Sunday, August 30, 2020

Hiyerarşide İletişimin Rolü

 

 

Hiyerarşide İletişimin Rolü

Communications’ Role in Hierarchy

Edwin Hutchins, Cognition_in_the_Wild

 


 

Bir karara ulaşmak için sık görülen bir çözüm belirli bir bireye gerçeğin doğasını belirleme otoritesini vermektir.  İlgili gerçeğin sosyal olarak tanımlandığı düzenlerde bunun görülmesi özellikle kolaydır– kanun gibi, olayların önemli bir durumu (suç veya masumiyet) yalnızca bir otoritenin (hakim) varlık nedeni ile vardır.

A common solution to the problem of reaching a decision is to grant to a particular individual the authority to declare the nature of reality. This is especially easy to see in settings where the relevant reality is socially defined - such as the law, where an important state of affairs ( guilt or innocence ) exists only because some authority (a judge) says it exists.

 

Fakat bu çözüm, belirli bir yoruma bağlanmanın zaman azlığı veya diğer etkenlerin gerektirdiği yerlerde,  fiziksel gerçeklere bağlı olarak ta benimsenir.  Bu ikinci durum iki çeşit olarak belirir: birincisi topluluğun diğer üyelerinin otoriteye delil sunması, ve ikincisi otoritenin yalnız başıma hareket etmesi.

But this solution is also adopted with respect to physical realities where time pressures or other factors require a commitment to a particular interpretation. This second case comes in two versions : one in which the other members of the community may present evidence to the authority, and one in which the authority acts autonomously.

 

İletişimsiz Hiyerarşi

Hierarchy without Communication

Bir grubun bütün elemanlarının bir yoruma ulaşmaya çalıştıklarını varsayın fakat bir alt ağın bütün grup üyeleri için gerçeğin doğasına karar verme otoritesine sahip olduğunu kabul edelim.

Suppose all members of a group attempt to form an interpretation, but one network has the authority to decide the nature of reality for all the members. 

Durumu yorumlamanın zihinsel işi, onaylama önyargısı olan tek bir kişinin araştırabileceği bir yorum uzayından daha çok alternatif araştırmaya izin verecek bir şekilde, sosyal olarak dağıtılabilir; yine de, eğer alternatif yorumlar hiç bir zaman karşı karşıya gelmezse, daha geniş araştırma hiç bir zaman gerçekleşmemiş olabilir.  Grup tarafından ulaşılan karar sadece tek bir kişinin kararıdır.

The cognitive labor of interpreting the situation may be socially distributed in a way that permits an exploration of more alternatives in the interpretation space than would be explored by -a single individual with confirmation bias; however, if the alternative interpretations never encounter one another, the wider search might as well have never happened. The decision reached by the group is simply the decision of an individual.

 

Bunun bir çeşit ‘kral’ veya ‘diktatör’ modeli olduğunu hayal edebilir kişi, fakat iletişim eksikliği bu özelliğe sahip olmaması gereken durumlarda bu modeli getirebilir.  Daha önce tarışılmış gemi kazası,

bir grup içinde doğru yorumun ortaya çıktığı fakat grubun gerçeğin hangi modeli etrafında davranışını düzenlemek zorunluluğuna karar verme otoritesine sahip şahsa, nasılsa hiç ulaşmadığı bir örnektir.

 One might imagine this as a sort of "king" or " dictator " model, but lack of communication can also bring it about in situations that are not supposed to have this property. The ship collision discussed earlier is an example of a case in which the correct interpretation of a situation arose within a group but somehow never reached the individual who had the authority to decide which model of reality the group must organize its behavior around.

 

İletişimli Hiyerarşi

Hierarchy with Communication

Bu durum simulasyonda ağların birinin(otorite sahibi olan) diğer herbirinden giriş aldığı, fakat diğerlerinin birbirlerinden harici girişler almadığı bir iletişim deseni ile modellenmiştir.

This situation is modeled in the simulation by changing the communication pattern so that one of the networks (the one in the position of authority ) receives input from all the others, but the others do not receive external inputs from one another.

 

Simulasyonda bu koşullarda, otorite olan ağ kendisine diğer ağlar tarafından sunulan delilin ağırlığını takip edecektir.  Diğer ağlar yorum uzayında hareket ettikçe, diğer ağlar tarafından sunulan delilin çekim ağırlığı da hareket eder.

In the simulation under these conditions, the network that is the authority will follow the weight of the evidence presented to it by the other networks. As the other networks move in interpretation space, the center of gravity of the weight of evidence presented by the other networks also moves.

 

Diğer ağların otorite ile iletişiminde  ikna ediciliğine bağlı olarak, otorite şu veya bu yoruma çekileblir, veya hangi yorumun daha iyi olduğu hakkında fikrini değiştirebilir.

