studiesroxxxy

ARENAOFCHANGINGTheGhostintheMachineBeingHumanintheAgeofAIandMachineLearningHenrikSkaugStra1Received:30July2018/Revised:7September2018/Accepted:10September2018/Publishedonline:19September2018AbstractHumanbeingshaveusedtechnologytoimprovetheirefficiencythroughouthistory.
Wecontinuetodosotoday,butwearenolongeronlyusingtechnologytoperformphysicaltasks.
Today,wemakecomputersthataresmartenoughtochallenge,andevensurpass,usinmanyareas.
Artificialintelligence—embodiedornot—nowdriveourcars,tradestocks,socialisewithourchildren,keeptheelderlycompanyandthelonelywarm.
Atthesametime,weusetechnologytogathervastamountsofdataonourselves.
This,inturn,weusetotrainintelligentcomputersthateaseandcustomiseevermoreofourlives.
Thechangethatoccursinourrelationstootherpeople,andcomputers,changebothhowweactandhowweare.
WhatsortofchallengesdoesthisdevelopmentposeforhumanbeingsIarguethatweareseeinganemergingchallengetotheconceptofwhatitmeanstobehuman,as(a)westruggletodefinewhatmakesusspecialandtrytocometotermswithbeingsurpassedinvariouswaysbycomputers,and(b)thewayweuseandinteractwithtechnologychangesusinwayswedonotyetfullyunderstand.
KeywordsArtificialintelligence.
Machinelearning.
Socialrelations.
Cognition.
HumanidentityIntroductionOurpresentageischaracterisedbyrapidprogressinthefieldofdigitaltechnology.
Techno-logicalchangehasalwaysbeenanimportantpartofhumandevelopment,buttoday'schangemaybedifferent.
Throughouttheages,wehaveexchangedhumanlabourwithothersourcesofenergy:animalpower,steampowerandthenelectricalpower.
Weareinthemidstofanothergreatalleviationofhumanlabour,butnow,wearereplacingtheneedforhumanmentalpowers,insteadofjustthephysical.
PartsofitinvolvescombiningartificialintelligencewithHumanArenas(2019)2:60–78https://doi.
org/10.
1007/s42087-018-0039-1*HenrikSkaugStrahenrik.
satra@hiof.
no1FacultyofBusiness,Languages,andSocialSciences,stfoldUniversityCollege,P.
O.
Box700,1757Halden,Norway#TheAuthor(s)2018physicalmachines—furtheringtheautomationofindustrialtaskspreviouslyperformedbyhumanbeings.
Otherparts,however,concernreplacingthehumanintellect.
Machineshelpusdrive,tradestocksinthestockmarkets,decidewhogetsloans,helpuschoosemovies,guideourmissiles,andsmartsystemskeeptrackofjustabouteverymovewemakeandmakesusmoreeffective.
Thegameofchesswasconqueredbythemachinelongago,andin2017,oneofthefinalfrontiersofhumansuperiorityingameswasbroken.
Google'sartificialintelligenceAlphaGoprovedabletobeatthebesthumansintheancientgameofGo.
Furthermore,machinesincreasinglycomplement,andevenreplace,humansassocialcompanions—evenlovers.
Ourchildrenbondwithsociablerobots,ourelderlygettherapyfromroboticsealsandourlonelyonescanfindcompanionshipinrobotsdesignedforintimacy.
Thelastpartofthisequationistheconnectionbetweenhowweemploytechnologyandthewaywegatherandusevastamountsofdata.
Allkindsofdataarecollected,andoureverymove—yes,eventhebeatsofourhearts—canbemonitoredandanalysed.
Inmodernsociety,itmayatthesametimebebothimpossibletohideanddifficulttoreallyconnecttootherpeople.
Bigdataletsmachineslearnwhatwelike,andevenwhatwearelikelytointhefuture.
Manynowpointoutthatthewayweconsumebothdataand,inasense,otherpeople,arechangingus.
Notjustchangethewayweviewourselvesandstruggletofindourplace,buthowwefunction.
Ourplasticbrainsadapttohowitisused,andthewayweuseithaschanged,bothinrelationtoinformationandotherbeings.
Inthisarticle,Iarguethatweareseeinganemergingchallengetotheconceptofwhatitmeanstobehuman,duetothedevelopmentsjustdescribed.
Firstly,whatitmeanstobehuman,whatdefinesusandmakesusspecial,maychangewhenwearesurpassedinvariouswaysbycomputers.
Secondly,whatitmeanstobehumanisbynecessityinfluencedbythefactthatwearechangedbythetechnologiesweemploy.
Understandinghowtechnologieschangeuswillbeimportantbothbecausewewillneedtounderstandtheconsequencesofsuchchanges,andwemaywanttopreventcertainchanges,oratleastproperlydebatethedesirabilityofthem.
ArtificialIntelligenceinToday'sSocietyThePrometheanmythhasacquiredanuglytwist:thegiantreachingouttostealthelightningfromthegodsisinsane(Koestler1967,p.
331).
Firstthingsfirst:whatisartificialintelligence(AI)InareportdetailingthepotentialmalicioususeofAI,theauthorsdefineitassystems"capableofperformingtaskscommonlythoughttorequireintelligence"(Brundageetal.
2018,p.
9).
Machinelearningreferstosystemsthatcan"improvetheirperformanceonagiventaskovertimethroughexperience"(Brundageetal.
2018,p.
9).
Arelatedconceptisbigdata,whichreferstotheuseofthevastamountsofdatathatwenowhaveavailableandhowweanalyseit(McAfeeetal.
2012).
AIhaslongexisted,andthequestforintelligentmachinesstartedwithAlanTuring'sComputingmachineryandintelligence,releasedin1950(Turing2009).
Inthisarticle,hedescribesatesttodeterminewhetherornotamachineisintelligent.
ThetestisknownastheTuringtestandispassedbyacomputerthatcanmakeahumanbeingthinkthatitisnotacomputerwheninteractingwithit(throughachat,forexample).
TuringhimselfworkedonTheGhostintheMachine61intelligentmachines.
DuringtheSecondWorldWar,heattemptedtocracktheGerman"Enigma"-codeandsucceeded,byusingacomputer(Copeland2014).
SinceTuring,machineshaveaccomplishedvariousdauntingtasks.
Forthegeneralpublic,amajormilestonewaswhenDeepBluebeatGarryKasparovinchessin1997(Campbelletal.
2002).
DeepBluewasacomputerdevelopedbyIBM.
Only5yearsearlier,Kasparovhad"scornedthepatheticstateofcomputerchess"(Kurzweil2015,p.
148).
Chessiswidelyconsideredaworthyintellectualpursuit,andwhenacomputernowprovedtobebetterthanthebestofhumans,itsintelligencecouldhardlybequestioned.
Itssuperioritytohumanbeings,however,couldstillbechallenged.
Afterall,chessischess,butwestillhadthegameofGo!
Goisanancientboardgame,widespreadinAsiainparticular.
WhenDeepBluebeatKasparov,evenanovicehumanGoplayercouldbeatthebestGocomputer.
Thisgamewasfarmorecomplex,andthebruteforceapproachofDeepBlueandearlyAIwasnomatchforthemoreadvancedcognitiveapproachofhumanbeings.
Then,in2016,thecomfortsomemayhavetakeninthefactthathumanintuitionbeatcomputerintelligenceinGowasshattered.
AlphaGobeatLeeSedol,widelyconsideredoneofthebestGoplayersofrecenttimes(Chouard2016).
AlphaGoisGo-playingadaptionofGoogle'sDeepMindAI,theself-titled"worldleaderinartificialintelligenceresearchanditsapplicationforpositiveimpact"(Google2018).
WhiletheseprominentpeaksofAIachievementsmaybeimportant,itisevenmoreimportanttoexaminethewidespreadandlow-keyproliferationoftheuseofAIandmachinelearningintoday'ssociety.
Wefaceitsapplicationseveryday,through"speechrecognition,machinetranslation,spamfilters,andsearchengines"(Brundageetal.
2018).
Computersare"flyingandlandingairplanes,controllingthetacticaldecisionsofautomatedweapons,makingcreditandfinancialdecisions,andbeinggivenresponsibilityformanyothertasksthatusedtorequirehumanintelligence"(Kurzweil2015,p.
148).
Carsdrivenbycomputers,machineshelpingdoctors,phonesthatrecogniseuswhiledronescanbeprogrammedtoplanttrees,giveaid,ordeliverbombs.
Inmanyrespects,thefutureisalreadyhere.
Ourchildrenareplayingwithcomputersfrominfancy,bothonscreen,andwithrobotslikeFurby,ZhuZhuPets,Tamagotchiandthelike;evenourelderlygetcarethroughroboticseals.
Whenonline,weoftengethelpfromchatbotsthatappearhuman,andpeopleeventalktotheirwatchesandphones.
Siri,theycallher,orsomethingelseiftheyuseanothersystem.
Evenloveisfoundinthecomputer.
Atleastthemoresuperficialkind,thoughsexrobotslikeRoxxxy,inadditiontoprovidingphysicallove,alsocarriesconversations(ScheutzandArnold2016).
Therearenowbrothelsstaffedbyrobots,lookingtogoglobal(Lockett2017).
Thepossibleapplicationsofintelligentmachinesarepracticallylimitless.
Whatwedecidetoletthecomputersdo,however,isstilluptohumanbeings.
Thusfar,itmaybearguedthatwearenotshowingmuchrestraint.
Inthefuture,"advancedAIholdsoutthepromiseofreducingtheneedforunwantedlabor,greatlyexpeditingscientificresearch,andimprovingthequalityofgovernance"(Brundageetal.
2018).
