Gum voting

Gum election in the public streets of Berlin – “Who sucks the worst? Vote with your gum“. Several weeks before the election on United States, this rather simple but extraordinary concept spread from NY city to San Francisco, from St. Louis to São Paulo, from Berlin to Sydney within a few days. This kind of remembers me one of my friend’s (Ivo et al’s) project – Stick Me!, due to some similar features. Even nowadays my own refrigerator has one Stick Me! sticker over it and I really enjoyed participating on it in the past via one very quick and humble “Stick Me Mate” proposal, while playing blitz chess with friends at a bar nearby my house.

A bunch of people (promoting Collective Intelligence?) is using the environment as a way to communicate (like over any chessboard). Communication is indirect, but still they communicate through the alterations and patterns they impose on the environment itself. Meanwhile, imposing a mark or sign somewhere, increases the probability of a second response later in time – a response to a stimulus (as ants put their pheromone marks on the ground). Though here however (on both projects) only positive feedback is used.

In fact, Mother Nature has conceived a very outstandingly simple and better strategy: their signs and cues vanish in time, simple as that! For instance, pheromone, a chemical substance segregated by ants and termites evaporates in time. Over here however, there is no evaporation at all working on (societal agents are not entitled to use negative feedbacks or using vanishing marks), which can curse it’s own dynamic – unless someone destroys the posters, of course. Amazon book recommendation system, works as well this way, that is by uniquely making use of positive feedbacks (people that bought this X book also as bought Y, etc). Unfortunately, Amazon system along with his wish lists could not integrate that someone who bought the X book did not bought Z (while others have done it), which basically leads to a snow-balling effect that does not self-organize in time (adapts) to new potential good-reading books. What you end up seeing is just the overall majority consensus, the “minimum common multiple” as I sometimes call it, who tends to over-look and underestimate some high potential new-coming solutions (over this precise context, good books coming in). Amazon should instead look carefully to some scientific works on collaborative filtering. Instead the consequences are this: check here for a real user feedback on what Amazon is suggesting, or in fact not suggesting at all.

Not only their system tends to adapt slowly, since the only thing it’s promoting is nothing else but memory (exploitation, which could be achieved by positive feedbacks), as he is not learning (exploration, which could be achieved by negative feedbacks), when we know that on the contrary, a delicate compromise between both is in fact of huge importance. The difficult but possible systemic trick is to remember the past as simultaneously innovating. If as a whole the system only remembers the past, no innovation is possible causing dramatic consequences when the “environment” changes. This could lead to stagnation. On the other hand, if too much systemic pressure is put on innovation itself, energy is lost, leading the system to explore the universe of possible solutions in a quite “”stupid” trial-and-error like random manner. Some dynamics between one thing (memory) and the other (learning) could be checked here (figs. 4,5,6,7 and 19), along with their speed.

After all a gum or a sticker is nothing else than a tag -as web blogging tags and internet tag clouds are. My question is – Could they vanish over time as I believe and propose they should? Having that question in mind, while looking at these precise public street projects, there are also other conceptual bridges we may found, as far as I recognize.

Let me refer at least 4, with the help of some passages below from other texts: (1) Hobo signs and codes (as well as the bottom-up like emergence of norms and ethical codes between them), (2) the role of Positive and Negative feedbacks briefly discussed above, (3) Swarm Intelligence and of course, (4) Stigmergy. In what specifically regards Hobo signs let me say that they are quite clever. Since they are done with chalk! So, rain and erosion could erase them, little by little, day by day. Thus, solutions that were good in the past, but no longer exist or that are partially vanished over time, tend to be replaced by new fresh ones, appropriated for the present, only loosing part of the whole systemic memory, serving us with new stimulus (we tend to respond to those fresh ones), allowing a continuous adaptation to reality. As I said in the past over a scientific invited lecture (not the right place to say it, though!), signs, quotes, delayed desynchronized dialogues and phrases over the doors of public bathrooms follow similar trends and tend to be stigmergic. In what regards the following four passages, I leave to you the connection between them (sorry for this now long food for thought post):

