Saturday, January 26, 2013

WaMu-ellujah! Forgetfulness and greed

The failure of Washington Mutual, and a new paradigm of corporate responsibility.

A blurb on the back of "The Lost Bank", by Kirsten Grind, calls it "entertaining". If having your teeth pulled is entertaining ... perhaps. It is tragedy of a classic sort, except that the main protagonist (the CEO) rides off rich as sin, leaving behind a landscape of smoldering properties, empty bank accounts, and angry mobs.

Washington Mutual began as a well-loved local bank of the northwest, focusing on customer service. It barely survived the S&L crisis, but thrived thereafter, buying up like-minded banks and integrating them into its low-key, family-oriented atmosphere. The key transition during this era was from a beloved CEO, Lou Pepper, to Kerry Killinger, a more MBA-style, somewhat inscrutable, and socially awkward banker. It turned out that he had enormous ambition and energy, however, and drove the bank to new heights of acquisition, all based on its retail banking prowess, honed through many years of takeovers and careful back-office integration.

Eventually, the company outgrew its small town values, became the sixth-largest bank in the US, hosted blowout parties featuring such inspirational messages as WaMu-ellujah! Along the way, the company acquired mysterious loan issuers in Southern California that reaped ever-increasing profits, while keeping the details murky. For a bank, loans equal assets. This means that if customers can be convinced to take out loans far beyond their ability to pay, the loan is, at least in the short term, booked as a higher asset for the bank.

Washington Mutual spent a long time torn between its addiction to these profits, and its underwriting standards, which gradually fell by the board. Risk officers came and went like ghosts in the night. Killinger became a rock star, by the standards of the business press, at least. As S&L crisis veteran Bill Black explains with regularity, the easiest way to rob a bank is to own one and run a control fraud on it. The money is guaranteed. (Short term, at least.)

The question that the author does not address is the most difficult one- how much of this was conscious, and how much unconscious? How much was the bubbly atmosphere where collapsing house prices were inconceivable, real estate assessors were bullied into marking to the desired price rather than to value, ratings agencies colluded with their bank customers to mark toilet paper as AAA, and real estate bundled-security investors trusted that someone else was minding the store?

Or how much was a conscious decision by Killinger and his lieutenants to throw the core banking concept of underwriting standards into the circular file and let the money roll in? How much was the negligence of auditors, investors, regulators, analysts, and journalists to connect the flow of profits to the sewer of underwriting that Washington Mutual was tending in secret?

We are not told, and it might take a psychoanalyst to address properly. In any case, we have heard a great deal about "moral hazard", a concept that applies here in spades. The institution of Washington Mutual was destroyed, absorbed into the ever-growing colossus of JPMorgan. But the executives responsible for all the bad decisions and betrayed standards and trusts ... they kept their loot and retired in comfort.

This is a problem- that corporations and their officers pay insufficient attention to the long term. Only corporations that thumb their noses at conventional business advice and analyst pressures (think Apple) have an inner culture strong enough to ignore the short term temptations and to focus on the long-term. The default criterion is the quarterly earnings, by which executives live and die, and get the bonuses and options whose time horizon is a few years at most. Take the money and run has become the normal way of business in the US, and it is damaging not only in rare crises, but in broader cultural terms.

Washington Mutual was mainlining the drug of greed- raking in money from its shady lending operations, and getting patted on the back in the bargain for extending credit to previously underserved communities(!) Internal risk assessment was subverted, long-standing cultural norms over-ridden. The regulatory establishment likewise lost its mind, likewise captured by criminal amnesia (especially considering the S&L crisis had blown through less than two decades before) and subject to the same Minsky cycle of forgetfulness as the bankers.

What to do? I think the answer is to create a societal mechanism to extend clawbacks deep into the ill-gotten gains of the executives and others profiting from this kind of toxic, antisocial activity. We need to think beyond simple criminal accountability. Corporations are organized to have limited liability. And as far as shareholders, that is perfectly fine. They takes their risks, and they gets their rewards. But when it comes to the general public interest, no such liability protection is appropriate. If corporations are to be people, they can not be immune from basic duties to the public interest, especially in view of their vastly expanded influence on that public interest, relative to the scope available to an individual citizen, due to their size and organization.

Just as lawyers are deemed officers of the court, and are held to some professional standards that go beyond fulfilling the letter of the law, corporate executives should be deemed officers of the state, with extra duties and liabilities that extend beyond the letter of the law. Prime among them would be general and long-term liability for all their gains beyond a base salary of perhaps the median for their company. All other monies would be subject to long-term review and surrender based on legislative findings of culpable irresponsibility and harm to the public good.