Reducingtheneedforlabour,whilepossiblyincreasingtheopportunitiesforloveandcompanionshipWhatisnottoloveConsideringtherapidproliferationoftheuseofAI,moreandmorenowhighlightthepotentialmalicioususeofAI.
Someofthesedangersarethreatstodigitalsecurity,physicalsecurityandpoliticalsecurity(Brundageetal.
2018).
Whenourlivesarethoroughlydigitised,computerhackerscancausegreatdamage.
Physicalsecuritymaybethreatenedbytheuseofautonomousweapons,oreventhedisruptionofvitalsocialservices.
AscaryexampleofthesekindsofthreatswasStuxnet,apieceofmalwareusedtoattackIraniannuclearfacilitiesin2007(Langner2011).
Thedangerstothepoliticalspherecanbeseenin"privacy-eliminating62Strasurveillance,profiling,andrepression,orthroughautomatedandtargeteddisinformationcampaigns"(Brundageetal.
2018).
Koestler(1967)wroteoftheGhostinthemachine—anallusiontoDescartesandthespiritwithinaphysicalbody.
Here,hediscussesthepossibilitythathumanitymaybeanevolution-arydeadend.
Inlightofthedevelopmentsherediscussed,itisinterestingtoseethisinlightofthepossibilitythatthetechnologicalprogresswepraisemaybepartofsomehuman"built-inerrorordeficiencywhichpredisposeshimtowardsself-destruction"(Koestler1967,p.
xi).
Bostrm(2014)haswrittenofonesuchpossibleerror:thepossibilitythatwebuildcomputersmorepowerfulthanourselves,whichweinturnbecomedependenton.
This,hesays,wouldbemuchlikehowgorillas—physicallysuperiortous—arenowmoredependentonusthanonthemselves(Bostrm2014,p.
9).
MüllerandBostrom(2014)interviewsagroupofAIexpertsontheirtakeonfuturedevelopmentsinthefield.
Whenaskedwhatyeartheythoughtitmightbea50%chancethata"high-levelmachineintelligence"(HLMI)exists,themedianresultsgivenarearound2040,withthemeansignificantlyhigheratbetween2072and2097forthevariousgroupssurveyed.
AHLMIisanAIthat"cancarryoutmosthumanprofessionsatleastaswellasatypicalhuman"(MüllerandBostrom2014).
Superintelligenceandthesingularityareoftendiscussedtogether.
Thesingularityis"afutureperiodduringwhichthepaceoftechnologicalchangewillbesorapid,itsimpactsodeep,thathumanlifewillbeirreversiblytransformed"(Kurzweil2015,p.
146).
Ultraintelligenceisanothertermsometimesusedtodescribe"amachinethatcanfarsurpassalltheintellectualactivitiesofanymanhoweverclever"(Good1965,p.
33).
Good(1965)goesontodescribehowthismachinewouldimproveuponitself,leadingtoasituationlikethesingularity,inwhicheversmartercomputersbuildeversmartercomputers,leavingmankindbehindinthedust.
Sciencefiction,yousayLetusreturntoAlphaGo.
WhenthisAIbeatLeeSedolin2016,ithadlearntwhatitknewbylearningfromhumanexpertsandreinforcementlearningfromself-play(Silveretal.
2017).
In2017,theteambehindAlphaGopublishedanarticlecalledMasteringthegameofgowithouthumanknowledge,wheretheyrevealAlphaGoZero(Silveretal.
2017).
ThisnewversionofAlphaGoisgivenonlytherulesofthegame,andlearnseverythingfromscratch,atruetabularasagoplayer.
TheresultAlphaGoZerospentjust3daysofteachingitselfthegamebeforeitbeattheversionthattheyearbeforehadmadenewsworldwidebybeatingthehumanchampion.
Itsrecordagainstthepreviousversion100–0(Silveretal.
2017).
Unhamperedbyhuman"expert"judgement,theAItakesoff.
Thesekindsofdevelopmentsmakeithardtocompletelydismissthenotionofexponentialgrowthandultraintelligentmachinesmakingortrainingeversmartermachines.
Chalmers(2015)arguesthatweneednotnecessarilyworry.
Wecouldintegrateourselveswithcomputersandharnessthebenefitstheygive:uploadingourselvesintothedigitalworldandbeingonewiththecomputers(Chalmers2015,p.
200).
DamasiocommentsonthispossibilityinhislatestbookTheStrangeOrderofThings,andsaysthatitis"animplausiblescenario"that"revealsalimitednotionofwhatlifereallyisandalsobetraysalackofunderstandingoftheconditionsunderwhichrealhumansconstructmentalexperiences"(Damasio2018,loc2936).
Apartfromintegration,Chalmersmentionstheotherpossiblestrategieswecouldtaketowardsthecomingofasingularity,namelyextinction,isolationorinferiority(Chalmers2015,p.
199).
Thefirstinvolvesusgoingextinct,thesecondthatwelimitourinteractionwithcomputersasmuchaspossible,whilethethirdmeansthatweinteractandthathumansareclearlytheinferiorpartyintheinteraction(Chalmers2015,p.
199).
Chalmersseesintegrationasthebest,orperhapsonly,choice(Chalmers2015,p.
200).
TheGhostintheMachine63WhatDoesItMeantoBeHumanFarmanimalswereasortofinferiorclass,reassuringthehumblestruralworkerthathewasnotattheabsolutebottomofthesocialscale,aconsolationwhichhisindustrialsuccessorwastolack(Thomas1983:50).
Asthe"humblestruralworker"indaysofyoreneededconsolationandreassurance,somaybethecaseoftheworkersoftoday.
Kurzweil(2015)referstopeoplethatarealarmedbytheriseofintelligentmachinesand"haveexpressedthewishtoremain'unenhanced'whileatthesametimekeepingtheirplaceatthetopoftheintellectualfoodchain"(Kurzweil2015,p.
167).
AmemberoftheEnglishHouseofLords'ArtificialIntelligenceCommittee,StevenCroft,pointedtothefactthatmore"mundanetasks"willbelosttoautomation,andthatthismightleadtoanidentitycrisis(Campbell2017).
Thatpersonalidentityisconnectedtojobs,andthefeelingofvalueconnectedtonotbeing"attheabsolutebottomofthesocialscale",isnottheendofthisstory.
Wemayimagineevenbiggerchallengestoouridentitycoming,aswearechallengedbymachinesthatrivalusinsomeoftheareasthatwemostvalue.
Anobviousexample:shouldtheprocessChalmers(2015)describedoccur,whereweintegrateourselveswithmachinesandbecomeuploadedselvesinadigitaluniverse,itisnotveryfar-fetchedtoassumethatwewillexperienceuncertaintyaboutthemeaningofbeinghuman.
Thepossibilityofverysmartmachines,hesays,"forcesustothinkhardaboutvaluesandmoralityandaboutconsciousnessandpersonalidentity",and"bringsupsomeofthehardesttraditionalquestionsinphilosophyandraisessomenewphilosophicalquestionsaswell"(Chalmers2015).
Human-computerinteraction(HCI)isafieldwheretherelationshipbetweenhumanandcomputesarestudied.
Norman(1991)isaproponentofthecognitiveapproachinthefield,andheexplainsthattherearetwodifferentviewofartefacts—devicesthat"maintain,displayoroperateuponinformation"inorderto,forexample,assistusincognitivetasks.
Thesystemviewinvolvesusseeingtheactor,thetaskandtheartificeasawhole,whereastheactor'scapacityisincreased.
Fromthepersonalview,however,whichistheviewoftheactor,hiscapacityisnotenhancedbytheartefact,butthetaskitselfischanged(Norman1991).
Ifwetakethisapproachtotheintroductionofartefactsingeneral,weseethatthedifferentperspectivesgiveusdifferentviewsonissuesofautomationandtheintroductionofAIinindustry.
Fromoneperspective,humansareempowered,butfromtheotherthetasksarechanged,andtheactormayevenfeeldiminished.
WhatisinterestingisthatthecapacityofAImaybesaidtohavegonebeyondtheroleofthecognitiveartefactsherediscussed.
AIsystemsareoftenlessofahelp,ortool,forhumanactors,butmoreofanautonomousreplacement.
ActivitytheoryisadifferentapproachtoHCI,inwhichbothofNorman'sviewsareconsideredpersonal(Kaptelinin1996).
Inactivitytheory,toolsareseentoempower,andevenchange,theactor.
Kaptelinin(1996)referstostudiesthatshowthatweoftengothroughthreephaseswhentoolsareusedtoassistusintasks.
First,wecannoteffectivelyusetotool,soperformanceofthetaskisthesamewithorwithoutthetool.
Inthesecondphaseweperformbetterwiththetoolthanwithout.
Thethirdphaseisthemostinteresting,andthatiswhenwecanperformtheoriginaltaskbetterthanbefore,evenwithoutthetool(Kaptelinin1996).
Usingthetoolsactuallychangesusandteachesusnewthings.
Itempowersus.
Iwillreturntothispossibilitylateron.
Penrose(1999)hasdiscussedthefactthathavingmachinesthatchallengeusmentallyissomethingdifferentthanthemchallengingusphysically.
Whilebeingrelievedofphysical64Stralabouriswelcomed,beingchallengedintellectuallyisworse,asthinking"hasbeenaveryhumanprerogative"(Penrose1999,p.
66).
Thequestionthatfollowsisoneofsuperiority,asif"machinescanonedayexcelusinthatoneimportantqualityinwhichwehavebelievedourselvestobesuperior,shallwenotthenhavesurrenderedthatuniquesuperioritytoourcreations"(Penrose1999,p.
