Visual Cortex: A Schematic Map October 22, 2007Posted by Johan in Neuroscience, Sensation and Perception, Social Neuroscience.
1 comment so far
I came across this figure in a review by Grill-Spector and Malach (2004). It condenses an already-dense 40-page review into a single figure, so I would have to write a post of similar length to explain it entirely in laymen’s terms. This may be one post to skip if you haven’t the slightest idea of visual perception.
Even if you know your vision, this figure isn’t entirely straightforward. Still, I think it serves as a useful reference for those dense vision papers. With one or two notable exceptions, vision scientists insist on ridiculous naming conventions (the motion sensitive area hMT+ being the case in point), so this might help you remember the plot.
This map is only schematic. However, it represents the rough relationships, as they stood in 2004. The areas are mapped onto the right hemisphere occipital lobe, which has been flattened so that the dark areas represent sulcii (grooves), and the light gyrii. The posterior-anterior axis is sort of bottom-left to top-right, so V1 is (predictably) at the very back of the brain, while the Parahippocampal Place Area (PPA) is on the ventral (bottom) side.Height in this picture represents hierarchy in the processing, as Grill-Spector conceives of it. In other words, the first area is V1, and then we move up the stairs to V2, V3, and so on.
The colours code specialization. In the early areas (V1-V3), this is represented by central versus peripheral mappings, where the cortical magnification factor ensures that the centre is largest, and the highest acuity. Helpfully, they are labelled P-D (down) for the superior end of each map, and P-U (up) for the ventral end (your retinal image of the world is upside down, and apparently the visual system has no need to reverse this representation in later areas).
In later areas, the areas are filled in with one colour, presumably for simplicity – Grill-Spector actually believes that these have some retinotopic organisation as well. The colours still reflect specialization though – we can see that areas such as the Fusiform Face Area (FFA) and the Lateral Occipital complex (LO) are based on central, high-acuity representations, while other areas such as the PPA are based on more peripheral, lower-resolution representations. The letters that are strewn over the areas are meant to approximate locations of sensitivity to certain object categories: places (Pl), objects (O), and faces (F).
Do note that the Superior Temporal Sulcus (STS) is treated as somewhat of a black sheep, placed out in the corner with no colouring or height. This is probably because it is relatively poorly understood. The STS responds to biological motion, such a Johansson figures (see a demo), but its activation also appears to be strongly modulated by the social significance of the stimulus. For instance, Pelphrey et al (2005) found that the STS response in normal controls was greater when a face looked away from an obvious object rather than when gaze was directed towards it, which suggests that the STS does more than merely detect biological motion. Interestingly, people with Autism failed to show the same modulation by expectation in the STS.
The poor understanding of the STS is in part because it responds so specifically to biological motion, which makes conventional retinotopy techniques impossible. Also, I suspect there is a deep-rooted fear in some vision scientists of anything that starts with “Social.”
Another thing to note is the chasm between the last V area and the STS. Presumably, the intermittent areas are also involved in vision, but we don’t know much about what they do yet.
Grill-Spector, K, and Malach, R. (2004). The Human Visual Cortex. Annual Review of Neuroscience, 27, 649-677.
Pelphrey, K.A., Morris, J.P., and McCarthy, G. (2005). Neural Basis of Eye Gaze Processing Deficits in Autism. Brain, 128, 1038-1048.
Is is rational to Do No Harm? September 23, 2007Posted by Johan in Abnormal Psychology, Behavioural Genetics, Emotion, Social Neuroscience.
1 comment so far
From the left: Josef Mengele, Rudolf Hoess, Josef Kramer, and unknown.
The picture above comes from a set recently added to the US Holocaust museum. The pictures have caused a stir since they depict staff at the Auschwitz concentration camp on what might in modern terms be described as corporate kick-offs and the like. I’m not sure why it’s surprising that the prison guards liked to relax and have fun now and then – clearly, they would not have taken up the position if they were not at least acceptant of the task at hand. There is little evidence that the third reich forced or ordered anyone to commit these atrocities, after all.
