瑪姬咕咕的生活指南 magigugu.com
返回   瑪姬咕咕的生活指南 > 節肢動物區 > 螞蟻

螞蟻
現在養一缸螞蟻當作寵物正在火紅,想要加入這個行列嗎?快來這裡認識螞蟻、尋找螞蟻、觀察螞蟻...一起討論關於螞蟻的種種新鮮話題。

回覆
 
主題工具
舊 2006-11-17, 10:12 PM   #1
野人
論壇管理員
 
野人 的頭像
 
註冊日期: 2004-05-16
文章: 10,029
野人 是一個將要出名的人 野人 是一個將要出名的人
預設

底下是知名螞蟻觀察家,史丹佛大學生物科學助教,黛柏拉‧M‧戈登博士(即"別和螞蟻拼命"作者)的一篇科普短文,介紹社會性昆蟲如何透過很簡單的行為互動,達到有效的分工。對於了解螞蟻、蜜蜂等社會性昆蟲的行為模式相當具有啟發性。原文來自一篇加大爾灣分校社會科學研究所的一篇PDF文件
為了便於譯介給有興趣的網友,我擅自把它偷下來,po在下面,之後慢慢有時間在逐段跟各位研討,一段時間後就會刪掉。

******************
The Organization of Work in Social Insect Colonies

DEBORAH M. GORDON

Task Allocation as an Ad Hoc, Dynamical Network

A social insect colony operates without any central control; no one is in charge,and no colony member directs the behavior of another. A worker cannot assess the needs of the colony. How do individual workers, using fairly simple, local information, in the aggregate produce the behavior of colonies? The dynamics of colony behavior results in task allocation [1]. Colonies perform various tasks, such as foraging, care of the young, and nest construction. As environmental conditions and colony needs change, so do the numbers of workers engaged in each task. For example, when more food is available or there are more larvae to feed, more foragers
may work to collect food. Task allocation is the process that adjusts the
numbers of workers engaged in each task in a way appropriate to the current situation.

社會性昆蟲的群體運作並沒有任何中央控制機制,之中沒有個體是奉命行事的,也沒有群體裡的成員會去指揮其他成員的行為。一個勞動者不能去判定群體的需求。那個別勞動者如何使用相當簡單、偏狹的資訊,去集成出群體的行為呢?工作配置中的群體行為動力學。群體去執行多樣的工作內容,比方覓食、照顧幼體,以及巢穴的建造。當環境條件及群體需求改變時,各類工作的勞動者數量也會跟著改變,例如,當可得的食物更多,或者需要餵養的幼蟲更多時,就可能有更多的覓食者去收集食物。工作配置是種以一方式依現況調整每類工作勞動者數量的過程。

註腳:
Task allocation is the process that adjusts the numbers of workers engaged in each task in a way appropriate to the current situation.


I study task allocation in harvester ants (Pogonomyrmex barbatus) [2]. Inside the nest, ants care for the brood (the preadult forms: eggs, larvae, and pupae); process and store seeds; construct and maintain chambers; and simply stand around doing nothing. The ants that work outside the nest are a distinct group, apparently older than the interior workers. I
divide the behavior I see outside the nest into four tasks: foraging, searching for and retrieving food; patrolling, assessing food supply and the presence of foragers from neighboring colonies; midden work, sorting the colony refuse pile or midden; and nest maintenance work, the construction and clearing of chambers inside the underground nest.

(野人註:harvester ants 收穫蟻,這戈登博士本身最引以為傲的研究領域;她曾研究這種螞蟻長達十七年的時間,可說是地道的收穫蟻通。)

作者以收穫蟻為"工作配置"過程的研究對象。在蟻巢內的,螞蟻對卵、幼蟲、預蛹、蛹的照料,處理及儲藏植物種子;建造及維護蟻窩;或者就只呆杵著沒事幹等。在蟻巢外的工作的螞蟻則又是一個顯著不同的群體,外表上看起來比巢內工作的老熟些。她把她所知的螞蟻在巢外的行為區分成四種工作類型:

覓食 (foraging)- 搜索並撿回食物。
巡邏 (patrolling)- 偵查食物供應及來自鄰近部落的覓食者。
堆糞 (midden work) - 分類(sorting)群體產生的廢棄物與糞便。
巢穴維護 (nest maintenance work) - 建造並清理蟻窩。

Tasks are interdependent; numbers engaged in one task depend on numbers engaged in another [3,4]. Ants switch tasks, though not all transitions are possible. In harvester ants, task switching funnels ants into foraging and away from tasks inside the nest [4]. An ant's decision whether to perform a task depends, first, on cues about the physical state of the environment: for example, if part of the nest is damaged, more ants do nest maintenance work to repair it. Task decisions also depend on
social cues arising from interactions with other ants.