67)Gillies(1996)discussestheseissuesandrelateshisdiscus-siontodebatesinthedecadesprecedinghisbookArtificialintelligenceandscientificmethod.
Herelates,amongotherthings,opinionsthatplacedsuperiorityinchesswellbeyondthescopeofanyfutureadvancesinAI.
Writingin1996,hehimselfstatesthat"iftheprogressofthelasttwentyorsoyearscontinues,thenitisnotunlikelythat,withinafewdecades,computerchessprogramswillexistwhichnohumaniscapableofbeating"(Gillies1996,p.
117).
Aswehaveseen,DeepBluebeatGarryKasparovin1997.
Similarly,Kurzweil(2015)predictsthat"withinseveraldecadesinformation-basedtechnologieswillencompassallhumanknowledgeandproficiency,ultimatelyincludingthepattern-recognitionpowers,problems-solvingskills,andemotionalandmoralintelligenceofthehumanbrainitself"(Kurzweil2015,p.
148).
Withoutgettingtootechnical,letus,forourcurrentpurposes,assumethatsuperhumanintelligenceinmachinesisafuturepossibility.
Intellectualinferiority,then,andapotentiallyverytroubledrelationshipwithwhatwehavemade.
MuchlikeinMaryShelly'sFrankenstein,thethingswemakemaycomebacktohauntus(Shelley2003).
WillthesuperintelligenceBostrm(2014)discussesbeourFrankensteinWillitcauseusgriefaswewatchitsurpassus,orwillwesimplywelcomethesenewentitiesthatcanperformevermoreimpressivetasksforusWemay,accordingtoactivitytheory,viewourselvesasempovered,andsimplyinneedofareorientationwithregardtoourrelationshipwithourtools(Kaptelinin1996).
ThesubtitleofShelley'sbookisTheModernPrometheus(Shelley2003).
ThisisareferencetotheGreekmythoftheTitanPrometheus,whoiscreditedwithcreatingmanfromclay.
Indoingso,Prometheusstealsfirefromthegodsandgivesittohumans.
Thisfirebringsnotonlylifetomanbut"alsothesubtlerfireofreasonandwisdomfromwhichallaspectsofhumancivilizationisderived"(Raggio1958).
Thegodswereangry,however,andsent"feveranddiseasetothehumanrace"whilecondemningPrometheustotorture(Raggio1958).
Therearemanyvariationsofthismyth,andperhapsmodernmanislivingoneofthem,makingnewformsoflifeandgivingthemreasonandwisdom.
Whilereachingforthatdivinefire,wemusthopewearenotoverreaching.
Iwillnotgointomuchdetailrelatingtothenatureofconsciousnessorthecognitionandmentallivesofrobotsinthispaper.
Instead,Ifocusonwhattheycandoandhowweseeandrelatetothem.
But,whatdoesallthismeanforwhatitmeanstobehumanWhatevergoesoninsidetheir"minds",smartrobotschangethewaywerelatetorobots,andtheyalsochangethewayweseeourselves.
AccordingtoGoldberg(2009)thisisatwo-waydeal,wheremuchofourknowledgeofthebrainhasbeenpresentedthroughcomputeranalogies,whilethecomputerswebuildaremodelledonwhatweknowofneurologyandbiology(Goldberg2009,p.
279).
Koestlerthoughtthatsciencewas"dizzywithitsownsuccess"and"forgottoaskthepertinentquestions"—questionsthatcouldnotbeanswered"solongasone'simageofmanisthatofaconditionedreflex-automaton"(Koestler1967,p.
xi-xii).
Turkle(2011)proposesthatthiswayofseeinghumansmaybepromulgatedbysmartrobots—creaturesthatwesomehowidentifywithorlikentoourselvesorotherbiologicalcreatures.
ReturningtoKoestler,thebehaviourismhesointentlycriticisedwouldbeagoodscienceforconflatinghuman-machineworking(Koestler1967).
Withbehaviourism,we"didawaywiththeconceptofmindandputinitsplacetheconditioned-reflexchain"(Koestler1967,p.
6).
TheGhostintheMachine65How,then,doweunderstandourselvesinrelationtothesenewentities—alreadysmart,butpotentiallyverysmartinthenotsodistantfutureAreWeReallySpecialThroughouthumanhistory,manyattemptsatansweringthequestionofwhatmakeshumanbeingsspecialhavebeenmade.
Whatisit,forexample,thatseparatesmanfromanimalsThisquestionhasgivenrisetomanyanswers,amongthemthatmanis"apoliticalanimal(Aristotle);alaughinganimal(ThomasWillis);atool-makinganimal(BenjaminFranklin);areligiousanimal(EdmundBurke);andacookinganimal(JamesBoswell,anticipatingLévi-Strauss)"(Thomas1983:31).
First,wewillhaveabrieflookatthephilosophicaldebate,beforemovingontosomepsychologicalperspectives.
ThePhilosophicalDebateThisdebatehasgottennewprominencetoday,asitisclearlyrelevanttothequestionofwhatintelligentmachinesare.
Ifreason,forexample,iswhatdefinesman,acomputerthatsatisfiesourdefinitionsofrationalitymaybeconsideredamanOr,ifourdistasteforthisconclusionistoostrong,dowereviseourdefinitionstoexcludemachines,orperhapssearchforotherdefiningaspectsofhumanityTurkle(2011)describesthisashappeningwhenprimitiverobotsstartedshowingwhatappearedtobeemotions.
Manwaspreviouslydefinedasarationalanimal,accordingtotheAristotelianmethodoffocusingonexclusivecharacteristics.
Then,whencomputerscouldnotbedeniedrationality,wesuddenlybecameemotionalanimals.
Cole(1990)proposedthatwhatsetshumansapartisourabilitytouseartefactstomodifyourenvironmentandtotransmitthesemodificationstosubsequenthumansthroughlanguage.
Theseconditions,too,cansurelybefulfilledbysmartcomputers.
Regardlessofhowweapproachthisissue,intelligentmachinesareclearlyachallengetothebeliefthatwearespecial,orinsomewaysuperiorto(all)otherthings.
Thischallengeislessofaproblemforthosethatdonotadheretosomeformofanthropocentrism.
Manyenvironmentalphilosophers,likeArneNss,adamantlydethronedmanandputhimsquarelyontheleveloftherestofthenaturalworld—promotingbiosphericalegalitarianismasanalternativetoanthropocentrism(Nss1999).
Thebiospheredoesnotincludemachines,though,sothesephilosophieswillalsohavetoberethoughtifwearetodealadequatelywiththepossiblemoralvalueofintelligentmachines.
Ascyberneticsandbiotechnologycontinuestoevolve,itwillbeevenhardertoexcludeobjectsthatarenotwhatwetraditionallyconsiderbioticfrommoralconsideration.
Thephilosophicaldebatesrelatingtotheissuesoftechnologyareoftenlinkedtothetermstranshumanismandposthumanism.
Transhumanismismainlyconcernedwiththepossibilityofenhancinghumanitythrough,forexample,"regenerativemedicinetonanotechnology,radicallifeextension,minduploadingandcryonics,amongotherfields"(Ferrando2013).
Posthumanismisaboutthe"decentering"ofmanand"raisesseriousquestionsastotheverystructuresofoursharedidentity—ashumans—amidstthecomplexityofcontemporarysci-ence,politicsandinternationalrelations(Braidotti2013,p.
2).
Wemayforgenewrulesthatexcludesmachinesbydefinition,butthatdoesnotprecludeapossibleidentitycrisisforhumanity.
Whatarewe,whenwearenotthemostintelligentormostpowerfulbeingsweknowofWhatarewe,whensomeofuswantstobewiththeirrobotsinsteadofotherpeopleOurinferiorphysicalpowerwasmostlikelyacknowledgedassoonas66Strawereflectedonthestrengthofthebeastswecameacross.
Thiswasperhapsnotthatmuchofadefeat,sincewealsorealisedthatwewerementallysuperior.
Wecoulduseourcunningtoovercomephysicaldisadvantage,andsoonweruledtheearth.
Ingenuity,intelligence,languageandcreativitymadehumansa"huge,dangerousandpowerful"force,andtheEarthnowsuddenlyappearstobebothsmallandfragile(Randers2012).
PsychologyandHumanityThetermconditionedreflex-automatonwasrejectedbyKoestlerasaproperdescriptionofman(Koestler1967)ButisitsuitableforAIsBehaviourismcanexplainalot,Koestlersays,butperhapsnot"scientificdiscoveryandartisticoriginality"(Koestler1967,p.
13).
WatsonandSkinnermayhavereducedustolittlemorethanrats,butGoldberg(2009)pointsoutthatourfrontallobeswasvastlydifferent.
Behaviourcanbeobserved,whereasthesoul,mindandconsciousnessarevagueterms;"[c]onsciousness,Watsonobjected,is'neitheradefinablenorausableconcept,itismerelyanotherwordforthe"soul"ofmoreancienttimes.
…Noonehasevertouchedasoulorseenoneinatest-tube.
Consciousnessisjustasunprovable,asunapproachableastheoldconceptofthesoul.
'"(Koestler1967,p.
15).
Thesetermsarerejectedbythoseseekinganobjectivescienceofpsychology.
Behaviourandphsyiologyareeasiertomeasureanddealwithinaseeminglyobjectivefashion.
RatomorphyisthetermKoestlerusesfordenyinghumansanyfacultiesnotpresentinrats—theoppositeoftheanthropomorphicfallacywherewegiveanimalshumanqualities(Koestler1967,p.
17).
Robotomorphymaybetheapplicableterminourcontext,ifweendupviewingmanaslittlemorethananadvancedcomputer.
