“Hershey’s major shareholder pledges to keep power”
http://news.yahoo.com/s/nm/20070401/bs_nm/hershey_cadbury_wsj_dc

The article I read was discussing the current situation of share ownership within the Hershey’s corporation. There had been speculation of a merger with a Cadbury in Britain, but recent information released by the majority shareholders makes that seem unlikely to happen. Hershey’s Trust who is the majority shareholder stated they were satisfied with current management and put forth a plan to gain majority ownership rivals in the future. The reason most of the speculation about combining with Cadbury came into light was the recent split of Cadbury into two seperate divisions. That aspect tied with a desire of Hershey’s to go global made a link sof the two seem possible and even probable to some. Hershey’s has had some financial downfalls lately but plans to refocus on traditional chocolate to boost sales.
The reason I felt this article was appropriate for discussion within the blog is the topic it covers. In class we have spent a good amount of time discussing different transactions and even the concept of power within social relations. The Hershey’s stock is traded within the New York Stock Exchange, which is one of the auctions that had been mentioned in class. This type of auction has multiple sellers and buyers, but as it could be seen within this article when one buyer or seller (or group of sellers and buyers like the Hershey’s Trust) has majority ownership of stocks they are in a greater position of power then the others. Another area power comes into play in the article is due to the trend of Hershey’s stock value falling lately. This shows that owners of the stock are losing power when compared to potential buyers. Their “good” is no longer worth as much so they have less power to barter. Another part to the article dealt with a new type of power that comes from outside the economic transaction. The article discussed the Hershey’s Trust thoughts about selling their majority as of last year and the negative reaction they recieved from stakeholders as a result of it. A stakeholder is anyone outside of an economic transaction that is affected by the economic transaction. In this specific case it was employees and communities worried about the loss of jobs that may possibly be a result of Hershey’s Trust selling the majority. It is an obvious display of power as Hershey’s Trust decided not to sell.

When it comes to athletic prowess, don’t believe your eyes.

The above link is New Yorker article written by Malcolm Gladwell on the highly esteemed Allen Iverson. Iverson is ranked as one of the best players of the game by coaches and fans alike. He is an exciting player to watch but is he worth all the press? Based on research by Berri et al. and a close look at his statistics, Iverson is far from one of the top 10 players in the game. Based on the algorithm by Berri, he averages a career ranking of 116 in the league. But we’d still rather see Iverson have the ball in the last seconds of the game instead of one of his statistically better teammates. How can this be?

Gladwell explains that, even around the business table, there is room for an unwarranted reputation. For example, Gladwell says, “Boards of directors vote to pay C.E.O.s tens of millions of dollars, ostensibly because they believe—on the basis of what they have learned over the years by watching other C.E.O.s—that they are worth it.” The truth of the matter is that we often become blind to all the stats and variables involved and it becomes too hard for us to judge without bias. We choose what we want to see, regardless of the actual statistics. We often look to the experts to see where they stand but Gladwell shows that even the experts aren’t perfect. “Basketball’s decision-makers, it seems, are simply irrational.” From this article, it seems to me like the next big basketball star is the next big herd.

Jumping on the Bandwagon

In this blog article, “Jumping on the Bandwagon,” Ian Cooper analyzes the rise and fall of the stock market, specifically, Black Monday, and how it occurred as a result of an information cascade among investors. A few investors may latch on to a bit of information and try to make a quick buck, buying an otherwise unimpressive stock, creating a snowball effect that drastically increases the price of the stock. Investors trying to make a short-term profit dump the stock as quicky as they had gotten it and send the stock plunging. And this is what happened on October 19, 1987, Black Monday.

The point of this blog article, is that these kinds of cascades can come from many sources, not just the outweighing of one’s own private information by the actions of two other people. A large investment can trigger this effect, as well as the dissemination of one small piece of information to a broad network of people by a weighted source, such as the media. With this knowledge, investors can try to profit from the market by trying to stay one step ahead of the cascades.

