Some implications of computer networks for psychology PDF BF03201778.pdf

Computer networking as a subject of psychological research has varied facets process, as well as the definitions of software, hardware, and communication

Our social-effects findings regarding position relative to the mean hold on plagiarized pairs, which by definition have the same textual quality Page 12 7 of 11

[PDF] psychology bed

C) Transfer of Learning - Concept, types, educational implications Unit 4: Factors 1 1 Introduction 1 2 Meaning and Definition of Educational psychology

[PDF] Finding “Meaning” in Psychology - Psychology - Northwestern

Following a review of research on this lay theories perspective in the field of social psychology, the implications of analyzing psycho- logical meaning for other

Some implications of computer networks for psychology

Computer networking as a subject of psychological research has varied facets process, as well as the definitions of software, hardware, and communication

Behavior ResearchMethods&Instrumentation

1976, Vol.8(2),218-222

Someimplicationsofcomputernetworks

forpsychology

WALTERA.SEDELOW. JR.

DivisionofMathematicalandComputerSciences,NationalScienceFoundation,Washington, D.C.20550 and TheUniversityofKansasatLawrence, Lawrence,Kansas660Jt5 Computernetworkingas asubjectof psychological research has varied facets. including the social psychology ofscientists.the role of group"labeling" practices.and theparadigmdevelopment process, as well as the definitions of software, hardware, and communication systems and interfaces that fit the researchstyles, aspirations, and work habits specifictopsychologists.

AlthoughI have beenprofessionallyinterestedin

theconstructionof social realitythroughlanguage morethan20 years ago. itcameas aconsiderable surpriseto me to find howgermanegeneralizations from thatresearchare tocomputernetworkingefforts onbehalfof the scientificdisciplines.

Itis verycommonly the casethat.if arubricor label

may betakento apply eitherto a physicalobjector to someothercomponentor complex out of reality,there isfrequentlyastrongdispositionfor the "concretized"option to be chosen.Whenwe are trying to deal with the factors involved incomputer networks,thereis awidespreadinclinationtosupposethatthereferentfor acomputernetworkis exclusively hardware.We live in a world in which the physical andtangiblehave decidedculturalpriority.Just as people in local schooldistrictshave often been more agreeabletospendingmoney forbuildingsthanfor supportofteaching -perse, so in the world of computernetworks many find it much more comfortableto concern themselves exclusively with the physical network,ratherthanwith the"full system" afunctioningnetworkimplies. Whilethere arelabelingtheory issues atstakein suchbehavior,it is also a verypragmaticproblemas well as one of theoreticalinterestto thesocial/behavioralscientist.

Anunderstandingof networksthatincludesthe

hardware,networksoftware, and special software and hardwarerequiredby specificapplicationsfoci for researchto beconductedon thenetwork, the modalitiesofrelated humancommunication,and the varioushuman/organizational/socialdimensionsof a totalnetwork'sfunctioning(as well as the systemic integration ofallthosesubsystems)is not easily gained.Apparently,partlyowing to socialinteraction dimensionsof working with people inother disciplines.it is often extremelydifficultto get the specialists who havesomethingtocontributeto an understandingof networks to conceive of network problemsin a"fullsystem"perspective.Those whothinkexclusively intermsof the physical network as the whole network do not include a concern for the otherfacets of anetwork.Even whenthereis some success inbringing thoseotherfacets within the purview ofconcern,aninterestin atotalsystems frameworkbuildingtoward onecoherentbody of theory and research(includinghuman,software, and communicationsandhardwarebehaviorin a single paradigm)is not an easy goal to achieve. The difficulty is, of course, verymuchamatterof calling resources to achievethem.

Theexperiencehere withcomputernetworks is also

verycharacteristicofotherexperienceswith man-machinesystems;i.e.,it is very like manyother issues andproblemsthatflow from our failure to develop anunderstanding of men andmachinesin onecoherentresearchframe.The absence of a genuine,comprehensivetheory ofman-machine systems does seem aremarkabledeficit in a society which. for at least 200 years, has been sodependent on advanced technology and science. One senses a flight from a towering reality for which, in the longer perspective ofoccidentalculturalhistory, the society was not prepared. Neither technology norsciencehas been as central to the cultural concerns of our world astheirimpactson our lives would seem towarrant; after,say,halfamillenium,highereducationhas still not come to terms with itsimportance. Therefore,as one seeks to develop research efforts relatingto anunderstandingof"full system" networks,the firstmajorbarrierto overcome is a too narrowdefinitionof the focus ofattentionfor network study. Oncethathurdleispassed.thereremains anotherofparticularinterestto the social scientist. In brief, onefeatureof the issues thatresearch managementconfrontsis theimplicationof which categories,withreferenceto thehumancomponents ofnetworks,are to betreatedin aggre gated form. and which indisaggregatedform, as well as what labels are to be used for theseaggregations anddisaggregations.For 218
morethanadecade,I haveobservedthejockeying of differenttypes ofacademicsandadministrators.The manifestin adefinitionof reality whichtreats membersofuniversityadministrationsin dis aggregatedform in thestudyof socialdimensionsof networkingfor the scientificcommunity.For example.computercenterdirectorshave been disaggregatedout of thetotaladministrativecadre.