各工作間是相互關聯的,螞蟻投入一類工作的數量取決於投入另一類工作的數量。螞蟻能改變工作類型,雖然不是所有工作都能夠相互過渡;在收穫蟻中,能從守門蟻工作轉換成覓食工作,從巢內工作轉換成巢外工作。一隻螞蟻選擇去執行一類工作,優先取決於環境中的物理狀態訊息:例如蟻巢部份破損,會有更多的螞蟻投入巢穴維護工作去修復它。工作類型的選擇也有賴於與其他個體間互動所帶來的社會訊息。

Workers from different task groups meet as they come in and out of the nest. The rate at which one ant encounters others influences its task decisions. The pattern of interactions among ants as they move around can be seen as a kind of ad hoc,dynamical network [5,6].

當不同工作類型的工蟻團隊在巢裡進出而相遇,個體與其他螞蟻接觸率會影響牠的工作選擇。這種發生在螞蟻行進間互動的模式,可以看成動態網路點對點通訊的型態。

When ants meet, they touch with their antennae. Antennae are the organs
of chemical perception. When an ant uses its antennae to touch the antennae or body of another, it can perceive the colony-specific odor that all nestmates share. We found that in addition to the colony odor, Pogonomyrmex barbatus ants have an odor specific to their task,
because the temperature and humidity conditions in which an ant works alter its cuticular hydrocarbon profile [7–9].For example, a forager makes long trips outside the nest in hot, dry air, and this increases the proportion of n-alkanes in its hydrocarbons. An ant may assess the task of an ant it meets using these task specific odors, so that an ant can evaluate
its rate of encounter with ants of a certain task. We are investigating how
patterns of interaction among workers contribute to the dynamics of task allocation in harvester ants. In laboratory studies, we found that an ant's decision whether to perform midden work depends on its recent rate of brief antennal contact with midden workers [10].

當螞蟻相遇時,牠們會用觸角去碰觸,觸角是化學感覺器官。當螞蟻藉由觸角去碰觸另一隻螞蟻的觸角或身體時,牠可以察覺到那種在同巢夥伴們共有的群體特有氣味。在Pogonomyrmex barbatus(一種在新墨西哥的螞蟻)蟻裡,除了群體氣味外,我們還發現個別工作類型特有的氣味,原因在螞蟻工作中的溫溼度條件會改變牠們表皮上的碳氫化合物前導物質;例如一隻覓食螞蟻長時間遊走在巢外又熱又乾的空氣裡,會導致牠表皮碳氫化合物中的烷比重增加,而螞蟻可能在接觸其他個體時,去判斷這些工作特有氣味。這樣,螞蟻便能就接觸率去評估一件確切的工作內容。而我們刻正探討收穫蟻間如何透過互動去促成工作配置動力學的模型,從實驗室研究,我們發現螞蟻是否選擇要進行堆糞工作,端視牠最近與其他擔當堆糞工作螞蟻簡短觸角接觸的接觸率而定。

In field studies, experiments suggest other decision rules based on encounter rate. One set of experiments shows how a forager's decision whether to go out and collect food depends on its interactions with other workers. We find that the rates of interaction with at least two types of workers influence a forager's activity: interactions with patrollers and with other foragers. Patrollers leave the nest each morning before foragers. First the nest patrollers go a few centimeters from the nest entrance and then turn back. The next set of patrollers go around the mound and then out on the trails. These trail patrollers choose the directions taken
later by the foragers [11], and foragers will ignore food sources not visited earlier by patrollers [12]. It appears that the first foragers prefer the directions in which they encounter most returning patrollers, and later foragers mimic the directions of the earlier foragers. The rate of interactions with patrollers determine whether foragers leave the nest.
Removal experiments [13] show that when nest patrollers do not return, activity outside the nest ceases; there is no further patrolling and foraging never begins. When trail patrollers do not return, outside activity ceases, and foraging never begins. Thus the patrollers influence an all-or-none decision, whether to forage or not on a given day.The return of the first, nest mound patrollers seems to inform the rest of the exterior workers, including foragers, that it is feasible to leave the nest that day. Nest mound patrollers may assess humidity and temperature. After the nest mound patrollers have gone back in, trail patrollers choose foraging directions, based on encounters with the foragers with neighboring colonies and perhaps on food availability [14,15]. Once foraging has begun, a forager's decision whether to go out to forage depends in part on its rate of contact with successful returning foragers [11,13].