(1) […] Synergy, from the Greek word synergos, broadly defined, refers to combined or co-operative effects produced by two or more elements (parts or individuals). The definition is often associated with the quote “the whole is greater than the sum of its parts” (Aristotle, in Metaphysics), even if it is more accurate to say that the functional effects produced by wholes are different from what the parts can produce alone. Synergy is a ubiquitous phenomena in nature and human societies alike. One well know example is provided by the emergence of self-organization in social insects, via direct (mandibular, antennation, chemical or visual contact, etc) or indirect interactions. The latter types are more subtle and defined by Grassé as Stigmergy to explain task coordination and regulation in the context of nest reconstruction in Macrotermes termites. An example, could be provided by two individuals, who interact indirectly when one of them modifies the environment and the other responds to the new environment at a later time. In other words, stigmergy could be defined as a typical case of environmental synergy. Grassé showed that the coordination and regulation of building activities do not depend on the workers themselves but are mainly achieved by the nest structure: a stimulating configuration triggers the response of a termite worker, transforming the configuration into another configuration that may trigger in turn another (possibly different) action performed by the same termite or any other worker in the colony. Another illustration of how stimergy and self-organization can be combined into more subtle adaptive behaviors is recruitment in social insects. Self-organized trail laying by individual ants is a way of modifying the environment to communicate with nest mates that follow such trails. It appears that task performance by some workers decreases the need for more task performance: for instance, nest cleaning by some workers reduces the need for nest cleaning. Therefore, nest mates communicate to other nest mates by modifying the environment (cleaning the nest), and nest mates respond to the modified environment (by not engaging in nest cleaning); that is stigmergy. […],

in Vitorino Ramos, Juan J. Merelo, Self-Organized Stigmergic Document Maps: Environment as a Mechanism for Context Learning, in AEB´2002 – 1st Spanish Conference on Evolutionary and Bio-Inspired Algorithms, E. Alba, F. Herrera, J.J. Merelo et al. (Eds.), pp. 284-293, Centro Univ. de Mérida, Mérida, Spain, 6-8 Feb. 2002.

(2) […] To cope with the difficulty of hobo life, hobos developed a system of symbols, or a code. Hobos would write this code with chalk or coal to provide directions, information, and warnings to other hobos. Some signs included “turn right here”, “beware of hostile railroad police”, “dangerous dog”, “food available here”, and so on. For instance: a cross signifies “angel food,” that is, food served to the hobos after a party. A triangle with hands signifies that the homeowner has a gun. Sharp teeth signify a mean dog. A square missing its top line signifies it is safe to camp in that location. A top hat and a triangle signify wealth. A spearhead signifies a warning to defend oneself. A circle with two parallel arrows means to get out fast, as hobos are not welcome in the area. Two interlocked humans signify handcuffs. (i.e. hobos are hauled off to jail). A Caduceus symbol signifies the house has a medical doctor living in it. A cat signifies that a kind lady lives here. A wavy line (signifying water) above an X means fresh water and a campsite. Three diagonal lines means it’s not a safe place. A square with a slanted roof (signifying a house) with an X through it means that the house has already been “burned” or “tricked” by another hobo and is not a trusting house. Two shovels, signifying work was available (Shovels, because most hobos did manual labor). […], in Hobo, Wikipedia.

(3) […] Swarm Intelligence (SI) is the property of a system whereby the collective behaviors of entities interacting locally with their environment cause coherent functional global patterns to emerge. SI provides a basis with which it is possible to explore collective (or distributed) problem solving without centralized control or the provision of a global model (Stan Franklin, Coordination without Communication, talk at Memphis Univ., USA, 1996). […] (here)

Hobo or tramp markings at Algiers entrance to Canal Street Ferry across Mississippi River, New Orleans. Ferry is free for pedestrians or on bicycle. "X" means "OK", slashed circle "Good way to go". (via Wikipedia above).