Mengele is an interesting character. As a leading physician in the camp, he (along with the other physicians) decided who would be sent to work and who would be sent to the gas chambers, as the captives disembarked their trains. He is also infamous for his cruel experiments on inmates. In particular, he collected twins, which were separated from the other inmates, and used to study the heredity of racial traits under much the same principles employed by modern-day twin studies in behavioural genetics, but without ethics committees or indeed basic human decency.
I came across an old NY Times article on Mengele, written by what I assume must be a psychoanalyst. Yes, the usual speculative attempts to explain the man’s behaviour as a function of repressed anxiety appear, but for the most part the article sticks to the story, re-telling the life and work of Mengele through witnesses.
Although this is probably not news to historians, I am struck by the contradictions and inexactness of the accounts, even though this is very recent history. Within the NY Times article, witness accounts frequently contradict eachother: Mengele is described as being an aloof person with no emotions, next he is playful, friendly (even in his role in deciding life and death on the ramp), and entertains his young twin subjects. When comparing the NY Times article to the Wikipedia entry or his entry at the Holocaust History Project, further discrepancies arise.
Yet, a clear picture emerges, and it is one of supreme rationalism and dedication to science (albeit science that turned out to be fundamentally flawed). These are terms that are usually considered positive in our society, so you may be excused if you think me a Nazi apologist for saying so. It’s quite the contrary, however. I think Mengele’s case highlights how the idealised image of the objective Academic, struggling only to further knowledge, can be a road straight to hell.
From a rationalist standpoint, it is relatively easy to understand Mengele’s decisions. As an anthropologist with an interest in heredity, he must have recognised that Auschwitz offered an unprecedented opportunity for all kinds of forbidden experiments. The NY Times article implies that this research opportunity may have been the very reason why he actively sought a position at the camp. From a moral standpoint, the prisoners at Auschwitz were going to suffer terrible suffering or immediate death regardless – one could then argue that Mengele only tries to make the best of the situation by adding to human knowledge, while killing and maiming. For Mengele, the suffering of the prisoners was likely a non-issue in any case, since the man was a rabid anti-semite, and considered his subjects less than human. In this sense, the work may have presented no more of a dilemma to Mengele than the dilemma a contemporary researcher faces in killing a macaque monkey after the conclusion of a single-cell recording study, in order to verify that the electrodes were placed in the right cortical location.
My point here is not to defend Mengele – he was clearly an appalling person who, apart from all other damage done, sullied the name of science. Still today, Mengele is the original template for the evil scientist, who seeks knowledge at any (human) cost. But it is unsatisfying to merely state that Mengele was “evil”, and thus did what he did. The NY Times article finally lets loose the full-on psychoanalysis towards the end, and this explanation proves no more satisfying:
His impulse toward omnipotence and total control of the world around him were means of fending off anxiety and doubt, fears of falling apart – ultimately, fear of death. That fear also activated his sadism and extreme psychic numbing.
I would prefer to invoke the behaviour of patients with damage to the ventromedial prefrontal cortex (VMPFC -the bit of brain just above and between the eyes). These patients display, among other symptoms, what can best be described as a lack of conscience. They are well aware of the conventions of society, but as the post linked above describes, their reasoning is defective. The defects do not appear in the expected, irrational direction, but rather in a rationality that is so extreme that is leads to horrendous decisions. These patients do advocate killing a crying baby in order to avoid the group’s capture and certain death at the hands of enemies. This is the most rational, utilitarian path to take – better the death of one than the death of all.
