caged neurotransmitter. The inactivating molecule has been specifically selected, however, such that in the presence of a specific wavelength of light it releases the neurotransmitter, thus allowing you to release a targeted dose of neurotransmitter as though you just activated a group of synapses by shining a laser onto the tissue.

What Dr. Lutz and his fellow researchers have done is extend the technique using optical techniques in holography. Up until now, experiments in optical stimulation have used a single column of laser light with various degrees of focus and targetting systems. Using a liquid crystal spatial light modulator, however, you can take a column of laser light and create multiple focus points, even at different focal lengths. Thus, Lutz was able to specifically stimulate along the length of a dendrite using a thin band of focused light without also activating the neurotransmitter farther away from the dendrite that the normal circular column of light would do (this extra neurotransmitter that is activated would then be free to diffuse through the local region, both weakly stimulating the neuronal membrane region being looked for an extended period of time after the laser light was turned off as well as possibly interacting with other nearby dendritic branches). By only stimulating the neurotransmitter directly along the length of the dendritic branch, you can more carefully localize the activated neurotransmitter to much more realistically simulate synaptic input. Alternatively, you are able to simultaneously focus light on multiple branches of a neuron's dendritic tree, allowing you to look at the interaction post-synaptic electric potentials generated to see how the signals interact.

Essentially, Lutz and his fellow researches have provided a novel application of well understood concepts in physics to design a much more powerful experimental technique for probing the properties of neurons. Since the computational power of a neuron rests in the electrochemical dynamics of its cell membrane, this expanded ability to probe the membrane's reaction to targeted chemical stimuli is likely to provide valuable information into the complicated world of neuronal computing.

Tomorrow: Robots with Organic Brains

Animal Intelligence Continued

I have been meaning to write an elaboration and comment response to my post on animal intelligence, but I kept putting it off for some reason. I think the main reason is because I find it to actually be a fairly daunting subject. There is a lot of nuance to it, as well as a lot of competing preconceptions and wildly differing interpretations by a lot of highly intelligent people. This makes it difficult to get a handle on the subject, but I think it is time I stop putting it off and give my attempt.

To start off with, I would like to briefly come to the defense of Noam Chomsky. Robert first brought him up saying that he "claims that humans are the only species that have the ability for language". Regan then adds his own comment in which he disparages Chomsky with a parting shot. While I think both Robert and Regan have the correct sentiment, I think there is a subtlety to Chomsky's claims that is being missed, thereby translating his claims into the territory in which they deserve the given disparagement. In the nature of full disclosure, I have not actually read a lot of Chomsky's work on linguistics, so much of my argument here will actually be based on Steven Pinker's book The Language Instinct. Throughout the book, Pinker draws heavily upon Chomsky's work and ideas, which is why I feel comfortable weighing in on Chomsky's defense. The thing is, as I understand it, Chomsky did not make the blanket claim that humans are the only species with the ability for language, but rather that we are the only species with a generative grammar based on an innate universal grammar. While I still think the statement that humans are the only species with this ability is still overly strong (there is evidence that some species of monkey are capable of a limited form of generative grammar, and our investigations into the cognitive abilities of other animals are far from complete), Chomsky's ideas are much more defensible than the patently false claim that humans are the only species with linguistic abilities (something which Regan rightly pointed out should immediately be rejected by anyone who has ever had a dog). Essentially, I think Chomsky is correct in pointing to our propensity for complex communication as one of the major cognitive advantages we have (though, like I said, I would modify his statements from we are the only species with these abilities to perhaps we are the most successful at utilizing them - much like one would correctly say a cheetah is the fastest land mammal instead of incorrectly saying the cheetah is the only land mammal to have mastered sprinting), but I do not necessarily agree with him on many of the subtler aspects of his analysis.

Speaking of unique cognitive abilities, this leads me to the apt question posed by jbrydle of whether or not our brains are fundamentally different from the brains of other creatures. The answer to this question is not straightforward, but my answer is that essentially we do not know for sure. I would now like to qualify that statement of ignorance with an extensive ramble of conjecture based on what we do know, and explain why that leads us to a state of ignorance. To start with, I would like to propose as an analogy the human hand. In plain appearance, we are capable of easily differentiating between a human hand and those of other species. However, in terms of function and arrangement, making the precise distinction becomes much harder. We are certainly not the only primate species with opposable thumbs, nor are primates the only mammalian species capable of gripping and manipulating objects with their hands (beavers, squirrels, and many other rodents have quite dextrous front paws). Likewise, elephants have adapted their noses rather than hands to the fine manipulation of objects, with the appendages on the end of their trunks capable of many of the same abilities as people. Outside of mammals, most birds have an opposable digit on their feet which allows them to manipulate and grip objects. Of course, this is quite a cursory and shallow view of the gross properties of the hand, but my point is that even something as basic and highly visible as the hand does not have a simple answer to what makes the hand of one species unique from others, especially when one is focusing on functional abilities rather than simple physical dimensions.

