DOCSLIB.ORG

The Challenges of Understanding Animal Minds

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
The Challenges of Understanding Animal Minds University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln Faculty Publications, Department of Psychology Psychology, Department of 11-2010 The challenges of understanding animal minds Jeffrey R. Stevens University of Nebraska-Lincoln, [email protected] Follow this and additional works at: https://digitalcommons.unl.edu/psychfacpub Part of the Psychiatry and Psychology Commons Stevens, Jeffrey R., "The challenges of understanding animal minds" (2010). Faculty Publications, Department of Psychology. 512. https://digitalcommons.unl.edu/psychfacpub/512 This Article is brought to you for free and open access by the Psychology, Department of at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Faculty Publications, Department of Psychology by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. SPECIALTY GRAND CHALLENGE ARTICLE published: 19 November 2010 doi: 10.3389/fpsyg.2010.00203 The challenges of understanding animal minds Jeffrey R. Stevens* Center for Adaptive Behavior and Cognition, Max Planck Institute for Human Development, Berlin, Germany *Correspondence: [email protected], [email protected] Comparative psychology is by nature an • Can animals use referential benchmark for parsimonious explanations. interdisciplinary science that lies at the communication? If a simpler, learning account can explain crossroads of psychology and biology but • How do animals track time and the data, then we should not invoke a more also draws from other fields in the natural, number? complex explanation. This stringent bench- social, and cognitive sciences. The study of • Do animals maintain a cognitive map mark has set a high bar for researchers in the the psychology of animals has been labeled of their environment? cognitive tradition to reject simple learning animal cognition, comparative cognition, • Do animals attend to the well being of explanations. Occasionally, the bar is raised animal learning, animal psychology, and others? higher and higher, sometimes to the extent animal intelligence. Here, comparative • How do animals categorize objects in that the learning mechanism becomes quite psychology is used interchangeably with their world? complex. How complex does a parsimoni- these terms, encompassing all fields that • Which emotional and motivational fac- ous explanation need to be before we accept explore the psychological mechanisms tors underly animal behavior? a representational one? How do we define underlying animal behavior, including the complexity of a mechanism? One solu- human behavior. Studying the psychology of other species tion is to sidestep the parsimony/complexity The primary goal of comparative psy- is not easy. With rare exceptions, we can- question completely. Rather than resorting chology is to understand the cognitive, not ask directly about their psychological to a scale of psychological complexity, we emotional, and motivational processes of states. Instead, we must make inferences can develop and test more precise models the animal mind. How do other animals about the psychology of animals based on of the phenomena of interest. perceive, learn about, and make decisions their behavior. Inferring internal mecha- Even after excluding simpler explana- in their worlds? From our pets to exotic nisms from external behavior results in a tions for phenomena, the question arises, animals portrayed in nature documenta- number of challenges for comparative psy- under what circumstances do animals ries, we are inherently curious about other chologists. Though we have met a number use the more complex form of cognition? animals. Comparative psychology both pro- of these challenges, several remain, slow- Possessing a particular psychological ability vides a window into their minds, as well as ing our progress in advancing comparative does not imply using it at every opportu- offers a unique perspective on the human psychology. To push forward, we must meet nity. In many cases, animals may use simpler mind. Which aspects of our psychology do these conceptual and practical challenges mechanisms, reserving the more complex animals share? Human uniqueness is con- head on. mechanisms for situations in which the sim- stantly challenged as we learn more about pler mechanisms do not work. The circum- the psychology of animal minds. Once dis- THE COMPLEXITY OF PARSIMONY stances under which animals use different tinctive human abilities – such as tool use, The difficulties of making inferences about mechanisms remains an open question in language, and mental time travel – appear, internal mechanisms has spawned two comparative psychology. at least to a degree, in other species. Though general approaches to studying cognitive other species exhibit elements of these abili- aspects of animal behavior. The animal INDIVIDUAL DIFFERENCES ties, the central question for comparing learning approach emphasizes the general A hallmark of data in comparative psychol- humans and animals remains, do humans learning principles, such as instrumental, ogy is variation across individuals. Despite and other animals share the same psycho- and Pavlovian conditioning, espoused by this, we have very little understanding of logical mechanisms? Hull, Spence, Tolman, and Skinner. The the sources of variation in psychological Comparative psychology explores many cognitive approach examines other forms mechanisms. Evolutionary models predict of the same topics as human psychology. of cognition such as perception, attention, individual differences in behavior (Wolf From learning and memory