Remote Communica-on in Wilderness Search and Rescue ...hcitang.org/papers/2020-group2020-wsar.pdf · Making information available in multiple formats (e.g., audio, pictures, videos)

PACMonHuman-ComputerInteraction,Vol.4,No.GROUP,Article10,Publicationdate:January2020.

Remote Communica-on in Wilderness Search and Rescue: Implica-ons for the Design of Emergency Distributed-Collabora-on Tools for Network-Sparse Environments

BRENNAN JONES, DepartmentofComputerScience,UniversityofCalgary,Canada ANTHONY TANG, FacultyofInformation,UniversityofToronto,Canada CARMAN NEUSTAEDTER, SchoolofInteractiveArts&Technology,SimonFraserUniversity,Canada

Wildernesssearchandrescue(WSAR)requirescarefulcommunicationbetweenworkersindifferentlocations. To understand the contexts fromwhichWSARworkers communicate and the challengestheyface,weinterviewedWSARworkersandobservedamock-WSARscenario.OurWindingsillustratethatWSARworkers face challenges inmaintaining a sharedmentalmodel. This is primarily donethrough distributed communication using two-way radios and cell phones for text and photomessaging; yet both implicit and explicit communication suffer.WSARworkers sendmessages forvariousreasonsandsharedifferenttypesofinformationwithvaryinglevelsofurgency.Thiswarrantstheuseofmultiple communicationmodalities and information streams.However, bringing inmoremodalitiesintroducestheriskofinformationoverload,andthusWSARworkerstodaystillprimarilycommunicateremotelyviatheradio.Ourworkdemonstratesopportunitiesfortechnologytoprovideimplicitcommunicationandawarenessremotely,andtohelpteamsmaintainasharedmentalmodeleven when synchronous realtime communication is sparse. Furthermore, technology should bedesigned to bring togethermultiple streams of information and communicationwhilemaking surethattheyarepresentedinwaysthataidWSARworkersratherthanoverwhelmingthem.

CCSConcepts:•Human-centredcomputing→Collaborativeandsocialcomputing;Empiricalstudiesincollaborativeandsocialcomputing

KEYWORDS Searchandrescue,outdoors,teamcommunication,distributedcollaboration,awarenessACMReferenceformat:BrennanJones,AnthonyTang,andCarmanNeustaedter.2020.RemoteCommunicationinWildernessSearch and Rescue: Implications for the Design of Emergency Distributed-Collaboration Tools forNetwork-Sparse Environments. In Proceedings of the ACM on Human-Computer Interaction, Vol. 4,GROUP,Article10(January2020),28pages.https://doi.org/10.1145/3375190

1 INTRODUCTION

Wildernesssearchandrescue(WSAR)involvesthesearchforandextractionofoneormorelostpeople(e.g.,hikers,skiers)fromawildernessarea.WSARisatime-criticaloperation

Permission tomake digital or hard copies of all or part of this work for personal or classroom use is grantedwithout feeprovidedthatcopiesarenotmadeordistributedforprofitorcommercialadvantageandthatcopiesbearthisnoticeandthefullcitationonthefirstpage.Copyrightsforcomponentsofthisworkownedbyothersthantheauthor(s)mustbehonored.Abstractingwithcredit ispermitted.Tocopyotherwise,orrepublish, topostonservers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions [email protected]/2020/January–10$15.00Copyrightisheldbytheowner/author(s).PublicationrightslicensedtoACM.https://doi.org/10.1145/3375190

10

10:2 Brennan Jones et al.


thatrequirescarefulcommunicationandcollaborationbetweenmanyworkerswhoarespreadoutinvariouslocations,includingatacommandpost(Fig.1,left)andinthefield(Fig.1,right).However,operationsusuallytakeplaceinareaswithpoorcellularcoveragewheregeographicfeaturessuchasmountainsandvalleyscanblockradiosignals,makingconsistentrealtimecommunicationsbetweenallteammembersdifficult.Teammembersdonotalwayshavedirectcommunicationlinkswitheachother,andsometimesfieldteamsspendlongperiodsoftimebeingunabletocontactanyoneelseintheoperationatall.Inthiswork,westudyhowthisaffectsremotecollaborationinWSAR.Researchershaveexploredtechnologiesforplanningandcollaborationinsearchand

rescue(SAR)[1,56],aswellasotheremergencydomainssuchasfirefighting[46,64],avalancherescue[21,22],andcrisisresponse[2,4,16,17,62,65].Inallofthesedomains,workersneedtobuildandmaintainanuancedsharedmentalmodel[10]oftheprogressmadeintheoperationandthestatusofteams,workers,andequipment;andprojectthatknowledgetounderstandwhatwillhappennextandmakefutureplans.Thissharedmentalmodeliscreatedandmaintainedthroughsharedsituationalawareness(SSA)[25],teamawareness[12],anddistributedcognition[57].Inourwork,wefocusonremotecollaborationinlarge-scaleWSARoperations,mainlyonthesearchphase,wheremultipleteamsaredeployedinthewildernesstocoveralargesearcharealookingforalostperson.Despitetherelatedliterature,wedonotknowwhatchallengesWSARworkersfaceasaresultofthewildernessandalackofrealtimecommunication.WeexplorethisthroughaCSCWandtechnology-designperspectiveinordertoinformthedesignoftechnologiestosupportdistributedcollaborationbetweenrespondersinhigh-stakesemergencysituationsinwhichrealtimecommunicationandawarenessarenotalwayspossible.Inourwork,wefocusonaWestern-CanadianperspectiveofWSAR,whichisanenvironmentcontainingmanyforests,mountains,rivers,andlakes.

Fig.1.WildernessSARinvolvescarefulcommunication,coordination,andinformationsharing

betweenmanagersatacommandpost(left)andsearchersinthefield(right).

Totackleourresearchgoals,weinterviewed13WSARvolunteers,includingfourmanagersandfivefieldteamleaders,andobservedafulldayWSARmock-searchtrainingactivity.OurfindingsindicatethatWSARworkerscommunicateremotelyforvariousreasonsandsharevariedinformationwithdifferentlevelsofurgency,warrantingtheuseofmultiplecommunicationmodalities(e.g.,voice,text,photos,videos,GPSlocations,etc.).However,communicationsbetweenthefieldandcommandtodaystilllargelyhappenviaasingleinformationstream:theradio.Thisreducestheopportunitiesforimplicitcommunicationandawarenessbetweenfieldworkersandcommand,whichcanleadtodifficultiesmaintainingteamawarenessandcognition[12,27,38],especiallyinlargesearchesthattakeplaceoveralongperiodoftime.