The rationale behind the quick buying and selling of a stock, and the resulting rise and fall of its value makes sense on the individual investor scale, whereas the investors in the cascade have no information of value other than previous actions. There is the added pressure of trying to buy the stock while it is still low, further speeding up the cascade. And just as easily as one rumor has investors clamoring for a stock, the stock can plummet when early investors decide to sell. This is slightly different, whereas in class we had discussed the effect of terminating/reversing a cascade at the present node, in the stock market, it is generally the earlier investors who started the cascade that also reverse it, by selling their stock while they’re ahead.

As long as individuals are concerned with their own self-interest, this type of market behavior is likely to continue as we all play the odds to find the next big thing.

http://www.cnn.com/2007/TECH/internet/03/29/no.facebook.lent/

Every year for lent, Catholics and other people who celebrate lent are adding new vices to their list of things to give up. Instead of the usual foods, drinks, or cigarettes, people are forfeiting their online social networks for the Lenten season. Online social networks such as MySpace and Facebook and Instant Messaging Services provided by AOL, MSN, and Yahoo, have been booming in the past 5-10 years, and a lot of people have been finding themselves spending too much time on these websites and services. With features such as listing interests and personal information, global and private chatting, and sharing other types of information, people can spend hours updating their own profiles or viewing others.

People who celebrate lent like to try to give up something that is really important in their everyday life and online social networks certainly falls into this category for many people. A lot of people who participate on these websites find themselves subconsciously visiting these sites when they open an internet browser, even if it wasn’t their original intent to go online. The choice of giving up online social networks may prove to be very beneficial when they find out how much more free time they have to allocate into other areas such as school or jobs. Nonetheless, these online services continue to attract and keep many people entertained with their ability to connect people from as far as around the world or as close as to the next room over.

Jumping On The Bandwagon Effect

American Venture Magazine has an interesting article (from a bit ago) on the effect of “Bandwagons” (in other words information cascades) on the world of venture capitalism. Venture capitalist investments tend to not be taken lightly as investing in this sense is a very high-risk business. A great deal of research and analysis goes into finding the next big investment, however investors can’t wait forever. VC is all about grabbing at the initial economic surplus before it is all gone, and so in order to be effective investors must move fast. This fast-paced nature leaves the investing world open to attack, so to speak, by what are essentially financial information cascades.

For instance, say a small technology company Stuff Co. has a stock price of $20 at the beginning of the day. They announce that they are preparing to release a new product that will revolutionize the Stuff industry. Investor X, a bigwig in the VC world, (unknowningly) makes a mistake and decides to pull his investment from the company as he was misinformed of the announcement by the company by one of his associates (who will later be fired). The next few smaller investors see both the announcement and X’s action and decide to pull out as well, as they assume X must know something they don’t since X is such a big name. Now more investors see that despite this huge announcement a number of large investments have been pulled and they decide similarly to withdraw so as not to be left out. Stuff Co., who had assumed their stock would skyrocket with the announcement of the revolutionary technology, is shocked to find out that their stock is now worth a total of $1 a share! Will this extreme downfall last?

There are two posibilities for the future of this once great company. In possibility 1, A new investor could come along, realize the overwhelming benefit of the new product, and decide to invest oodles of money. Subsequent investors would see this action, which supports the initial announcement, and decide to invest as well, thereby reversing the cascade. Soon after, Stuff Co.’s stock has risen again (though whether or not it meets or exceeds the intial stock price depends on the situation). In possibility 2, before the market has a chance to begin to balance out, Stuff Co. goes under due to the tremendous, unexpected loss of capital. In this example, let’s assume that the later takes place. Now this new, revolutionary technology may be lost to society forever.

These kind of cascades often result in “bumps” in the overall value of a stock. The stock will briefly shift a large amount (either positive or negative depending on the initial actions), then quickly move in the other direction, often ending up where it started. Because of this, it’s quite possible for big events, such as the announcement of a new technology or product, may be met with decidedly less-than-expected results. A new product announcment could even potentially tank what could be a great thing. It’s a pretty scary thought that the future of many potential innovations may never see the light of day due to a very small number of misguided mistakes, and the instinctive human need to follow the crowd.