Seniorcentraladministration

personnel-university cal1ed "users"has grownup-s-significantly,ahighly aggregatedcategory.Thesetermsare used so consistently andare such forcefulsurrogatesfor stronginterests thatrenderingthemproblematicis a verydifficulttask,perhapsnot so muchconceptually for theindividualsocialscientist, butcertainlyin generalfor theaffectedpublic.

Itwouldtakea vast

make a good try at swinging thevocabularyaround, sothat,withreferenceto networks for science, at a minimumone would have theaggregatedcategoriesof "scientists"on the one sideandagroupof"servers" on theother.In asocialworldofcomputercenter directors.thelikelihoodofself-designationsas servers seems extremely unlikely. Similarly. eventhoughthe word"scientist"should befamiliartocomputer seners isextremelyunlikely. Aconsiderabletangleof scientistsbydisciplinealongsidea setofservers, computercenterpersonnel.andothers.If you find it less label.administratorsas servers. you have a kmd ofinformalcalibrationto theproblemsinvolved ingettingpeople tothinkaboutdefiningnetworks from auser'sperspectiverather thanthe so-called managerialperspectiveof thecomputercenterand partsof the universityadministrationwhichtendsto hepreponderant.

In thissituationthereareclearanalogieswith the

world of universitylibraries.Computercenter personnel, those in both universities and large research facilities, come toconstitutea group of such numbers and expense that they readily seek, and in some measure succeed in obtaining, a significant autonomy-anautonomyvis-a-vistheneeds ofgroups theirexistencein the first place.Academiclibrarians are often verymuchthesameway. I amremindedin this connection of a story of a well-known research laboratoryon the East Coast whose directing scientists decided that they would never hire professionally trainedlibrarians.

Thathas come to seem asomewhat

dubiousdecision inrecentyears. But for a long time it

NETWORKSFORPSYCHOLOGY219

served toinsure thatthelibrarianswould not be developingpractices andproceduresthatwere consistentwithwhatthey hadbeentaughtinlibrary school (asdistinctfrom serving the objectives of scientists,for whosebenefitthelibrarywascalledinto being). Such often is the case with computer networks, even when they are meant to serve, at least ostensibly, the interests of particular groups of scientists. Very frequently the computer professionals in the organization in effect prescribe not only the facilities in their general design, but also their day-to-day realization in terms congenial to themselves.

Sometimes so topsy-turvy does this reality become

thatone canpointtoscientists(seniorscientists,at that)who arediffidentaboutexpressingtheircurrent sense ofcomputerrequirements,lest indoingso they offendcomputerprofessionalson whom they are dependentfor acertaingrade andcharacterof day-by-dayservice.AlthoughI wassurprisedatthe extentto which the 1975AmericanSociological Asso ciationsessions on thefunctioningofcomputer centersinuniversitieswere focused on thesocial powerrelationshipsamongthe people involved, the generalideacouldscarcely fail to seem likely to one who has looked closely at thecurrentproblemsin bringingaboutcomputercapability thatdirectly servesresearch andteachingneeds.

Ingeneral,one would

startwith theperceived whosharecomplementarycomputerneeds, supplementedbyspecificationsfrom computer perhapspointabetterwayto theachievementof goals set by theacademicdisciplinesto beserved.Thenone would want to define the "user"interface,including bothhardwareandsoftware,in suchfashion thatit was generallycongenialfor thescientists,rightdown todetailing thelocationand types ofterminals.Only afterthespecificationof goals andcongenialityof interface would it seem appropriate to go on to the question of what was to be embedded insoftware and what inhardware andwhichoptionswere to betaken to achieve thoseconfigurationalrequirements. What onespeaksof here is anetworkin which thescientist and his needs haveprimacy;thenetworkindesign andoperationissubordinatedto hisconvenienceand hisaspirations.Thereis anemergingset of constituenciesin thiscountryforspecialized,and probablyvirtual,network access. To date, amongthose who have hadthegreatestopportunities to developtheirrequirementsforspecializednetwork computing.

We look to psychology asanotherscience within

whichnetworkneeds may be wellspecified. The computationalmaturityof thedisciplineandwhat we hope can belearnedfrom thespecialconcernsthat

220SEDELOW

psychologistswouldbringtothedefinitionof a networkcould beappliedbygeneralizationto the needsofotherscientists.

Perhapsin

partbecauseofthenatureoftheirown of the poorquality,especiallytheunduecomplexity,of thescientist(user)interfaceswithpresent computer systems(much less withnetworksystems).There seems to be aninterest amongsomepsychologistsin undertaking theproductionofaninterfacethatwould be far morecongenial.Fromapsychologist's professionalorientationwould come a broader definitionof theinterfaceproblem.Adefinitionofthe interfaceproblemcongenialto apsychologistmight includetheterminals(from a humanfactors standpoint)andpropertiesof theprogramming languages, thedatastructures,theoperatingsystems, andcertainlythe accessrules,i.e.,whatevera scientistwouldencounterwhen he attemptedtomake use ofcomputingpower over anetwork.Psychologists mightwish to give acertainpriorityto theintegration (to the"systematicness")possibleinresearchon the interfaceto provide for abetterimpedancematch with thepropertiesof the humanbeingasinformation processor.