to commu- memory, categorization, navigation, timing, et al., 2007), but few models explore why nication and decision making, the field number, communication, decision mak- cognition and emotion should vary across investigates a number of key questions, for ing, and social cognition. Though learning individuals. A number of questions arise example: mechanisms often are considered simpler when developing a theoretical understand- explanations, some cognitive mechanisms ing of individual differences. Is there really • How do animals understand causal are more complex, requiring an organism variation in animal psychological abilities relationships in their environments? to generate a mental representation. Since or only in the expression of these abilities? • Can animals represent the perceptions, Morgan’s (1903) admonition to opt for the To what extent is this variation adaptive? To intentions, and beliefs of others? most parsimonious explanation supported what extent does the variation result from • Do animals plan for the future? by the data, the learning tradition has set the environmental influences? Are there general www.frontiersin.org November 2010 | Volume 1 | Article 203 | 1 Stevens Understanding animal minds reasons for variation in psychology or do share similar psychological mechanisms? that tap natural behaviors and situations we need to consider each ability separately? What kinds of evolutionary pressures can yield more valid studies of compara- Researchers of animal behavior investigate shape psychological mechanisms across tive psychology. individual differences under the headings of species? Does the social environment play animal personality, behavioral syndromes, a special role in shaping psychology? How STRONG INFERENCE and temperament. We must now extend this can we test whether species share actual Much of comparative psychology is based study of variation at the psychological level mechanisms rather than just general psy- on existence proofs: Do animals have theory and explore the sources of such variation. chological abilities? Comparisons across of mind, metacognition, episodic memory, species pose difficulties and require great empathy, or other-regarding preferences? To PUttING THE COMPARATIVE BACK care regarding the species tested and the further mature, however, the field needs to INTO COMPARATIVE PSYCHOLOGY experimental methods used. Nevertheless, develop more sophisticated models of how Despite the name and origins of the field, the fruits of this labor will yield valu- psychological mechanisms work, not just comparative psychology lacked many com- able insights into the nature of animal whether they are present and what influ- parative studies during much of the twenti- psychology. ences them. Weisman (2008) offers an eth century. Only a handful of species acted important challenge to the young research- as the workhorses of the field: the pigeon, ECOLOGICAL RELEVANCE ers in behavioral and cognitive science: the rat, and the rhesus macaque. Focusing Simon (1990) argued that a serious study embrace Platt’s (1964) notion of strong on a few species offers a more in-depth and of cognition must explore both the mind inference. Strong inference emphasizes the cumulative method for understanding the and the environment in which the mind development of multiple hypotheses and psychology of those species. This restric- interacts. He likened this to two blades the design of critical experiments allowing tion, however, limits our broader under- in a pair of scissors that cannot function for alternative outcomes that will exclude standing of psychological mechanisms in unless both blades are present. This pro- some of the hypotheses. Currently, many animals. Comparative psychologists have vides an important lesson for comparative behavioral scientists rely on testing a single lamented this lack of diversity for decades psychology as well. Studies
Load more

Recommended publications
  • Beyond Clever Hans: Learning from People Without Their Really Trying
    Beyond Clever Hans: Learning From People Without Their Really Trying by Vivek Veeriah Jeya Veeraiah A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Computing Science University of Alberta c Vivek Veeriah Jeya Veeraiah, 2017 Abstract Facial expressions and other body language are important for human commu- nication. They complement speech and make the process of communication simple and sustainable. However, the process of communication using existing approaches to human-machine interaction is not intuitive as that of human communication. Specifically, the existing approaches to human machine inter- action do not learn from whatever subtle non-verbal cues produced by a user. Many of the existing approaches map body language cues to instructions or re- wards and use them to train a machine. These mappings are not learned from ongoing interactions and are assumed to be defined by some external source. As a consequence, the communicative process through these approaches can cause significant cognitive load on the user. This is an important problem that needs to be addressed if we are to amplify our existing cognitive and physical capabilities through intelligent machines. Towards addressing this, we intro- duce our idea that allows machines to learn from whatever subtle cues people produce during the process of interaction. Particularly, the agent learns a value function that maps the user’s non-verbal cues to later rewards. By using this value function, the user can teach their agent to complete a task according to their preferences, and consequently maximize their satisfaction. We demon- strate this by training an agent using facial expressions.