The paper Imperfect Learning and Error Cascades in Sequential Guessing Games: An Experiment describes an experiment in which students played Chinos, a Spanish parlor game. In this game each player hides a number of coins or other small items in his or her hand. Then each player must guess the number of items in each person’s hand. This is done sequentially such that if you are player t, you have also heard the guesses of (t-1) other players. This resembles an information cascade because you have your own private information (your instinctual guess) and you know the guesses of those who have guessed before you (who based their information instinct and on those who have guessed before them).

Four sessions of a Chinos-like-game was played with 12 students per session. Each session contained 20 rounds. Each session had four groups of three such that through the course of the experiment 16 groups of 20 rounds were observed. Players were anonymous to each other but consistent throughout the game (guessing order was kept constant). Students were paid to participate and there was a small monetary prize for each round.

The results of the experiment were averaged together and show that as player position (t) increases the probability of a correct guess increases. It was also noticed that these probabilities were not as high as the theoretical probabilities. This abnormality suggests players deviated from equilibrium strategies and made errors in judgment. These error cascades were noticed in several rounds. The authors of this paper describe an error cascade as “the situation where the deviation from learning by a player increases with respect to the deviations from learning by preceding players.” Section 6 of this paper describes the modeling of error cascades.

It was concluded that if preceding players started to deviate from the correct guess, there would be imperfect learning and this deviation would continue to cascade through the game. The authors argue that imperfect learning results in error cascades. As more rounds of the game are played, the first guesser in the sequence realizes that he or she has a low probability of winning; therefore it is less costly for that player to deviate from optimal behavior. I imagine myself playing this game as the first guesser. I would probably get upset or frustrated and want to either make different guesses to try to win (compared to following my instinct which has caused me to be wrong so far) or make absurd guesses in order to make other players loose.

The IEEE article “Overview of Sensor Networks” (Aug 2004) provides a concise summary on what wireless sensor networks are, their applications, and the challenges in their design.  Wireless sensor networks are collections of small computing devices with some sensing ability constrained by low computing capability and low energy restrictions.  Each device is ineffective on its own, but the overall interactions and behavior of the sensor network can produce results far superior to a single, high-performance device.

The primary application of wireless sensor networks is the monitoring of environmental processes.  This could range across anything from environmental monitoring to security to equipment maintenance.  The example given in the article is the monitoring of the ecosystem that exists in and around one redwood tree.  A network of sensors was positioned along the length of tree in various locations, measuring temperature, barometric pressure, and humidity, along with several other environmental factors.  Every five minutes, the sensors would collect data on these various environmental conditions and then transmit it to the base node.  The base node would process it and transmit it to the outside world, where the researchers could do further analysis on the data.

Several of the challenges involved in sensor networks include the complexity of the individual interactions among all the nodes, the sheer number of nodes, and the limited computing and power capabilities of each node.  The high number of nodes in the sensor network, along with limited communication among them, require that the nodes be self-organizing.  There is no central control unit that maintains and tends to every individual node.  The nodes must communicate among each other to organize themselves in order to accomplish their overall task.  In the example above, there was in fact a central processing node that handled the computations and managed the sensor nodes, but in a more advanced sensor network, the network would be large enough such that a central node would not be feasible.  Another challenge is the power limitations and computing limitations of each node.  Each node must be inexpensive and small, which restricts the individual capability of each node.  Thus, they will end up relying on each other to make up for their individual limitations.  For example, nodes in a sensor network will generally be in some sort of “sleep” mode to conserve power, and only activate when required, as communicated by neighboring nodes.

Wireless sensor networks are part of an emerging field in wireless information networks that are made possible because of the miniaturization of electronics along with advances in algorithms designed to handle networks.  Although this course is mostly concerned with social networks, the same principles and ideas surrounding social networks, along with their challenges and obstacles, can be applied to wireless sensor networks.

http://www.reason.com/news/show/33643.html

This article talks about information cascades in military intelligence, which is partly the reason that certain events could not have been predicted or “wrong” decisions were made – when thinking ex-post. A few examples the author uses are the inability to prevent 9/11 with vague warnings, the situation with “yellowcake uranium,” and deadly aluminum tubes Sadam was going to use for centrifuges. Robert Anton Wilson, in the SNAFU principle, says that the reason for this is because subordinates tend to tell their superiors what they like to hear. The higher up you go in the hierarchy, the more distorted information gets. A problem with “accountability” in information work is that you will get more noticed if you dissent from the majority when the majority is correct than when you dissent from the majority when they are wrong. Therefore, playing follow-the-leader becomes a rational strategy. Since we all have limited time and information, following the majority in some cases makes sense, but sometimes you have to rethink of whether it is a good idea.