Remote Communication in Wilderness Search and Rescue 10:3


ThesefindingsdemonstratethatWSARworkers,teams,andagenciescouldbenefitfrommultiplechannelsofcommunicationandawareness,bothsynchronousandasynchronous,andeachtailoredtotheirspecificpurposesandneeds.Makinginformationavailableinmultipleformats(e.g.,audio,pictures,videos)frommultiplechannelscouldhelpbettersupportbuildingasharedmentalmodel,asitcouldallowfordifferentwaystointerpretandunderstandwhatishappening.However,thesestreamsofinformationshouldbepresentedinwaysthataidWSARworkersratherthanoverwhelmingthem,presentinginformationthatisactionable.Furthermore,giventhattheworkofCommandisheavilyreliantoninformationcominginfromfieldteams,designshouldfocusmoreoncommunicationandawarenessbetweenthefieldandCommand.Thisworkmakesthefollowingcontributions:(1)astudyrevealinginsightsaboutwildernessSARcommunicationcontextsandgoals,commoncommunicationchallenges,anddesignneedsforeachcontext;and(2)discussionofpotentialdesignsolutionsandrecommendationsforfuturework.

2 WILDERNESS SEARCH AND RESCUE

InatypicalCanadianWSARscenario,anagencyfromanearbycommunityiscalledtorespondtoareportofamissingperson[42,43](oftencalledthesubjectofthesearch).TheSARmanagerondutyfortheagencysendsacallouttovolunteermembersoftheagencytomeetataspecificlocationnearthesearcharea.Whenthemembersarrive,theyformoneormorefieldteamsthatsearchassignedpartsofthesearcharea.Themanagerandtheirmanagementteamworkfromacommandpost(amobile-officetrailerparkednearthesearcharea)inadvanceofthemembers’arrivals.Here,theyplanspecificsearchassignments(e.g.,whichareastosearch,whichsearchtechniquestouse,etc.)andgiveassignmentsbasedonthemembersavailableandtheirskillsandbackgrounds.Searchplansaredrawnupbasedontheinformationtheagencyhasonhandaboutthesubject,suchastheirlastknowlocation,theirdirectionoftravel,andotherknowledgeabouttheirbehaviour.Probabilitymapsarecreatedbasedonthisinformationusingstatisticsandyears-worthofdatadescribinghowlostpeopletypicallybehaveinthewilderness[48].TheCommandteamisresponsibleforcoordinatingalargenumberofpeople,keepingtrackofvastamountsofinformation,andusingthatinformationtoconstructfutureplansofaction.Formoreseriousincidents(e.g.,wheremultiplepeoplearelost),multipleagenciesand

theirmembersarecalledtomutuallyassisttheagencyreceivingtheinitialcallout.Whenthishappens,eachagencyhasitsownsecondarymanagerandmanagementteam,whilethemanagerfromtheteamreceivingtheinitialcalloutremainstheprimaryhigher-levelmanager.Eachagencythensendsoutitsownfieldteamsbasedontheadviceoftheprimarymanagementteam.Fieldteamstraverseassignedlocations,searchingforthelostsubject(s)throughcareful

listeningandobservation.Searchtechniquescanrangefromaprecisesearch,inwhichasmallerareaiscoveredinalotofdepth,toabroadersearch,inwhichalargerareaiscoveredinashortamountoftimebutwithlessprecision.Workersarespeciallytrainedinnavigatingtoughenvironmentssuchasrapidrivercrossings,steepmountainclimbs,anddeepsnow.FieldteamsreporttoCommand,viatwo-wayradiocommunications,anyinformationthatmaybeimportantinfiguringoutwherethesubjectis(e.g.,footprints,objectsleftbehind).Lastly,ifateamfindsthesubject,theyreportthisbacktoCommandviatheradioandbegintheprocessofsafelyextractingthemandmovingthemtosafety.WemainlyfocusonthesearchaspectofWSAR.Assoonasthesubjectisfound,the

dynamicchangesfrommainlyinformationcollectionandplanning(akintopuzzlesolving),torecoveryandensuringthesafetyofthesubject(s).Additionally,duringalargesearch,teamsarescatteredindifferentlocationsandCommandneedstokeeptrackofallofthem;



whereasinarescue-onlyoperation,ifthelocationofthesubjectisalreadyknown,Commandtypicallyonlyhastosendoneteamouttoextractthem.Duringsuchoperations,thissingleteamhasfargreaterabilitytocommunicatewithoneanotherviaeasiermeans;e.g.,yellingatoneanother,usingradiosthatdonotfacereceptionissuesduetoworkersbeingclosertogether,etc.Ontheotherhand,alargesearchoperationthatturnsintoarescueoperationisdifferent,inthatCommandstillneedstodealwithalloftheotherteamsoncethesubjecthasbeenfound.Theyeitherhavetogivethemrolestoplayinthetreatmentandextractionofthesubject,recallthembacktobase,ortellthemtobeonstandby.Iftherearemultiplesubjectstoextractindifferentlocations,thenitissimilarlycomplicated.WhileweareinterestedinavarietyofWSARoperations,bothsearchandrescue,weputthebulkofourfocusonoperationsthatstartaslargesearches,astheyarecomplicatedresponsesinvolvingmultiplepersonnelandalotofresources,inanactivitythatisakintodistributedcollaborativeproblemsolving.Wedothiswiththeideathatourfindingsandinsightscouldalsobeappliedtosimplerresponsesinvolvingfewerpeopleandlessresources.

3 RELATED WORK

3.1 Team Cogni-on and Awareness

WSAR,especiallylargeoperations,requiresalotofstructuredteamwork,collaboration,andcoordinationacrossdistances.Earlyresearchhaspromotedtheideathatasharedmentalmodel[10],orsharedunderstandingandawarenessoftheactivityandthethingsthatcompriseit,helpsmaintainstrongteamcognition[5,10,51,53].Asharedmentalmodelisimportantforlargecollaborativeactivitiesinvolvingmanypeopleandresources,includingWSARresponses.Thus,workprocedures,thetechnologyinuse,andWSARworkers’behavioursaroundsuchtechnologiesshouldbeinserviceofsupportingasharedmentalmodel.Tosupportasharedmentalmodel,itisalsoessentialtosupportteamcognition[27,38]acrossdistances.TeamcognitionhasbeenstudiedextensivelyinCSCWresearch[3,5,10,12,24,27,33,38,53],andisthesharedknowledgeandawarenessofteammembers,workprocesses,tasks,andtheworkspace;andtheabilitytocoordinateandacttogetherbasedonthatsharedknowledge.Itisoftensupportedbydistributedcognition[38,39,58],whichistheideathatknowledgeisdistributedintheworkspaceandtheartifactscontainedwithinit.Awareness,whichisimportantforteamcognition[33],isdifficulttoestablishandmaintainoverdistance[12,25,26,34].InWSAR,membersgenerallygothroughthesametrainingandusethesamelanguageandcommunicationprotocolswhileworkingtogetherandpassingalonginformation[42,43],althoughthesestandardscouldvaryslightlyacrossjurisdictions.TheIncidentCommandSystem(ICS)[11,35,69],nowpartoftheUSNationalIncidentManagementSystem(NIMS)[66,69],isastandardemergencyprotocoldevelopedandusedextensivelyinemergencymanagementoverthelastfewdecadesaroundtheworld,anditisthecentralunderpinningofhowrolesandtasksareassignedandhowinformationflowsinWSARinCanada[42,43].ICSisacorepartofthesharedmentalmodelofWSARandotheremergencydomains,giventhatithelpsemergencyresponderscommunicateandcollaborateusingsharedlanguageandprotocolstheyagreeon[6].WSARstandardsandprotocolsallowmultipleagenciestousethesamesharedlanguagewhen,forexample,oneneedstoprovidemutualaidtoanotherperson[42,43].Inaddition,ICSallowsemergencyrespondersfrommultipledomains(e.g.,EMS,firefighters,police,etc.)tocollaborateusingacommonlanguage[66,69].MorerecentresearchinCSCWhasarguedthatsharedmentalmodelsarenotalways

