So let me share another link. This one’s on the teenage brain and the adaptive account of adolescence:

National Geographic: Teenage Brains

It’s a good read, which lists the major insights of studies that examined the adolescent brain. Still, I feel that I should add a bit of disclaimer to stories like these.

The development of the adolescent brain has been instructive to cognitive neuroscience. Adolescents are healthy, but show some weird behavior, which is nice if you want to correlate behavior to physical processes in the brain. As such, the studies have been worthwhile to neuroscience as a whole.

Yet, we shouldn’t accept any of the neuroscience stories as a full explanation for the behavior of adolescents. It’s true that if the rewiring and maturation of the brain is part of some inescapable, timed cascade of chemical reactions, this program could be at the foundation of adolescent behavior. In every age of history, the unfolding brain would have been noticeable through different sorts of behavior, but its biology would have been the same.

However, we don’t know what drives the rewiring and maturation of the brain. It may be that these carefully documented rear-to-front brain alterations are happening in brains that are encultured in our society, and in our society alone. It may be that the way we treat kids is reflected in their brain development just as much as their own behavior is. Perhaps the brains of adolescents in some faraway tribe, where 14-year-olds are supposed to be fully mature men, tend to “ripen” a bit more quickly.

Without knowledge on the mechanisms through which brain maturation works, it cannot be accepted as a causal explanation for adolescent behavior. It just allows us to take all these psychological observations that we’re familiar with – lack of planning, novelty-seeking – and rephrase them as neural processes. That’s useful to science, but doesn’t answer any questions about how to raise your child, or what kind of society fits best with universal human needs.

Neuroscientists don’t examine kids from faraway tribes. In fact, they stick to college undergrads, usually from the psychology department. That’s no biggie, but it’s something to keep in mind when you’re reading stories about how brain science explains why we are who we are.

http://kenanmalik.wordpress.com/2011/06/01/morality-on-the-brain-2/

Now before I do so, I ‘d like to start up something that I always wanted to do with this blog, but was never able to: doing psychological experiments online. The first test I have in mind is a classic with a minor twist – some of you may know it already as I used it in earlier attempts to run online tests. The experiment is called the Wason Task, and if you haven’t done it before, you should definitely try it now. Since embedding the test isn’t working, I’ll just have to point you to this link to offer you the test. I hope to discuss the results and their relationship to the modularity debate as soon as the amount of responses is adequate (which will probably take a few weeks).

I am curious if it does. For those curious what it means, it is the average percentage score of a large group of students, on a multiple choice quiz they made for their introductory cognitive neuroscience course. Supposedly, it will increase as I update the underlying sheet.

“Thanks to the proliferation of the many electronic devices that we use on a daily basis, from cell phones to Global Positioning Systems and the Internet, that capture everything from our communications to our whereabouts, the complex system that we are most likely to tackle first in a truly quantitative fashion may not be the cell or the Internet but rather society itself.”

A quote from Albert-László Barabási in his article Scale-Free Networks: A Decade and Beyond, with which I wholeheartedly agree. This is the time to go study sociology, as Twitter, Foursquare and all those other services monitoring our daily lives are turning the study of society in an exact science. Graph theory and data-mining are required skills for the 21st century social scientist!

1. Modularity of the brain

I know modularist accounts of the brain mostly from evolutionary psychology. There, they serve the adaptationist’s aim to explain different cognitive faculties as having evolved separately, to deal with different types of problems imposed by the environment. I was somewhat surprised to learn that a number of psychologists accuse cognitive neuroscience as a discipline of considering the brain in modular terms. Truth is, I mostly see a highly flexible organ that has structure, but also has the fascinating capacity to adapt to differences in sensory input or cognitive demands. I know I am not alone in rejecting the “Swiss Army Knife” view from evolutionary psychology for exactly this reason. Reconciling this perspective on the brain with findings in psychology on, say, moral modules is one of the more interesting challenges for cognitive neuroscience, in my view.

Still, it’s true that I think the brain somewhat constrains which neurons performs which function, a view that builds on the comparability of different brains: although individual differences exist in functional regions, it takes surgical interventions to shove typical primary visual cortex function into auditory cortex, for example (Sharma, Angelucci & Sur, 2000). Whatever terms may best capture such a function is a different matter – and it is here that systems-level neuroscientists become somewhat confusing, as they use placeholders to make preliminary sense of functional networks. I think such placeholder modularism is responsible for a lot of the criticisms that are aimed at cognitive neuroscience

Yet, careful use of placeholder modularism is not enough. There’s a particular brand of anti-modularism that really needs to be considered by neuroscience, as it goes beyond well-established findings like the memory role of sensory cortex. In a meta-analysis of BOLD fMRI coactivations, Michael Anderson found that regions throughout the brain are being involved in virtually all task domains he identified. This “neural reuse” as he calls it was present in both subcortical and cortical structures and only showed variation across the posterior-anterior axis, which he used as a rough proxy of evolutionary age (Anderson, 2010). That is, despite heterogeneity at the cytoarchitectonic, receptor and microcircuit level, no real functional specialization was found using the terminology that is common in cognitive science (e.g. visual perception or emotion). Neuroscientists should really have a look at his work to see whether it demands for a reconceptualization of the brain.