Depending on the persuasiveness with which the other networks communicate with the authority, it may be pulled to one interpretation or another, or even change its mind about which is the better interpretation.

 

Otorite böylece özel bir mekanizma haline gelir; zaman içinde her an kavramsal düzlemde bütün topluluğun ağırlık merkezi haline gelir.

The authority thus becomes a special kind of cognitive apparatus ; one that tracks the center of gravity of the entire community in conceptual space at each point in time.

 


 

Friday, August 21, 2020

Communications’ Role in Hierarchy

 

 

Communications’ Role in Hierarchy

Edwin Hutchins, Cognition_in_the_Wild

 

A common solution to the problem of reaching a decision is to grant to a particular individual the authority to declare the nature of reality . This is especially easy to see in settings where the relevant reality is socially defined - such as the law , where an important state of affairs ( guilt or innocence ) exists only because some authority (a judge) says it exists.

 

But this solution is also adopted with respect to physical realities where time pressures or other factors require a commitment to a particular interpretation . This second case comes in two versions : one in which the other members of the community may present evidence to the authority , and one in which the authority acts autonomously .

 

Hierarchy without Communication

Suppose all members of a group attempt to form an interpretation , but one network has the authority to decide the nature of reality for all the members.

 

The cognitive labor of interpreting the situation may be socially distributed in a way that permits an exploration of more alternatives in the interpretation space than would be explored by -a single individual with confirmation bias; however , if the alternative interpretations never encounter one another , the wider search might as well have never happened . The decision reached by the group is simply the decision of an individual .

 

 One might imagine this as a sort of "king" or " dictator " model , but lack of communication can also bring it about in situations that are not supposed to have this property . The ship collision discussed earlier is an example of a case in which the correct interpretation of a situation arose within a group but somehow never reached the individual who had the authority to decide which model of reality the group must organize its behavior around .

 

Hierarchy with Communication

This situation is modeled in the simulation by changing the communication pattern so that one of the networks (the one in the position of authority ) receives input from all the others, but the others do not receive external inputs from one another .

 

In the simulation under these conditions , the network that is the authority will follow the weight of the evidence presented to it by the other networks (figure 5.5). As the other networks move in interpretation space, the center of gravity of the weight of evidence presented by the other networks also moves.

 

Depending on the persuasiveness with which the other networks communicate with the authority , it may be pulled to one interpretation or another , or even change its mind about which is the better interpretation (figure 5.5c).

 

The authority thus becomes a special kind of cognitive apparatus ; one that tracks the center of gravity of the entire community in conceptual space at each point in time .




Friday, August 14, 2020

Grup Zekasının Ortak Akıl Yapısı

 

Grup Zekasının Ortak Akıl Yapısı

The Commonsense Architecture of Group Intelligence

Edwin Hutchins, Cognition_in_the_Wild

 


 

Genellikle bir grubun performansını iyileştirmenin en iyi yolunun üyeleri arasında iletişimi iyileştirmek olduğu varsayılır, veya diğer yandan, bir grupta eksik olanın iletişim olduğu...

It is often assumed that the best way to improve the performance of a group is to improve the communication among the members of the group, or, conversely, that what is lacking in groups is communication.

 

Simülasyonlarımız  bu soruyu cevaplayabilecek bir yol sağlıyor bize. Bunlar daha çok iletişimin her zaman daha az olandan daha iyi olmadığını belirtir.  Bazı koşullarda, iletişimin zenginliğini arttırmak grup seviyesinde istenmeyen özelliklerle sonuçlanabilir.

Our simulations provide us -with a means to answer this question.  They indicate that more communication is not always in principle better than less. Under some conditions, increasing the richness of communication may result in undesirable properties at the group level.

 

Eğer sosyal yapı ikna edicilik artırılarak, tekrar tekrar başlatlırsa,kavramsal uzaydaki ağların daha da hızlanır.  Var olan yorumlara ulaşmak için acele ederler (bazıları birine, bazıları diğerine), ve bir kere daha çevredeki delillere çok az yanıt verirler.

If the community is restarted again and again, with the persuasiveness increased each time, the velocity of the networks in conceptual space increases even more.  They hurry in groups to the available interpretations (some to one, some to the other ), and once there, they respond only a little to additional evidence from the environment.

 

Asimov’un tarif ettiği gibi bir megazihinin onaylama önyargısına karşı herhangi bir bireysel zihinden daha zayıf olacağı açıktır.  Bu, yorumlara acele ile ulaşan bir zihin olurdu, bir yoruma takılınca mutlak bir onaylama önyargısı ilan eden.  İletişimin gerçek bir insan sosyal yapısında bu çeşit bir grup hastalığına yol açacak kadar zenginleşmesi mümkün müdür? Belki.