necessaryforsuccessfulcollaborationduringteamactivities,ascollaboratorsdonotnecessarilyneedtohavetheexactsameknowledgetocompleteatasktogether[12].For



example,ifworkisdecoupled[59]andeachteammemberisplayingadistinctroleortakingonauniquetask(buteachtaskandrolebuildsuptoformthelargeractivity),theneachteammemberwillhaveuniqueknowledgeandauniqueperspective.Inthiscase,sharingknowledgemaynotalwaysbebeneficial,andinsomecasesifitisirrelevantknowledge,itmaybedistractingtothetaskathand[12].Rather,justhavingasharedsenseoftheteam’sgoalsandobjectivesandawarenessandunderstandingofwhoontheteamknowswhatandwhateveryone’scapabilitiesaremaybesufficientenough.Asaresult,itisalsousefultounderstandothertheoriesthatcancontributetoanunderstandingofhowteamcollaborationcanbesuccessful.Theseincludeteamawareness[12],situationawareness[25],andworkspaceawareness[32,34].Teamawarenessisone'sawarenessandunderstandingofthepresence,activities,andcharacteristicsoftheirteammembers,aswellasthelargermakeupoftheteam[12].Itinvolvesanunderstandinggroundedintheteam’sculture,reachingcommonground[18],andunderstandingone’sroleasanindividualwithintheteam[12].Situationawarenessistheabilitytoperceiveandunderstandone'sowncurrentsituationandusethatunderstandingtomakeproperdecisionsandprojectone'sfuturestatus[25].Workspaceawarenessisanunderstandingofwhois'in'theworkspaceandwhatishappeningandhashappenedwithinitstemporalandphysicalbounds[34].Itisanunderstandingofwhoisandhasbeenpresentinthe'workspace'(oractivityathand),whattheyaredoingwithinit,whattheircontributionshavebeen,andhowthestateoftheworkspace,includingtheartifactsandinformationwithinit,havechanged[34].Collaborativecouplingisthedegreetowhichteammembers’workisreliantonone

another[59].Collaborativeworkistightlycoupledifteammembersrelyonotherteammemberstocompletetheirworkandlooselycouplediftheydonot[53,59].Ratherthanworkbeingexplicitlytightlyorlooselycoupled,mostcollaborativeworkactuallyfallswithinaspectrumroughlysomewherebetweentightlyandlooselycoupled[34].Insomecases,collaboratorsswitchbackandforthbetweenindividualandcollaborativework[24,29].Whiledoingso,itoftenhelpsforworkerstobeawareoftheirteammembers’activities,asithelpsthemunderstandwhentheyareavailableforcollaborationandassistance[36].Inourwork,weaimtounderstandhowWSARworkersandteams,throughtheirdomain-

specifictraining,trytobuildandmaintainteamcognition,awareness,andasharedmentalmodelinlargesearchoperations.Wefocusonthisthroughoutourfindings.Wealsofocusonwheretheirworkpractices,aswellasthetechnologiestheyuse,arefailingthem.BeyondjustWSAR,someofourfindingscouldalsoapplytothecircumstancestothebasicsituationthatWSARintroduces;i.e.,intermittentandunreliableconnectivity,notonlybetweenthefieldandCommand,butalsobetweenthedifferentfieldteams.Thesefindingscouldcontributenuanceintohowweshouldthinkaboutsupportingsharedmentalmodels,awareness,andteamcognitioninnetwork-sparseconditions.

3.2 Search and Rescue

SARinvariouscontexts(e.g.,urban,wilderness)hasbeenextensivelystudiedbyresearchersinHCI,Human-RobotInteraction(HRI)[31,54],andCSCW[1,21,22,56].HRIresearchershavestudiedtheuseofrobotssuchasdrones[19,23,30,41,52,70]andlandrovers[8,13,47,60]forSAR,andfoundthatcontrolinterfacesforsuchrobotsshouldbedesignedtoprovideasmuchspatialandsituationawarenessofthesurroundingenvironmentaspossible.RobotshavebeenusedinrealSARincidentssinceatleastthe9/11attacks[13].Manyrobotshavecapabilitiesthatgobeyondthoseofhumansandareabletosearchthroughareasorfromperspectivesthathumanscannoteverreachbythemselves.Theobviousexampleisdrones[19,23,30,41,52,70],whichcaninspectlargeswathsof



wildernessenvironmentsfromoverheadperspectivesinarelativelyinexpensiveway(comparedtohelicoptersandairplanes)[19,30].CooperandGoodrich[19]exploredthedesignofinterfacesforWSARfieldworkersto

controlandviewinformationfromadrone.Throughtheirexplorations,theauthorsfoundthatcontrolandinformationpresentationshouldbesimplified,astoomuchinformationandcontrolcouldincreasethetimeandoverheadincompletingatask.Thehigh-stressnatureoftheoperationcanfurtherexacerbatethis.Desjardinsetal.[22]foundthesamethingintheirexplorationsofco-locatedcollaborationaroundbeaconsduringavalancherescues.Bothoftheseworksalsofoundvalueinpresentinginformationinrelationtothelayoutandphysicalmakeupoftheoutdoorspace.Furthermore,Alharthietal.[1]foundthatasizeableamountofplanninganddiscussionofplansiscentredaroundmaps.Theyrecommendmixingindividualandteammaps,mixingdigitalandphysicalmaps,andprovidingtheabilitytomodifyandpopulatemaps.Weareinterestedinseeingifsuchdesignpracticescouldbeextendedtoremotecollaborationactivities,suchasinformationsharingbetweengroundsearchteamsandCommand.WSARplanninginvolvesbuildingprobabilitymapsbasedonthecharacteristicsofthe

lostpersonusingstatisticsgroundedinyears-worthofdatadescribinghowlostpeopletypicallybehaveinthewilderness[48].Planningmapsindicatethelastknownpositionofthesubjectaswellascolouredzonesindicatingtheprobabilitythatthesubjectwouldbefoundinthatparticulararea[42,43,48,49].Fromtheseprobabilitymaps,SARmanagersdrawoutsearchpathsanddefinethesearchtechniquesthatthefieldteamswouldcarryout[43].Inmanycases,thesearchareagrowslargerbytheminute[42,43,48,49],soWSARworkersmustrespondquicklyandefficientlytoincreasetheirchancesoffindingthesubjectalive.Searchplansareoftenconstructedthroughprovenlost-personbehaviourmodelssuchasthosefromKoester[48]andKoopman[49].Therealsoexistcomputerprogramsandmobileapps(e.g.,[71])thathelpSARteamsbuildandfollowsearchplansbasedonprovenWSARmodels.ThisliteraturetellsusabitaboutthenatureofWSAR.Ittellsusthatmanagersand

plannersneedtokeeptrackofwhathasandhasnotbeensearched,whereeveryoneislocated,howresourcesarebeingallocated,andvariousotherthings;andusethattoplanfutureactionsbasedonprovenWSARprobabilitymodels.Theyneedstrongteamcognition,awareness,andasharedmentalmodelforthis.Wewanttounderstandhowtheygettheinformationthattheyneedfortheseactionswhiledistributedinnetwork-sparseenvironments.Furthermore,weareinterestedinunderstandingthenuancesofhowWSARremotecollaborationaroundtheinformationcollectedandtheuseofprobabilitymodelstakesplace,andhowitcanbebettersupportedwithnewertechnologies.