    [Show full text]
  • Clever Hans, by Oskar Pfungst 1
    Clever Hans, by Oskar Pfungst 1 Clever Hans, by Oskar Pfungst The Project Gutenberg EBook of Clever Hans, by Oskar Pfungst This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org Title: Clever Hans (The horse of Mr. Von Osten): A contribution to experimental animal and human psychology Author: Oskar Pfungst Translator: Carl L. Rahn Release Date: October 11, 2010 [EBook #33936] Language: English Character set encoding: ISO-8859-1 *** START OF THIS PROJECT GUTENBERG EBOOK CLEVER HANS *** Clever Hans, by Oskar Pfungst 2 Produced by Audrey Longhurst, Linda Hamilton and the Online Distributed Proofreading Team at http://www.pgdp.net [Illustration: CLEVER HANS] CLEVER HANS (THE HORSE OF MR. VON OSTEN) A CONTRIBUTION TO EXPERIMENTAL ANIMAL AND HUMAN PSYCHOLOGY BY OSKAR PFUNGST WITH AN INTRODUCTION BY PROF. C. STUMPF, AND ONE ILLUSTRATION AND FIFTEEN FIGURES TRANSLATED FROM THE GERMAN BY CARL L. RAHN Fellow in Psychology in the University of Chicago WITH A PREFATORY NOTE BY JAMES R. ANGELL Professor of Psychology in the University of Chicago [Illustration] NEW YORK HENRY HOLT AND COMPANY 1911 COPYRIGHT, 1911, BY HENRY HOLT AND COMPANY PREFATORY NOTE [BY JAMES R. ANGELL] The University of Chicago It gives me great pleasure to accept the invitation of the publishers to write a word of introduction for Mr. Rahn's excellent translation of "Der Kluge Hans", a book which in the original has been but little known to American readers.
    [Show full text]
  • Doing Away with Morgan 'S Canon
    Doing Away with Morgan’ s Canon SIMON FITZPATRICK Abstract : Morgan ’ s Canon is a very widely endorsed methodological principle in animal psychology, believed to be vital for a rigorous, scientifi c approach to the study of animal cognition. In contrast I argue that Morgan ’ s Canon is unjustifi ed, pernicious and unnecessary. I identify two main versions of the Canon and show that they both suffer from very serious problems. I then suggest an alternative methodological principle that captures all of the genuine methodological benefi ts that Morgan ’ s Canon can bring but suffers from none of its problems. 1. Introduction Morgan ’ s Canon is a very widely endorsed methodological principle in animal psychology. Proposed at the end of the 19th century by the British comparative psychologist Conwy Lloyd Morgan, it states that: In no case may we interpret an action as the outcome of the exercise of a higher psychical faculty, if it can be interpreted as the outcome of the exercise of one which stands lower in the psychological scale (Morgan, 1894 , p. 53). This principle has played a central role in subsequent scientifi c and philosophical work into the nature of non-human animal minds. So central in fact that it has been described as ‘ possibly the most important single sentence in the history of the study of animal behaviour ’ ( Galef, 1996, p. 9; cited in Radick, 2000 , p. 3). Morgan ’ s Canon has attained this eminent status because it has been seen as a necessary component of a suitably rigorous approach to explaining animal behaviour. According to a recent textbook, ‘ Morgan ’ s Canon … has enabled us to approach the analysis of behaviour sensibly and to avoid the superfi cial anthropomorphism which led to many absurdities in the past ’ ( Manning and Dawkins, 1998 , p.