Moving from the hand to the immensely more complicated brain, therefore, one can see how the question is easily bogged down by nuance and qualifications. When one looks at the gross anatomy of the brain, the human brain does tend to have a more wrinkled outer exterior, known as the neocortex, than most other mammals (this increases the surface area of the cortex, since the functional setup of the cortex is a thin set of cellular layers. There are other anatomical differences as well, but I am going to try to avoid making this post into one long rendition of specific anatomical details). The neocortex is usually just referred to as the cortex, but there is more than one area of cortex in the brain as there is also a cerebellar cortex forming the outer layer of the cerebellum, which I have discussed in some detail before, so the neocortex is the term used to specifically speak about the more recently evolved cortical regions that comprise the outer shell of the cerebrum. It is also sometimes called the cerebral cortex, and it is only found in mammals. In non-mammals with advanced cognitive abilities (like many birds), quite different neural structures have evolved, often through the extension of primitive structures shared with our own brain. An example of this is in the different forms of visual perception between birds and mammals. In primates, most of our visual processing takes place in the visual cortex of the occipital lobe (the lobe at the back of one's head). There is also small midbrain structure called the superior colliculus (it is kind of like a little knob on the anterior dorsal surface of the midbrain), however, which plays an important role in visual perception. The superior colliculus provides a map of the visual field and is primarily used in spatial navigation. Its existence in the human brain allows for a condition known as type one blindsight, in which damage to a person's visual cortex has left them consciously blind. They are often, however, capable of fairly competent spatial navigation (such as walking through a room with scattered furniture without a large number of collisions) due to their superior colliculus. One interesting side-note to this condition is that because of the loss of cortical visual processing, a person suffering from blindsight is no longer consciously aware of the presence of light in their environment. Therefore, while they will be able to navigate an environment with the lights on, if you turn the lights off and ask them to walk back through the room, they will be unable to avoid collisions in the same manner (and will be rather confused about why they are suddenly having a much greater degree of difficulty).

Why am I focusing so much on the superior colliculus? Because in non-mammals it is called the optic tectum and it is particularly well developed in several predatory bird species such as owls, eagles, and hawks. Researchers have found, for example, that crows can visually differentiate between specific individuals among both crows and people. The ability to visually distinguish individuals is quite a complex task, and requires far more visual processing than that available in the standard mammalian superior colliculus. Thus, this rather long-winded example illustrates the first of many difficulties for anyone trying to definitively differentiate between the cognitive abilities of species. Because the brain is essentially a computational device providing behavioural control for an animal based on the interaction of its neurons, many different architectures can yield similar computational power. Thus, even if we have evolved novel cognitive abilities beyond those of our other ape relatives, it is extremely hard to definitively say whether or not those same abilities have independently evolved in other species.

This has grown quite long already, so I think I might stop it here. Hopefully this provides at least a little bit of explanation and feedback for the comments to my earlier post, as well as prompts some more thought and input on the subject.

Top-down Processing in Visual Perception Part IV: Ramifications

This is the final instalment of my series on top-down processing in the visual system (links to part I introducing the topic, part II discussing faces and anthropomorphizing, and part III discussing artificial edges). While I find the topics of vision and optical illusions to be fascinating in their own right, I think the analysis of perception and cognition is also vitally important. This is by no means an original outlook, as David Hume made the statement in the introduction to his A Treatise of Human Nature:

'Tis evident, that all the sciences have a relation, greater or less, to human nature; and that however wide any of them may seem to run from it, they still return back by one passage or another. Even Mathematics, Natural Philosophy, and Natural Religion, are in some measure dependent on the science of Man; since they lie under the cognizance of men, and are judged by their powers and faculties.

While couched in somewhat archaic English, Hume's statement strikes me as remarkably astute. In many ways, our brains function as vast pattern-matchers. Understanding the underlying cognitive tricks we use to analyse perception is an important endeavour for making sense of our own observations, and avoiding mistakes in our interpretation of experimental results. Of course, the most pertinent application of perceptual understanding is in automated sensory processing applications (like machine vision which I have discussed before), but as Hume pointed out, it also matters in the way our thought processes interact with every other endeavour. We must be wary of our tendency to anthropomorphize, or to view causal connections that are not actually there. Realising our tendency to perform processing without being consciously aware of it helps reinforce the necessity of mathematical, logical, and statistical tools on which to rest one's theories.

Animal Intelligence

As I mentioned in my last post, things haven't gone particularly well for me recently. As part of an attempt to ease myself back into the hustle and bustle of not feeling shut down, I picked up one of the popular science books on my shelf that I have been meaning to read for over a year now: Synaptic Self by Joseph LeDoux. It didn't hold my attention for long for a variety of reasons (some of them not entirely its fault), but one thing that quite bothered me about LeDoux's style was his habitual grouping of human beings in one set and all other animals in another. Such a species-centric view is quite widespread within general discourse, but it is also unfortunately rife within the field of neuroscience where there really is no excuse. As far as I can tell, it is a carry-over from the western theistic philosophers (Descartes and his ilk) that continues to pervade our thinking for no good reason. In the same way that angry evolution-deniers splutter indignantly that their ancestors "weren't no ape", there seems to be a general antipathy toward the suggestion that human beings share their realm of cognitive functions with other animal species.

The reason such an attitude bothers me more when it comes from a neuroscientist than a layman (though it still irks me coming from those without a neuroscience background) is because your average neuroscientist really ought to know better. The vast majority of our knowledge of neurophysiology comes from non-human species which we then extrapolate to ourselves. For example, we know more about the visual cortex of the macaque monkey than we do about the human visual cortex. For such an extrapolation to work, however, we must necessarily share the same domain as those species from which we take our starting data. Of course there are cognitive differences between species, but I think the quintessential human ingredient, the nature of humanity if you will, that people have been searching for in literature and the sciences for centuries only exists if one is willing to attach reams and reams of caveats, addendums, and qualifications. Claiming sole ownership of an ethereal conscious soul that imparts a whole new level of cognitive function for the human race is quite simply unsupported specieistic bullshit.