3.3 Emergency Response and Control Rooms

Inotherteamsituationssuchasfirefighting[46,64],thereareanumberofworkers‘ontheground’respondingtoanemergency,communicatingcarefullywitheachother,andansweringtoanIncidentCommander(themainpersoninchargeoftheincidentresponse[69])atabase(e.g.,acommandpostorafireengine)[42,64].Thereisbothexplicit[28]andimplicit[64]communicationtakingplace,andworkershavetobehighlytrainedincommunicating[28,42],coordinating[42,64],andusingtheradio[28,42].ToupsandKerne[64]foundthatfirefightersuseimplicitcoordination,i.e.,communicatingin-personviawordsandnon-verbalgestures,andseeingeachother’sactionsandintentions,toplanandguidetheiractionsasateam.InlargeWSARoperations,workersandteamsaredistributedacrosslargedistances,thusreducingopportunitiesforface-to-faceinteractionsthatonemightseeinfirefighting.WeexplorehowWSARworkerstrytoperformsimilar



coordinationactionsasfirefightersyetoverdistancestoday,whatroletheseactionsplayintheoveralloperation,andwhatopportunitiesexistfortechnologytoprovidethissupportinWSAR.Todates,studieshavenotuncoveredthisknowledge.Onemajorchallengeinemergencysituationsis“howdoyougettherightinformationto

therightpeopleattherighttime?”Actionabilityreferstotheideathatcertaininformationisrelevant,oractionable,toacertainpersondependingontheroletheyplay,thecontexttheyarein,andthetime[67].Informationthatisusefultosomepeopleincertaincontextsandtimesmaybeuselesstoothers,atdifferenttimes,and/orindifferentcontexts[67].InWSAR,Commandhasalotofinformation,mostlycomingfromfieldteams,butoccasionallyfromexternalsourcessuchaspolice,thevictim’sfamily,and/orweatheragencies[42,43].Thisinformationisaggregated,processed,andmadesenseofovertime.Commandworkersusetherawinformationtothenprocessnewinformationsuchastaskassignments,pathstotraverse,andprobabilitycharts(i.e.,theprobabilitythatthelostsubjectwillbefoundinacertainarea)[42,43,48].Thisinformationthenneedstobepassedontotherelevantworkersinthefieldattherighttime.Relevanceinthiscasecoulddependon,forexample,one’srole,taskassignment,skills,searchstrategy,andlocationinthefield.Todate,researchhasnotexploredhowWSARcommandworkersdeterminewhichinformationisactionabletowhichpeople,howcommandworkerspassonthisinformationtotherightpeople,andhowthiswholeprocesscanbeimprovedand/orexpedited;thisisourfocus.ThecollectiveactionsofWSARworkersmaybesimilartothecollaborativecontrol-room

environmentsthathavebeenstudiedinpreviouswork—forexample,metro-systemcontrolrooms[37],space-shuttlemission-controlcentres[55],andair-trafficcontrolrooms[50].Implicitandconsequentialcommunicationarecommoninthesetypesofsettings.Thesestudieshaveshownthatbeingabletoconstantlyobserve[37],hear[55],andreadtheintentionsof[37]colleagueshelpsinproperplanninganddecisionmakingforcomplexgroupactivitiesrequiringhighamountsoffocus,collaboration,andcoordination.WhilethereiscollectiveactioninWSAR,workersinlargepartdonotoperateasasinglecollective.Rather,becausetheyarespreadoutandmoreoftenthannotdisconnectedfromeachother,teamsareoperatingsemi-autonomouslyanddonotnecessarilyknowwhateveryteamhasseen,norhowthatrelatestotheirsituation.OurfocusisonhowthisaffectsWSARdistributedcollaboration,andhowwecantacklethisthroughtechnologydesign.

4 STUDY METHOD TheresearchgapweaimtofillisanunderstandingofthechallengesthatWSARpersonnelfaceintryingtobuildandmaintainasharedmentalmodelwhilemovingbetweenvaryingstatesofradioconnectivity.Thiscouldalsocontributetoahigher-levelgoalofattainingabetterunderstandingofthenatureofcollaborativeworkinnetwork-sparseenvironments;asmuchofwhathasbeenexploredinCSCWresearchtodatehasinsteadtouchedonco-locatedwork(e.g.,[37,50,55])ordistributedworkwhereconnectivityisatleastmostlyreliable(e.g.,[64]).Toworktowardthisunderstanding,weconductedastudyofWSARvolunteers,looking

mainlyatworkpractices,thechallengesthatworkersfaceasaresultofthewildernessandalackofrealtimecommunication,andtheirworkaroundstosuchchallenges.Thisstudywasapprovedbyourresearchethicsboard.Thisstudywasconductedintwocomponents:(1)aninterviewcomponent,and(2)anobservationcomponent.

4.1 Interviews



First,weconductedextensiveone-on-oneinterviewswithWSARworkers.ThepurposewastohaveparticipantsreflectmorebroadlyontheirexperienceswithWSARacrossmultipleincidents.Participants.Weinterviewed13WSARvolunteers(11men,twowomen),includingfourSARmanagersandfivefieldteamleaders,inordertobetterunderstandthechallengesfromboththefieldandcommand-postperspectives.WerecruitedinterviewparticipantsfromvolunteerWSARagenciesinWesternCanadathroughsocialmedia,wordofmouth,andbycontactingindividualagencies.Theagenciesthatourparticipantsvolunteeredforservedvariouscommunities,smallandlarge,aroundWesternCanada,allnearwildernessregionscontainingmountains,lakes,rivers,andforests.Ourinterviewparticipantswerebetweentheagesof32and65(M=49,SD=13),andhadbetweenfourand21yearsofexperienceworkinginWSAR(M=10,SD=7).Thoughweaimedforasmuchdiversityinourparticipantsaspossible,thegenderimbalanceofourparticipantsstemsfromthefactthat,inCanadaatleast,therearemoremenservingasWSARvolunteersthanwomen.Method.Theinterviewswereapproximatelyone-hourlong,semi-structured,andtookplaceoverthephone,viavideocalling,orin-person(dependingontheavailabilityandlocationoftheparticipant).Weaskedparticipantsabouttheircommunicationpractices,theirneeds,howtheyusecurrenttechnologytocommunicateduringWSARoperations,thechallengestheyface,andhowtheyovercomesuchchallenges.Wemainlycentredourquestionsaroundstoriesofrealscenariosbyaskingparticipantstorecountpastincidentsandfocustheirtellingoftheincidentsonthecommunication,collaboration,andinformation-sharingpracticesandchallenges.Samplequestionsandpromptsthatwegaveparticipantsincluded“tellastoryaboutasituationyou’veexperiencedinwhichcommunicationtoCommandbrokedown”and“tellastoryaboutasituationyou’veexperiencedinwhichitwasdifficulttomakesenseofincominginformationfromfieldteams.”