Andifwedo,aclassicalquotefromSirCyrilBurtmaybeappropriate:Theresult,asacynicalonlookermightbetemptedtosay,isthatpsychology,havingfirstbargainedawayitssoulandthengoneoutofitsmind,seemsnow,asitfacesanuntimelyend,tohavelostallconsciousness(Burt1962,p.
229).
AntonioDamasioisofinterestwhendiscussingthedifferencesbetweenmanandmachine.
InhisfamousbookDescartes'Error(1994),hediscussesthenatureofrationality,andarguesthatitcannotbeseparatedfromemotions.
Letussaythatapersonhas"knowledge,attention,andmemory",languageandlogic,andtheabilitytoperformcalculations.
Damasiomentionsapatientlikethis,seeminglywellequippedtoforrationalthought,yetstillsomehowunabletoachieveit.
Thepatienthadabraindamagerelatingtotheprocessingoffeelings,andthiscaused"flawed"reasoning.
ToDamasio,itbecameclearthat"feelingwasanintegralcomponentofthemachineryofreason"(1994,p.
xii).
Heacknowledgesthatemotionsandfeelingsalsocanhavenegativeinfluencesonrationalthinkingbutmakesthepointthattheirabsencealsohasdramaticconsequences(Damasio1994,p.
xii).
InLookingforSpinoza(2003),heupdateshistheorieswithnewevidence,andconsidersfeelingsthemselves,notnecessarilyrelatingtodecision-making.
Hisviewisthat"feelingsaretheexpressionofhumanflourishingorhumandistress,astheyoccurinmindandbody"(Damasio2003,p.
6).
Damasioismostfamousforhissomatic-markerhypothesis,discussinghowrationaldeliberationactivates"gutfeelings"thatguideusintheprocessofdeliberation(Damasio1994,p.
173).
Thesomaticmarkers"forcesourattentiononthenegativeoutcometowhichagivenactionmaylead,andfunctionsasanautomatedalarmwhichsays:Bewareofdangeraheadifyouchoosetheoptionwhichleadstothisoutcome"(Damasio1994,p.
173).
Inarecentpaper,MaldonatoandValerio(2018)discussthesametopic,andsaysthat"emotionsarecrucialinTheGhostintheMachine67moraldecisionsandthattheirunderstandingmayhelpustoavoidmistakesintheconstructionofhybridorganismscapableofautonomousbehavior"(MaldonatoandValerio2018).
Isthishowweareabletokeepthebarbedwirefencebetweenmanandotherthingsintact,whilekeepingthecriteriaofrationalityMachineshavejustabouteverythingrequiredforreason,butnotemotionsDamasiodiscusseswhatisrequiredforemotions,andstartswiththerequirementthat"anentitycapableoffeelingsmustbeanorganismthatnotonlyhasabodybutalsoameanstorepresentthatbodyinsideitself";plantshave"bodies",butnowaytomentallyrepresentorunderstandtheirbodies,andthusnoemotions(Damasio2003,p.
109).
Secondly,bodystatesmustbemapped,andtheorganismmustbeabletotranslatethesebodystatesinto"mentalpatternsorimages"(Damasio2003,p.
110).
Thirdly,hedemandscon-sciousnessinreturnforemotions;"inplainterms,wearenotabletofeelifwearenotconscious"(Damasio2003,p.
110).
Lastly,"thebrainofanorganismthatfeelscreatestheverybodystatesthatevokefeelingsasitreactstoobjectsandeventswithemotionsandappetites.
Inorganismscapableoffeeling,then,thebrainisadoublenecessity"(Damasio2003,p.
110).
Itmustbothcreatethe"bodymaps"thatrelatetovariousfeelings,anditmustalso"commandorconstructtheparticularemotionalbodystatethatendsupbeingmappedasafeeling"(Damasio2003,p.
110).
Heconcludesthat"[m]ostanimalswithcomplexbrainssatisfytheseconditions,inallprobability"(Damasio2003,p.
110).
Cominellietal.
(2018)picksuponDamasio'stheoryofmindanddiscussesthepossibilityofartificialintelligencewithsocialandemotionalintelligence.
Theypresenta"SEAI(SocialEmotionalArtificialIntelligence),acognitivesystemspecificallyconceivedforsocialandemotionalrobots.
Itisdesignedasabio-inspired,highlymodular,hybridsystemwithemotionmodelingandhigh-levelreasoningcapabilities"(Cominellietal.
2018,p.
1).
Theydescribetheirmodelindetail,andwhiletheyconcludethatthereare"stillsomeshortages",itseemsAIcannotbeexpectedtostayunabletoapproachsomeofthewayshumanbeingswork,andthewayemotionsguideourrationality.
Theyclaimtoprovide"cleardemonstrationofhowSEAIandthechosen'understandingbybuilding'approachleadtoanimportantconfirmation:withSEAI,robotscanbenefitfromtheirownartificialemotionsfortakingdecisionsandtreasuretheirpastinteractions"(Cominelli2018,p.
18).
So,consciousandemotionalrobotsareontheirway,andasthesescientistsalsoconclude,"ethicalissueswillbecomeextremelyrelevantandcritical"(Cominelli2018,p.
18).
Inasimilarvein,MaldonatoandValerio(2018)claimthatthe"encounterbetweenHumansandverypowerfulAIwilllead,inthenearfuture,toorganismscapableofgoingoverthesimulationofbrainfunctions:hybridsthatwilllearnfromtheirinternalstates,willinterpretthefactsofreality,establishtheirgoals,talkwithhumansand,especially,willdecideaccordingtotheirown'systemofvalues'"(MaldonatoandValerio2018).
Scheutz(2016)echoesthesesentimentsandmakesthecaseforthedevelopmentofrobotswith"moralcapabilities".
Thisisnecessary,hesays,because"anyordinarydecision-makingsituationfromdailylifecanbeturnedintoamorallychargeddecision-makingsituation"(Scheutz2016,p.
515).
Damasio(2018)commentsonthesedevelopments,butisscepticalofitsprospects.
Wemayhaverobotswithsomethinglikefeelings,"arisingfromanartificialsubstrateprovidedtheywouldbereflectionsof'homeostasis'",but"thereisnoreasontoexpectthatsuchfeelingswouldbecomparabletothoseofhumansorthoseofotherspecies"(Damasio2018,loc2973).
Turklesupportsthesesentiments,andstatesthatrobots"hasnofeelings,canhavenofeelings,andisreallyjustaclevercollectionof'asif'performances"(Turkle2011,p.
6).
Authenticityforherrequires"theabilitytoputoneselfintheplaceofanother,torelatetotheotherbecauseofasharedstoreofhumanexperiences:weareborn,havefamilies,andknowlossandthe68Strarealityofdeath"(Turkle2011,p.
6).
Damasioismoretechnicalandnames"thesubstratethatfeelingsactuallyuse"asthecrucialmissingfactor(Damasio2018,loc2973).
Theseunder-lyingsubstratesisalsowhatmakesDamasiorejecttheperspectivethatmanisnothingbutalgorithms(Damasio2018,loc2951).
Substrates,andthelivingbody,asheexplicitlystatesthat"tobuildfeeling,ontheotherhand,werequirealivingbody"(Damasio2018,loc3974).
Damasioalsohasanotherargumentagainstthealgorithmicviewofman,inthatthatit"doesnotadvancethehumancause",somethingmanyotherswouldarguethatweshouldnotrequiredirectlyfromscience(Damasio2018,loc3974).
WesawGillies(1996)predicttheriseofacomputerchessmasterwithindecades,anditarrivedtheverynextyear.
Similarly,thetimingofthearrivalofintelligentcomputersthatsatisfyvariousconceptionsofconsciousnessandemotionishardtopredict.
Whatseemsclearisthatifourdefinitionsrevolvearounddescriptionoffunction,andnotbiologyofphysiology,computerswillgetthere.
Probablysoonerthanweimagine,ifwearetobelieveKurzweil's(2015)descriptionoftheexponentialprogressweareinthemidstof.
Whatisevenmoreimportanttousthanthesetechnicalaspectsisthatmosthumanbeingsconsidersmartrobotsasaliveenoughtowarrantbothemotionalreactionsandethicalconsideration(Turkle2011,ch.
2).
Whenwealreadyconsiderthemtobesomesortoflife,andpeopleinexperimentsexperiencediscomfortwhenatoyFurbyappearstobehurt,wecanassumethattheproblemsofmeaningfullyseparatingourselvesfromsmartcomputerswillbeanincreasinglytallorder(Turkle2011,p.
45).
Wemay,ofcourse,choosetotaketheapproachthatcomputersarenotachallengetous,butsomehowanextensionofus—ourtoolsthatempowerus.
Thiscan,forexample,bebasedonactivitytheory(Kaptelinin1996).
Suchanapproachwouldnot,however,changethefactthatmanystruggletoperceivethemeaningfuldifferencebetweenusandthesetools—intellectu-ally,emotionallyandmorally.
Afurtherchallengeinthisrespectwillbethemergingofthetechnologicalandthebiological,atopicIwillnotdiscussinthispaper.
DigitalTechnologyandHumanRelationshipsTechnologyisseductivewhenwhatitoffersmeetsourhumanvulnerabilities.
Andasitturnsout,weareveryvulnerableindeed.
Wearelonelybutfearfulofintimacy(Turkle2011,p.
1).
Ascomputersbecomemoreintelligent,weinteractwiththeminnewways.
Theinteractionsthemselvesarenotnecessarilynew,buttheywerepreviouslyreservedforrelationsbetweenhumansandotherlivingthings.
Thingslikefamily,friends,petsandotherbiologicalbeings.
Thisdevelopmentisnotentirelynew,butasAIbecomesincreasinglypowerful,thepartthattechnologycanplayinchangingoursociallivesdramaticallyincreases.