2. Language use

Philosophers take some offense at the use of intentional language by neuroscientists. They fear it fosters categorical fallacies and offers a false sense of explanation. That risk does exist, although I feel it is being a bit exaggerated by the critics and probably limited to talks and papers that exist at the interface of disciplines or are aimed at the general public. Still, those are important places to remain accurate. It might be best to avoid intentional language as much as possible: speaking of atoms as “wanting” the noble gas configuration did not give rise to much confusion, but stating that neural subsystems “want” drug intake is a bit more tricky. It is best to express neural functions in neural terms, especially when engaging people outside of the neuroscientific field.

On the other hand, philosophers will also need to accept that some intentional actions have processing homonyms. Peter Hacker objected to use of the word “selection” in a neural of model of language production, although such nomenclature has a long tradition in physics (e.g. filters in signal processing or sieves) and biology (e.g. natural selection for traits) and can be said to have been “deintentionalized” by these fields. Still, cross-disciplinary communication is likely to benefit of clear language use. As natural scientists live in a culture where sloppy (natural) language use is okay “as long as you get the point”, they will need to adapt a bit if they want to communicate with psychologists and philosophers.

This adaption also requires pointing out bad uses of natural science terminology by social scientists. To me, doing this really feels like splitting hairs, but it is important, if only because of the bias that exists against people being non-accurate. For example, one of the conference members made a valid argument about the difference between PET and BOLD neuroimaging, but referred to “glucose versus oxygen metabolism”. As a chemist, this sounded off to me because you shouldn’t directly compare the reductive agent in one reaction to the oxidizing agent in another. As a sort-of-biologist, I also took exception to the use of the word metabolism, which refers to the set of all biochemistry in a given species: calling different catabolic pathways different “metabolisms” belies this fact.

I know from experience that some of my colleagues would readily turn such sentiments in outright dismissal of the actual message, despite its validity. I also think the sloppy use can lead to wrongful conceptualizations (e.g. ignoring that the “oxygen metabolism” measured by BOLD fMRI also includes glucose transport and consumption or that glucose consumption results in a haemodynamic response). The fact that such terms are used in neuroimaging and radiology drives a wedge between them and the molecular biological field, even though collaboration is vital at this point in time.

3. The network approach

Neuroscientists do functional and effective connectivity analyses at the microcircuit and systems level. They use single-cell recordings, anatomical tracers, functional imaging and computational methods to do so. They improve signal analysis methods to deal with the idiosyncracies of different data types and levels of description. Apparently, this is not widely known outside the field, judging by the amount of people who ensured me neuroscience does not look at connectivity. This is a real shame, given the success that the network approach has had the past 15 years: current understanding of sustained activity or the mechanistic role of hubs like the anterior cingulate cortex comes from these analyses and underlies conceptualizations of constructs that are of interest to psychology and philosophy, such as mood regulation, sleep, learning and memory. An effort should be made to export this knowledge – it is strange that obscure findings such as mirror neurons have found their way into the humanities, while the perspective on whole-brain function is apparently not known.

4. The effect of brain science on society

It’s nice that people like reading about brain science, but they might be doing so for the wrong reasons. Although I am reluctant to take responsibility for a chain of causality from the moment it crosses other responsible individuals, neuroscience as a whole profits from the interest, so brain scientists should be honest about what they offer. Personally, I think societal interest stems from the quest for identity, which is especially prevalent in western countries. The role of the west is unclear, as is the role of individuals in a globalized society. These types of questions deserve way better answers than neuroscience can provide. This is something brain scientists need to make clear to society at large – even though that may decrease book sales – to prevent people from acting misguidedly.

5. A new cognitive ontology as the goal of cognitive neuroscience

I have called this the “periodic system of cognition” in the past, but cognitive ontology is the better term. Insight in informational processing within the brain can lead to the emergence of a new vocabulary to describe cognition. For some applications, this vocabulary may not be useful, or even counterproductive, much like civil engineers tend not to describe their work in terms of configuration space, lest they want buildings to collapse. However, in the same way as quantum mechanics has paid off technologically without affecting language use, a brain-based cognitive ontology could pay off through new therapies, without changing the language with which we describe our behaviour.