I could go on at length about this topic, and I am actually fairly surprised I have not mentioned it before since it is something that has been on my mind since the very beginning of this blog (my selected internet pseudonym, after all, is intended as a somewhat sarcastic allusion (hidden within the Russian language) to the apparent love affair a predominant number of neuroscientists seem to have with the human brain). Despite the temptation to ramble on, however, I really should be studying tonight, so I am going to end my rant here. I will most likely pick it up again in the future, particularly if readers take exception to any of the unqualified vitriol I have haphazardly spewed here (for example, I know a lot of people seem to hold Descartes in quite high esteem). In the nature of full disclosure, I did make up the word specieistic, and I apologize for my unimaginative cussing. It may still be a while before I am back to my usual self, so my writing for the next little while might be a little cumbersome.

Note: This discussion was subsequently expanded upon here.

"You, sir, are a mouthful"

There are two general 'facts' people know about the German language: it is harsh sounding, and it has extremely long words. I would actually disagree with the first part, or at least I think German tends to get a harsher representation than it deserves. This is because most German in popular culture is from war movies, and if people are running, shooting, and worried about killing other people or being killed themselves, they tend to be yelling rather harshly (especially when cast on the villainous side). There is a lot more to the German language than angry men shouting "Schneller! Schneller!" Perhaps Kari can weigh in here with her opinion (if she's still around...), as she has been living in Austria for almost a year.

That said, they do have some ridiculously long words. In their defence, that makes their sentences a lot less wordy, because the reason the words are so long is because German tends to simply stick words together to make new ones. Take, for example, the word for speed limit:
Geschwindigkeitsüberschreitung
That is a pretty long word. However, what if you want to talk about the maximum speed limit?
Hoechsgeschwindigkeitsbegrenzung
Those are pretty impressively long, but I came across a term while studying for my neuroanatomy exam that seems to give them a run for their money. It is the pontomesencephalotegmental complex. Why does it have such a ridiculous name? The answer, basically, is to describe where it is. The ponto part means it is located within the pons, while the mesencephalo part means it is within the midbrain (so it is located at the border between the pons and the midbrain), and the tegmental part means that it is located near the midline (within the tegmentum). The thing is, though, that people are fairly lazy. So, while the pontomesencephalotegmental complex is an informative name, nobody wants to have to say it (except perhaps when one is trying to be impressive at parties). It therefore is usually shortened to PMTC. Of course, this laziness is not unique to anatomy, but happens all over the sciences. People who write a lot of proofs get used to the fact that wrt = "with respect to", ow = "otherwise", and a small coloured in square = QED = Latin for "I'm done". Likewise in anatomy, people get sick of saying "dorsal" and "ventral" all the time so they become D and V, respectively.

This kind of shortening doesn't usually bother me, except when physiologists and anatomists get so comfortable with their acronyms that they forget to define them. I have had several lectures in physiology courses where I have had only a vague idea of where in the brain we might be talking about because everything is just an ugly jumble of capital letters. For example, SN is the subthalamic nucleus, but how is one supposed to know that it doesn't stand for the substantia nigra if one doesn't already know that substantia nigra is usually abbreviated SNr? Therefore, if there are any physiology professors out there who read my blog, I urge you to doublecheck your lecture slides and see if you use any undefined acronyms. You might not even care if your students know that structure specifically, but I would bet you that somewhere out there is a student who doesn't know that you don't care and is therefore wasting a great deal of time trying to figure out what that small collection of letters means.

Note: I seem to have misplaced my German-English dictionary and my German is rather rusty, so the German word examples were pulled from this site.

More Musings on Computational Neuroscience Paradigms

A couple months ago I posted a brief description of two overarching paradigms in theoretical computational neuroscience. During the course of writing my final project report, I addressed the same subject in slightly more detail. Since I seem to have strayed from my purported task of publishing pertinent computational neuroscience posts, I thought I would reproduce the two paragraphs in question here. I aready sent them to a friend of mine in biophysics who I know from one of my physiology courses, and he mentioned that I didn't address a couple things that I had never heard of before... so please keep in mind that this is all relatively new content for me, and the paragraphs I post here might simply be the pedestrian musings of an undergraduate amateur. Of course, they could also be brilliantly insightful, but I think the amateur option is a little more likely.

Anyway, here are the paragraphs:

Within the field of theoretical computational neuroscience, there are two general forms in which the problem of cognitive function is mathematically cast: as an adaptive control system and as a dynamical system on the edge of chaos. As with many competing fields of academic thought, disdain from adherents of one mode is often expressed for the ideas of those in the other camp. Fundamentally, the two interpretations are quite similar, as an adaptive controller functions on a dynamical system. However, proponents of the view that the brain functions as a system on the verge of chaos argue that the well-behaved systems generally analysed within the context of control theory fail to take into account the entire activity of the brain and therefore fall short of the goal of generating an accurate physiological model for cognitive function. These proponents also point to the efficacy of mathematical techniques from chaotic and dynamical system analysis to interpretations of electroencephalogram (EEG) readings, which serves as support for the near-chaotic dynamical system interpretation of the brain.

I would argue, however, that while an adaptive control experiment such as the one being implemented here seeks to isolate and investigate a specific cognitive task irrespective of the rest of the neuronal activity (or, in the case of the simulated robots used in this study, assuming no other neuronal activity), such a blinkered approach is not necessarily done out of ignorance of the larger issues of overall cognitive interconnectivity. Rather, I posit that the near-chaotic nature of the global brain behaviour arises out of the necessity of having many simultaneous well-behaved and sometimes contradictory control loops operating as one. The phase transitions apparent in EEG readings could arise from the necessity of transitioning from one set of precedent control loops to another, and a full understanding of the underlying control loops themselves can thus still further our overall understanding of cognitive function. While admittedly ad hoc, I hope this reasoning may serve to at least somewhat mollify those detractors who would dismiss adaptive control as a convenient tool of engineering misapplied to neuroscience. Continued exploration of adaptive control and implicit supervision can therefore have benefits for the field of theoretical computational neuroscience in addition to direct practical benefits in robotics.