4.2 Observa-on

Tocomplementourinterviewsandtoseefirst-handsomeoftheexperiencesthatWSARworkerstoldusabout,weobservedasimulatedWSARexercisewherevolunteersfromvariousnearbyWSARagenciessearchedforfictionallostsubjectsinaforestedandmountainouswildernessareanearamedium-sizedcityinWesternCanada.Thisday-long(eighthour)mockWSARoperationwasorganizedandhostedbyalocalSARagencytotrainvolunteers.TheeventsimulatedtheentireexperienceofanormalWSARoperation,includingamissingpersonphonecall,calloutstothevolunteers,settingupamobileofficetraileronsite,organizingandsendingoutsearchteams,andsoon.Onlytop-levelorganizers,whowerenotinvolvedinthesimulationasparticipants,knewallofthedetailsofthesimulation,somanagersoperatedwithoutnecessarilyevenunderstandingwheretheboundsofthesearchareawere.Over100WSARvolunteersfrom14localSARagenciesnearbysearchedfor15fictional

lostsubjects(whothemselveswerevolunteersfromanearbycommunity)inthisexercise.Theleadresearcheronourteamobservedthemocksearchinitsentiretyfromtheoperationsvehicleatthecommandpost(withpermission).Theresearcheractedasafly-on-the-wallintheoperationsvehicle,observingandtakingnotesonthecommunicationscominginfromthefieldteamsviatheradio,thereactionsofthepeopleinthecommandvehicletotheincomingradiocommunications,theoutgoingcommunicationsfromCommandtothefieldteams,andtheco-locatedcommunicationshappeningwithinthecommandvehicle(e.g.,fieldteamshandingindebriefformsastheyreturntobase,membersofthemanagementteamsharinginformation,etc.).Whenevervolunteerswerenotbusy,weaskedaboutthethingstheydid,whytheydidthem,whatworkedwellintheirduties,and



whatchallengestheyfacerelatedtoinformationsharing.Wewereunabletogetaresearchertoobservefromthefieldperspectiveduetosafetyandliabilityconcerns.

4.3 Data Collec-on and Analysis

Theinterviewswereaudiorecordedandtranscribed.Fortheobservationcomponent,aresearchertookthoroughnotesofwhathappenedinthecommandvehicle.Weusedopen,axial,andselectivecodingtoanalyzeboththeinterviewandobservationdataandrevealhigher-levelthemes.Opencodesincludedthingslikelocationawareness,sendinginformation,andrecordinginformation,whileaxialcodesincludedcategorizationsoftheopencodessuchasawarenessandinformationsharing.Ourselectivecodesandthemesincludedcommunicationgoals,communicationchallenges,andworkaroundstocommunicationchallenges.Theleadresearcheronourteamdidmostofthecoding,butthecodesandwhattheypointedat(e.g.,whattheydescribedshorthand)werereviewedcollectivelyanditerativelybytheothertworesearchersonourteamthroughoutthedata-collectionandcodingphases.WelookedatthesegoalsandchallengesfromtheperspectivesofbothfieldandCommandworkers,tounderstandthesimilarities,differences,andtensionsbetweentheirneedsandcircumstances.Wenowdiscussourfindings.Interview-participantquotesarelistedwith‘P#’indicating

whichparticipantgavethequotefollowedbythemainrolethatshe/heindicatedtakingon(FieldWorker,FieldTeamLeader,orSARManager).Forquotesfromthemock-searchactivity,weindicatetheroleplayedbythepersonbeingquoted.

5 FINDINGS

5.1 Documenta-on, Logging, and Awareness

AtCommand.WefoundthatSARmanagersrelyheavilyonwrittenforms,physicalartifacts,andtheirpositionsinthecommandoffice(Fig.2)inmaintaininganoperationalpictureandsharedmentalmodel.Thisinformationisusefulforplanning,recordkeeping,andmaintainingsituationawarenessofthestatusesofteamsandresources,especiallywhenunabletodirectlyobservethem.Throughtheseforms,commandworkerscanstillbeawareofwhateveryoneissupposedtobedoingandmakepredictions(basedonhowmuchtimehaspassed)ofhowmuchprogresseachteamhasmadeintheirassignments.Indirectaccordancewithdistributedcognition,knowledgeiscontainedintheartifactsspreadoutacrossthecommandvehicle.WhenthefirstSARmanagersarriveatthesceneandaftertheyhavesetupthestaging

areawiththecommandvehicle(s),theybeginfillingoutforms.Weobservedthattheseformsinclude(butarenotlimitedto):asubjectdescription,ICSformsindicatingtherolesandtaskstobeassigned,andmapsofthesearchareaincludingprobabilitiesandsearchpaths.Asvolunteersarriveatthestagingarea,theyreporttothemanagerondutyandsignin.Oncethemanagementteamfinishescreatingaresponseplan,theycallallthevolunteersonsiteanddoaninitialbriefing.



Fig.2.Theinsideofthecommandvehicle,wheretheSARmanagerandhermanagementteamoverseetheoperationandcoordinatefieldteams.Thecommandpostisfilledwithwrittenformsandphysical

artifactsplacedoverwalls,desks,andwhiteboards.

Duringthesimulatedresponseweobserved,themanagementteamfirstgavethevolunteersatasknumber,followedbydescribingthescenario.Inthistrainingscenario,itwasahelicoptercrashintheforest,andtherewere15crewmembersonboardwhowerenowmissing.Theygavethetimeofthecrash,descriptionsofthesubjectsandtheaircraft,theestimatedlocationofthecrashbasedoneyewitnessreportsofwherethehelicopterwasflyingbeforeitcrashed,andthetypesofcluesthatthevolunteersshouldbeonthelookoutfor.Volunteersweregivenashortdescriptionofthesearcharea,includingwhatfeatures(hills,streams,boulders,etc.)werecontainedwithinit.Theywerealsotoldsafetyinformationsuchastheweatherforecast,therecommendedclothing,andamountofwaterneeded.Whileallofthisinformationwasbeingtold,weobservedvolunteerstakingnotesinpocketnotebookswhichtheythencarriedwiththemintothefield.Volunteerswerethenorganizedintoteams,andeachteamwasassignedoneexperienced

SARworkertobetheteamleader.Theteamswerethengiventheirtaskassignments.Foreachteam,amanagerbriefedtheteamleader,whowouldthenbrieftherestoftheteam.Commandgaveeachteamasetofpapermapsandformswiththesubjectdescription,andnotesonthetypeofsearchtoconduct,whichlocationstosearch,andwhattobeonthelookoutfor.Fieldteamsweretorelyonthisinformationwhenoutinthefieldperformingtheirduties.BothCommandandthefieldteamhadacopyofthesametaskandroleassignmentforms.Photocopiesweremadeimmediatelyaftertheformswerefilled,sothatCommandcouldkeepacopytoplaceonthewall.Thisallowedthemtokeeptrackofwhowasdeployedinthefieldandwhattheyweredoing.EachteamcarriedatleastoneGPS-equippedradiotransceiverwiththemforcommunicatingwithCommand.Whenteamsreturned,theywoulddoapost-assignmentdebrief,theirformswouldbereturnedtomanagement,andmanagement’scopiesoftheformswouldbemovedtoasectionofthewallshowingthecompletedassignments.