(4) […] – Positive feedback, f+: in contrast to negative feedback, positive feedback generally promotes changes in the system (the majority of SO systems use them). The ex-plosive growth of the human population provides a familiar example of the effect of positive feedback. The snowballing auto catalytic effect of f+ takes an initial change in a system (due to amplification of fluctuations; a minimal and natural local cluster of objects could be a starting point) and reinforces that change in the same direction as the initial deviation. Self-enhancement, amplification, facilitation, and auto catalysis are all terms used to describe positive feedback. Another example could be provided by the clustering or aggregation of individuals. Many birds, such as seagulls nest in large colonies. Group nesting evidently provides individuals with certain benefits, such as better detection of predators or greater ease in finding food. The mechanism in this case is imitation : birds preparing to nest are attracted to sites where other birds are already nesting, while the behavioral rule could be synthesized as “I nest close where you nest“. The key point is that aggregation of nesting birds at a particular site is not purely a consequence of each bird being attracted to the site per se. Rather, the aggregation evidently arises primarily because each bird is attracted to others. On social insect societies, f+ could be illustrated by the pheromone reinforcement on trails, allowing the entire colony to exploit some past and present solutions. Generally, as in the above cases, positive feedback is imposed implicitly on the system and locally by each one of the constituent units. Fireflies flashing in synchrony follow the rule, “I signal when you signal”, fish traveling in schools abide by the rule, “I go where you go”, and so forth. In humans, the “infectious” quality of a yawn of laughter is a familiar example of positive feedback of the form, “I do what you do“. Seeing a person yawning , or even just thinking of yawning, can trigger a yawn. There is however one associated risk, generally if f+ acts alone without the presence of negative feedbacks, which per si can play a critical role keeping under control this snowballing effect, providing inhibition to offset the amplification and helping to shape it into a particular pattern. Indeed, the amplifying nature of f+ means that it has the potential to produce destructive explosions or implosions in any process where it plays a role. Thus the behavioral rule may be more complicated than initially suggested, possessing both an autocatalytic as well as an antagonistic aspect. In the case of fish, the minimal behavioral rule could be “I nest where others nest, unless the area is overcrowded“. In this case both the positive and negative feedback may be coded into the behavioral rules of the fish. Finally, in other cases one finds that the inhibition arises automatically, often simply from physical constraints. Since in SO systems their organization arises entirely from multiple interactions, it is of critical importance to question how organisms acquire and act upon information. Basically through two forms: a) information gathered from one’s neighbors, and b) information gathered from work in progress, that is, stigmergy. In the case of animal groups, these internal interactions typically involve information transfers between individuals. Biologists have recently recognized that information can flow within groups via two distinct pathways – signals and cues. Signals are stimuli shaped by natural selection specifically to convey information, whereas cues are stimuli that convey information only incidentally. The distinction between signals and cues is illustrated by the difference ant and deer trails. The chemical trail deposited by ants as they return from a desirable food source is a signal. Over evolutionary time such trails have been molded by natural selection for the purpose of sharing with nestmates information about the location of rich food sources. In contrast, the rutted trails made by deer walking through the woods is a cue, not shaped by natural selection for communication among deer but are a simple by-product of animals walking along the same path. SO systems are based on both, but whereas signals tends to be conspicuous, since natural selection has shaped signals to be strong and effective displays, information transfer via cues is often more subtle and based on incidental stimuli in an organism’s social environment. […], in Social Cognitive Maps, Swarm Collective Perception and Distributed Search on Dynamic Landscapes.

Stick Me! sticker in plain nature over Aljezur, Algarve (South of Portugal). Unknow author. "Copyrigthed nature" or a way of saying "I was here. I am connected. You could also be connected. We are all connected" ?!

On Financial Markets

Image source: ITGO.COM (large size)

In what concerns Social Psychology, just check this out: Milgram et al.(1) found that if one person stood in a Manhattan street gazing at a sixth floor window, 20% of pedestrians looked up; if five people stood gazing, then 80% of people looked up.

As latest fraud facts told us last week in cascade manner (Freddie Mac, Fannie Mae, Lehman Brothers, AIG), essentially, the current financial crisis results from the fact that in order to be a truly self-organized, self-regulated adaptive evolving system as Adam Smith envisioned (back in 1776 – The Wealth of Nations), financial markets need not only to adopt Positive Feedback as they do (e.g., the one Milgram founded, above), but mainly – though in some proportion – Negative Feedback features as well, that is, some form of a priori outside Regulation – in other words, the context and environment of the entire “game” (check here for some basic features of Self-Organization -(2).  Ill or badly defined environments among complex systems, lead to chaotic or unprecedented biased evolutionary pressures (e.g., many of the entities – read it as some companies -, cheaters playing economic games such as the Iterated Prisonners Dilemma IPD over Bounded Rationality – that should naturally “die”, in fact proliferate), turning to be counterproductive for the whole system. Self-Organization occurs in precise very-subtle-narrow regimes (“at the edge of chaos” – Christopher Langton, Santa Fe Institute), not over entire entropic regimes as the one we were facing, neither over the entire spectrum (Order, Semi-Order, Edge of Chaos, Chaos – check Stuart Kaufmann’s I, II, III, IV phases). They occur near them, not in them. In order to do so, among several other things (2), also Negative Feedback is necessary. As we know from complex systems, nature, as well as artificial intelligence, the system becomes too greedy and instable.