I have removed the references, but if anyone is interested in what I am basing the discussion on, let me know and I will send you the appropriate articles.

Visceral Emotions

I have had a request to write about the physiological basis for emotional reactions to be felt in other parts of the body, namely the sensation of heartache and gut feelings, both of which fall under the category of visceral emotions (perhaps not officially, but within my own individual vernacular they at least do). There are several possible reasons for such sensations, so what I am going to discuss in this post is a great deal of conjecture and, though I expect it is fairly reasonable, might of course be refuted by careful scientific study.

There are three main aspects of our physiology that I think best explain the sensation of visceral emotions. The first is that our body is essentially formed segmentally (an evolutionary throwback to the days of worms) with a series of dermatomes (I've linked to the Wikipedia article on them for those that don't know what I'm talking about). While dermatomes are fairly well defined (though with some overlap) on the surface of our skin, things get somewhat less organized when it comes to our insides. Organs might form embryologically with one segment of skin but then get pushed and shifted around to end up somewhere else. Combined with the fact that many of our somatic sensations are not nearly as well localized as we sometimes assume they are (if you have ever done the test where, while blindfolded or looking away, someone simultaneously pokes you on the back or leg with a pair of pointy sticks separated by only a few centimetres you might know what I mean. It is very difficult (provided the poke comes simultaneously) to tell whether you have been poked in one spot or two), this can sometimes lead to a misallocation of sensation. One common example of this (at least for men) is the horrifying sensation in the pit of one's stomach after a blow to the testacles. Also, while I've never experienced them myself, other examples of misleadingly localized visceral pain include appendicitis and a hernia. Even heartburn is a misallocation of indigestion to pain in the heart.

The second major aspect of our physiology which leads to visceral sensation of emotions is the widespread autonomic responses we experience corresponding to shifts in our mental state. As we enter states of alertness, our sympathetic responses tend to be recruited and our heart-rate increases. The sympathetic nervous system is most famously known for its role as the 'fight or flight' response system, but it is engaged by other stimuli as well. Thus, while you have no intention of fighting a pretty girl or handsome guy whom you would like to go on a date with, the presence of your crush still elevates one's alertness and results in many of the same responses that result from stress and fear. I'm not actually sure the physiological reason, but acute action of the sympathetic nervous system can sometimes lead to vomiting and nausea (if anyone has ever been in an exam room in which a test taker was so nervous he lost his lunch, you have an idea of what I am talking about), which might help explain the (much more pleasant but similar) sensation of 'butterflies in the stomach'. A fairly cute psychology study was done a number of years ago in which an attractive lady stopped men for a survey in two different situations. In the first, she waited on a foot-bridge over a rather severe drop, and in the second she simply stopped men on the street. Part of the survey asked for takers to follow-up with a phone call to the researcher. Significantly more of the men who encountered the attractive lady over the gorge made the follow-up call. This was interpreted (debatably, of course, like a lot of psychology research) to mean that the men who encountered the lady in a dangerous situation found her more enticing due to a conflation of their physiological response to the fear (quickened heart and elevated alertness) with a similar response to an attractive member of the opposite sex. While I don't think the study was in any way conclusive, I bring it up now because I do think it provides some supporting evidence for the fact that there is not a unique set of physiological responses to every emotion. Rather, there is a messy and often confused interplay.

The third aspect of our physiology involves the setup of our reward pathways and their strong connection to our viscera. After all, when you think of people as survival and propagating machines, obtaining and consuming sustenance and having sex are pretty much the two main functions (we are, of course, slightly more complicated than that, but those are integral aspects of our species). One of the favourite "brain facts" espoused by clever people who sometimes like to repeat relatively inane facts in lieu of conversation is that eating chocolate provides some of the same pleasurable sensations as sex. While true, the statement is still rather misleading because virtually all food does (particularly when one is excessively hungry), since food and sex are both largely driven by the reward pathways of the limbic system. Chocolate just happens to be an especially rewarding food, and thus sounds better than, say, broccoli and cheese. Chocolate also seems to be a favourite with women, and thus using chocolate in the sentence has a greater chance of garnering a wry smile and snide response, "Oh, I think it's better," from a lady in the group as she gives her boyfriend obviously coy eyes, at which point everyone gets a good chuckle (except, perhaps, for the poor fellow with the slighted sexual prowess).

Taking these three physiological aspects together, I think we may now make a reasonable conjecture as to the nature of both heartache and gut feelings. I will start with heartache, which I am interpreting for the purposes of this post to mean a deep ache felt in the lower chest following a break-up, loss of a loved one, or some other sense of emotional loss. This sensation, I believe, mainly combines aspects of the first two physiological facts discussed. Emotional loss can be deeply distressing, thereby vastly increasing a person's stress levels and forcing a powerful response from the sympathetic nervous system. Unlike in the case of the attractive lady on the bridge discussed above, there is no positive stimulus upon which one can distract and project their feelings, and thus they are interpreted as wholly unpleasant. With an increase in heart rate and mild nausea from increased stress, I surmise that the brain interprets the sensation as an ache centred upon the heart.