So did Mengele have an undetected lesion to the VMPFC? I don’t think so, and there is absolutely no way of finding out. Such an account would be as speculative as the psychoanalytic drivel cited above. I only wish to raise the possibility that sometimes, a behaviour as complex as immorality or a lack of conscience may arise from relatively simple lesions. Repressed traumas and anxiety may well cause such cruel behaviour too (who knows?), but a blow to the head strikes me as the more parsimonious explanation, if we’re going to speculate about it anyway
I don’t think Mengele was mad, evil, or suffered from repressed anxiety. He was a dedicated and supremely rational scientist. This is why he caused so much harm.
How useful is a group-level truth to the individual? September 18, 2007Posted by Johan in Applied, Economics, Math & Stats, Social Neuroscience.
add a comment
Today’s post is about the tension between the often-times weak predictions of psychology and other sciences, and the decisions of the individual. My blog stats might serve as an example. The figure below plots the number of visitors over the last 30 days.
You would be hard pressed to make any predictions regarding how many visitors the blog will have tomorrow. Nor would you be able to say what the overall trend in the data is. The story is similar when we look at the number of visitors per week:
You can now predict that the blog should receive somewhere between 400 and 700 visitors this coming week, but it looks like the blog is receiving pretty much a constant stream of visitors. This is unlikely, considering how little blogging has taken place lately. Let’s look at the number of visitors by month instead.
At this level, the trend is very clear. The blog was growing steadily up until June, which is almost exactly when I went on a summer break.
The point here is that trends may emerge from a set of data that looks thoroughly chaotic on a lower level. This is a familiar story to psychologists, economists, and others seeking to predict human behaviour: predictions are only valid on the group level (at best!). In the example above, I averaged over time to produce a trend, while predicting human behaviour usually involves averaging over many data points (i.e., people) instead. The common principle of patterns emerging from noise holds, however.
So far, so obvious. However, the corollary of this is that when it comes to the individual case, most predictions that scientific psychology offers are so weak that they are next to useless.
For example, while I and other scientifically-oriented psychologists mock the various psychoanalysts and other psychotherapists who come up short compared to cognitive behavioural therapy in the research, the differences between the different therapies in terms of success rates are rather subtle. For a government welfare program where it is likely that thousands of clients will use the service, it makes a lot of sense to go for the most supported remedy, as the subtle differences between them become quite noticeable here. However, as a depressed individual, the difference in efficacy between the different therapies is trivial.
I think this is part of the reason why there is a considerable rift between clinical psychology and research psychology. The researcher looks at the group statistics and sees clear advantages for one remedy or another, while the clinician meets one individual after another, and is thus exposed fully to the amazing variability in how well a given approach works.
Similar concerns arise in epidemiology, the branch of medicine that deals with prevention rather than cures. The NY Times has an excellent article on the methodological difficulties that epidemiology faces, which I will only sample from. The ideal randomised double-blind control trial (where patients randomly receive treatment or placebo, and neither patient nor researcher knows who receives what) is very expensive, and thus much epidemiological research relies on weaker association studies, where patients usually fill out questionnaires about various health practices (e.g., how often do you exercise?), and the researchers then see how well the sample gets on. The goal is to correlate diagnoses or mortality rates with the questionnaire scores in order to discover what causes the disease. Or rather, what behaviours are associated with the disease.
The biggest feather in epidemiology’s hat is the link between cancer and smoking. However, as the NY Times article points out, this effect was truly magnificent, with smokers experiencing an increase in the risk of certain cancers by thousands of percent compared to non-smokers. Current studies into the health effects of hormone replacement therapy, vitamins, and omega3 fish oils deal with far smaller effect sizes, and thus other explanations for the observed associations become more probable.
The NY Times article does a good job explaining why the finding that people who take vitamins live longer doesn’t necessarily mean that vitamins keep you alive. I want to emphasise a different aspect of these studies: because the change in mortality or disease incidence is usually counted in the tens of percent at best, following this advice really isn’t going to make a dramatic difference to the individual. Much like the depressed patient choosing therapies, the effects of making these “fine tuning” health choices are so small that you might as well not bother (see also a previous post on the link between birth order and IQ).