    [Show full text]
  • BROWN, BARNUM (1873–1963) Dinosaur Fossil Collector Amed After the Great Showman P
    BRoWn, BaRnUM (1873–1963) Dinosaur Fossil collector amedafterthegreatshowmanP.T.Barnum,indefatigabledinosaurdiggerBarnum Brownassembledhisownversionof“The Greatest Showon Earth”:aparadeofgiant n dinosaurfossilswrenchedfromthecliffsandarroyosoftheAmericanWest.Brown’s -lastingcontribution—hundredsoftonsofdinosaurfossils—formedthenucleusoftheAmeri .canMuseumofnaturalHistory’sworld-famouscollection during the 1960s, Brown, nearly 90, could still be seen leading visitors around the crammeddinosaurhalls,announcing,“Here’sanotheroneofmychildren,”ashepointed outthebonesofasauriangiant.Butwhenhebeganhiscareerin1897,themuseumhad .notasingledinosaur AsachildinCarbondale, Kansas,Browncollectedfossilsfromfreshlyplowedfields.He -attendedtheUniversityof Kansas,thenmovedto New YorkCity,wherehestudiedpaleontol ogyatColumbiaUniversityandbeganworkingatthemuseumwhilehewasstillagraduate student. Forhisfirstfieldassignment,themuseum’sdirector,HenryFairfield Osborn,sentBrown BONE hUNTER barnum brown, when almost ninety, toComoBluff,Wyoming,toprospectitsrichJurassicdeposits.Brownandhiscolleagues supervised construction of discoverednewbedscontainingenormousquantitiesoffossils,includingtheApatosaurus .sinclair oil’s lifesize dino- (thencalled Brontosaurus)thatstilldominatesoneofthemuseum’shugedinosaurhalls saur models, shown below However, yale paleontologist othniel C. Marsh was furious about his former sites being being barged down the hud- .workedandbeganabitterfeudwithosbornthatlastedtotheendofhislife son river to the 1964 new york World’s fair.
    [Show full text]
  • THE TRUE FAIRYTALE of CLEVER HANS of Course We Don’T Want to Tell Any Fairy Tales Here …
    THE TRUE FAIRYTALE OF CLEVE R HANS THE TRUE FAIRYTALE OF CLEVER HANS Of course we don’t want to tell any fairy tales here … … Even though the story of how a three-man business founded by Hans Grohe in 1901 in the little Black Forest town of Schiltach became a world-wide enterprise with over 3,200 employees might seem to some to be more like a fairy tale… What follows is the story of the firm of Hansgrohe, spanning more than a century of growth and development. And the story of Hans Grohe himself, the handyman, the craftsman, the visionary pioneer in the field of sanitation, without whom the company of Hansgrohe could never have come into being. And the story of his descendants who right up to the present day lead and guide the firm of Hansgrohe SE. It’s not a dry-as-dust story, and doesn’t claim to be complete in itself or to be a scientific account; it is told here sometimes tongue in cheek, with great joy in the telling, giving deep insights into the world of Hansgrohe and into the family firm, founded by Clever Hans. And if the story of “Lucky Hans” (“Hans im Glück”) as once told by the Grimm Brothers in their collection of fairy tales is unfamiliar to you, just read on. Truth is stranger than fiction, as they say … 2 Hans had served his master for seven years and now he had an urgent desire to go to Schiltach. So he asked his master to give him his wages.
    [Show full text]
  • “Clever Hans Effect” You Are Going to Listen to a Scientific Radio Show
    Kammerhofer/Zangerl Going for Finals B2, Audio-CD – Transcripts GOING FOR FINALS B2,-AUDIO-CD – TRANSCRIPTS 1 The “Clever Hans Effect” You are going to listen to a scientific radio show about an experiment with an animal. First you will have 45 seconds to study the task below, then you will hear the recording twice. While listening, complete the sentences (1–9) using a maximum of 4 words. Write your answers in the spaces provided. The first one (0) has been done for you. After the second listening, you will have 45 seconds to check your answers. Hello and good evening, dear listeners. Today I’m talking about the “Clever-Hans-Effect”. This term refers originally to a trained horse, named “Clued Hans”, referred to in the literature as “Clever Hans” … who was taught by his master, William Von Osten, to do math and to report on this math … So William would ask the horse, ask Hans, to add 3 + 2 for example and Hans would tap his hoof five times. This may have seemed as if the horse understood not only the spoken word, but understood math, to mean that it was able to do mathematical calculations on its own, which was unheard of in the ani- mal kingdom. Several scientists reviewed this demonstration and declared it genuine. However, when proper con- trols were put into place it turned out that, if the people who knew the correct answer to the question Hans was being asked were not within sight of Hans, Hans could not come up with the correct answer.