InmovieslikeHerfrom2014,thelovebetweenamachineandamanisportrayed.
Movies,however,aremovies,andreallifeisreallifePeoplesurelydonotformthosekindsofbondstoartificialentitiesDavidLevywoulddisagree,andheargueshiscaseinLoveandSexwithRobots(Levy2008).
Robotswillteachusmanythings,heargues,andtheyareinmanywaysbetterthanhumanmatesastheyarebothfaithful,caringandabletoteachusandinteractwithusinwayshumanbeingscannot(Levy2008).
Unlimitedattention,patienceandaccommo-dation—whatisnottoloveinsuchapartnerTurklediscussesthepossibilityofbeingagainsthuman-computerrelationsasequallyreactionaryinthefutureastoday'scurrentopponentstoTheGhostintheMachine69same-sexmarriage,etc.
(Turkle2011).
WhenareportercallsheraboutLevy'sLoveandSexwithRobots(Levy2008),shediscussesherreservationsandissurprisedwhen"thereportersuggestedthatIwasnobetterthanbigotswhodenygaysandlesbianstherighttomarry"(Turkle2011,p.
7).
LessobjectionabletomostmaybetheideaofusingAIforcompanionshipandentertain-ment.
Relationshipsthatarenon-intimate,thatis.
Lessobjectionable,becausewealreadyhavethis,andwehavehadthemforalongtime.
Fongetal.
(2003)describes"sociallyinteractiverobots"as"robotsforwhichsocialhuman-robotinteractionisimportant"(Fongetal.
2003,p.
1).
Fromtheearlybeginninginthelate1940s,wehavegottenveryadvancedrobots,whosemainfunctionistointeractsociallywithhumans,likeMIT'sKismetandCog(Fongetal.
2003,p.
4).
Astepbackfromtheseadvancedrobots,however,wehavehadseveralverypopularsocialcomputers(notalwaysembodied)inoursocieties.
TheTamagotchiwasperhapsthefirst,whenitsweptacrosstheworldinthe1997TheTamagotchihadnobodybutlivedonascreeninanegg.
Itsuserhadtofeedit,cleanupafterit,andentertainit,andpeopleeverywherewerethrilledtoseetheirTamagotchis"growup"withpersonalitiesshapedbytheir"parent's"actions.
Thethrillwasreal,andsincepeoplecaredforthem,theyalsogrievedwhentheydied(Turkle2011,p.
34).
AnotherbigsuccesswastheFurby,releasedin1998.
TheFurbydidhaveabody,requiredcareandattention,andevenseemedtolearnEnglishfromitsowners.
TheFurbysandTamagotchiswere"thefirstcomputersthataskedforlove"(Turkle,2011,p.
30).
Furbyscouldnothearanddidnot"learn"inthatway,butthekidsthatusedthemthoughttheydid(Turkle2011,p.
38).
MyRealBaby,alsoaHasbroproduct,launchedin2000,wasanintelligentbabydoll,goingthroughthedevelopmentfrom0to2yearsintheirowner'scare(Turkle2011,p.
67–8).
Itwasnotahugesuccess,butitshowedsomeofthepotentialinthesekindsofmachines.
ThelastexampleweshouldmentionisAiboandParo.
Aibotakesusbeyondtheworldoftoysandintotheworldofpetreplacement(Sony2018).
Itissoldas"fullofcharm"with"[r]ound,alluringeyeswithapowerfulpull,acute,roly-polyform,movingaroundwithinfectiousenergy,andanidentity…"(Sony2018).
Peoplebondwiththese"pets"—feelingguiltywhennotentertainingthemandworryingabouttheirhealthandwell-being(Turkle,2011,p.
57).
Ifwearetobelievetheproducers,Aiboloveshumans,too,as"[b]eingwithpeopleiswhataibolovesbest"(Sony2018).
Paroisperhapsthemostinteresting,asitissoldasa"therapeuticrobot",andemployedinmanyeldercarefacilitiesaroundtheworld(ParoRobots2018).
Paroprovidesthe"docu-mentedbenefitsofanimaltherapy",evenifitisnotactuallyananimal(ParoRobots2018).
Paroremembersthenamepeoplegiveit,andevenrememberstheactionsitdidbeforebeingpetted—tryingtorepeattheseactionswiththegoalofbehaving"thewaytheuserprefers"(ParoRobots2018).
Stressandlonelinessisrelievedintheelderlyanddemented,timeisfreedforthecaretakers,andthefamilymembers'conscienceisabitclearerwhenleavingtheirelderswithasmileandasealontheirlap.
Win,win,winTheproblemis,however,thatwhiletheserobotsarenotdesignedforthatmoresuperficialkindofintimacy,theydocreatebondsthatcaneasilybeconstruedasintimate.
Turkle(2011)tellsmanystoriesofhowevensimplecomputerscreateemotionalresponsesinus,andhowwegrowtocareforthem.
Weevengrievewhentheydie,andarereluctanttoletthemgo,orleavethem.
Evenifweknowthattheyarenotrealandhavenoproperfeelings,theylookatus,askfornurtureandprovidecompanionship.
Thismakesthem"aliveenough"forpeopletobond,whichisthemainpointTurkle(2011)emphasisesassheasksustothinkaboutwhatwearedoingwiththesemachines.
Thequestion,then,iswhathappenstousWhathappenstoourrelationshipswhenwebondwithcomputers,andevenreplacehumancontactwithcontactwithmachines70StraTurkleisconcerned,asapsychoanalyticallytrainedpsychologistvaluing"relationshipsofintimacyandauthenticity",abouttheprospectsofintimaterelationshipswithcomputers.
Computers"havenofeelings,canhavenofeelings,andisreallyjustaclevercollectionof"asif"performances,behavingasifitcared,asifitunderstoodus"(Turkle2011,p.
6).
Damasiosupportsthisview,whenstatingthat"lifelike,humanlike"are"anabsurdandnon-existingmyth",andthat"[h]umanshavelifeandhavefeelings,andsuchrobotshaveneither"(Damasio2018,loc3058).
Thisgoesbacktothefundamentalquestionofwhatmachinesreallyare,andwhatmakesforauthenticity.
Afterall,whatarefeelings,apartfromchemicalreactionsinourbodiesIsthisreallyfundamentallydifferentfromthe"feelings"oractionsofanAI,influencedbyimpulsesandreactingonthemLearning,evolving,anddevelopingnewreactionsasitgathersexperienceIfweseehumansthisway,thereislittledangerininteractingwithotherautomata.
If,however,weseeourselvesasmore,asTurkle(2011)does,theinteractionsweincreasinglyhavewithcomputermaydiminishus.
Perhapsitwillevenmakeusmorelikeautomata,justbyinteractinginsuchawayandacceptingthatourcompanionsdo.
ForTurkle,anauthenticrelationship"involvescomingtosavorthesurprisesandtheroughpatchesoflookingattheworldfromanother'spointofview,shapedbyhistory,biology,trauma,andjoy"(Turkle,2011,p.
6).
Sheconcludesthatrobotscannotdothis.
We,however,mayarguethatthisconclusionispremature,Metzleretal.
(2016)doeswhentheycommentonTurkle'spositionandendsupsayingthatrobotsmightbecomeauthenticcompanions.
Theprogressonthisfieldissorapidthatitisdifficultnottoimagineafuturewheresomewillarguethatrobotsareshapedbytheirexperiences,andwehavealreadyseenthatcomputersarenowbuiltwiththeintimationsofemotions(Cominelli2018).
Furthermore,aswehaveseen,itisalsopossibletoviewhumanbeingsaslittlemorethanmachinesreactingtoexternalimpulses—muchthesameasrobots.
Turkle(2011)callsouracceptanceofthesesocialmachines"theroboticmoment"andthat"wedon'tseemtocarewhattheseartificialintelligences'knows'or'understands'"(Turkle2011,p.
9).
Whilechildrenpreviouslyprojectedthemselvesuponinanimateobjectslikedolls,wetreatthesemachinesassubjects—"Rorschachgiveswaytorelationship"(Turkle2011,p.
95).
Itistemptingtoproposetheargumentthatweneverreallyknowifotherpeoplereallyknoworunderstandus,either.
FollowingDescartes,wecouldnotsaythatallIreallyknowisthatImyselfexistThemainquestionTurkleseemstoaskisthis:"Whatifarobotcompanionmakesusfeelgoodbutleavesussomehowdiminished"(Turkle2011,p.
6)Companionshipfeelsgood,andwedonotseemtoobjecttorobotsreplacingloneliness–orevenhumanbeings.
"Peoplearelonely",Turklesays,butinalwaysfindingsomethingtoletusescapethatloneliness,"wemaydenyourselvestherewardsofsolitude"(Turkle2011,p.
3).
WhatrewardsaretheseOneistobeable"toreflectonone'semotionsinprivate"(Turkle2011,p.
176).
ErichFromm(1994)describestheneedforrelatednesstotheoutsideworld,andthefactthat"mentaldisintegration"followslonelinessandfeelingsofisolation(Fromm1994,p.
17).
Butwhatif"lonelinessisjustfailedsolitude"(Turkle2011,p.
288)ThereareothersthatagreewithTurklethatsolitudehasitsrewards,bothfortheindividualandsociety.
Solitudehasbeenatopicinphilosophysinceearlymodernity,especiallysincepopulationgrowthleadmoreandmorepeopletoliveindenselypopulatedareas(Thomas1983,p.
268).
ForThomasHobbes,forexample,solitudewashorrible,whilephilosopherscomingafterhim,likeRousseauandJ.
G.
Zimmermann,wroteextensivelyonthebenefitsofTheGhostintheMachine71solitude(Thomas1983,p.