It is clear that a megamind such as that described by Asimov would be more prone to confirmation bias than any individual mind. It might be a mind that would rush into interpretations and that, once it had lodged in an interpretation, would manifest an absolutely incorrigible confirmation bias.  Is it possible for communication to ever be rich enough in a real human community to lead to this sort of group pathology?  Perhaps.

 

Buckhout (1982) bir suçta şüphelilerin birleşik tariflerini şahitlerden istedi ve ‘grup tariflerinin bireysel tariflerden daha tam fakat beklenti hatalarına daha hassas olduğunu: bir dizi yanlış ve basmakalıp ayrıntılar içerdiğini’ raporlamıştır.

Buckhout (1982) asked groups to produce composite descriptions of a suspect in a crime that all had

witnessed and reported that " the group descriptions were more complete than the individual reports but gave rise to significantly more errors of commission : an assortment of incorrect and stereotyped details."

 

Yorumlama Çeşitliliği Üretme

Producing a Diversity of Interpretations

Onaylama önyargısı ile ilgili sorun bir canlıyı daha geniş olası yorumları araştırmaktan alıkoymasıdır.

İlk karşılaşılan yorum en  iyisi olabilse bile, olası yorumlar uzayındaki ek bir arama eldeki delillere daha iyi uyabilecek bir ikincisini çıkarabilir.

The problem with confirmation bias is that it prevents an organism from exploring a wider range of possible interpretations.  Although the first interpretation encountered may well be the best, a search of the  interpretation space may reveal another one that better fits the available evidence.  

 

Bu arayış nasıl tamamlanabilir?  Daha önce belirttiğim gibi, iletişimin eksikliğinde bireysel ağlar tarafından oluşturulan yorumlar – herbiri kendi onaylama önyargısını gösterdiği için – bireysel ağları temsil eden üç değişkene dayanır: altta yatan sınırlayıcı etkenler yapısı, çevredeki delillere erişim imkanı, ve başlangıçta harekete geçirici olgular dizisi.

How can this search be accomplished ?  I have already shown that in the absence of communication the interpretations formed by the individual networks - as each exhibits its own confirmation bias- depend on the three parameters that characterize the individual networks : the underlying constraint

structure, the access to environmental evidence, and the initial pattern of activation.

 

Eğer bir topluluk, bir sosyal yapı bu değişkenlerin en az biri cinsinden birbirlerinden farklı bireylerden oluşursa, bu topluluğun çok sayıda elemanının farklı yorumlara ulaşmaları daha olasıdır.  Böylece, topluluğun elemanları arasındaki iletişim çok zengin olmadığı takdirde, yorumların çeşitliliği daha kolay oluşur.

If a community is composed of individuals that differ from one another in terms of any of these parameters, then various members of the community are likely to arrive at different interpretations.  Thus, diversity of interpretations is fairly easy to produce as long as the communication among the members of the community is not too rich. 

Saturday, August 08, 2020

The Commonsense Architecture of Group Intelligence

 

The Commonsense Architecture of Group Intelligence

Edwin Hutchins, Cognition_in_the_Wild

 

 Hearings at European Parliament

It is often assumed that the best way to improve the performance of a group is to improve the communication among the members of the group, or, conversely, that what is lacking in groups is communication.

 Our simulations provide us -with a means to answer this question.  They indicate that more communication is not always in principle better than less. Under some conditions, increasing the richness of communication may result in undesirable properties at the group level.

 If the community is restarted again and again, with the persuasiveness increased each time, the velocity of the networks in conceptual space increases even more.  They hurry in groups to the available interpretations (some to one, some to the other ), and once there, they respond only a little to additional evidence from the environment.

 It is clear that a megamind such as that described by Asimov would be more prone to confirmation bias than any individual mind. It might be a mind that would rush into interpretations and that, once it had lodged in an interpretation, would manifest an absolutely incorrigible confirmation bias.  Is it possible for communication to ever be rich enough in a real human community to lead to this sort of group pathology?  Perhaps.

 Buckhout (1982) asked groups to produce composite descriptions of a suspect in a crime that all had witnessed and reported that " the group descriptions were more complete than the individual reports but gave rise to significantly more errors of commission : an assortment of incorrect and stereotyped details."

 Producing a Diversity of Interpretations

The problem with confirmation bias is that it prevents an organism from exploring a wider range of possible interpretations.  Although the first interpretation encountered may well be the best, a search of the interpretation space may reveal another one that better fits the available evidence.  How can this search be accomplished ?  I have already shown that in the absence of communication the interpretations formed by the individual networks - as each exhibits its own confirmation bias- depend on the three parameters that characterize the individual networks : the underlying constraint

structure, the access to environmental evidence, and the initial pattern of activation. If a community is composed of individuals that differ from one another in terms of any of these parameters, then various members of the community are likely to arrive at different interpretations.  Thus, diversity of interpretations is fairly easy to produce as long as the communication among the members of the community is not too rich.