“We use whiteboards and erasable pens. We also use the physical team assignments, so we'll write the team assignment, what they're doing, where they're going, and then once they check in that they're in the field, then we move [the form] to a different location [on the wall], so now we know they're in the field.” – P4 (SAR Manager)



“The team leader will take a photocopy of that task sheet with him/her out into the field and the original will stay at Command. When the team comes back in, the team leader will […] do a written record of what they did, what they encountered, what they found, all that stuff. And then [the] team leader will sit down with someone from Command and go over it.” – P6 (Field Worker)

Weobservedthatthemanagementteamcarefullyloggedallradiocommunications.Duringthesearchphase,allradiocommunicationsbetweenthefieldteamsandCommandtookplaceonasingleradiofrequency.TheSARmanagerassignedonepersontobethecommunicationsofficer.ThispersonwasinchargeoftalkingtoteamsinthefieldontheradioonbehalfofCommand,andloggingalloftheradiocommunicationsthattakeplace.Weobservedthatthisloggingisprecise,complete,andextensive,withthegoalfortheretobeasmuchaccuracyaspossible.Thecommunicationsofficerdidnotnecessarilylogeachmessageword-by-word,butratherloggedeachhigher-level‘communicationevent’inarowonaspreadsheetwiththefollowingfourcolumns:(1)whothemessagewasfrom;(2)whothemessagewasto;(3)thetimeofthemessage;and(4)ahigher-levelsummaryofthemessage,event,orreasonforthecommunication.ThereweretworeasonsCommandwasloggingthisinformation:(1)tokeepatimelogofkeyeventstopossiblyrefertolater,and(2)toprotectthemselvesforliabilitypurposes.

“If we're walking along the road and I find a gold bracelet with a cross on it on the side of the road, I'll radio Command […] They'll log that. […] Then let's say the next day the family members come up and say ‘oh we forget to tell you that little Susie has a gold bracelet with a cross on it.’ We can go back and look at that record and say ‘oh we found this bracelet at this location at this time.’” – P11 (Field Worker)

ThisallowstheknowledgetobestoredinanartifactaftertransmissionfromthefieldtoCommand,andusedatalatertime,possiblybyotherworkers.Inaccordancetodistributedcognition,thispieceofknowledgeisthuscontainedintheworkspaceandbelongstotherespondingorganization,ratherthanjustafewindividualswithinit.IntheField.Fieldteamsalsoreliedondocumentation(consistingofbothformstheyandmanagementfilledout,aswellashand-writtennotes)torecallwhattheyweresupposedtodoandwheretheyweresupposedtogo.However,incontrasttothemanagementteam,whohadanawarenessofthebiggerpicture,fieldteamshadamore-focusedlower-levelpictureofthings,inrelationtotheircurrenttaskassignments.AccordingtosomeSARworkers,thislevelofdetailwasusuallyenoughtocompletetheirduties.

“[Field teams] know who they're looking for and the generals, but they don't need to know what the big picture of the search is. They're going to go to the area that they're told to, [but] they may not even know why.” – P4 (SAR Manager)

ThebestcasemeantthatconditionsinthefieldwereasCommandexpectedthemtobe,andthefieldteamdidnotrunintoanyunusualorunanticipatedchallenges.However,thiswasnotalwaysthecase.Sometimesafieldteamwouldrunintoanobstaclethatpreventedthemfromconductingtheirsearchassignmentexactlyastheywererequestedto.Forexample,inthesimulatedsearchweobserved,ateamradioedCommandtoinformthemthattheywereobservingmoretrailsaroundthemthanwereshownonthemapthattheyweregiven.Laterintheoperationduringtherescuestage,anotherteamthatwassearchingformoresubjectsontheirwaytothecrashsiteencounteredbouldersalongtheroutetheyweresupposedtotake,andtheycouldnotpassthrough.Thus,theyhadtotakeanalternateroute.TheyinformedCommandontheradioandalsosharedthelocationoftheboulders,sothatCommandcouldkeeptrackandperhapsinformotherteamsinthearea.



5.2 Consistency, Agreement, and Control

OuranalysisshowedthatWSARworkersuseddocumentationandcommunicationstomaintainasharedmentalmodelonkeyaspectsoftheoperation(e.g.,sharedagreementandcommongroundonwhatwastobedone).CommanddidnotwanttheretobediscrepanciesinwhatateamisdoingandwhatCommandexpectsthemtobedoing.Theywantedconsistency,clarity,control,agreement,andsharedunderstanding.Whenateamdidsomethingthatwasunexpected,Commandneededtoknowwhy.Asingleteamdeviatingcanaffecttheentireoperation,andCommandmayneedtoadjusttheiroperationplanaroundchangingconditionsinthefieldorateamdeviating.

“As a non-military organization, we sometimes face the issue of giving a field team leader a task, only to realize they decided to do something they thought was more ‘important’.” – P13 (SAR Manager)

Therearesometimesvalidreasonsforfieldteamstooccasionallydeviatefromtheirassignments,andCommandoftenknowsthis.Duringanoperation,Commanddoesnotexperiencetheconditionsofthefieldfirst-hand,sotheyarenotalwaysabletojudgeforthemselveswhetherornotacourseofactioncanbetaken.Theygenerallytrusttheteamleadertomakethatcall:

“[We're] content to just let them do their job. […] We don't need to micromanage. We can trust the team leader to make sure that things are being done correctly. But having said that, we still need constant radio communications or equivalent because of changing information.” – P10 (SAR Manager)

“All kinds of things can happen. Maybe [there are] some natural features there, it all looks great and flat and clear on a map in the command post where I'm warm and dry, but out in the field, the reality [could be different].” – P4 (SAR Manager)

Nevertheless,Commandstillneededfieldteamstoinformthemofwhentheydeviate,andforwhatreason.Whentheydidthisthough,thiscommunicationwasnotalwaysclear,especiallyovertheradio:

“The helicopter pilot dropped me in the wrong place […] and I let command know, [but] they didn’t quite understand. When I got back with my team, it’s like ‘Where did you go?’ ‘Well, here.’ ‘Why did you go there?’ I’m like ‘Because the helicopter couldn’t land there, so we searched the area we were in.’” – P4 (SAR Manager; describing an incident when he played the role of Field Team Leader)

Inordertomakesurethatteamsareontrack,Commandgenerallywantstomaintainconsistentradiocommunicationswithteamswhentheyareinthefield.WeobservedthatteamscontactedCommandtogiveroutinestatusupdatesatleastonceeveryhour,andsometimesmorefrequentlyiftheyencounteredthingsthattheyfeelareimportanttosharewithCommand.Commandwantedtoensurethattheteamsweresafe,thattheywereperformingtheirassigneddutiesproperly,andtheywantedtounderstandwhatchallengesteamsfaced.Commandwantednewinformationtokeepflowinginsothattheirmentalmodelkeptupdating.ThisalsokepttheteamsintouchwithCommand,incaseCommandneededtosendthemnewinformation.