在田野研究,經由實驗知道其他基於接觸率的工作選擇規則,其中一組實驗顯示覓食螞蟻對於外出並收集食物取決於跟其他工蟻的互動,我們發現這種互動至少有兩種影響覓食螞蟻行動的形式:與巡邏螞蟻的交互作用及與覓食螞蟻的互動,每天早晨巡邏螞蟻在覓食螞蟻之前離巢,首先巢穴巡邏螞蟻(nest patrollers)會出去到距蟻巢入口幾公分外的地方然後掉頭回去,接著下一群巡邏螞蟻會來到土丘周圍,然後循著蹤跡走去。這些循跡巡邏螞蟻(trail patrollers )所選定的方位,會被之後的覓食螞蟻採納,而且覓食螞蟻會忽略那些之前巡邏螞蟻未探訪的食物源。覓食螞蟻傾向採納接觸到最多返回巡邏螞蟻的方位。之後的覓食螞蟻會仿照(mimic)先前巡邏螞蟻採納的方位。與巡邏螞蟻互動的比率決定覓食螞蟻是否離巢。

移除實驗顯示,當巢穴巡邏螞蟻(nest patrollers)有去無回時,巢外的活動便會中止,不會有更遠的巡邏活動,而且覓食活動不會開始。當循跡巡邏螞蟻(trail patrollers)有去無回時,外頭的活動中止,且覓食活動不會開始。就這樣,巡邏螞蟻帶來了不是全部就是都沒有的決定,不管那天是不是覓食的好日子。首先回來的,巢穴土丘巡邏螞蟻似乎通報給靜待在外頭的工蟻,包括覓食螞蟻,那天可以離巢的消息。巢穴土丘巡邏螞蟻或許也評估溼度和氣溫。巢穴土丘巡邏螞蟻回來後,接著循跡巡邏螞蟻便會基於鄰近蟻群覓食螞蟻的遭遇次數,甚或食物的易取得性來選擇覓食方位,一但覓食活動展開,覓食螞蟻就會經由與成功返回的覓食螞蟻接觸率決定該往哪邊去覓食。

Experiments like these show how an ant's moment-to-moment decision
about which task to perform and whether to perform it actively depends
on its interactions with other workers. Interactions between workers of some task groups apparently provide negative feedback, whereas others provide positive feedback. It appears that what matters to an ant is the pattern of interactions it experiences, rather than a particular message or signal transferred at each interaction. Ants do not tell each other what to do when they meet, but the pattern of interaction each ant experiences
influences the probability it will perform a task. Each ant uses a set
of rules such as, “I'm a forager and if I meet a returning patroller every so often, I remain likely to go out.” Evidence for such a rule is that if the forager does not meet a returning patroller, the probability it will go out diminishes.

Research on social insects is only now beginning to unravel the local rules
that influence individual behavior. Social insect research is a small and young field. Most of the thousands of social insect species have never been studied at all. Honeybees have been more intensively studied than any other social insect species, because they have been agriculturally important for agriculture throughout human history. Even for honey bees many of the details of individual task decisions are not known. There is no reason to expect that the details of task allocation will be the same in all social insect species. In fact, because social insects thrive in such diverse
environments, it is likely that different species have evolved very different
social interactions. For example, the relation of patrolling and foraging in
harvester ants allows for very slow adjustment to changes in food availability, which is appropriate in the desert where food availability changes slowly. More opportunistic ant species, which quickly take advantage of small bursts in food supply, probably operate very differently.