But of course most of us do bother, myself included. No one wants to die any earlier than they necessarily have to. Unfortunately, due to the weak nature of the evidence upon which we base these decisions, it is not outside the realm of possibility that we aren’t just wasting money, but also damaging our health. The case in point would be hormone replacement therapy (HRT), which eases the strain of menopause. As the NY Times article outlines, HRT was originally found to have protective effects on mortality, if anything, and the drugs became massively popular. Later research has instead found that HRT may in fact increase mortality, only to settle on an uncomfortable juxtaposition where the experts deem that HRT is harmful if you start later in life, but helpful if you start as you go through menopause. There are probably smaller risks associated with vitamins and other supplements, but it isn’t necessarily the case that you stand nothing to lose by keeping up with the latest health craze.
The NY Times article advices the reader to simply ignore most health advice (including that regularly published in the Times, one wonders), unless the effect sizes are very large (as in smoking and cancer), or if there is what the author terms a “bolt from blue sky” effect (as in the link between asbestos and certain types of cancer), where few alternative accounts of the association are plausible.
So far, the conclusion is rather glum: psychology, epidemiology, and other sciences not covered here may have little firm advice to offer the individual. It’s worth remembering, though, that this conclusion is only valid for the individual, and at a single point in time. For example, there are studies to show that mirroring the poses and postures of the person you’re speaking to tends to result in that person liking you better. This helps me little in that one crucial employment interview since the effect is subtle, and thus unlikely to prove a deciding factor in whether I get the job or not. However, if I were Machiavellian enough to make posture mirroring my habit, over time it might work out in my favour, much like how patterns emerge in the blog stats above, if you average over long-enough time periods.
“Feel the heat… skips a beat” or the other way around? August 16, 2007Posted by Johan in Emotion, Neuroscience, Social Neuroscience.
1 comment so far
Can’t go with my heart when I can’t feel what’s in it
Red House Painters – Katy Song
The common-sense understanding of emotions and their physical expressions is that the causal arrow goes one way only: if you feel upset, the feeling will cause your cheeks to flush and your heart to beat faster. However, the opposing notion is also possible: we experience physical arousal, and interpret this arousal as emotion. This is referred to as the James-Lange theory, or simply as feedback theory. This alternative idea is not much younger than the common-sense notion – William James was probably the first to propose it, in the 19th century. More recent simulationist accounts such as Damasio’s Somatic Marker Hypothesis could be said to build on this notion.
A story in the news (Washington Post via Gizmodo) seems to offer some anecdotal evidence for the role of physical arousal in the experience of emotion. Peter Houghton had a “ventricular assist device” installed, which is a developed version of the artificial heart. Unlike the more established Pacemaker, the Jarvik 2000 that Houghton had implanted not only paces the heart but moves blood for it, exchanging the familiar thuds with an even mechanical whir (as an aside, the Jarvik 2000 only helps the right ventricle, since the left ventricle is often left unharmed in cardiac illness – in other words, there is a bit of heartbeat left).
While this is impressive in itself as an example of the advances in medicine, the side-effect is what captures attention and perhaps imagination: Houghton claims to experience less emotion. For instance:
[...] he’s become more “coldhearted” — “less sympathetic in some ways.” He just doesn’t feel like he can connect with those close to him. He wishes he could bond with his twin grandsons, for example. “They’re 8, and I don’t want to be bothered to have a reasonable relationship with them and I don’t know why,” he says.
Judging by the Post story, the emotional flatness seems to occur mainly with regards to other people. This is a little surprising since heart rate is considered more of a marker for fear (which is associated with a raised pulse) or interest (which can lower the pulse) than for happiness or attachment. It is possible that these emotions are also affected, but this went unnoticed in a modern world where there are many opportunities to experience emotional attachment, but few to experience genuine fear.