A gut feeling, on the other hand, seems to involve the third aspect more than the other two. For me, at least, gut feelings are not particularly localized in the gut, but are rather a sensation that something feels like the right solution from the core of one's being. One thing which many people do not realise about the brain is that the emotional parts of the brain (which tend to be concentrated around the limbic system) actually do quite a bit of information processing and decision making (it is not all rational thought processing and planning from our frontal cortex). This is actually one of the favourite topics discussed by Jonah Lehrer, particularly since it is the subject of his latest book. I plan to write a post about it myself going into more detail, but I hope the aspects of our physiology I discussed in this post can at least give one a general idea for a possible physiological basis for gut feelings and viscerally felt emotions in general.

A Course I'd Like to Take

The other day while I was walking to school my mind was wandering between the uncertain future and reflections on my undergraduate education. As I find myself more and more drawn toward an interest in robotics and computational models of intelligence with my neuroscience background serving in a more supplementary role, I was thinking about how the neuroscience courses I have taken have served my educational development. At a university there tends to be a few aspects of a subject which are more popular with the majority of professors, and this tends to be reflected in the available courses. Here at the University of Toronto (U of T), there are approximately three ways to study the brain: behavioural psychology, microbiology and genetics, and systems neurophysiology. Of those three, I find I prefer the systems approach despite the fact that it tends to be less research-oriented than the microbiology approach (as one may guess, behavioural psychology I have the least time for). The reason I prefer the systems approach is that it tends to take a more global look at the brain and understand how it performs (though it tends to come at this from a more clinical diagnostic perspective than a theoretical modeling one), while the microbiology approach I find frustrating in its excessive detail. Thus, while the microbiology approach tends to be more research oriented, it is in avenues of research which I find to themselves be far more clinically oriented (not that clinically oriented biomedical research is a bad thing - in fact, I am expecting it at some point to likely save my life. It is simply I find the research itself mostly tedious and uninteresting).

I have gotten myself off on a tangent, however. What I intended to do was outline a course which does not exist (as far as I know) but which I would have found fascinating to take. As I mentioned, I find the systems approach to be the most appealing, but most of that approach is done at U of T with a clinical mind. When non-human animals are discussed, it is almost always in the case of a specific study with a mind to extrapolate the information to that which is applicable to understanding and diagnosing deficits in the human brain (despite the fact that we understand many of the widely used model organisms' nervous systems far better than our own). What I would find fascinating would be a course on comparative neurophysiology. For example, our cerebral cortex is, as I understand it, a mammalian novelty (and this is where most of our higher brain functions are found). Despite the avian lack of a neocortex, many birds have an odd similarity to primates in terms of cognitive function (with many extremely visual and social species). A course that examined in detail how the visual system, for example, of predatory birds compared to that of primates might be extremely illuminating in understanding visual processing techniques. Likewise, there are many non-humans which show remarkable manual dexterity and spatial reasoning (elephants with their trunks and confounding cephalopods come to mind). While I would guess that the elephant motor cortex would likely closely resemble our own due to our shared mammality, looking at the motor control mechanisms of invertebrates as dexterous as an octopus could be quite fascinating. So, if any professors happen to be reading this and know someone who might be interested in setting up a course like that, I think it would be quite worthwhile ( I just hope there are other students out there who would find the same thing if someone goes through the trouble of setting it up).

Note: I made the word mammality up. Is there an actual word that means what I was trying to say? Mammalianity?

Naughty humour with the cranial nerves

Warning: This post has some racy humour in it. Reader beware.

Wednesday I have my first (and hopefully only) university bellringer exam. This is a daunting examination technique popular in medicine and anatomy in which students are sent into a room around which multiple stations have been set up. Each station has a bit of tissue (either preserved or fresh) or some sort of medical image (an fMRI slide or some other such image) that has been marked in some way (a toothpick sticking out of the tissue or an arrow sticker on the slide) and then the student is asked to either simply name the structure or answer some sort of clinical question about the structure (for example, a blood clot in what artery would lead to loss of function in this structure?). One of the things which I have to learn for this exam is my old nemesis - the cranial nerves. For some reason, remembering the names and locations of those twelve stupid nerve fibres coming out of the brainstem eludes my memory more than any other part of neuroanatomy. I am clearly not the only person who has this problem, as there are dozens of mnemonic devices designed to help people remember the cranial nerves. These range from the rather tame, "On old Olympus' towering top a finely vested German vaults and hops" to the decidedly disturbing, "Oh, oh, oh, to touch and feel virgin girls' vaginas and hymen". Of course, while the latter mnemonic might be more memorable simply due to its shocking lewdity, it does run its own special risks as one girl in my neuroanatomy class related in our last lab. You see, the cranial nerve corresponding to the word vagina is the tenth cranial nerve called the vagus nerve. It thus shares the first three letters with its mnemonic counterpart, dramatically increasing one's risk when writing quickly on a test of starting with the 'vag' part and doing a mental flip to finish off the answer with an 'ina' rather than an 'us'. While the girl relating this story realised her mistake and rectified it, the moral of the story is clear. Filthy humour may be a wonderful memory aid, but you use it at your own risk.