Now, all businesses are in the more or less culpable in cutting corners, transforming public goods into private gains, imparing markets, and the like. This proposal wouldn't be aimed at typical businesses. On the other hand, there are many corporate endeavors that are socially positive but are not sufficiently rewarded in the market. Those who invent the lasers, microchips, diagnostic tests, and other great things do not always get their due, and might be beneficiaries of some of the funds gathered from the malefactors held to account by the above process.

Obviously, this is a highly problematic type of proposal. Now that corruption is standard practice in the revolving door between business and government, how could we possibly expect this nakedly political process to work any better than the criminal process that has already so obviously failed to bring financial criminals to account? Would this new process of accountability not skirt the protections of the rule of law and lead to hyper-political battles that leave the country even more corrupt, and well-connected but disastrous businessmen even better off than before?

I do not have good answers. All I know that it is fundamentally unjust and wrong to see someone like Mr. Killinger hide his ill-gotten gains behind a curtain of legalism, when the whole idea of law is to create a just society, and the whole idea of business is to render people useful to each other, not only for a New York minute, but over the long term.

Saturday, January 19, 2013

How do we get five digits?

Hox genes control the number of digits (fingers and toes) in an interesting way.

Once the trail of animal development research was blazed in the fruit fly, mammalian investigators eagerly followed, using similar methods and looking at related genes. Some of the most interesting have been the Hox genes, which control patterning at a very high level- the identities of segments in flies, and the identities and numbers of related body areas in mammals (vertebrae, ribs, limbs, digits, etc.)

Genomic diagram of Hox genes in various organisms. A rough phylogeny is on the left, and diagrams of where some of the genes are expressed and have their effects is on the right. The middle shows the clusters of Hox genes, where the entire clusters has been quadruplicated in the vertebrate lineage, creating A, B, C, and D clusters of genes 1 to 13, though a few are missing here and there due to later loss. Note the general rule of linear expression of Hox genes from front to back in the organism, coordinate with genomic position.

The wiki page on Hox genes supplies this graphic of the Hox gene clusters of various species, related by a rough phylogenetic tree (left; tetrapods are us). Each colored box represents one protein-coding gene, positioned roughly as it appears in the genome (not to scale). Note that vertebrates picked up a quadruplication of the entire Hox cluster, after which a few individual genes were later lost. This expanded the body plan repertoire of this lineage substantially- a significant evolutionary event. The original Hox cluster was incidentally already the result of long-ago duplication of a single gene encoding a transcription regulatory protein. All Hox proteins have very similar structures.

Hox stands for homeobox, which stands for homeotic transforming transcriptional regulator protein containing a diagnostic protein sequence that binds to DNA and was called a "box" of sequence, due to its appearance in sequence alignments. And homeotic? That is not kinky at all, but refers to genetic effects on the body plan, i.e. the transformation of one part of the body into one "like" another via mutation, from the Greek for similar. Hox proteins all have their effects by binding to other genes and controlling their expression, though unfortunately little is known about these details, at least in mice, since this end of things rapidly becomes extremely complex.

Another part of the story of digit control is a different DNA-binding protein, Gli3. When mutated, Gli3 is known to cause polydactyly- the production of typically a sixth digit- as well as many other malformations (see image at bottom, left side). Gli3's activity (details of which are largely unknown) is the effect of a gradient of another protein, called sonic hedgehog (Shh) in the developing limb bud, and which at last is a protein that forms an actual physical gradient in the tissues of a developing limb.

Shh protein forms gradients that help direct development of body patterns during early embryonic times. But it can't do its job alone.

A recent paper showed that these two systems, the Shh/Gli3 system and the Hox system (specifically Hoxa13 & Hoxd11-13, the last genes in the tetrapod clusters above, dark blue) interact to generate the five-finger pattern. The last ingredient believed to be involved is another gradent forming protein, fibroblast growth factor 1 (Fgf1), presumed to be downstream of the various Hox regulators. The researchers speculate that these two gradients, of Fgf1 and Shh, are controlled by different genetic inputs, and interact to create patterns. In this case it is fingers, but in other organisms, similar processes are thought to make zebra stripes, wing patterns, shells, etc..

One molecular gradient can provide some information about where to put things, but probably not anything very consistent or detailed. The interesting part of this story, though the actual biology is unfortunately not yet well developed, is that the combination of two molecular gradients generates far more interesting possibilities. This was, intriguingly enough, pointed out by one of the greatest mathematicians of all time, Alan Turing, who, taking time off from inventing the computer, provided the mathematical foundations of a two-component chemical system which, across a gradient field of both chemicals, which react at different rates as they go forth, can create amazingly stable and interesting patterns, somewhat counter-intuitively.

Abstract model of a Turing wave 2-component system, resolving itself over time spontaneously from a homogenous solution into a complex binary pattern.