268).
OneclearstatementonthebenefitsofsolitudecamefromJohnStuartMill:Aworldfromwhichsolitudeisextirpatedisaverypoorideal.
Solitude,inthesenseofbeingoftenalone,isessentialtoanydepthofmeditationorofcharacter;andsolitudeinthepresenceofnaturalbeautyandgrandeur,isthecradleofthoughtsandaspirationswhicharenotonlygoodfortheindividual,butwhichsocietycouldilldowithout(Mill2004,p.
692).
Solitudegivesusthe"spacetothinkourownthought"andit"refreshesandrestores"(Turkle2011,p.
204,288).
AccordingtoAnthonyStorr"byfarthegreaternumberofnewideasoccur"duringstatesakintosolitude(Storr2005,p.
198).
Butsolitudeseemshardtocomebythesedays.
Youngpeopleexperiencediscomfortwhentheirphonesarenotaround,andthatmanymodernpeople"trainedbytheNet"cannotfindsolitudeeveninnature(Turkle2011,p.
289).
If"[l]onelinessisfailedsolitude"peopleseemtobefailinginrecordnumbersifonelooksatreportedlonelinesscombinedwiththepurporteddifficultyoffindingsolitude(Turkle2011,p.
288).
Onafinalnote,itisinterestingtoseethattherearedevelopmentsinTurkle'sowndisciplineofpsychologytowardsrobottherapists.
ByrobottherapistsIamnottalkingabouttherapeuticrobotslikeParo,butrobotsthatcommunicatewithwords.
Whilecommunicationtechnologyletstraditionaltherapybeprovidedovergreatdistance,andwithoutsynchronicity,wenowgoevenfurtherandremovethehumantherapistaltogether(Ebertetal.
2015).
PracticeMakesPerfect,orMachineEvolved/ManDevolvedWehavetoloveourtechnologyenoughtodescribeitaccurately.
Andwehavetoloveourselvesenoughtoconfronttechnology'strueeffectsonus(Turkle2011,p.
243).
Machinelearningreferstothefactthatcomputerslearnandevolvethroughtrainingandexposure.
WhatifhumanbeingsactuallylearninasimilarfashionWhatifwelearnbybeingexposedtointellectualchallenges.
Ifthisisso,whataretheimplicationsoflettingmoreandmoreofthemostchallengingintellectualworkbeingdonebycomputersOnceuponatime,pupilswererequiredtolearnthemultiplicationtable.
Theywereevenabletodomanualcalculations.
Perhapstheystilldo,butnowadaysouryoungmaybetheonlyonesskilledatmanualcalculation,astheyaretheonlyonesforcedtoperformitonaregularbasis.
Therestofushavemostlylosttheskill,duetotheubiquitouscalculator—everpresentonourcomputer,phoneorwatch.
Infact,IcanjustaskSiritoperformthecalculationsIneed.
Researchshowsthatchildren'smathematicalabilitiesarenotnegativelyaffectedbycalcula-tors,butmanualcalculationabilitieswouldbeexpectedtosuffer(Barton2000).
Thisgoeswellwithresearchontheeffectsof"braintraining",wherethereismuchevidenceonthepossibilityofimprovingperformanceoncertaintasksthroughtraining,butthattransfertoothertasksishardtoprove(Nouchietal.
2013;Simonsetal.
2016).
Ifthisprocessisindicativeofhowourbrainworksinotherwaysaswell,wemaybepoisedforasomewhatsadfuture.
Ifweleavetheadvancedintellectualchallengesandthinkingtocomputers,whatwillhappentoourcognitiveabilitiesWemay,accordingtoactivitytheory,becomeempoweredandlearntoperformthetasksbetterafteraperiodofusingthetools(Kaptelinin1996).
Thisdoesnotseemlikelytooccurifweusecomputerstoautonomously72Strareplaceus,insteadofmerelyassistinguswiththetasksperformed.
Norman's(1991)ideafromthepersonalviewofcognitiveartefacts,thattaskschangeandthatthecapacitieshumansmustthenuseisaffected,seemsmorerelevantinthiscase.
Thismeansthatwewillpracticeotherabilitiesthanbefore,andtheneteffectmayofcoursebebothpositiveandnegative.
Whatisimportanthereistomakesurethatweunderstandhowwearechanged,andthatwedonotletourselvesloose,ordiminish,abilitiesthatwevalueandconsiderimportant.
AdifferentbutrelatedtopiciswhathappenstoourbrainsifwechangeoursociallivesandreplacehumanswithrobotsTheneuropsychologistandcognitiveneuroscientistGoldberg(2009)askswhether"socialstimulationistothedevelopmentofthefrontalcortexwhatvisualstimulationistothedevelopmentoftheoccipitalcortex"(Goldberg2009,p.
174).
Thequestionbeing:willourbraindevelopdifferentlyifwechangethewaywesocialiseHere,Turkle'sstatementthatwhen"youpracticesharing'feelings'withrobot'creatures,'youbecomeaccustomedtothereduced'emotional'rangethatmachinescanoffer",becomesrelevant(Turkle2011,p.
125).
Thefrontalcortexisalsocalledtheexecutivebrainandislinkedwithissuesofself-controland"intentionality,purposefulness,andcomplexdecisionsmaking"(Goldberg2009,p.
4).
Furthermore,thefrontallobesaresaidtobewhat"makeushuman"and"theorganofcivilization",asnootherspeciesdeveloptheselobeslikeushumans(Goldberg2009,p.
4).
Itwouldbeparadoxical,then,ifweusedourevolvedexecutivebrainstodeveloprobotcompanions,whosecompanionshipinturnpreventsthedevelopmentofthispartofthebrain.
This,again,couldbytheverysamelineofreasoningleadustotheoutrageouspropositionthatsocialisingwithrobotsmayleadtounintelligenceanddecadence.
Evenmorality,Goldberg(2009)argues,islinkedwiththefrontallobes(Goldberg2009,p.
17).
Doesthisleadtoasituationinwhichweretardourexecutivefunctions,muchlikethedegenerationofourabilitytodomanualcalculationsafterthearrivalofcalculatorsCarr(2011)haswrittenontheeffecttheInternetonourbrains.
Hecitesthescepticswhosaythatournewwayofconnectingandconsumingmedialeadstoanew"darkageofmediocrityandnarcissism"(Carr2011,loc116).
Turkle(2011)echoesthesesentiments.
Shedescribesaselfthatisdeveloped"inthecacophonyofonlinespaces"andinthisprocesstempted"intonarcissisticwaysofrelatingtotheworld"(Turkle2011,p.
179).
Carr(2011)describeshisownfeelingsofchange:Nowmyconcentrationstartstodriftafterapageortwo.
Igetfidgety,losethethread,beginlookingforsomethingelsetodo.
IfeellikeI'malwaysdraggingmywaywardbrainbacktothetext.
Thedeepreadingthatusedtocomenaturallyhasbecomeastruggle(Carr2011,loc158).
Hethengoesontodescribethatthisfeelingofchangeisechoedbyalotofpeopleintoday'ssociety.
Oneofhismainargumentsisthatourbrainsareplastic(adultonestoo,eveniftheyarelessplastic),andthatthewayweusethemchangesthewaytheyfunction:"Ourwaysofthinking,perceiving,andacting,wenowknow,arenotentirelydeterminedbyourgenes.
Noraretheyentirelydeterminedbyourchildhoodexperiences.
Wechangethemthroughthewaywelive—and,asNietzschesensed,throughthetoolsweuse"(Carr2011,loc595).
Cabdrivers'brainschange,becausetheycontinuallystoreandusesomuchgeographicalinformation;whentheystartusingGPSthisdoesnothappen,andthey"losethedistinctivementalbenefitsofthattraining"(Carr2011,loc3594).
Anotherinterestingstudyshowsthatpeoplethatphotographedobjectsinamuseumactuallyrememberedlessofwhattheysaw—asifthebrainunderstoodthatitneednotbotheraboutthethingswecanuseothertoolsfor(Henkel2014).
vanNimwegen(2008)callsthisexternalisation,andreferstoTheGhostintheMachine73situationsinwhich"thenecessitytoremembercertaincharacteristicsofoperatorsiseliminatedorlessened"(vanNimwegen2008,p.
123).
Hetestedtheseeffectsinanexperimentwheretwogroupsweregivendifferentinterfacesforatask—onewhichprovidedvariousformsofassistanceandonethatdidnot.
Theusersoftheinterfacewithlessassistance"imprintedrelevanttaskandruleknowledgebetterandwherenotaffectedbyasevereinterruptionintheworkflow"(vanNimwegen2008,p.
125).
Professionalsarealsoaffected,asastudyshowsthatdoctor'suseofclinicalguidelinesandelectronicmedicalrecordsleadstoaperceivedandexperienced"deskilling",involving"decreasedclinicalknowledge,decreasedpatienttrust,increasedstereotypingofpatients,anddecreasedconfidenceinmakingclinicaldecisions"(Hoff2011).
So,researchshowsthatbraintrainingisnotaneffectivemeanstoimprovinggeneralcognitiveabilities,butthattheyareclearlyeffectiveforthespecificskillswetrain(Danieletal.
,2016).
Ina2010articleOwenetal.
(2010)statesthesame,andthatwhile"improvementswereobservedineveryoneofthecognitivetasksthatweretrained,noevidencewasfoundfortransfereffectstountrainedtasks,evenwhenthosetaskswerecognitivelycloselyrelated"(Owenetal.
2010,p.
775).
Thereasoningisthatthesesortsofmechanismsapplytomostareasofourlives.