Note: For the record, the cranial nerves are:
I. Olfactory
II. Optic
III. Occulomotor
IV. Trochlear
V. Trigeminal
VI. Abducens
VII. Facial
VIII. Vestibulocochlear
IX. Glossopharyngeal
X. Vagus
XI. Accessory
XII. Hypoglossal

Top-down Processing in Visual Perception Part III: Artificial Edges

This was supposed to be the concluding chapter in my series on top-down visual processing started in part one and continued in part two, but it got quite a bit longer than I expected and will thus be expanded in an upcoming fourth part. In the first installment I introduced the concept of top-down and bottom-up processing and gave a low-level example of top-down processing in the Necker Cube. In the second part I discussed faces in the context both of anthropomorphizing objects through the visualization of faces as well as a preponderance of optical illusions involving faces.

In this installment I am going to discuss another artifact of top-down visual processing which I am going to call artificial edges (I'm not actually sure if there is a better or more technical term for it, so if anyone knows of one, feel free to let me know). I find this phenomenon interesting from both a physiological aspect (in terms of providing evidence for top-down processing) as well as a machine vision aspect (in terms of duplicating our object recognition abilities). The basic idea is that our brain is fairly good at joining edges which belong to the same object but which have been in some manner obscured (either through occlusion, camouflage, or illumination problems). What is interesting, however, is that our visual processing system is so good at this that we can actually create edges and object boundaries that are not there. Two classic examples of this are shown in figures 1 and 2.

Figure 1

Figure 2

There is not actually a white square in figure 1, but the pieces removed from the black circles give the illusion of a white shape occluding them. Our mind then fills in the boundaries of the square to separate it from the white background to the point where we are able to discern boundary lines that are not actually there (of course, those lines disappear when one focuses on them since there isn't actually a change in hue). Figure 2 shows a somewhat more complicated version of the same phenomenon in which the image of a dalmatian is hidden within the scattered ink blots. The artificial edges created in these images are a consequence of our ability to group objects together and differentiate foreground from background. Part of our ability to do this rests in our expectations of what objects are most likely to appear in an image as well as our expectation of how such objects might be arranged. If figure 2 were shown to an individual who had never seen a dog (or even a dalmatian or similarly coloured dog), he would most likely have a very difficult (if not impossible) time spotting it.

That said, I am going to make a brief digression and point out one of the things that makes psychology such an annoying subject - differences in individual processing. Just as some people mentioned they had a difficult time spotting the old lady in the previous part, my girlfriend told me she does not see four circles occluded by a white square in the first image, but rather she sees four Pac-men biting a square. At the least the square is still there so she doesn't completely spoil my premise.

This ability to group contours and blobs together into expected objects is a massive advantage when it comes to image understanding, and it is one of the biggest problems in machine vision. Outside of tightly controlled circumstances, object contours rarely display consistent properties. This is hard for people to even spot, because our mind automatically accentuates valid contours and minimizes invalid contours as shading and texture.

Figure 3

To demonstrate this, I have included a picture I pulled off the internet of a chrome stapler (figure 3). As one can see, there are plenty of strong and weak edges in this image (when I speak of the strength of an edge, I mean roughly the rate at which pixel intensity changes. For a more thorough explanation, see the Wikipedia article). People have no difficulty picking out the stapler in this image and could easily outline the object if one were to ask, despite the fact that this is a monochromatic image and several of the boundary edges are much weaker than internal edges caused by shadow and geometric variations in the object's surface. For example, if you look at the two rearmost edges of the stapler, you can see that the posterior edge is virtually nonexistent while the lateral edge starts fairly strong near the bulb at the front of the stapler but fades as one moves toward the posterior. Our minds have no problem mentally accentuating that lateral edge along its entire length, however, and recognizing that it is a continuous edge despite its vast variation in edge strength. If you looks at the opposite side, however, you see a continuous dark band that extends the length of the stapler, forming two powerful edges. Neither of these strong edges actually depicts one of the object's boundaries, rather they are an artifact of the object's geometry, the view angle, and lighting. Thus, even a computer system whose sole purpose is to determine if one has a picture of a stapler or not would have a great deal of difficulty with that task without some pretty hefty processing on top of the edge detection (even then it would highly unlikely to be as reliable as a person, and we can recognize far more objects than just staplers) or some ability to constrain the view angle, lighting, and object variation. A great deal of these concepts are actually discussed in Gestalt psychology (if you follow the link to the Wikipedia article, you will see some familar images too. It looks like I could have just acquired links from there rather than searching randomly through the internet if I had looked at Wikipedia earlier).

I had planned on speaking about the ramifications of what I have discussed, but I have already been working on this post on and off for several weeks and it is starting to get cumbersome in length. I will therefore be expanding this series into a fourth post to be published in the not too distant future.

Continue reading in Part IV: Ramifications.

It is unfortunate to "propulgate" ignorance

As I mentioned might happen in my previous post, the urge to procrastinate has gotten the better of me. That is not entirely my fault, as whatever sickness I have seems to have redoubled its efforts today and made today an unfortunate combination of non-productivity and unpleasantness. To pass the time and hopefully gear myself out of a television watching, herbal tea drinking, sniveling state of feeling sorry for myself, I have decided to write about one of the most persistent fallacies about the brain I encounter. Perhaps engaging my brain at least vaguely upon the topic of neuroscience will motivate me to study neuroanatomy. Before I get to that, however, I would like to note that the title of this post, while not all in quotation marks, includes the magical set of quotations around the fake word "propulgate", making it fair game for the reference game. So far, no one has made an attempt at the other most recent addition to the game, which should have been an easy one.