“We always want to know where our teams are in the field and how they're doing.” – P10 (SAR Manager)

“You're listening to the radio all the time, so you're getting an idea of what teams are encountering, what they're doing.” – P4 (SAR Manager)



“For me as a team leader, I'm updating command with what our next move is, if we found the subject, or noticed any along the trail that could impede the other teams. But [we] don't […] make a big deal out of it. It literally is just to see that you're alive and that everybody is doing [well].” – P5 (Field Team Leader)

InWSARresponses,therearealsotimeswhenSARmanagersneedtoinformteamsofupdatedinformationthatisrelevanttotheirtasks.Forexample,duringthemocksearch,oncethecrashsitelocationwasfoundbyoneofthefieldteams,Commandinformedalloftheotherteamsviatheradiothatthesitehadbeenfound.Commandthencommunicatedthelocationofthesitetotheteams(asasetofUTMlocationcoordinates).Afterward,Commandradioedsomeoftheteamsindividually(bycallingtheirteamnumber;e.g.,“Team102,comein.”)andaskedthemtomaketheirwaytothecrashsitelocationwhilebeingonthelookoutforvictimsalongtheway.Anotherexamplefromoneofourinterviewparticipants:

“At some point, […] we need everybody to know about [new information]. […] ‘Okay, now they're barefoot.’ That's important to know if people are looking for tracks, […] that would be the kind of collective knowledge that would need to be spread out.” – P4 (SAR Manager)

Duringthemocksearch,weobservedthatthecommunicationsofficerandothersaroundherinthecommandvehiclewouldoccasionallywatchoutthewindowsinfrontofthemtoseebitsofwhatishappeningoutsideofthecommandtrailers.Thisviewallowedthemtoseeteamsleavingtoandreturningfromtheirassignments.Atonepointintime,thepersonnelatCommandwatchedasavehicleleftonitsown:

[Vehicle starts to leave the staging area.]

Operations Manager: Which vehicle was that? […] Were they asked to go?

[Vehicle goes out to the field on their own, without instruction from Command.]

Operations Manager: Unacceptable!

Assoonasateamorvolunteerselfdeploys,theirmentalmodelofwhattheyaredoingbecomesinconsistentwiththatofCommand.ThisleavesCommandwithanincompleteunderstandingoftheoperationandthepeople/resourcesavailable,hinderingtheirabilitytomakeproperdecisionsandkeepeveryonesafe.

“If we're not asked to [go] in, then it's not our job to go in. We don't self-deploy. That's management. They have the big picture. We have our focus picture.” – P1 (Field Team Leader)

ThisevengoessofarthatCommandgetsupsetwhenfieldteamsdonotproperlycheckinoroutwhentheyaresupposedto.Inthemocksearch,weobservedCommandpersonnelscoldingfieldteamswhoreturnedwithoutinformingCommandoftheirreturn,evenifworkersatCommandwereabletoobservethemreturningthroughthewindowsofthecommandvehicle.

5.3 Recep-on Gaps

RemotecommunicationbetweenCommandandfieldteamsprimarilyhappenedviatheradio.WeobservedthatitistheprimarywaythatCommandkeptintouchwiththeteamsandupdatedtheirmentalmodeloftheoperation,astheywereunabletodirectlyobserveorcommunicatewithteamsface-to-facewhendeployed.Thechallenge,however,wasthatradioreceptionwasoftenunreliableinthewilderness.Receptiongapsinthewildernesscontributedtoanasymmetryofworkers’knowledgeofdetailsoftheoperation,anda



breakdowninthesharedmentalmodel.Asaresultoftheseradiogaps,fieldteamsconstantlymovedinandoutofradiocoverage,leadingtoCommandnotalwayshavingconstantcommunicationwiththeteams.Further,fieldteamssometimesdidnotknowiftheywereoutofradiorange,orevenwhatinformationtheyaremissingiftheywereoutofrange.

“[The radio] usually doesn't work well. […] I can't actually think of a situation where I've been happy with [communications] ever.” – P3 (Field Team Leader)

“Very often radio communications [are] quite sketchy. […] We'll be in areas where we just can't communicate with Command and they can't communicate with us. We have to keep going until we get to a point where we can communicate. That happens all the time out here.” – P11 (Field Worker)

Wefoundthatunreliablecommunicationwiththeteamscouldincreasethetimeittakesfornewinformationtoflowin,leadingtoCommand’smentalmodeloftheoperationbeingincomplete.Ifcommanddidnothavethemostup-to-dateinformation,theywerenotalwaysabletomakethebestdecisionsgiventhepresentsituation.Commandworkersareconcernedwithreceivingasmuchinformationfromthefieldaspossibleinasshortofatimeaspossibleandkeepingintouchwiththefieldteamstoknowwheretheyare,howtheyare,andwhattheyaredoing.

“What [you’re doing as a] SAR manager is […] trying to make sure that all your teams are proceeding as directed and that everyone's safe. It's nice to know that at all times. When you don't have communication, you make that assumption, but until you've got communication again you don't know that. It's hard when you have dead spots when you can't get in touch with people.” – P8 (SAR Manager)

WhenCommanddoesnothaveconsistentcommunicationswithfieldteams,theycanonlymakeassumptionsandpredictionsoftheirstatusesandprogress,basedontheirlast-knownstatus.Theycanonlymakeaninferenceorpredictiononwheretheymightbe,basedontheirlaststatusupdate,theirgiventaskassignment,knownconditionsinthefield,andanyotherknowledgetheymighthaveinthecurrentmoment.Additionally,informationfromCommandmaynotalwaysgetpassedontoallofthe

necessaryteamsinthefieldrightaway,iftherearesometeamsoutofradiorange.Ifateamdoesnotanswertheradiorightaway,Commandcannotwaitforthemtobecomeavailable,asmanagementmustattendtotheirowndutiesandrespondtoincomingcommunicationsfromotherteams.

“We were searching for maybe 40 minutes longer than we had to be, because we were in a radio dead zone. We didn’t get the update that the person had been found.” – P7 (Field Worker)

Radiogapsalsoresultedinfieldteamshearingonlypartsofamessageorconversation.Additionally,theydidnotalwayshearmessagesfromeveryoneiftheywereinanareawheretheywerewithinradiorangeofsometeamsbutnotofothers.

“If we're all on the same channel, you could hear the conversations on that. […] You don't [always] hear both ends. But if you've got teams that are in your area, […] you'll hear what they're chatting about. Like [for example] the other day, when the [subject] was located, that team [that located her] was close to us. And so immediately we heard [the team say on the radio]: ‘subject responded’.” – P1 (Field Team Leader)

Hearinggapsinradiomessagesmadeitdifficulttoputtheradioconversationsintocontextwhenafieldteamonlyheardsomeofit.