Naturally, these researchers made mutations to look at the effects of these genes. The complete deletion of Hoxa13 turns out to be lethal in early embryos. On the other hand, they find that the Hox code is rather complicated, such that deletion of Hox members d11 to d13 causes added effects that mimick or accentuate deletion of a13. In early embryos, they can see the hand region setting up dramatically more fingers (marked by staining for the protein Sox9, a marker of pre-cartilage/bone formation) as they delete either of the gradient-forming or responding genes Gli3 and Hox*.

Early embryonic (day 12.5) limb-buds stained for finger primordia, from mice mutated for various genes, as noted. "+" is wild-type for one or both genes, while "-" is deleted or otherwise inactivated. While deletion of Gli3 alone has some effect on finger number, only with the added deletion of the Hox genes do finger numbers become truly uncontrolled.

There is something amazing going on here. By the time all these genes are deleted (-, or other non-"+" variants) for Gli3, Hoxa13, Hoxd11, Hoxd12, and Hoxd13, there are no individual digits left. The whole zone has turned into a smooth non-digital mess. Lesser amounts of the Hox genes in particular lead to dramatically rising numbers of digits.

Mice mutated for various genes as noted, now at birth, stained for cartilage (blue) and bone (red).

In roughly the same amount of tissue, many different numbers of digits can develop, based on a few genetic alterations.  Clearly the researchers are hot on the trail of how this pattern develops and will be looking for the particular components downstream of the Hox genes that carry out its regulatory directions, especially the protein or other chemical that forms the counter-gradient to Shh. It is a common theme in biology, that most of the action lies in complex layers of regulation (i.e. management) so that the ultimate actors can toe their lines with precision even in a variable genetic and external environment.

"It reports recent cases such as the US iconic firm Caterpillar which “reported record profits last year” but “insisted on a six-year wage freeze for many of its blue-collar workers”. That is not an isolated case. Indeed it is the norm and is one of the defining characteristics of the neo-liberal era that has dominated economic policy making over the last three decades.
... Everybody should benefit from productivity growth – that is what we call a society."

Saturday, January 12, 2013

Our parents give us meaning

We crave meaning in the approval of the parent, real and imagined.

How many times do you hear.. if only my father had said I was doing OK, if only my father said he loved me, if only I had a chance to show my mother how well I was doing before she died...? We grow up competing for our parent's attention and utterly dependent on it. If there is one sure influence from childhood, it is the frame of reference and attitudes of the parents. Sometimes this happens in reverse, by rebellion, but inevitably we later become our parents, so enmeshed in their world that leaving entirely is not an option.

And when they are gone? What then? The Romans, Japanese, and many other cultures made cults of their parents (in patriarchial fashion, just the male line). The parent is called to an alter where their judgement, forgiveness, boons, and advice are sought. Their gifts to us and ongoing effects are recognized. Their utter absence is so inconceivable that prayer starts to make sense. After all they are so much a part of us that even if we are mumbling to ourselves, we speak to them too.

But why keep a different flame at every alter and hearth? They are ultimately the same supervisory concept, and a culture gains solidarity from giving them the same name. God. It is funny how, no matter the theological complexity and reasoned mystery of one's god, it is never "it", but always "Him" (or in outré cases, "Her").

The model never strays far from the father/mother model, which makes it immensely powerful- as a way to acculturate children with concepts the actual parents are not strong enough to convey, as a way to sanction whatever the reigning powers want to do, as a way to comfort and soothe adults who remain children deep inside. It goes to the extreme of denying death itself, as if putting our heads under the covers will make the horror go away.

And of course, it gives the deepest meaning to those who believe most "deeply". Who see the universe as a machine to give them meaning through the imagined directives of the invisible father, who gives them the most arduous tasks, attends to their most minute needs, and gives them the most glorious rewards. It almost makes you wonder just how far that great principle of neoteny can go- how far humans can go by refusing to grow up. For creating meaning is the true task of the adult.

Saturday, January 5, 2013

The music of working memory

An analysis of brain oscillations binding content between the frontal and parietal cortices.

I'll admit it- I am fascinated by brain waves. They seem to provide dynamic structure to the activity coursing through the static anatomy of the brain, organizing all those things we love about it, like consciousness. There is also a deep connection to music, as the the various "bands" of the neural oscillations, going in very rough terms (delta ~2Hz, theta ~4Hz, alpha ~10Hz, beta ~20Hz, gamma ~40Hz) can be seen as octaves in an internal, ongoing symphony.

But knowledge about how these oscillations work has been very slow in coming. It is obviously a very complicated system, and our scientific tools are still woefully unequal to the vast scale of tens of billions of neurons and trillions of synapses. A recent paper describes at a very gross level the function of oscillations between the frontal cortex and the posterior parietal cortex as carrying some kinds of working memory.