    [Show full text]
  • ST~~YI~~ ?F ~F~F Artifacts and Ethics in Behavioral Research
    ?f ~P~f ST~~YI~~ ?f ~f~f Artifacts and Ethics in Behavioral Research Ralph L. Rosnow Temple University Robert Rosenthal Harvard University II w. H. FREEMAN AND COMPANY NEW YORK ACQUISITIONS EDITOR: Susan Finnemore Brennan We are as sailors who are forced to rebuild their ship on the open sea, PROJECT EDITOR: Christine Hastings without ever being able to start fresh from the bottom up. Wherever COVER AND TEXT DESIGNER: Victoria Tomaselli a beam is taken away, immediately a new one must take its place, ILLUSTRATION COORDINATOR: Bill Page PRODUCTION COORDINATOR: Maura Studley and while this is done, the rest of the ship is used as support. In this COMPOSITION: W. H. Freeman Electronic Publishing Center/Susan Cory way, the ship may be completely rebuilt like new with the help of the MANUFACTURING: RR Donnelley & Sons Company old beams and driftwood-but only through gradual rebuilding. 1 Library of Congress Cataloging-in-Publication Data Rosnow, Ralph L. People studying people: artifacts and ethics in behavioral research/ Ralph L. Rosnow, Robert Rosenthal. p. cm. Includes bibliographical references and indexes. ISBN 0- 7167-3070-7. -ISBN 0-7167- 3071-5 (pbk.) 1. Psychology-Research. 2. Social sciences-Research. I. Rosenthal, Robert, 1933- II. Title BF76.5.R645 1997 150'.7'2-dc21 96-48978 CIP © 1997 by W. H. Freeman and Company No part of this book may be reproduced by any mechanical, photographic, or electronic process, or in the form of a phonographic recording, nor may it be stored in a retrieval system, transmitted, or otherwise copied for public or private use, without written permission from the publisher.
    [Show full text]
  • A History of Modern Psychology, 10Th
    This page intentionally left blank This page intentionally left blank A History of Modern Psychology TENTH EDITION DUANE P. SCHULTZ University of South Florida SYDNEY ELLEN SCHULTZ Australia • Brazil • Japan • Korea • Mexico • Singapore • Spain • United Kingdom • United States This is an electronic version of the print textbook. Due to electronic rights restrictions, some third party content may be suppressed. Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. The publisher reserves the right to remove content from this title at any time if subsequent rights restrictions require it. For valuable information on pricing, previous editions, changes to current editions, and alternate formats, please visit www.cengage.com/highered to search by ISBN#, author, title, or keyword for materials in your areas of interest. A History of Modern Psychology, © 2011 Wadsworth, Cengage Learning Tenth Edition ALL RIGHTS RESERVED. No part of this work covered by the copyright Duane P. Schultz and Sydney Ellen herein may be reproduced, transmitted, stored or used in any form or by Schultz any means graphic, electronic, or mechanical, including but not limited to photocopying, recording, scanning, digitizing, taping, Web distribu- Senior Publisher: Linda Schreiber-Ganster tion, information networks, or information storage and retrieval sys- Executive Editor: Jon-David Hague tems, except as permitted under Section 107 or 108 of the 1976 United Editorial Assistant: Sheli DeNola States Copyright
    [Show full text]
  • People Studying People: Artifacts and Ethics in Behavioral Research
    People studying people: artifacts and ethics in behavioral research The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Rosnow, Ralph L. and Robert Rosenthal. 1997. People studying people: artifacts and ethics in behavioral research. New York: W.H. Freeman Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:32300329 Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA ?f ~P~f ST~~YI~~ ?f ~f~f Artifacts and Ethics in Behavioral Research Ralph L. Rosnow Temple University Robert Rosenthal Harvard University II w. H. FREEMAN AND COMPANY NEW YORK ACQUISITIONS EDITOR: Susan Finnemore Brennan We are as sailors who are forced to rebuild their ship on the open sea, PROJECT EDITOR: Christine Hastings without ever being able to start fresh from the bottom up. Wherever COVER AND TEXT DESIGNER: Victoria Tomaselli a beam is taken away, immediately a new one must take its place, ILLUSTRATION COORDINATOR: Bill Page PRODUCTION COORDINATOR: Maura Studley and while this is done, the rest of the ship is used as support. In this COMPOSITION: W. H. Freeman Electronic Publishing Center/Susan Cory way, the ship may be completely rebuilt like new with the help of the MANUFACTURING: RR Donnelley & Sons Company old beams and driftwood-but only through gradual rebuilding. 1 Library of Congress Cataloging-in-Publication Data Rosnow, Ralph L.