For instance, on social insect societies know to be self-organized, Positive Feedback (PF) could be illustrated by pheromone reinforcement on trails, allowing the entire colony to exploit some past and present solutions. Generally, as in the above cases, positive feedback is imposed implicitly on the system and locally by each one of the constituent units, whereas Negative Feedback is imposed explicitly mainly by environmental “pressure” conditions, promoting counter-balanced innovative solutions. Fireflies flashing in synchrony follow the PF rule, “I signal when you signal”, fish traveling in schools abide by the rule, “I go where you go”, and so forth. In Humans, the “infectious” quality of a yawn of laughter is a familiar example of positive feedback of the form, “I do what you do”. Seeing a person yawning, or even just thinking of yawning, can trigger a yawn. There is however one associated risk, generally if Positive Feedback acts alone without the presence of Negative Feedbacks, which per si can play a critical role keeping under control this snowballing effect, providing inhibition to offset the amplification and helping to shape it into a particular pattern (2). Indeed, the amplifying nature of Positive Feedback means that it has the potential to produce destructive explosions or implosions in any process where it plays a role. Thus – several biological studies express it – the behavioral rule may be more complicated than initially suggested, possessing both an autocatalytic as well as an antagonistic aspect (2).

The fundamentalist ultra-liberal strategy of “no regulation at all”, followed in recent years, believing ideologically – like a fatwa – that markets are a truly Darwinian CAS (Complex Adaptive System) where invisible hands operate every day in a perfect situation, lead us to a chaotic situation, where shimmering waves of panic proliferate through the world (check video below), constraining states to intervene in a a posteriori manner (the ongoing Paulson Plan). Though, the ultimate best option was to do it a priori, ceasing nations to sleep much of their time earlier in face of many disastrous – out of control – innovative as well as cannibalistic financial products invented in the last decade (many of them being nothing else than financial pyramid schemes, though sophisticated), without reasoning of possible and profound dramatic social costs. Alternatively, a priori smart intervention seems to be the only resource to avoid the current ongoing privatization of profits, and immoral massive losses nationalization, being payed by tax contributors across USA and Europe. Keeping in reasonable shape the wealthy resource that finance markets really are and could be for all of us: promoting robust and innovative companies.

{ [VIDEO] Shimmering Giant Honeybees from Science News on Vimeo.
For some, Financial Markets in crisis could be seen as shimmering waves across the globe, however they are far from being self-organized as honeybee colonies. The invisible hand metaphor originates with Adam Smith in The Wealth of Nations (1776). Bernard Mandeville made a similar point with his Fable of the Bees (1705), which fancifully describes human society as a wondrously productive bee hive, even though each bee is as selfish as can be (3). – Video and article (see 8). }

Washington is right now asking for help and money to China, as well as indulging many other countries in the world to step in. At the same moment, in UK, in the aftermath of the Northern Rock takeover, some big and many medium companies are now selling themselves for short, in Belgium, FORTIS is on the verge, and in Ireland – once the first example of economic bloom in Europe -, the first technical ressence in many years is now fully recognized. France seems to go next. Partial-nationalizations are now occuring from Iceland to the sunny mediterranean Gibraltar. While, back in the US, a very recent LA Times-Bloomberg poll revealed that 60% of Americans claim for some sort of state intervention. In face of this facts, ultra liberals appeal to two typical arguments: (first) that the current phenomena is inevitable (no comments on my side on that, since 1929 till now nothing of this dimension happened before), and (second) markets evolve, some die and some prosper, forgetting however that markets are far from being perfect, and truly self-organized. The question is not if some die (which they should), the question is why the current system is not being able to self-control the proliferation of cheaters and pyramidal schemes on the entire pool of economic agents, in contrast to what happens in truly evolvable economic agents playing IPD or other economic-like games, showing profound traces of self-organized features. Sadly, among many of these if not all, ultra market liberalism evangelists, self-organization is wrongly recognized as self-interest (check 3). Self-interest taken to this limits, is not only their repeated mantra, used to quote “there ain’t those things as a free lunch” (NFL) over and over again as their blindness. NFL after all, is instead broadly recognized as being connected to robustness in computational search and optimization areas. Unfornately, they will never recognize that even under some complex co-evolutionary domains (nature is full of them; where explicit targets or interests were not embedded on the evolutionary algorithm), indeed free-lunches were found (5).