"We tested the hypothesis that neuronal synchronization across the fronto-parietal network carries content-specific information that contributes directly to visual working memory. The pattern of fronto-parietal synchronization should thus vary as a function of the object held in memory."

Their memory test (done in macaque monkeys) was: first, an image is displayed, then turned off, and then after a pause of one to three seconds, two images are displayed, one of which matches the first one. The pause is the important part, during which the researchers hunt through their electrode recordings for the ghosts of working memory. The monkeys finally get a reward for correctly tracking their eyes to the matching image of the pair that are displayed later. It's an extremely simple test, and one could imagine that it hardly touches the capabilities of these monkeys for remembering things.

They have inserted a series of electrodes into the macaque's brains, in the areas putatively involved in keeping the first image "in mind". These locations are the side of the prefrontal cortex, and the posterior of the parietal cortex:
Macaque brain, marked for the areas studied- lateral frontal cortex (right), and posterior parietal cortex (left).
The idea is that the first image needs to be kept in mind to rapidly match it in the test, so the researches are hunting for neural correlates of this mental content. They know it is visual content, which narrows things down slightly. They also know about prior work which has observed some brain wave synchronization between these two brain areas during focussed attention and task performance, and which involves the parietal cortex closely with visual working memory.

Their observations take the form of correlations between the oscillations detected in one location compared with those in the other. An example is shown below, where a sample electrode trace from each area is shown in B, and from that data, a graph of correlation (coherence) with respect to time and frequency (C) is developed. The correlations are detectable at 20Hz

The visual test is diagrammed in A, while a pair of local field electrode recordings is in B. dPFC = dorsal prefrontal cortex, PEC = posterior parietal cortex, caudal section of arbitrary area E.

Fine, so correlation is not so hard to establish. But causality between the two brain areas and content of what these oscillations are encoding? That is a taller order.

To start, they sift through many trials on one monkey, choosing electrode pairs that show the kind of correlation seen above, to see whether the identity of the displayed shape or its location had additional effects on the oscillation correlations. This would indicate that the relationship was related to the visual content, not just part of the more global attention system or something similar. They develop a metric called coherence selectivity index (CSI), related to mutual information theory, that can be graphed:

Coherence selectivity index showing the degree to which correlations (of electrical oscillations) between the anatomical locations are also correlated with the content of the visual image, either of location or of image identity.

This supports the idea that the oscillations being observed as related between the two brain locations also are related to the tested content- the images of the memory test, during the time when the images are not visible. I have to say that this is pretty meagre stuff, both in terms of the statistics being used to ferret signals out of messy data, and conceptually, in that the connection between this kind of wave correlation and something called "working memory" is pretty faint. I am not saying that they are misrepresenting anything, just that the technical means for figuring out what is on someone's "mind" by electrodes is still awfully primitive.

Lastly they ask questions about which parts of the brain lead these synchronizations, and which follow. They use another statistical method, of Granger causality, which allows them to classify sub-regions where they have electrodes- paired and showing image-identity correlations as discussed above- as either senders or receivers of signals. That is to say, the oscillations (and perhaps detailed spike trains) are reliably offset in one or the other direction.

The conclusions are mixed, that while the parietal areas they measure are more often "senders" than "receivers", causality goes in both directions. This follows a growing theme in brain science, where most higher-level activity happens in recurrent networks, not in linear processing streams. For example, attention is believed to be an interaction between top-down selection and bottom-up processing streams to form coalitions of synchrony. But of course, the result could also just reflect muddy data.

They also note that the relative phases of the presumptively correlated oscillations are not firmly related, which is quite different from the highly structured phase relations of neural signalling and oscillations in the memory storage system of the hippocampus. Again, this may reflect their muddy data, or perhaps suggests that such detail can not be retained in this much longer-range, yet still functional, inter-cortical synchronization.

WGC is Wiener-Granger causality, PPC is posterior parietal cortex,  PFC is prefrontal cortex, and time is as above, relative to the image memory task. Inthe lower panels, various sub-areas of the parietal cortex area are colored blue, and sub-areas of the prefrontal cortex are red. Parietal are

"Our findings demonstrate that fronto-parietal synchronization during visual working memory is widespread, task-dependent, and content-specific during the delay period."

This seems a bit strong, though perhaps if one adds in all the prior work that has been done in this network and how visual memories are handled in the parietal cortex, it makes more sense. From what I see in this paper, however, the conclusions are more preliminary than final. At any rate, it is clear that scientists are gaining (slowly) on this significant problem, and pinning down just what constitutes our various memories and thoughts.