An important question about task allocation in harvester ants arises from
intriguing results on the effects of colony age. Colonies live 15 to 20 years,
founded by a single queen who produces all the workers. Colonies begin
with 0 ants and grow to a size of about 10,000 ants when the colony is about 5 years old and begins to reproduce [14]; it then stays at about this size for another 10 to 15 years. The behavior of older, larger colonies, of about 10,000 ants, is more stable to perturbation and more homeostatic than that of younger, smaller ones of about 3,000 ants [3].

Because individual ants live only a year, this cannot be due to the experience of older ants. The simplest hypothesis is that individual decision rules are the same in young and old colonies, but such rules have a different outcome in a small and large system. For example, interaction rates in a small colony might be lower than those in a large one, because in a young, small colony, each ant has fewer nestmates it could meet. The dynamics of the interaction network seems to depend on its overall
size.

Several types of models have been used to model task decisions in social
insects (reviewed in [16]). Most of these model the behavior of workers engaged in one task, such as foraging, or in several related tasks, such as nest construction in wasps, which involves collecting both paper and water. These include self-organization models (e.g., [17]), which have mostly been used to predict the formation and shape of foraging trails (e.g., [17]). Versions of these models have been applied to more general
AI problems such as the traveling salesman problem [18]. Nest construction by wasps has also been studied theoretically
and empirically [19].

There have been few attempts to model formally the allocation of workers
among different tasks. One approach to modeling the allocation of
workers among tasks is the “foragingfor-work” hypothesis [20] that an individual's decision whether to perform a task may depend on whether it finds itself in a location where that task is required. The threshold model of Robinson and Page [21] is an informal model of honeybee behavior based on genotypic differences among workers. It supposes that each genotype has a threshold stimulus at which it will perform a task. Empirical studies of honeybees support the threshold model with regard to some tasks but not others [22,23].

So far there have been two theoretical models of task allocation in harvester ants. The first [14] involves a parallel distributed process, such as a neural network. In this model, individual decisions are based wholly on interactions with nestmates. The second model [24] is an analytic model that uses differential equations to describe more deterministically the dynamics of task allocation. In this model, an individual's decision about which task to perform and whether to perform it actively is based on two kinds of stimuli:
(1) environmental cues that determine whether the ant is “successful” at its task, e.g., whether a forager finds food and gets it back to the nest, whether a midden worker manages to carry a dead ant to the midden and dump it there, and so on, and (2) interactions with other individuals. This model is more realistic than the first in that it incorporates environmental as well as social stimuli. Following on this, we investigated how the robustness and sensitivity of the system depends on colony size and on the type of feedback between tasks [25]. Currently we are developing
an agent-based simulation to model task allocation in harvester ants using
empirical data to set the parameter values.

It seems likely that in all social insects, both environmental and social
cues contribute to an individual's task decisions. Picnics provide an example
of an environmental cue; if ants did not respond to changes in food supply, we would not see ants at picnics. Response to environmental stimuli is a component of several previous models (e.g.,[26]). Numbers of interactions, rate of interaction, transfer of material, waiting time to transfer material, or time elapsed since the last interaction are a component of several recent models of the organization of some aspect of social insect behavior (e.g., [17,27,28]).The most important outstanding questions about task allocation in social insects are probably similar to the outstanding questions about any complex biological system: How much do the attributes of the individual components (in social insects, the workers) contribute to the dynamics of the system? For example, is it reasonable as a first step to assume that all workers are alike? How do the internal dynamics of the system react to and bring about changes in the environment? The colony is not a closed system: it absorbs materials from, takes cues from, and in turn modifies its environment. Finally, how does the size of the system determine its dynamics? In social insects, this is a developmental question because colonies grow larger as they grow older. Though similar methodological questions apply to all complex systems,
the answers will be in the details, and thus will certainly differ among
systems.

註腳:
Research on social insects is only now beginning to unravel the local rules that influence individual behavior. Currently we are developing an agent-based simulation to model task allocation using empirical data to set parameter values.