In fact, in order to emerge as a truly self-organized system, self-interest, should constitute just one among many of the ingredients over the entire financial system, and not the isolated unique ingredient. Self-interest promotes amplification and positive feedback, which is – as I recognize – necessary. However, left alone, promotes instead dramatic snowballing drifts over chaotic regimes, due to it’s intrinsic amplification. It’s necessary to promote negative feedbacks as well. This is recognized for some time in neurosciences, neurocomputation and learning (check LTD – long term synaptic depression) (6,7):

• Learning by reinforcement good responses (Positive Feedback) is a process that by definition never stops. There is not an explicit rule that ends the reinforcement whenever the goal has been reached. On the other hand, if learning proceeds only by correcting mistakes it implies a process that stops as soon as the goal is achieved. This prevents formation of “deep holes“, i.e. highly stable states from which adaptation to new rules is difficult and slow, requiring, perhaps, a significant amount of random noise.
• If an adaptive system is placed on a new environment, or otherwise subjected to learning something new, the likelihood of making mistakes is generally larger than the chance to be initially right. Therefore, the opportunity to shape synapses is larger for the adaptive mechanism that only relies on mistakes, leading to faster convergence.

Of course, being arrived here at this turmoil chaotic stage, we are nowadays assisting – ironically – to a dramatization of Bush’s Administration speeches. Bush words “Markets are not working properly” are indeed surprising from what we are used to expect from him, his office, as his known to be quite unusual words within his neo-liberal Haliburton centered – pro oil pro preemptive war – social networks (at VisualComplexity), however, subliminally he’s nothing else then reinforcing McCain’s election over Obama, as if great part of the actual crisis did not derived from the Republican past “I see nothing, I hear nothing and I say nothing” political strategy. Lying saying the truth, again ironically, this is the quickest formula to maintain things as they were before, the entire status quo going on, while markets in despair applaud –awkwardly – state interventions for the first time (or second to be precise – 1929; also check October 1907’s actions under J.P. Morgan). Being a liberal I have always believed that some ground-smart rules are always necessary. Let’s face it: even when we drive our car over a highway.

Take the following example. For some moments image yourself to puzzle out how to create a mathematical-algorithmic model on how a flock of birds fly in collective formation. You could on one hand try to model the dynamics of each part, using differential equations, in order to achieve somehow the global behaviour – however, the phenomena is so complex and intricate that differential equations could not handle it. On the other hand, you could try to observe the phenomena innumerable times while envisioning a set of rules, based on the behaviour of the whole system, however as is typical in Self-Organized complex phenomena’s there is no pre-commitment to any particular representational scheme: the desired behaviour is distributed and roughly specified simultaneously among many parts, and there is minimal specification of the mechanism required to generate that behaviour, i.e. the global behaviour mainly evolves from the many relations of multiple simple behaviours. Parts and wholes behave differently. Relations are the key. Surprisingly, and having self-organization theory in mind, you could envision 3, and just 3 simple generative rules following positive and negative feedback features that are able to precisely model this complex phenomena (check Boids): (one) Separation: steer to avoid crowding local flockmates, (two) Alignment: steer towards the average heading of local flockmates, and (three) Cohesion: steer to move toward the average position of local flockmates. Rather, non-linear phenomena are most appropriately treated by a synthetic approach, where synthesis means “the combining of separate elements or substances to form a coherent whole’. In non-linear systems, the parts must be treated in each other’s presence, rather than independently from one another, because they behave very differently in each other’s presence than we would expect from a study of the parts in isolation (4). In order to form the complex coherent whole, antagonistic measures are needed. Realistic counter-powers on the entire global economic TIC-TAC.

That’s why also, being a liberal, I defend a priori over a posteriori interventions. States, after all, were not created or envisioned to be omnipresent firefighters, specially to those few that under the 80-20 Pareto rule umbrella, profited before, month after month, with the current aftermath ([…] And so, my fellow Americans: ask not what your country can do for you – ask what you can do for your country […], John F. Kennedy). Not only Size is important (1) (critical mass) as well as Time, that is, over when should we lay down initial conditions in order to emerge the complex whole to flourish on the precise and desirable Self-Organized domain. Even better than nothing, as I believe, the current and late a posteriori state intervention will only endure the collective illusion for a while.