References
Anderson, M. L. (2010). Neural reuse: A fundamental organizational principle of the brain. Behavioral and Brain Sciences, 33(04), 245-266.

Sharma, J., Angelucci, A., & Sur, M. (2000). Induction of visual orientation modules in auditory cortex. Nature, 404(6780), 841-7. doi: 10.1038/35009043.

Since it’s not just a poorly composed shot, but also pretty hard to read the bullet points, I’ll summarize them below:

• Require cognitive neuroscientists to have theories and make predictions (don’t accept “wanting to look what the brain does”
• Ask what the functional implications of the findings are!
• Ask whether the function of a brain module is exclusive
• Ask how the module [operates]
• Ask which input it requires and which output it produces!
• Don’t accept causal statements about brain-cognition relations
• Ask whether the module can change through experience!
• Ask how!
• Remind cognitive neuroscientists (and other scientists) of the societal/political implications of their conclusions

The list is amusing, because modularity and conflation of causality with description level jumps are things that my particular in-crowd refers to as “being a psychologist”, but I will readily acknowledge that disciplines are ill-defined in this particular field of study and I’m sure there’s modularist neodualists out there who can be called cognitive neuroscientists.

But what is neodualism, I hear you think. Is it the stuff that creeps into neuroscience because our language and other intellectual heritage still presupposed mind-body dualism? Nope, it’s something different altogether. It is the idea that brain causes mind, instead of holding some identity relationship with it (I say ‘some’, because I think the identity relationship needs more elaboration to make sense, but I’m not willing to write about that now). This is reverse Cartesian dualism, says Bernhard Hommel, whose copyright I violate in the picture and whose bullet points I stole.

Posted from WordPress for Android

A group is stable if the preferences of individuals are satisfied, whether or not those preferences include maximizing the amount of offspring. If optimizing satisfaction sounds familiar, that’s because it should be: this is exactly the leading principle of VNM-rationality. A normative view of morality can therefore be given as the optimization of the collective VNM utility function. A moral agent in this sense would, upon contemplation on the morality of a decision, estimate how his decision would affect the expected utility of others in the group. If net utility rises, the choice would be moral, while it would be immoral to lower net utility.

Notice the words “normative account here”. This proposed “VNM-morality” cannot be descriptive because firstly, it presupposes VNM-rationality, which is not how humans really act and secondly, it’s completely utilitarian and humans do not roll that way. I’ve argued it to be normative by claiming the function of morality is pleasing the group to stabilize the community (and your position in it). I also claimed that deviations from normative morality are typically heuristics that tend to lead to group-pleasing in a Pleistocene setting.

Of course, I don’t really know if they are heuristics – I should prove so for every known deviation from VNM-morality before expecting anyone to believe me. Yet let’s just suppose, for the sake of the argument, that this is the case. Would this then mean that utilitarianism, being the normative morality, somehow trumps deontological moral theories? The answer to this is a bit tricky: I am indeed arguing that utilitarianism is “the better morality” according to some functionalist perspective, but at the same time I am admitting that there are some heuristics – which I guess Kant would call maximes – which are part of the human condition. A utilitarianist would have to them into account when determining the morality of an action, as violating the heuristics decreases net utility. Using VNM-morality in a normative manner still leads to practically deontological behaviour in this indirect way.

In the past, there have been attempts to settle the morality debate by looking at the neural basis of moral decision-making. For example, William Casebeer has argued that the brain seems to take a virtue-ethical approach to morality, because whole-brain engagement is seen in moral reasoning and behaviour, which corresponds to the whole-character development that is central to virtue ethics (Casebeer, 2003).

One of the weaknesses in Casebeer’s argument is that even Kantian moral reasoning is likely to engage all types of systems. Kant himself may have considered his categorical imperative to be purely reason-driven, but a few of his maxims could be neurally implemented by using automated, motivational systems. For example, harm avoidance is likely to depend on emotional empathy mediated by the limbic system. To fully engage this system, relevant inputs from the mnemonic system may also be necessary. As much the same reasoning applies to utilitarianism, the fact that a variety of functional systems is involved in human morality does not argue in favour of one or the other.

In fact, at the level of the brain the different philosophies seem to dissolve completely. As said, some of Kant’s maxims may be processed by the limbic system. Other rules could be encoded by the lateral regions of the prefrontal cortex (Bunge et al., 2003), causing the deontological machinery to consist of at least two distinct subsystems. As damage to the ventromedial prefrontal cortex has shown, there is some sort of utility calculator in the brain that survives the damage and renders subjects more “utilitarian” in a number of tasks, presumably be due to decreased input from the limbic pathway (Koenigs et al., 2007). The moral brain is therefore best conceptualized as a set of subsystems that perform philosophically orthogonal computations. Actual moral action is the result of their integration, which likely took place differently in the head of Jeremy Bentham than it did in that of Immanuel Kant.