Further-more,Carr(2011)statesthatelectronicmediaisparticularlyeffectiveatchangingournervoussystems,muchbecausetheyfunctioninsimilarways(Carr2011,loc3618).
Itmayseemthatpractice,ifitdoesnotnecessarilymakeperfect,atleastimprovesus.
Theuseofcomputersissomekindofpractice,butitseffectsarenotnecessarilyonbalancebeneficial,as"ourabilitytolearncanbeseverelycompromisedwhenourbrainsbecomeoverloadedwithdiversestimulionline"(Carr2011,loc3635).
ThisisalsodescribedbyTurkle(2011),whodescribesoureverbettermultitaskingabilities.
Oncehailedasagreatbenefitofthedigitalrevolution,wenow"knowthatmultitaskingdegradesperformanceoneverythingwetrytoaccomplish"(Turkle,2011,p.
242).
Shealsodescribeshowthedigitalconsumptionofnewsandothermedia"alwaysinvitesyouelse-where"(Turkle,2011,p.
242).
Theold-fashionedbooksinvited"daydreamsandpersonalassociations"withreaderslookingintothemselves,whilethenewmediaisusuallyfleetingandconstantlydisruptedbythingslikemessagesandFacebook(Turkle,2011,p.
242).
Longingforapastthatneverexisted,youmaysay,buthermusingsabouthowweconsumetextsechoesCarr'sfindings,thatmedia"aren'tjustchannelsofinformation"butsomehow"shapetheprocessofthought"(Carr2011,loc174).
CarrdislikesthereshapingtheNethasbeendoingtohim,asheseesitas"chippingawayatmyconcentrationforconcentrationandcontemplation"(Carr2011,loc174).
AsTurkleputsit,"[a]swetrytoreclaimourconcentration,weareliterallyatwarwithourselves"(Turkle2011,p.
296).
InthewordsofCulkin:"Weshapeourtoolsandthereaftertheyshapeus"(Culkin1967,p.
70).
Carrrewritesthistobecome"weprogramourcomputersandthereaftertheyprogramus"(Carr2011,loc3633).
Whileactivitytheorymayseeusasempoweredwhenweuseournewtoolstoperformbetteractions,wemustkeepinmindthatwearealsochanged.
ItseemsthatweareonlyatthesecondstagethatKaptelinin(1996)described—thestagewhereweperformtasksmoreeffectivelywiththetools.
Thethirdstage,whereweachieveincreasedperformanceofthetasksevenwithoutthetools,maybehardertoachieveoncethetoolsweusebecomeaspowerfulastheyaretoday.
Conclusion:aWorldfortheBraveWehaveseenthatAIpermeatesallaspectsofmodernsocietyandthatthedevelopmenthasbeenrapid.
WealsoknowthatmuchoftheapplicationsofAItonewfieldsaredoneby74Stracorporationsandindustriousindividuals,oftenintheabsenceofanylawofregulationtocontrolorprohibitit.
Whywouldweprohibitit,whenitgrantssomanybenefitsWehaveseenthatthereareseveralreasonstothinkthisthrough,aswearerapidlychangingbothwhatitmeanstobehuman,andhowhumanbeingsfunction.
Thechallengetohumanidentityisachallengefromtwosides:automationanddisplace-mentandachallengeformentalsuperiority.
Ihavenotfocusedonjobslosttosmartcomputers,butchosentodiscussindepthhowtheprogressofartificialintelligencechallengestheverynotionofwhatitmeanstobehuman.
WhatseparatesusfromthesesmartcomputersFirst,itispossibletoviewhumanbeingsasmachinesreactingtotheirenvironments—emotionsandallbeingnothingbutchemicalreactions.
Thisviewportraysmanassomesortofmachine,andthefiringofsynapsesinthehumanbrainislittlemorethantheelectricalimpulsesthatworksitsmagicinarobot.
Secondly,weseethattheyevenmakeartificialsynapsesnow,andemotionalrobotssatisfyingprettystrictconditionsconcerningwhatisrequiredforsomethingtobeconsideredconscious(Schneideretal.
2018).
Withthisdevel-opment,itisbecomingincreasinglyhardertosettheboundariesbetweenmanandsmartmachine,unlessonehappenstobereligiousorspiritual.
Thesoulissomethingcomputerscientistshavenotyetclaimedtohaveproduced.
Fortherestofus,however,wemighthavetocometotermswithourcreations,whichnowrivalusintellectuallyinmanyways,beingconsideredlifeofsomesort.
Oneinterestingfactisthatwealreadyconsidersocialrobotsofvariouskinds"lifeofsomesort"—meaningfulenoughtolove,tomourn,toconsiderasmoralbeingsthatshouldnotbeneedlesslytortured,etc.
WhenthishappenswithTamagothisandFurbys,thereislittletosuggestthatwewouldnothaveincreasinglystrongrelationshipswithcomputers—evenrelationsofthemostintimatekind.
Inthefinalpartofthispaperdiscussedhowwemustbeawareofthefactthattrainingactuallymatters,alsowhenitcomestomattersofthebrain.
Thebraindevelopsasitisused,whichmeansthatitwilldevelopdifferentlyifwesetawayvarioustaskstocomputers.
Firstly,thedevelopmentofourbrainisinfluencedbyhowweinteractsocially.
Thelatterpartwasdiscussedinrelationtohowweexchangehumansforrobotsinsocialsettings,andhowGoldberg(2009)showsthatthesesocialrelationsareimportantforthedevelopmentofour"executivebrain"—thefrontallobes.
Impaireddevelopmentofourexecutivebrainswouldnegativelyinfluencenotonlymosthigh-levelcomplexcognitivefunctionsbutalsomoralevaluations.
Secondly,ourcognitivefunctionschangeaccordingtohowweemploythem.
Wehaveseenthattrainingofspecificcognitiveskillsimprovesthem,butthatthereislittletosuggestmuchtransferorgeneraleffectsfromsuchtraining.
Thesemechanismspointtowardsincreasedspecialisation,ascomputertoolshelpuswithmanyofthegeneralskillsthateveryonepreviouslyhadtoknow,lettingustrainandbecomeproficientatmoreexpertskills.
Ifso,weareinfactempoweredbythesetools,inaccordancebothwithactivitytheoryandthesystemviewofthecognitiveapproachinHCI.
ThefinalpointIhavediscussedisthegeneraldevelopmenttowardsbettermultitaskingabilities,atthecostofperformanceonsingletasks.
Thiswouldseemtogoagainstsomeofthebenefitsofspecialisation,andmightputaquestionmarkaftertheideathatweareempowered,atleastifCarr(2011)andothersarerightaboutourdecreasedattentionspansandabilitiestofocusoncomplexandchallengingtasks.
Itmightalsobeworthwhiletoconsiderthelong-termeffectsonhumanintelligenceifweleavethemorechallengingintellectualtaskstocomputers.
Beingscaredofsomeevolutionaryretardationdowntheroadmayseemfar-fetched,butcombinedwiththecognitivechangesTheGhostintheMachine75described,wearealreadyseeingthatthepathwewereonmightnotbeheadingexactlywherewewouldlikeittoo.
Thestoryofthecreatorthatgetsintotroublewithhiscreationshasbeentoldmanytimes.
Itusuptoustobeawareofwhatwemake,howweemployit,andnotleast:howitaffectsus.
ShouldwemanagetoavoidthepitfallsofPrometheusandFrankenstein'screator,goingalittlebitsloweronourwaytothefutureseemslikeasmallpricetopay.
OpenAccessThisarticleisdistributedunderthetermsoftheCreativeCommonsAttribution4.
0InternationalLicense(http://creativecommons.
org/licenses/by/4.
0/),whichpermitsunrestricteduse,distribution,andrepro-ductioninanymedium,providedyougiveappropriatecredittotheoriginalauthor(s)andthesource,providealinktotheCreativeCommonslicense,andindicateifchangesweremade.
ReferencesBarton,S.
(2000).
Whatdoestheresearchsayaboutachievementofstudentswhousecalculatortechnologiesandthosewhodonot.
InFife(Ed.
),ElectronicProceedingsoftheThirteenthAnnualInternationalConferenceonTechnologyinCollegiateMathematics.
Atlanta:AddisonWesley.
Bostrm,N.
(2014).
Superintelligence:Paths,dangers,strategies.
Oxford:OxfordUniversityPress.
Braidotti,R.
(2013).
Theposthuman.
Cambridge:PolityPress.
Brundage,M.
,Avin,S.
,Clark,J.
,Toner,H.
,Eckersley,P.
,Garfinkel,B.
,Dafoe,A.
,Scharre,P.
,Zeitzoff,T.
,Filar,B.
&Anderson,H.
(2018).
Themalicioususeofartificialintelligence:Forecasting,prevention,andmitigation.
arXivpreprintarXiv:1802.
07228.
Burt,C.
(1962).
Theconceptofconsciousness.
BritishJournalofPsychology,53(3),229–242.
Campbell,C.
(2017).
Theywilltakeourjobs!
RisingAItothreatenhumanjobsandcausemajoridentitycrisis.
SundayExpress.
Retrievedfromhttps://www.
express.
co.
uk/news/uk/846434/Robot-AI-bbc-sunday-Artificial-Intelligence-Committee-Steven-Croft-Edward-Stourton-musk.
Campbell,M.
,HoaneJr.
,A.
J.
,&Hsu,F.
H.
(2002).
Deepblue.
ArtificialIntelligence,134(1–2),57–83.
Carr,N.
(2011).
Theshallows:WhattheInternetisdoingtoourbrains.
NewYorkCity:WWNorton&Company.
Chalmers,D.
J.
(2015).
Thesingularity:Aphilosophicalanalysis.
InS.
Schneider(Ed.