Information cascades that we learned about in class seem to be a big factor of the “mess ups” of the government. It is because after enough people make a choice, you begin to question your own choice because if enough people choose something you think they all can’t be wrong. Therefore, you ignore your own information and begin to switch choices - if that is where the majority lays. This becomes the problem in military intelligence, but their reasoning also involves the “accountability” factor of getting something wrong. If the majority also gets it wrong, you hardly get notice. Therefore to keep yourself under the radar, it is “better” to just follow-the-leader. Like the author said, it is not always bad to ignore your own signals and enter an information cascade. For common things like picking restaurants or automobiles, playing the follow-the-leader game will not get you in trouble. It is when the situations become uncommon and more serious, is when information cascades become no trustable.

The brain is one of the most profound networks we can study and implements information cascading in a unique framework. The brain is a network of many neurons that provide one-way signals to each other, similar to the linear model presented in class. Neurons signal other neurons to fire by releasing neurotransmitters at synapses to other neurons upon reaching an action potential, causing affected neurons to repeat this procedure to continuing neurons. Brain activation occurs in groups rather than in a straight line as presented in class. Therefore the first neuron activated in a cascade may activate several neurons and continue to activate other proximate neurons. As the cascade continues the growth of activated neurons may slow down as less neurons in a particular group are left to activate. The brain’s cascading is similar to the linear cascade in class in the sense that the longer the cascade lasts, the more number of nuerons become active. Our cerebral cascading explains how a certain smell or song may cascade into a larger population of neurons that encode for a larger concept or memory. For instance the smell of our first date’s perfume or the music playing during that car ride can invoke that experience long after its completed.

The Neural Cascade of Attentional Control provides some interesting evidence for cerebral cascading. The experimenters told the subject to focus on a computer screen in which the subject was presented with a left, right, or neutral signal (used as a control). Then a dot is then presented on either the left, right, or not presented according to the respective signals. This separated the subject’s spatial queue processing (expecting the dot to the left or right) from general queue processing (waiting for the left, right, or neutral signal to appear). The left and right signals created a spatial orienting signal in the brain, which the research has confirmed to be located in the more medial portions of the frontal cortex. The lateral parts of the frontal cortex first activate in response to the general queue processing with the subject waiting for the signal. Research then found that about 400ms later the medial parts of the brain would become more active with the subject utilizing spatial attention to expect the dot to the left or right. The brain imaging data reveals that after repeated trials the activations become stronger, suggesting there is some pathway in between the two brain activities that utilizes a cascade. This research illustrates a cascading principle not outlined by the linear model. The more often a cascade is used, the stronger it becomes. Professor Easley presents a framework for this idea with the umbrella example in class. If the people are only correct part of the time then we may not bring an umbrella every time we see people with umbrellas. This parallels the idea that if the subjects do not see a left, right, or neutral signal when expecting to, then the entire cascade may fall apart. However if the people with umbrellas are reliable then the cascade becomes very strong and we would be inclined to follow it more often.

Traumatic Brain Injury illustrates the necessity of cascading activation in the brain. Neurons need to fire and interact to maintain their cascading exchange of resources within the brain. Physical damage to the brain often kills several brain cells that may release toxic chemicals and kill surrounding cells, causing a negative cascade and severe brain damage. At the same time the loss of neurons in previous cascades hurts further neurons dependant on these intermediaries. This illustrates one implementation of The Tipping Point, where a certain selection or number of influential neurons may cause systemic brain damage if destroyed through physical damage. On a lighter note, researchers have recently found a potential treatment for the destructive cascading of brain damage: progesterone. The hormone prevents further propagation of toxic chemicals released from brain damage, and also develops new circuits and fixes old connections to repair the previous cascading paths in the brain needed to maintain a healthy level of activation.