REFERENCES
1. Gordon, D.M. The organization of work in social insect colonies. Nature 1996, 380, 121–124.
2. Gordon, D.M. Ants at Work: How an Insect Society is Organized. Free Press: Simon and Schuster, 2000 paperback; W.W. Norton: 1999.
3. Gordon, D.M. Group-level dynamics in harvester ants: Young colonies and the role of patrolling. Anim Behav 1987, 35, 833–843.
4. Gordon, D.M. Dynamics of task switching in harvester ants. Anim Behav 1989, 38:194 –204.
5. Adler, F.R.; Gordon, D.M. Information collection and spread by networks of patrolling ants. Am Nat 1992, 40, 373–400.
6. Albert, R.; Barabasi, A-L. Statistical mechanics of complex networks. Rev Modern Phys 2002, 74, 47–97.
7. Wagner, D.; Brown, M.J.F.; Broun, P.; Cuevas, W.; Moses, L.E.; Chao, D.L.; Gordon, D.M. Task-related differences in the cuticular hydrocarbon composition of
harvester ants, Pogonomyrmex barbatus. J Chem Ecol 1998, 24, 2021–2037.
8. Wagner, D.; Tissot, M.; Cuevas, W.; Gordon, D.M. Harvester ants utilize cuticular hydrocarbons in nestmate recognition. J Chem Ecol 2000, 26, 2245–2257.
9. Wagner, D., Tissot, M.; Gordon, D.M. Task-related environment alters the cuticular hydrocarbon composition of harvester ants. J Chem Ecol 2001, 27,
1805–1819.
10. Gordon, D.M.; Mehdiabadi, N. Encounter rate and task allocation in harvester ants. Behav Ecol Sociobiol 1999,45, 370–377.
11. Gordon, D.M. Behavioral flexibility and the foraging ecology of seed-eating ants. Am Nat 1991, 138, 379–411.
12. Gordon, D.M. The relation of recruitment rate to activity rhythms in the harvester ant, Pogonomyrmex barbatus. J Kansas Entomol Soc 1983, 56(3), 277–285.
13. Gordon, D.M. The regulation of foraging activity in red harvester ant colonies. Am Nat 2002, 159, 509-518.
14. Gordon, D.M.; Goodwin, B.; Trainor, L.E.H. A parallel distributed model of ant colony behaviour. J Theor Biol 1992, 156, 293–307.
15. Gordon, D.M.; Kulig, A.W. Founding, foraging and fighting: Colony size and the spatial distribution of harvester ant nests. Ecology 1996 ,77, 2393–2409.
16. Hirsh, A.E.; Gordon, D.M. Distributed problem solving in social insects. Ann Math Artif Intel 2001, 31, 199–221.
17. Deneubourg, J.L.; Goss, S.; Franks, N.; Pasteels, J.M. The blind leading the blind: chemically mediated morphogenesis and army ant raid patterns. J Insect
Behav 1989, 2, 715–729.
18. Dorigo, M.; Maniezzo, V.; Colorni, A. Ant system: Optimization by a colony of cooperating agents. IEEE Trans Systems, Man, and Cybernetics-Part B 1996, 26,
29–41.
19. Theralauz, G.; Bonabeau, E.; Deneubourg, J. L. The origin of nest complexity in social insects. Complexity 1998, 3, 15–25.
20. Tofts, C. Algorithms for task allocation in ants (a study of temporal polyethism theory). Bull Math Biol 1993, 55, 891–918.
21. Robinson, G.E.; Page, R.E., Jr. Genetic basis for division of labor in an insect society. In: The Genetics of Social Evolution. Breed; M.D., Page, R.E., Jr., Eds.;
Westview Press: Boulder, CO, 1989.
22. Winston, M.L.; Katz, S.J. Foraging differences between cross-fostered honey bee workers (Apis mellifera) of European and Africanized races. Behav Ecol
Sociobiol 1982, 10, 125–129.
23. Robinson, G. E.; Page, R.E., Jr. Genotypic constraints on plasticity for corpse removal in honey bee colonies. Anim Behav 1995, 49, 867–876.
24. Pacala, S.W.; Gordon, D.M.; Godfray, H.C.J. Effects of social group size on information transfer and task allocation. Evol Ecol 1996, 10, 127–165.
25. Pereira, H.; Gordon, D.M. A trade-off in task allocation between sensitivity to the environment and response time. J Theoret Biol 2001, 208, 165–184.
26. Jeanne, R.L. Regulation of nest construction behaviour in Polybia occidentalis. Anim Behav 1996, 52, 473–488.
27. Jeanne, R.L. Group size, productivity and information flow in social wasps. In: Information Processing in Social Insects; Detrain, C., Deneubourg, J.L., Pasteels,
J.M., Eds.; Birkhauser, Basel, 1999; pp 3–30.
28. Anderson, C.; Ratnieks, F.L.W. Worker allocation in insect societies: coordination of nectar foragers and nectar receivers in honey bee (Apis mellifera) colonies.
Behav Ecol Sociobiol 1999, 46, 73–81.
野人 目前離線   回覆時引用此篇文章
舊 2006-11-18, 05:22 PM   #2
野人
論壇管理員
 