Not only have we recognized that markets are not perfect as Smith’s Invisible Hand metaphor seems to be dead wrong (3).  As David Sloan Wilson tackles it:

[..] I hope that our economy recovers, but the time has come to declare its guiding metaphor dead. This is the metaphor of the invisible hand, which makes it seem as if the narrow pursuit of self-interest miraculously results in a well-functioning society. […] The collapse of our economy for lack of regulation was preceded by the collapse of rational choice theory. It became clear that the single minimalistic principle of self-interest could not explain the length and breadth of human behavior. […] Mandeville could not have been more wrong about actual nature of bees. There is a difference between self-organization and self-interest. Beehives and other social insect colonies are indeed self-organized. There is no single bee commanding the troops, certainly not the queen. Each bee plays a limited role in the economy of the hive, just as a single neuron plays a limited role in the economy of the brain. The intelligence of both can be found in the interactions among the parts, which have been shaped by natural selection operating over countless generations. But bee behavior cannot be reduced to a single principle of self-interest, any more than human behavior. There are solid citizens and cheaters even among the bees, and the cheaters are held at bay only by a regulatory system called “policing” by the biologists who study them. […] We can argue at length about smart vs. dumb regulation but the concept of no regulation should be forever laid to rest. […]

Somehow, within the middle of countless wrecks, affecting innumerable millions of people thorough out the planet (even those not playing at the stock-exchange), via energy, tax, food and life cost raisings, we are still assisting at truly interesting phase-transition times. The question is: will we learn from it, or will we maintain the recent blind faith that markets, by themselves, will drive us all – similarly to communism – to the land of milk and honey?

1. in Milgram, Bickerman and Berkowitz, “Note on the Drawing Power of Crowds of Different Size“, Journal of Personality and Social Psychology, Vol 13(2), Oct 1969, pp. 79-82. Abstract: Reports on the relationship between the size of a stimulus crowd, standing on a busy city street looking up at a building, and the response of passersby. As the size of the stimulus crowd was increased a greater proportion of passersby adopted the behavior of the crowd. Data included 1424 pedestrians. The results suggest a modification of the J. S. Coleman and J. James model of the size of free-forming groups to include a contagion assumption.
2. in V. Ramos et al., “Social Cognitive Maps, Swarm Collective Perception and Distributed Search on Dynamic Landscapes“, 2007. Abstract: Swarm Intelligence (SI) is the property of a system whereby the collective behaviors of (unsophisticated) entities interacting locally with their environment cause coherent functional global patterns to emerge. SI provides a basis with wich it is possible to explore collective (or distributed) problem solving without centralized control or the provision of a globalmodel. To tackle the formation of a coherent socialcollective intelligence from individual behaviors, we discuss several concepts related to Self-Organization, Stigmergy and Social Foraging in animals. Then, in a more abstract level we suggest and stress the role played not only by the environmentalmedia as a driving force for societal learning, as well as by positive and negative feedbacks produced by the many interactions among agents. Finally, presenting a simple model based on the above features, we will adressthe collective adaptation of a socialcommunity to a cultural (environmental, contextual) or media informational dynamical landscape, represented here – for the purpose of different experiments – by several three-dimensional mathematical functions that suddenly change over time. Results indicate that the collective intelligence is able to cope and quickly adapt to unforseensituations even when over the same cooperative foraging period, the community is requested to deal with two different and contradictory purposes.
3. in David Sloan Wilson, “The Invisible Hand is Dead. Long Live (Smart) Regulation“, in Axis of Logic, Sep. 2008.
4. in V. Ramos, “On the Implicit and on the Artificial – Morphogenesis and Emergent Aesthetics in Autonomous Collective Systems“, in ARCHITOPIA Book, Art, Architecture and Science, INSTITUT D’ART CONTEMPORAIN, J.L. Maubant et al. (Eds.), pp. 25-57, Chapter 2, ISBN 2905985631 – EAN 9782905985637, France, Feb. 2002.
5. in Wolpert, D.H., and Macready, W.G. (2005) “Coevolutionary free lunches,” IEEE Transactions on Evolutionary Computation, 9(6): 721-735.
6. in Chialvo, D.R., Bak, P., “Learning from Mistakes“. Neuroscience, Vol. 90 (4), pp. 1137-1148, 1999.
7. in  Bak, P., Chialvo, D.R., “Adaptive Learning by Extremal Dynamics and Negative Feedback“, Phys. Rev. E., Vol. 63, p. 031912, 2001.
8. in Susan Gaidos, “Honeybees do the Wave“, in Science News, Web edition, Sep. 2008.