This integrated-systems view at the neural level only loosely constrains moral philosophy. While any ethical system needs to take human nature seriously, the different subsystems allow for a wild variety. The same lateral regions that encode some of Kant’s maxims can probably be encultured in such a way that allows the utilitarian subsystem to thrive. Similarly, we could maybe program our brain to compute the categorical imperative each time we face a moral choice. Cognitive neuroscience cannot say which is the better option – it can only say in what neural milieu the cultivation needs to take place.

References
Bunge, S. A., Kahn, I., Wallis, J. D., Miller, E. K., & Wagner, A. D. (2003). Neural circuits subserving the retrieval and maintenance of abstract rules. Journal of neurophysiology, 90(5), 3419-28. doi: 10.1152/jn.00910.2002.

Casebeer, W. D. (2003). Moral cognition and its neural constituents. Nature reviews Neuroscience, 4(10), 840-6. doi: 10.1038/nrn1223.

Koenigs, M., Young, L., Adolphs, R., Tranel, D., Cushman, F., Hauser, M., et al. (2007). Damage to the prefrontal cortex increases utilitarian moral judgements. Nature, 446(7138), 908-11. doi: 10.1038/nature05631.

But genetics? One of the biggest news stories in science in the last few years has been that, despite the fact that just about everything is heritable, there do not appear to be genes “for” traits. The human genome project failed to find genes or even sets of dozens of genes that account for more than a few percent of the variance in any target disease or trait. Even for physical height, which has a heritability of 0.9 and can be measured with nearly perfect accuracy, nobody can find a gene or a set of genes that explain why some people are taller than others. The most successful of these genome-wide association scans identified 27 genes that, when combined, explained just 3.7% of the variance in height. What hope, then, is there for finding genes “for” reciprocity, loyalty, or authority?

It’s part of a well-written response by Jonathan Haidt and Craig Joseph to a critique of their work by Patricia Churchland and Christopher Suhler. Both articles are available as uncorrected proof, you can find them at the following links:

Suhler & Churchland (2011) Can Innate, Modular “Foundations” Explain Morality? Challenges for Haidt’s Moral Foundations Theory
Haidt & Joseph (2011) How Moral Foundations Theory Succeeded in Building on Sand: A Response to Suhler and Churchland

The rest of the talk focused on language, but as the colloquium was held by the Institute of Logic, Language and Computation, this part used some vocabulary too esoteric for me. Since I was an intruder who did not do his homework, I was reluctant to ask for clarification, which probably caused me to miss out on some important points. It seemed to me that Stenning’s argument was closely connected to the social brain hypothesis and Robin Dunbar’s thoughts on encephalization (Schultz and Dunbar, 2010). He argued morphogenetic changes occurred in a positive feedback cycle with sociality and led to cognitive faculties such as language. He then elaborated on the language faculty by stressing it involved delay of planning. Although humans are pretty bad at this, they’re probably better than animals that do not charge interest rates at all. Indeed, delaying actions depends on brain regions that expanded in the recent evolutionary past of primates.

Stenning listed a number of the problems that pop up when speculating on evolutionary history, but he skipped Tinbergen’s point on ultimate versus proximate levels of explanation. That proved to be a shame: one of the first comments made after his talk was about language being spoken because it’s fun to do so – a proximate explanation of why humans engage in speech that is independent of the ultimate explanation. To put it differently, the type of answer Stenning is looking for is not what motivates people to speak, but what evolutionary pressures put this motivational system into place – this is a wholly different kind of answer. Although I think the type of systems involved in motivation can instruct us on its evolutionary origin (e.g., the evolutionarily relevant connection between group bonding, pair-bonding and parental care becomes apparent at the neural level), the reason why an organism engages in a behaviour should not be confused with why that behaviour exists in the first place.

Doing so sounds easy enough, but in practice it can be very difficult to keep the two levels of explanation separated. I wouldn’t be surprised if this in itself is testament to the social origins of our cognition – questions of causality are quickly reinterpreted as questions of intentionality, attracting any conclusions on the origins of some behaviour towards a motivational account.

References:
Schultz, S. and Dunbar, R. (2010). Encephalization is not a universal macroevolutionary phenomenon in mammals but is associated with sociality. Proceedings of the National Academy of Sciences. doi:10.1016/S0140-6736(08)61345-8