Now, back to the unfortunate neuroscience fallacy. For some reason (mostly when people are trying to justify the pseudoscientific concepts of telephathy or telekenisis in bad science fiction), the idea that we only use x% of our brain (where x usually equals somewhere from 5 to 15) continues to persist in modern culture. As far as I can tell, the idea originated from the realization in the early 1900s that large sections of a person's brain could be destroyed (or lobotomized, as the medical 'treatment' was called) without causing death. However, even given that line of evidence, I have no idea how this contributed to the notion that only x% of the brain was used by the average person. Lobotomy patients, though they usually survived the procedure, were still severely changed. Their personalities were irrevokably altered, which should indicate a profound change on a neurological level. Since I have such little understanding of where this claim has its grounding, I'm not even sure how to go about properly debunking it, other than saying there is pretty clear evidence for some use of every part of the brain along several different lines of reasoning. There is the loss of function experienced by stroke patients, functional imaging which allows one to view changes in metabolic rates throughout the brain, and the simple evolutionary question of why a brain would evolve that was only utilized to a small degree. While I am a big fan of trivia and the dissemination of knowledge, one should be careful what knowledge one chooses to propagate. I know that in my precocious youth I often repeated things from questionable sources (I believe that might even have included this very 'fact'), but I would like to think such a habit has reduced as I have (hopefully) matured.

A Brief Introduction to Computational Neuroscience Paradigms

Within computational neuroscience there seem to be two main theoretical paradigms. In the first, the brain is viewed as an elaborate and nested control system. This branch of investigation tends to use many of the same mathematical models as those used in the engineering discipline of control, albeit with an eye on the biological feasibility and possible neuronal configurations necessary for attaining such a control system. In the second, the brain is viewed as a dynamical system on the edge of chaos, and thus utilizes the mathematical tools found in dynamical system analysis. I have to admit that the latter of these two paradigms I am rather fuzzy on, despite having taken (and done rather well in) a course on Chaos, Fractals, and Dynamics. I am not sure if my inability to fathom what a 'dynamical system on the verge of chaos' means is due to a lack of intellectual capacity on my part or a lack of substance underlying the fancy terms being thrown around on the part of those championing the dynamical system interpretation. My guess is that the two paradigms are not as entirely exclusive as some claim them to be, but I think I will have to gain a better understanding of the application of dynamics to physiology before I can be sure. In the meantime, the control systems approach speaks quite clearly to the (former) engineer in me, and I find the control theory approach rather appealing. It is simple, elegant, and powerful.

Before I continue in this vein, however, I should mention a brief caveat. There is a third branch of thought which I have not included in this description known as machine learning. While it could also be argued to be a paradigm of computational neuroscience (or at least my interpretation of what computational neuroscience ought to be), I have not included it in this discussion because, to me, it is much more a branch of traditional approaches to artificial intelligence. Machine learning tends to focus more on function modeling through stochastic methods. While this provides many powerful tools (some of which are even utilized within the control systems approach), there is a lack of emphasis on physiological feasibility which might provide for a general theory of intelligence. Of course, I think many of the mathematical tricks used in machine learning (like principle components analysis (PCA)) will likely have neuronal correlates found in which our brains somehow provide a system to achieve similar results, machine learning does not tend to be devoted to uncovering methods of cognition as its primary goal.

Now that I have rambled about machine learning, I shall return to control theory. A control system is essentially any system designed to control a variable through time. The actual form the control system takes can be quite varied, including electronic control systems, mechanical ones, and, as I surmise our brains might be, electrochemical. They usually utilize some form of feedback (most often negative), since an open control system (as those without feedback are called) are not really much good at controlling anything. However, I will go into more detail about control theory in another post. This post was simply meant to introduce the idea of the different paradigms, as well as the fact that I am currently more focused on control theory.

Pronunciation

I sometimes find it interesting how a mispronunciation can propagate through academic circles. For example, the Greek letter Φ often gets called 'fai' by English speakers (and virtually every professor I have ever had) when it is technically the letter 'fee'. I have had two professors not do so, the first being my intellectual and erudite linear algebra professor from first year who is vastly well read and interested in a huge variety of fields and made a conscious effort to unlearn the 'fai' pronunciation, and the other was my third year dynamics professor who had quite poor English and therefore clearly had originally learned the correct pronunciation. Unfortunately, the poor fellow was so self-conscious about his poor English (which, admittedly, was quite poor, often rendering his tests and problem sets entirely incomprehensible or quite poorly worded, such as a ball pissing across a plane rather than passing) that he ended up changing his pronunciation when he noticed his students said 'fai' instead of 'fee'.

This post is not about the Greek letter, though. Today I sat through a neuroanatomy lecture and cringed every time our professor said Wernicke's area. Wernicke's area is one of the more well-known and famous areas of the brain due to its uses in language comprehension. Located toward the posterior end of the lateral fissure and surrounding the primary auditory cortex, a person who has suffered damage to Wernicke's area (such as through stroke) suffers a form of aphasia known as either Wernicke's aphasia or fluent aphasia. That person can speak fluidly and continuously, but their speech is mostly nonsensicle. Their own comprehension of others is often likewise impaired, with the appearance of listening but very little apparent processing. As one might surmise by the name of the area, it was first described in detail by a man named Carl Wernicke. The thing is, he was a German physician. Thus, while an English speaker might be tempted to pronounce his name "Were-nick-ee", a much more correct and appropriate pronunciation would be "Ver-nick-eh" (where 'eh' is an 'uh' sort of sound, not the Canadian 'a'... I wish I knew how to do more symbols in html, but I've got to run soon so there is no time to look them up right now). Anyway, I know anatomy has a lot of strange names and it is hard to know how to pronounce them all, but this one is a major one. It just worries me that her pronunciation of all the other parts that I don't know the proper pronunciation of is also wrong, and I will have no way of knowing this until years later when I embarrass myself at a party.

