Hugendubel.info - Die B2B Online-Buchhandlung 

Merkliste
Die Merkliste ist leer.
Bitte warten - die Druckansicht der Seite wird vorbereitet.
Der Druckdialog öffnet sich, sobald die Seite vollständig geladen wurde.
Sollte die Druckvorschau unvollständig sein, bitte schliessen und "Erneut drucken" wählen.
E-BookEPUBDRM AdobeE-Book
552 Seiten
Englisch
Elsevier Science & Techn.erschienen am29.07.2011
Advances in the Study of Behavior was initiated over 40 years ago to serve the increasing number of scientists engaged in the study of animal behavior. That number is still expanding. This volume makes another important contribution to the development of the field by presenting theoretical ideas and research to those studying animal behavior and to their colleagues in neighboring fields.

Advances in the Study of Behavior is now available online at ScienceDirect - full-text online from volume 30 onward.mehr
Verfügbare Formate
BuchGebunden
EUR124,00
BuchGebunden
EUR131,00
BuchGebunden
EUR116,00
E-BookEPUBDRM AdobeE-Book
EUR112,00
E-BookEPUBDRM AdobeE-Book
EUR124,00
E-BookPDFDRM AdobeE-Book
EUR124,00
E-BookEPUBDRM AdobeE-Book
EUR112,00
E-BookEPUBDRM AdobeE-Book
EUR124,00
E-BookEPUBDRM AdobeE-Book
EUR124,00

Produkt

KlappentextAdvances in the Study of Behavior was initiated over 40 years ago to serve the increasing number of scientists engaged in the study of animal behavior. That number is still expanding. This volume makes another important contribution to the development of the field by presenting theoretical ideas and research to those studying animal behavior and to their colleagues in neighboring fields.

Advances in the Study of Behavior is now available online at ScienceDirect - full-text online from volume 30 onward.
Details
Weitere ISBN/GTIN9780080915494
ProduktartE-Book
EinbandartE-Book
FormatEPUB
Format HinweisDRM Adobe
Erscheinungsjahr2011
Erscheinungsdatum29.07.2011
Seiten552 Seiten
SpracheEnglisch
Dateigrösse9375 Kbytes
Artikel-Nr.2742435
Rubriken
Genre9200

Inhalt/Kritik

Inhaltsverzeichnis
1;Cover;1
2;Advances in The Study of Behavior;4
3;Copyright page;5
4;TOC$Contents;6
5;Contributors;10
6;Preface;12
7;CH$Chapter 1: Using Robots to Understand Animal Behavior;14
7.1;I. Introduction;14
7.2;II. Behavior and the Physical Interface;19
7.3;III. Completing the Mechanism Description;37
7.4;IV. Toward the Complete Cricket;49
7.5;V. Conclusions;55
7.6;References;58
8;CH$Chapter 2: Social Foraging and the Study of Exploitative Behavior;72
8.1;I. Why Study Foraging?;72
8.2;II. The Advent of Social Foraging Theory;74
8.3;III. The PS Game;78
8.4;IV. Rate-Maximizing PS Model;81
8.5;V. Stochastic, Risk-Sensitive Models;89
8.6;VI. State-Dependent Dynamic PS Game;95
8.7;VII. PS Information Games;97
8.8;VIII. Projecting Down to Individual Behavior;98
8.9;IX. Implications for Population Effects;103
8.10;X. Relevance of PS Games for Non-Food Resources;107
8.11;XI. Conclusions;110
8.12;Acknowledgments;111
8.13;References;112
9;CH$Chapter 3: Social Processes Influencing Learning in Animals: A Review of the Evidence;118
9.1;I. Introduction;118
9.2;II. Classification of Processes Involved in Social Learning;119
9.3;III. Empirical Evidence for Social Learning Processes;135
9.4;IV. Conclusions;169
9.5;Acknowledgments;170
9.6;References;170
10;CH$Chapter 4: Function and Mechanisms of Song Learning in Song Sparrows;180
10.1;I. Introduction;180
10.2;II. Studies of Social Factors in Song Learning;185
10.3;III. Developing Theories of Song Learning;187
10.4;IV. Song Function and Song Learning in Song Sparrows;189
10.5;V. Discussion;213
10.6;VI. Summary;227
10.7;Acknowledgments;229
10.8;References;229
11;CH$Chapter 5: Insights for Behavioral Ecology from Behavioral Syndromes;240
11.1;I. Introduction;240
11.2;II. A Brief History of the Idea;241
11.3;III. Clarifying the Definition of a Behavioral Syndrome;244
11.4;IV. Understanding Variation in Behavioral Syndromes;247
11.5;V. Beyond the Usual Behavioral Syndromes;261
11.6;VI. Future Prospects;278
11.7;VII. Summary;283
11.8;Acknowledgments;284
11.9;References;284
12;CH$Chapter 6: Information Warfare and Parent-Offspring Conflict;296
12.1;I. Introduction;296
12.2;II. Parent-Offspring Conflict as a Selective Force in Nature;297
12.3;III. Information from Offspring to Parents;315
12.4;IV. Interactions Among Siblings;325
12.5;V. Information from Parents to Offspring;332
12.6;VI. Conclusions;338
12.7;Acknowledgments;339
12.8;References;339
13;CH$Chapter 7: Hormones in Avian Eggs: Physiology, Ecology and Behavior;350
13.1;I. Introduction;350
13.2;II. Physiology;351
13.3;III. Effects of Yolk Androgens;358
13.4;IV. Variation Within Clutches;371
13.5;V. Differences Between Females;374
13.6;VI. Comparative Studies;381
13.7;VII. A Mechanism for Sex-Ratio Adjustment?;391
13.8;VIII. Egg Cocktails;393
13.9;IX. Conclusions and Future Directions;397
13.10;Acknowledgments;399
13.11;References;399
14;CH$Chapter 8: Neurobiology of Maternal Behavior in Sheep;412
14.1;I. Introduction;412
14.2;II. Expression of Maternal Behavior in Sheep;414
14.3;III. Neurobiology of Maternal Responsiveness;420
14.4;IV. Neurobiology of Maternal Selectivity;430
14.5;V. Conclusion;436
14.6;List of Abbreviations;441
14.7;References;441
15;CH$Chapter 9: Individual Odors and Social Communication: Individual Recognition, Kin Recognition, and Scent Over-Marking;452
15.1;I. Introduction;452
15.2;II. Individual Discrimination and Recognition;456
15.3;III. Discrimination and Recognition of Kin;473
15.4;IV. Individual Advertisement and Competition by Scent Marking;485
15.5;V. Scent Over-Marking;489
15.6;References;507
16;IDX$Index;520
17;Contents of Previous Volumes;542mehr
Leseprobe