As the Post article is quick to point out, it remains possible that the emotional symptoms are secondary to general depression, brought on by the ordeal of a life-threatening cardiac illness. However, one of the physicians hints at previous reports of these emotional symptoms occurring following implants. I haven’t been able to find any references – feel free to give me a shout if anyone should find some.
Generally speaking, I’m quite open to the idea that the removing one physical marker of emotion (ie, changes in heart rate) could produce a corresponding reduction in the experience of certain emotions. However, the complexity of the deficits in this particular case is perhaps more consistent with depression. The loss of a physical marker should – if the feedback has merit – result in more global deficits in the experience of basic emotions like fear or interest.
N170 face controversy continues June 27, 2007Posted by Johan in Face Perception, Neuroscience, Sensation and Perception, Social Neuroscience.
1 comment so far
To recap: the Thierry et al (2007a) paper is interesting because it challenges the notion that a specific component of the EEG waveform called the N170 (since it is negative, and occurs at 170 ms) is specific to faces. Thierry et al found that the N170 disappeared when they controlled for inter-stimulus perceptual variance (ISPV), that is, the fact that faces tend to be presented in portraits, while other stimuli are often shown at various angles and sizes.
As the comments on the Neurocritic’s post suggest, some investigators were not entirely convinced by Thierry et al’s (2007a) demonstration. Now one of the critical comments has made its way into the latest issue of Nature Neuroscience, along with a reply from Thierry et al (2007b).
Bentin et al (2007) point to previous research that shows how controlling for ISPV does not in fact explain the N170 specificity to faces. They’ve packed a bit too much information into this figure than necessary, but it’s worth a look:
The black line indicates the mean pixel-wise correlations within each stimulus group, that is, which is an indicator of how much ISPV there was. The grey bars show the N170 amplitude. If Thierry et al (2007) is correct in their argument that the N170 reflects ISPV rather than face specificity, you would expect the black line (indicating ISPV) and the grey bars (indicating N170 amplitude) to match up reasonably well. This isn’t really happening. To make this point even clearer, they show in the supplements that Thierry et al’s (2007) own data seems to indicate the same thing. Bentin et al (2007) also take Thierry et al (2007a) to task for failing to note how their conceptualisation of the N170 contradicts a bulk of well-known effects in previous literature.
So how could Thierry et al (2007a) get such contradictory results? Bentin et al (2007) suggest that one reason may be Thierry et al’s (2007) choice of EEG recording sites, which differ from those generally used by other investigators.
In their reply, Thierry et al (2007b) argue that, among other things, the recording site explanation doesn’t hold, since the same pattern of results appeared across all electrodes. They also contend Bentin et al’s (2007) notion that there is a generally agreed standard for electrode selection.
As for the lack of similarity between the pixel-wise correlations and the N170, this is explained (if you can call it that) by arguing that pixel-wise correlations may not be a perfect measure of ISPV – individual pixels may have very different effects depending on their location, which is not taken into account with such an analysis.
Confused yet? I don’t think there are any clear answers at this point. At the most basic level, the findings of Thierry et al (2007a) contradict previous findings, something that they appear to have failed to mention themselves. There is also some disagreement over what type of analysis is appropriate for this kind of research. Personally, I would like to see an independent replication of the results before I make any further attempts to understand what’s going on.
Bentin, S., Taylor, M.J., Rousselet, G.A., Itier, R.J., Caldara, R., Schyns, P.G., Jacques, C., and Rossion, B. (2007). Controlling interstimulus perceptual variance does not abolish N170 face sensitivity. Nature Neuroscience, 10, 801-802.
Thierry, G., Martin, C.D., Downing, P., & Pegna, A.J. (2007a). Controlling for Interstimulus Perceptual Variance Abolishes N170 Face Selectivity. Nature Neuroscience, 10, 505-511.
Thierry, G., Martin, C.D., Downing, P., & Pegna, A.J. (2007b). Is the N170 sensitive to the human face or to several intertwined perceptual and conceptual factors? Nature Neuroscience, 10, 802-803.