),Sciencefictionandphilosophy:Fromtimetraveltosuperintelligence(pp.
171–224).
Chichester:Wiley-Blackwell.
Chouard,T.
(2016).
TheGoFiles:AIcomputerwrapsup4-1victoryagainsthumanchampion.
NatureNews.
Retrievedfromhttps://www.
nature.
com/news/the-go-files-ai-computer-wraps-up-4-1-victory-against-human-champion-1.
19575.
Cole,M.
(1990).
Culturalpsychology:Aonceandfuturediscipline.
InNebraskaSymposiumonMotivation1989:Cross-culturalperspectives.
UniversityofNebraskaPress.
Cominelli,L.
,Mazzei,D.
,&DeRossi,D.
E.
(2018).
SEAI:SocialemotionalartificialintelligencebasedonDamasio'stheoryofmind.
FrontiersinRoboticsandAI,5,6.
Copeland,B.
J.
(2014).
Turing:Pioneeroftheinformationage.
Oxford:OxfordUniversityPress.
Culkin,J.
M.
(1967).
Aschoolman'sguidetoMarshallMcLuhan.
SaturdayReview,51-53,70–72.
Damasio,A.
(1994).
Descartes'error:Emotion,reason,andthehumanbrain.
NewYork:Quill.
Damasio,A.
(2003).
LookingforSpinoza:Joy,sorrow,andthefeelingbrain.
Orlando:HarcourtInc.
Damasio,A.
(2018).
Thestrangeorderofthings.
NewYork:PantheonBooks.
Ebert,D.
D.
,Zarski,A.
C.
,Christensen,H.
,Stikkelbroek,Y.
,Cuijpers,P.
,Berking,M.
,&Riper,H.
(2015).
Internetandcomputer-basedcognitivebehavioraltherapyforanxietyanddepressioninyouth:Ameta-analysisofrandomizedcontrolledoutcometrials.
PLoSOne,10(3),e0119895.
Ferrando,F.
(2013).
Posthumanism,transhumanism,antihumanism,metahumanism,andnewmaterialisms.
Existenz,8(2),26–32.
Fong,T.
,Nourbakhsh,I.
,&Dautenhahn,K.
(2003).
Asurveyofsociallyinteractiverobots.
RoboticsandAutonomousSystems,42(3–4),143–166.
Fromm,E.
(1994).
Escapefromfreedom.
NewYork:HenryHoltandCompany.
Gillies,D.
(1996).
Artificialintelligenceandscientificmethod.
Oxford:OxfordUniversitypress.
Goldberg,E.
(2009).
Thenewexecutivebrain:Frontallobesinacomplexworld.
Oxford:OxfordUniversityPress.
Good,I.
J.
(1965).
Speculationsconcerningthefirstultraintelligentmachine.
AdvancesinComputers,6,31–88Elsevier.
76StraGoogle.
(2018).
Solveintelligence.
Useittomaketheworldabetterplace.
Retrievedfromhttps://deepmind.
com/about/.
Henkel,L.
A.
(2014).
Point-and-shootmemories:Theinfluenceoftakingphotosonmemoryforamuseumtour.
PsychologicalScience,25(2),396–402.
Hoff,T.
(2011).
Deskillingandadaptationamongprimarycarephysiciansusingtwoworkinnovations.
HealthCareManagementReview,36(4),338–348.
Kaptelinin,V.
(1996).
Computer-mediatedactivity:Functionalorgansinsocialanddevelopmentalcontexts.
InB.
A.
Nardi(Ed.
),Contextandconsciousness:Activitytheoryandhuman-computerinteraction(pp.
45–68).
Cambridge:MITPress.
Koestler,A.
(1967).
Theghostinthemachine.
NewYork:TheMacmillanCompany.
Kurzweil,R.
(2015).
Superintelligenceandsingularity.
InS.
Schneider(Ed.
),Sciencefictionandphilosophy:Fromtimetraveltosuperintelligence(pp.
146–170).
Chichester:Wiley-Blackwell.
Langner,R.
(2011).
Stuxnet:Dissectingacyberwarfareweapon.
IEEESecurityandPrivacy,9(3),49–51.
Levy,D.
(2008).
Loveandsexwithrobots:Theevolutionofhuman-sexrelationships.
NewYork:HarperPerennial.
Lockett,J.
(2017).
World'sfirstbrothelstaffedentirelybyrobotsexworkersnowlookingforinvestorstogoglobal.
TheSun.
Retrievedfromhttps://www.
thesun.
co.
uk/news/4131258/worlds-first-brothel-staffed-entirely-by-robot-sex-workers-now-looking-for-investors-to-go-global/.
Maldonato,M.
,&Valerio,P.
(2018).
ArtificialentitiesormoralagentsHowAIischanginghumanevolution.
InMultidisciplinaryapproachestoneuralcomputing(pp.
379–388).
Cham:Springer.
McAfee,A.
,Brynjolfsson,E.
,Davenport,T.
H.
,Patil,D.
J.
,&Barton,D.
(2012).
Bigdata:Themanagementrevolution.
HarvardBusinessReview,90(10),60–68.
Metzler,T.
A.
,Lewis,L.
M.
,&Pope,L.
C.
(2016).
CouldrobotsbecomeauthenticcompanionsinnursingcareNursingPhilosophy,17(1),36–48.
Mill,J.
S.
(2004).
Principlesofpoliticaleconomy.
NewYork:PrometheusBooks.
Müller,V.
C.
,&Bostrom,N.
(2014).
Futureprogressinartificialintelligence:Apollamongexperts.
AIMatters,1(1),9–11.
Nss,A.
(1999).
kologi,samfunnoglivsstil.
Oslo:Universitetsforlaget.
Norman,D.
A.
(1991).
Cognitiveartifacts.
InDesigninginteraction:Psychologyatthehuman-computerinterface(Vol.
1,pp.
17–38).
Nouchi,R.
,Taki,Y.
,Takeuchi,H.
,Hashizume,H.
,Nozawa,T.
,Kambara,T.
,Sekiguchi,A.
,Miyauchi,C.
M.
,Kotozaki,Y.
,Nouchi,H.
,&Kawashima,R.
(2013).
Braintraininggameboostsexecutivefunctions,workingmemoryandprocessingspeedintheyoungadults:Arandomizedcontrolledtrial.
PLoSOne,8(2),e55518.
Owen,A.
M.
,Hampshire,A.
,Grahn,J.
A.
,Stenton,R.
,Dajani,S.
,Burns,A.
S.
,Howard,R.
J.
,&Ballard,C.
G.
(2010).
Puttingbraintrainingtothetest.
Nature,465(7299),775.
ParoRobots.
(2018).
ParoTherapeuticRobot.
Retrievedfromhttp://www.
parorobots.
com.
Penrose,R.
(1999).
Theemperor'snewmind:Concerningcomputers,minds,andthelawsofphysics.
Oxford:OxfordPaperbacks.
Randers,J.
(2012).
2052:Aglobalforecastforthenextfortyyears.
InTheFutureinPractice:TheStateofSustainabilityLeadership.
UniversityofCambridge.
Raggio,O.
(1958).
ThemythofPrometheus:Itssurvivalandmetamorphosesuptotheeighteenthcentury.
JournaloftheWarburgandCourtauldInstitutes,21(1/2),44–62.
Scheutz,M.
(2016).
Theneedformoralcompetencyinautonomousagentarchitectures.
InFundamentalIssuesofArtificialIntelligence(pp.
517–527).
Cham:Springer.
Scheutz,M.
,&Arnold,T.
(2016).
Arewereadyforsexrobots.
InTheEleventhACM/IEEEInternationalConferenceonHumanRobotInteraction(pp.
351–358).
IEEEPress.
Schneider,M.
L.
,Donnelly,C.
A.
,Russek,S.
E.
,Baek,B.
,Pufall,M.
R.
,Hopkins,P.
F.
,Dresselhaus,P.
D.
,Benz,S.
P.
,&Rippard,W.
H.
(2018).
UltralowpowerartificialsynapsesusingnanotexturedmagneticJosephsonjunctions.
ScienceAdvances,4(1),e1701329.
Shelley,M.
W.
(2003).
Frankenstein.
NewYork:Barnes&NobleClassics.
Silver,D.
,Schrittwieser,J.
,Simonyan,K.
,Antonoglou,I.
,Huang,A.
,Guez,A.
,Hubert,T.
,Baker,L.
,Lai,M.
,Bolton,A.
,&Chen,Y.
(2017).
Masteringthegameofgowithouthumanknowledge.
Nature,550(7676),354.
Simons,D.
J.
,Boot,W.
R.
,Charness,N.
,Gathercole,S.
E.
,Chabris,C.
F.
,Hambrick,D.
Z.
,&Stine-Morrow,E.
A.
(2016).
Do"brain-training"programsworkPsychologicalScienceinthePublicInterest,17(3),103–186.
Sony.
(2018).
aibo.
Retrievedfromhttps://aibo.
sony.
jp/en/.
Storr,A.
(2005).
Solitude:Areturntotheself.
NewYork:FreePress.
Thomas,K.
(1983).
Manandthenaturalworld:ChangingattitudesinEngland1500–1800.
Middlesex:PenguinBooks.
TheGhostintheMachine77Turing,A.
M.
(2009).
Computingmachineryandintelligence.
InParsingtheTuringTest(pp.
23–65).
Dordrecht:Springer.
Turkle,S.
(2011).
Alonetogether:Whyweexpectmorefromtechnologyandlessfromeachother.
NewYork:BasicBooks.
vanNimwegen,C.
(2008).
Theparadoxoftheguideduser:Assistancecanbecounter-effective.
UtrechtUniversity.
78Stra