Top-down Processing in Visual Perception Part II: Faces

I started this series of posts a couple months ago with Part I on the definition and role of top-down processing. When I originally wrote that post, I had meant to expound upon the topic in a more timely fashion, but I clearly became distracted and forgot about it. If there are other topics which you think I have been neglecting of late, please do not hesitate to leave a comment and I will endeavour to correct such lapses.

As I was saying in the first part of this series, optical illusions are a great way to get one thinking about how one's perceptual system works. In one particular vein of optical illusions are those that 'jump' between interpretations, the most basic being that of the Necker Cube mentioned in the previous post. I think it is particularly revealing about our visual system, however, that when one surveys a large number of optical illusions of that nature, the vast majority are devoted, in at least one of their interpretations, to faces. One of the classic examples of this is shown in figure 1, in which both the back and side of a young lady's face are visible along with the direct side profile of an old woman.

Figure 1: Old lady and young woman

There is good reason for our focus on facial perception, as it is our primary method for recognizing other individuals in social interaction. The ability to differentiate between individuals is an exceedingly important aspect of social intelligence, as there would be, for example, no way without it to differentiate between cheaters and trustworthy members of a tribe. The supreme prevalence of our nuanced ability to analyse faces, however, is often discounted by people. Interestingly, there is a condition known as prosopagnosia in which sufferers lack the ability to distinguish individual faces. There is no problem with the person's sight, but rather faces look as indistinguishable from each other as any other body part (for example, if you could only see peoples' torsos, it would be quite difficult to correctly identify others. There would of course be certain indications like weight and muscle tone, but telling the difference between a pair of scrawny teen boys or flabby middle-aged business men would be awfully difficult). The fact that something can be so selectively lost is rather indicative of quite specialised neuronal processing involved in the identification and distinguishment of faces (although it may be that there are other cognitive impairments that are less obvious). Our predisposition to seeing faces in ambiguous images or in anthropomorphising objects most readily with the appearance of a face (figure 2, 3, , and 4) indicates just how greatly our brain tries to match incoming sensory data with the expecation of seeing a face.

Figure 2

Figure 3

Figure 4

Of course, there are other forms of anthropomorphism available, but the appearance of a face resounds more greatly within us and accentuates the illusion of life. The next post in this series will look at another area of top-down processing as well as some of the ramifications.

Continue reading in Part III: Artificial Edges.

Top-down Processing in Visual Perception Part I: Introduction and Some Examples

One of the subjects I have written about before is machine vision and the incredible difficulty of developing a robust visual processing system that can equal the robustness of our own visual system. It shouldn't be entirely surprising, though, that our visual system is as incredibly powerful as it is, since a huge proportion of our brain is utilized primarily for visual processing. One of the interesting debates in perception psychology and neuroscience is whether the brain performs bottom-up or top-down processing. As with most things (especially in psychology), neither one is entirely correct and your brain utilizes a combination of the two. Optical illusions and trick images are one relatively simple way to explore the way our brain processes visual information, and they are also fairly fun to look at.

Bottom-up processing basically means your brain reads in the raw visual information captured by the retina and gradually figures out what it means as one moves farther along the processing chain that is your cerebral cortex. Top-down processing means you start with an idea of what you ought to be seeing (most likely determined by recent sensory information, other sensory clues, and your past experience). Your brain clearly does some bottom-up processing, since you react to raw changes in the visual stimuli even if there was no reason to expect that change. What is fairly surprising, though, is top-down processing is also clearly involved in visual processing. Effectively introducing top-down processing into artificial visual systems, however, is quite difficult, and it would seem that the top-down algorithms instituted by our brains (and their handy parallel architecture) are what keep us currently so far ahead of computers.

One example of top-down processing that is fairly easy to demonstrate is the blind spot. In your retina you have a small area devoid of receptors where nerves and blood vessels enter and leave your eye. This is normally not a problem since the blindspot of each eye falls on a different area of your visual field, so the sensory perceptions of one eye can compensate for the other. Also, your eyes are almost constantly performing saccades (small jumps around to focus on different regions of the visual field). However, if you close one eye and keep your other eye locked on a specific target, your blind spot becomes anchored in place. You do not realise this, though, because your brian manages to fill in that area of your visual field with its best guess as to what is there. A quick way to demonstrate this is to take a piece of scrap paper and put two X's on it about eight centimeters apart. Then close one of your eyes and stare at the opposite mark with your open eye (for example, if you closed your left eye, look at the left X with your right eye). Hold the paper about half an arm's length in front of you and gradually move it closer. At a certain point, the X on the periphery of your vision should disappear. When it does, it is sitting in your blind spot, and your brain fills in that area with it's best guess (in this case, blank white paper).

Another example that occurs slightly higher up in your visual processing is the Necker Cube, shown here.

This simple drawing forms a three dimensional clear cube. It is ambiguous, though, whether it is intended to be in one of two possible orientations: are you looking slightly down onto the cube, or slightly up at it (in other words, are the bottom two corners corners on the front or back face of the cube)? For most people, there is a default orientation when they first see it. However, after staring at the cube for a few moments, they can cause it to 'flip' into the other orientation. At no point, though, can both orientations be held in one's head at once (at least, I cannot manage to do that). It would seem that your brain takes the visual information provided about the cube's edges and then tries to fit an interpretation on it. Since more than one interpretation is possible, your brain alternates between them. However, whenever one particular interpretation is selected, the others are suppressed to avoid conflicting interpretations of a visual scene.

Continue reading in Part II: Faces.