Chapter 1
Using Robots to Understand Animal Behavior

Barbara Webb    Institute for Perception, Action and Behaviour, School of Informatics, University of Edinburgh, Edinburgh EH8 9AB, United Kingdom

Publisher Summary

This chapter discusses the use of robots to understand the animal behavior. A fundamental difference of robotic implementations from computer simulations of animal behavior is that there must be explicit means of transducing relevant signals and of materially affecting the surroundings. Animals cannot respond to signals for which they have no sensors. While the nature of physical interaction importantly shapes behavior, it is nevertheless true that most interesting behaviors of animals, such as crickets are also dependent on nontrivial neural processing that connects sensory and motor systems. Robotic implementations of models have several benefits while adding the advantage of bypassing the need to simulate (or mathematically represent) the interaction with the environment, instead using real interaction to represent itself. Clearly, the biorobotic approach is not ideal for every problem. The areas where it is likely to be most productive are those where a complete behavioral loop can be closed, that is, with some knowledge or plausible hypotheses about the nature of each of the intervening mechanisms.

I Introduction

What does it mean to have "understood" or "explained" animal behavior? Tinbergen's (1963) four questions are often cited: What is the function, how did it evolve, how did it develop (in the animal's lifetime), and what are the immediate internal and external causes? Of course, as Tinbergen himself realized, these questions are not independent. They can be paired in at least two ways, as concerning ultimate (functional and evolutionary) or proximate (developmental and immediate) causes, and as concerning historical (evolutionary and developmental) or mechanistic1 (functional and causal) accounts. More recently, the concept of mechanism has been more explicitly developed in the philosophy of science as a general account of the nature of explanation (Garber, 2002; Machamer et al., 2000). According to this view, a scientific explanation is the description of a mechanism, that is, of an actual physical system (rather than the more general sense of "mechanism" as some sequence of causal events) consisting of parts or components, their operations and their organization, which interact to produce the phenomena of interest.