Onemightexpect

thatpsychologists,maybeeven more thanothersocialscientists.would beconcerned with thesocialproblemattributesthatafailureto

Computersinnetworksseem to be moreintimidating

to thegeneralpublic than(objectionablethoughthey sometimesseem)stand-alonecomputers. If accessibility to networkcomputersis notcomfortable for people who are notcomputerprofessionals.a deeper andmorewidespreadgeneralfear of evolve. Such a fearmightbeevaluatedas highly dysfunctional,given our currentsocial andecological condition.

Itmay beargued,then,thatresearchon

thecomputernetworkinterface(to makethat rendera social service inproducingagenerallymore congenialcomputerinterfacefor thenonprofessional.

Onemotivethatpsychologistsprobablyhave for

attemptingresearch thatcould lead toimproved the more to thepointwithreferencetonetworking, capabilitiesavailablein any givenpresentconsortium ofcomputerusers are too haphazardandpiecemeal, i.e.,toosegmentalfor the fullrangeof computer-basedresourcesto whichpsychologists mightwish access. Forexample,psychologistsmay well beinterestedinnetworks,not only for access to majorempiricalresearch databasesand/ortosubtle analyticalprograms and/ortocomplexsimulation andmodelingresources, butalso forsimplerforms ofinformationexchangeandinterpersonalprofessional communication.

J.C.R.Lickliderhelpedpopularizea

notion ofresearchstimulusandinteractionunderthe aspectof anon-lineintellectualcommunityis access to its own scientific andintellectualaccomplishments. more have the physicists) to use thecomputerin conjunctionwith themanagementofbibliographical informationrelatingtoresearch.Onecapabilityof a virtualnetworkfor psychologists thatwould be of interesttootherresearchareaswould be a more over acomputernetwork,includingaccess tobothfull

Theinterestof manypsychologistsinnational

networkingcertainlyalsoincludesnetworkaccess to artificialintelligenceprojects; thatinterestisshared bymanycomputerscientistswho work onmachine intelligence.Artificialintelligence(AI) projects characteristicallyarebothexpensive anddifficultto develop andmaintain,as istrueof variouskindsof psychologicalmodeling.Many of theseelaboratedAI effortsinvolvecomponentsnotamenableto being transported,including special hardware-dependent developments, speciallanguages,and operating systems. Currently it could be contended that the development of AI research is held back by the psychological community's inability to gain easy access to extant AI systems. When such accessis available, it must be provided in a way that protects the integrity of each AI system as it operates at any givensite. At the same time, the system must be availableto those who wish to critically inspect details of it with a view to independent validation.

Ithas been

contendedthat the specifically scientific character of psychology is in some measure being prejudiced by the isolation within which much artificial intelligence and computermodeling of psychological functions is done. That same issue emerges, of course, with respect to other disciplines' network needs; for instance, it has been argued that greateraccessibility to the details of certain general circulation models in the field of meterology might havecontributedtotheir improvement.

Thus,amajor

argumentfornetworkingin a num ber ofareasofresearchspecializationis that,given computer-basedresearchof a highly elab oratedsort,thetraditionalsoundmecha nisms in science forcorrectionthroughcritical access to what is going on at specialized sites is not readily available.

Ithas even been said that at some

major research sites the researcher's conduct begins to somewhat approximate that of a hieratic"priesthood," rather than to approximate the normative ideal of a scientist. No matter how sound theirintentions,when network accessis not available,their scientific peers are notina position to examine critically what otherwise is presented tothem,necessarily, in anabbreviated, interpretivemode.

Perhaps

italsoshouldbeemphasizedthatthe increasingmathematizationofpsychologyis an inherentlypowerful argumentforpsychologists' interestinnetworking.

Thatisthecase inpart

mathematicalsoftwareis not likely to berealizedin manyinstancesexcept throughthe use ofthenetwork and over anetwork.Statisticallearningtheory, psychologyonsound mathematics,soundstatistics, soundalgorithms, andtheiradequaterealizationin computersoftware.

Not somanyyears agotherewas far moreoptimism

cannedroutinesinprocessingresearch data.There have always beennonstandardexperimentsand deemedacceptable,by andlarge.Now we are not only moreaware thanpreviously ofinadequaciesin such so ftwarebutare alsocognizantof thefurther problemsthatare going to be posed by themareready computersatresearchsites.

Thelimitationswe know of inavailablesoftware

will grow to be a moreseriousproblemwhen thatsoft ware is runon agreatvarietyofminicomputers.Net works areneededforsomeaspectsofquotesdbs_dbs17.pdfusesText_23

[PDF] Some implications of computer networks for psychology

[PDF] Broader implications: sociology/social psychology - Cornell CS