    [Show full text]
  • Unmasking Clever Hans Predictors and Assessing What Machines Really Learn
    Unmasking Clever Hans Predictors and Assessing What Machines Really Learn Sebastian Lapuschkin1, Stephan W¨aldchen2, Alexander Binder3, Gr´egoireMontavon2, Wojciech Samek1y and Klaus-Robert M¨uller2;4;5y∗ 1Department of Video Coding & Analytics, Fraunhofer Heinrich Hertz Institute Einsteinufer 37, 10587 Berlin, Germany 2Department of Electrical Engineering and Computer Science, Technische Universit¨atBerlin Marchstr. 23, 10587 Berlin, Germany 3ISTD Pillar, Singapore University of Technology and Design 8 Somapah Rd, Singapore 487372, Singapore 4Department of Brain and Cognitive Engineering, Korea University Anam-dong, Seongbuk-ku, Seoul 136-713, Republic of Korea 5Max Planck Institut f¨urInformatik Campus E1 4, Stuhlsatzenhausweg, 66123 Saarbr¨ucken, Germany yCorresponding authors. E-mail: [email protected], [email protected]. Abstract Current learning machines have successfully solved hard application problems, reaching high accuracy and displaying seemingly \intelligent" behavior. Here we apply recent techniques for explaining decisions of state-of-the-art learning machines and analyze various tasks from computer vision and arcade games. This showcases a spectrum of problem-solving behaviors ranging from naive and short-sighted, to well- informed and strategic. We observe that standard performance evaluation metrics can be oblivious to distinguishing these diverse problem solving behaviors. Furthermore, we propose our semi-automated Spectral Relevance Analysis that provides a practically effective way of characterizing and validating the behavior of nonlinear learning machines. This helps to assess whether a learned model indeed delivers reliably for the problem that it was conceived for. Furthermore, our work intends to add a voice of caution to the ongoing excitement about machine intelligence and pledges to evaluate and judge some of these recent successes in a more nuanced manner.
    [Show full text]
  • Beyond Clever Hans: Learning from People Without Their Really Trying
    Beyond Clever Hans: Learning From People Without Their Really Trying by Vivek Veeriah A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Computing Science University of Alberta c Vivek Veeriah, 2017 Abstract Facial expressions and other body language are important for human commu- nication. They complement speech and make the process of communication simple and sustainable. However, the process of communication using existing approaches to human-machine interaction is not intuitive as that of human communication. Specifically, the existing approaches to human machine inter- action do not learn from whatever subtle non-verbal cues produced by a user. Many of the existing approaches map body language cues to instructions or re- wards and use them to train a machine. These mappings are not learned from ongoing interactions and are assumed to be defined by some external source. As a consequence, the communicative process through these approaches can cause significant cognitive load on the user. This is an important problem that needs to be addressed if we are to amplify our existing cognitive and physical capabilities through intelligent machines. Towards addressing this, we intro- duce our idea that allows machines to learn from whatever subtle cues people produce during the process of interaction. Particularly, the agent learns a value function that maps the user's non-verbal cues to later rewards. By using this value function, the user can teach their agent to complete a task according to their preferences, and consequently maximize their satisfaction. We demon- strate this by training an agent using facial expressions.
    [Show full text]
  • From Our Readers
    From Our Readers The letters column is a forum for views multiplied beyond necessity." But when on matters raised in previous issues. Let­ we translate this statement into "the ters are more likely to be published if simplest explanation for an observation they are brief and typed double-spaced. is most likely to be the correct one," then They may be edited for space and clarity. I pause. In the first place, if the simplest explanation is most likely to be correct, then it is not always correct. Then how Occam's Razor do we know when it is correct? A law with exceptions is of no use unless we I agree with Elie Shneour's results in his have a rule defining the exceptions. article on Occam's Razor (SI, Summer As a matter of fact, there is no reason 1986), but disagree with the method he to believe that the simplest explanation uses to arrive at them. If we take the for anything is the correct one. There is meaning of Occam's Razor to be "the no scientific or logical basis for this belief. simplest explanation for an observation It is simply a convenient prejudice. is most likely to be the correct one," what Counterexamples abound. Consider John is the observation that he is trying to Dalton's belief that all elemental gases explain? Is he trying to explain the bib­ are composed of molecules, each with a lical account of the Flood? Where is there single atom. This was the simplest theory an observation of a flood or an ark? The that explained the observations of the only thing we observe is a story about a day.
    [Show full text]