Importantly, this description can specify the function of the parts at different levels (i.e., not requiring a full reduction to physical mechanics) relative to the interests of the scientist. For example, a population biologist might describe a system made up of replicators with different survival rates, without being concerned precisely how replication comes about; a geneticist, on the other hand, might want to understand how the particular replication mechanism of DNA operates. While both would consider the system they describe to be ultimately grounded in basic physical principles, it is not considered necessary to describe the system down to this level to have provided a scientific account of the phenomenon. Moreover, the explanation at the higher levels might be the same for systems that differ at lower levels-the same population dynamics can result from different replication mechanisms. Or, to take a more neuroethological example, the fact that vertebrate photoreceptors signal increases in light intensity by hyperpolarization, and invertebrate photoreceptors by depolarization, due to completely different transduction mechanisms, "appears to be trivial" (van Hateren and Snippe, 2006) from the point of view of understanding visual processing algorithms, since the response characteristics of the sensory cells to varying input (under daylight conditions at least) are sufficiently similar to be treated as equivalent.

The idea of explanation as mechanism description implies that we could evaluate our explanations by building the machines so described and seeing if they produce the relevant phenomena. In this review, I will describe just such a literal approach to understand some of the mechanisms responsible for animal behavior. The discussion above suggests that we can attempt to replicate the relevant mechanisms, at least with respect to a certain level of explanation, without necessarily having to use the same fundamental material basis, for example, we can use novel electronic transduction mechanisms as the front end for vision to replicate some biological mechanism of visual processing. Of course, it remains a matter of hypothesis that the lower level mechanism is not essential to understand the higher level function. There may well be conditions of testing that reveal the difference, for example, in range, sensitivity, efficiency, adaptation, and recovery properties of photoreception. Nevertheless, if we can, for example, fabricate an electronic system that responds to visual motion by adjusting turning torque to successfully stabilize its trajectory (Harrison and Koch, 2000), then this can be considered a potential mechanistic explanation for the optomotor reflex seen in flies (Warzecha and Egelhaaf, 1996)-although, as with any explanation, it remains possible that the precise explanation encapsulated in this device is incorrect. Perhaps more importantly, if we think we have the correct explanation but, on building the described mechanism, find that it does not produce the expected phenomena, it is evident that our explanation is flawed or incomplete.

Building machines that replicate animal capabilities as a means of understanding how they work is an old idea, with a history stretching back to the automata of the Greeks. However, until the last century, technological limitations severely restricted the scope of such devices. In 1912, Hammond and Miessner (cited in Cordeschi, 2002) designed and constructed an "electric dog" which exhibited phototropism by connecting two light sensors via relays to a drive motor and a steering wheel. The design was explicitly influenced by Loeb's descriptions of tropisms in animals, and Loeb (1918/1973) wrote that:



The best proof of the correctness of our view would consist in the fact that machines could be built showing the same type of volition or instinct as an animal going to the light … the actual construction of a heliotropic machine not only supports the mechanistic conceptions of the volitional and instinctive actions of animals but also the writer's theory of heliotropism, since this theory served as the basis in the construction of the machine (pp. 68-69).



A fascinating account of similar early machines is provided by Cordeschi (2002). An important motivation for these machines was to demonstrate that "biological" capabilities such as goal directedness, learning, variety of response, and intelligence could be replicated (and hence accounted for) mechanistically, and did not require some unique or vitalist force. Hull (1943), for example, explicitly outlined a "robot approach": "Regard … the behaving organism as a complex self-maintaining robot [that could be] constructed of materials as unlike ourselves as may be …" and argued for the development of "psychic" machines to illustrate that the principles of learning and goal directed behavior could be mechanized. Hull's ideas inspired the robot rat of Ross (1935), which was built to illustrate that:



it may be possible to test the various psychological hypotheses as to the nature of thought by constructing machines in accordance with the principles that these hypotheses involve and comparing the behavior of the machine with that of intelligent creatures (Ross, 1935, p. 387).



But note that this work was (perhaps of necessity) imitation at a high level-



this synthetic method is not intended to give any indication as to the nature of the mechanical structures of physical functions of the brain itself, but only to determine as closely as may be the type of function that may take place between 'stimulus' and 'response' (ibid).



More recently, interest in building machines that reproduce specific behaviors of animals has revived, this time often aimed at replication at a deeper level of similarity, including the "nature … of physical functions of the brain itself." This has been motivated by the recognition that we are still unable to build machines that have the capability and flexibility of animals to interact intelligently with real environments, despite huge advances in technology, particularly computational power. With potential robotic applications in mind, it is believed that imitating biological systems could be a good way to discover effective solutions (Ayers et al., 2002; Beer et al., 1997; Paulson, 2004). However, a problem quickly revealed when trying to imitate biology to build better robots is that our understanding of the underlying mechanisms of the biological systems is rarely good enough to enable a direct translation into hardware and software. The problem is more fundamental than just continuing limitations in the available technology for implemention. It is frequently found, when replication is attempted, that supposedly "complete" descriptions of a biological system turn...
mehr

Autor