野人 的頭像
 
註冊日期: 2004-05-16
文章: 10,029
野人 是一個將要出名的人 野人 是一個將要出名的人
預設

戈登博士的研究其實在科景有一篇簡短的介紹,可以參考
螞蟻巢中簡單的複雜系統
野人 目前離線   回覆時引用此篇文章
舊 2007-05-02, 10:20 PM   #3
野人
論壇管理員
 
野人 的頭像
 
註冊日期: 2004-05-16
文章: 10,029
野人 是一個將要出名的人 野人 是一個將要出名的人
預設

上面的翻譯,我翻得很彆扭,誰能幫忙校對一下?
因為這篇研究跟我觀察到的"評價系統"議題有關;可能能夠建立一個簡單的模型,解釋蟻群如何用一個的機制,有效率地達到複雜的團隊分工效果。

有關評價系統的假想,請參見去年底我在飼育日誌的探討
這次因為觀察東京巨山蟻,發現似乎有不同的作法(也算是評價系統的一種"變形應用");但我個人時間比較零碎,設備跟基礎知識沒辦法讓自己更加系統化地深入探討,希望提出一些片面而簡略的個人觀察,讓有興趣追究下去的朋友有個起點。
野人 目前離線   回覆時引用此篇文章
舊 2007-11-03, 06:46 PM   #4
野人
論壇管理員
 
野人 的頭像
 
註冊日期: 2004-05-16
文章: 10,029
野人 是一個將要出名的人 野人 是一個將要出名的人
預設

真的難翻~尤其是那段一會兒巢穴巡邏螞蟻,一會而循跡巡邏螞蟻的地方(最後那段有翻成中文的地方)...

做這種事真的要有很大的熱情驅動~
剩下的改天吧。
野人 目前離線   回覆時引用此篇文章
舊 2017-12-20, 09:14 PM   #5
野人
論壇管理員
 
野人 的頭像
 
註冊日期: 2004-05-16
文章: 10,029
野人 是一個將要出名的人 野人 是一個將要出名的人
預設

哇~ 經過十年了...

來看一則研究報導:
ScienceDaily - An algorithm that explains how ants create and repair trail networks

以前探討過,我認為螞蟻互動就像拍賣網站的評價系統一樣,會在很短的時間內讓東西更好吃,或對外來威脅的仇恨值迅速暴增。這則螞蟻行為研究也證實了這件事:

Ants of C. goniodontus choose which route to take at a junction by following pheromone laid by the ants that recently crossed that junction. The pheromone evaporates, so the path that recently had the most ants is the most attractive one. By marking the ants with nail polish, Gordon found that the same ants tend to go along the same trails from a nest.

現在可以更清楚,在沒人發號施令的情況下,螞蟻是如何藉由個體自發性行為,形成整個團隊有序的合作,這就是自組織理論(Self-organizing Theory)!

報導中的 Cephalotes goniodontus 是有一種生活在中美洲,有著許效舜方腮頭特徵的螞蟻。牠們工蟻特化的頭可以拿來當門,負責抵住樹樁上巢室的洞口,防止宵小進入。這種螞蟻也曾被證實,蟻后在婚飛時可以跟多隻雄蟻交配,然後儲藏他們的精液,產生遺傳多樣性高的子代。
野人 目前離線   回覆時引用此篇文章
回覆

書籤

主題工具

發文規則
不可以發表新主題
不可以發表回覆
不可以上傳附件
不可以編輯自己的文章

啟用 BB 代碼
論壇啟用 表情符號
論壇啟用 [IMG] 代碼
論壇啟用 HTML 代碼



所有時間均為 +8。現在的時間是 11:12 AM


Powered by vBulletin® 版本 3.8.4
版權所有 ©2000 - 2018,Jelsoft Enterprises Ltd.