5.4 Communica-on Priori-za-on and Discre-on

Thefactthatallradiotransmissionsweresentandplayedinthesamewaymeantthatallinformation,whetheritwascrucialormundane,wasbeingheardontheothersideinthesameway.Thisintroducedthepotentialfortherecipienttomisunderstandthetrueprioritylevelorurgencyofsomemessage.Notallinformationmayneedtobecommunicatedorreceivedinthesameway,andindeedsometimesfieldteamsusedothercommunicationmodalities,bothsynchronousandasynchronous,togettheirmessageacross.TheserangedfromSMS,picturemessaging,andlocationdata.Iftheinformationwasnoturgent,sometimesteamsevenwaiteduntiltheyreturnedtoCommandbeforesharingit,ortheysentitasamessagewithoutneedinganimmediatereply.Asanexample,itcanbechallengingforafieldteamtogetCommandtounderstandthe

importanceofsomepieceofevidencefromthefieldovertheradio:

“Often […] you'll have something that maybe you think is of particular relevance and Command does not seem to be taking it as seriously as you think they should. Very often Command will just tell you to stand by and you don't hear anything.” – P6 (Field Worker)

Workerstriedtoaddprioritytoradiomessagesthroughtheuseofstandardterminology:

“We can get priority for a radio transmission via the use of either, depending on severity, ‘no duff’ or the next step up ‘pan pan pan’ or the max urgency ‘mayday mayday mayday’.” – P2 (Field Worker)

EvenwhensomethingwasimportanttosharewithCommand,ateammayhavesometimesneededtowaittosendit,asCommandmighthavebeentoobusytoanswerinthemoment.Thiscouldhavebeenduetotherebeingmultiplepeoplecominginandoutofthecommandvehicletoexchangeinformation,ormultiplefieldteamswantingtosendinformationtocommandviatheradioatthesametime.

“Sometimes you might have a situation where there's 10 or 12 teams out on the field, all wanting to communicate with command at the same time.” – P6 (Field Worker)

Duringthemocksearchweobserved,oncethecrashsitehadbeenlocated,theradiochannelbecamesooverloadedwithcommunicationsfromthefieldteamsthattheradiooperatorcouldnothandleallofthemessagesbyherself.Atthispoint,anothermemberofthemanagementteamtookontheroleofasecondradiooperatorandopenedupanotherradiochannel;thusleavingonechannelopenforteamsstillsearchingforsubjects,andanotheropenforteamswhowereextractingsubjectsthatwerealreadyfound.Toavoidwaitingontheradio,interviewparticipantshavetoldusthatsomeagencies

allowmemberstosendtextmessagesforlessurgentmessagesormessagesthatdonotneedtobereadimmediately.Someworkerscoupledthesewithphotosiftheyneededtoshowsomething.

“Cell phones may work, if not then [satellite] phones [or the] text-message capabilities of new [satellite-phone] devices.” – P2 (Field Worker)

Inothercases,participantstoldusthatteamssometimeswaituntiltheyreturntoCommandbeforegivingthemsomeoftheinformationtheyhavecollected,toavoidclutteringtheradiowithchatter.Forexample,weobservedthatwhenateamreturnstocommand,theteamleaderwouldgivetheteam’sGPS-equippedradiotoamanager,whowouldthenuploadtheGPSrecordtoacomputerthatwouldthendisplaythepaththeteamtookoverlaidonamapofthesearcharea.Overtime,thisdigitalmapwouldpopulatewith



thepathstakenbyteamsandpinsshowingkeylocations,suchaswhereclueswerefound,thusconstructinganinformationpictureofthestatusandprogressofthesearch.

“Teams come back with their GPS. It's uploaded, and you gradually build a map of what teams have been where, and you have your different colours. So the dog team's been through here, we got a hasty team through there, […]” – P4 (SAR Manager)

Thisgaveteamsanopportunitytoexplaintheroutetheytook,theactionstheyperformed,andthethingstheyfoundwhileconductingthein-persondebriefwithaSARmanager.ThiswaseasiertodoatCommandthanwhiletheteamwasinthefieldandinthemidstoftheirtaskassignment.Thelackofclarityofsomemessagesovertheradio,thedifficultyofusingtheradiowhiletrudgingthroughthewilderness,andradioreceptiongapsmadeitmorefeasibletosharethislargeramountofinformationatCommand,afterithadalreadybeenloggedautomaticallybyadevice.

“If it's not super urgent […] it just kind of clutters up the radio chatter trying to tell them where we are. So, we just mark [where the clue is] on our GPS, so when we go back to Command, they can download it.” – P3 (Field Team Leader)

ThedecisionofwhetherornottopasssomepieceofinformationontoCommandwasusuallylefttothediscretionofthefieldteam,andmostoftentheteamleader.ThisleftopenthepotentialforCommandtomisssomecrucialinformationtheyneeded,simplybecausetheteamdidnotthinkitwasimportanttosendthatinformation.Inparticular,ifateamhasalessexperiencedteamleader,orevenonethatdoesnot

sharethesameperspectiveorknowledgeofCommand,theymightnotknowwhatinformationisimportanttopasson:

“People are human and sometimes maybe it's a team [without] a particularly assertive team leader and they may be like, ‘we kind of found this thing but we don't really think it's important,’ so they don't push it.” – P4 (SAR Manager)

Lastly,inorderforateamtotalktoCommandontheradio,theymuststopwhattheyaredoingandfindalocationwheretheycangetradioconnectivitywithCommand.Thisoftendistractsthemfromtheirfocusontheirassigneddutiesandusesuptimethatcouldbebetterspentsearchingforthesubject.

“If you are going and stopping and trying, it takes away […] There's decreasing returns if you put a whole lot of effort into finding a place where you can get a hold of base and tell them, ‘nothing to report’” – P2 (Field Worker)

Forthissamereason,SARmanagerswerealsohesitanttopagefieldteamsforstatusupdateswhentheyweresilentontheradioforawhile.

“Every time you radio [a field team] ... they're skiing along, they have to stop, they have to take their gloves off, depress the radio [button] ... it slows them down. It's a hindrance to them.” – P8 (SAR Manager)

SomenewerradioscomeequippedwithGPStransceivers,whichallowcommandtoobtainthelocationofanyfieldteamwithinradioreachinrealtimethroughtheclickofabutton.Thiswasthecaseforthemocksearchweobserved.SARmanagershaveexpressedthatobtaininginformationinthiswayishelpful,giventhatitcomesimmediatelyandupdatesasfrequentlyastheywantitto,andtheyhavealsosaidthatitiscomfortingtoknowthattheycangetthisinformationwithoutbotheringthefieldteams.

5.5 Awareness, Distrac-on, and Level of Detail in the Field



FieldworkerssometimeswanttomaintainasharedmentalmodelthatisconsistentwithCommand’smentalmodelandhigher-levelawareness.Inparticular,field-workerparticipantshavetoldusthattheywantworkspaceawarenessoftheenvironmentandotherfieldteams’activities,andhigher-levelsituationawarenessofthebiggerpictureofthesearchstatus,totheextentthatitisrelevanttothem.Theywanttoknowhowthebigger-pictureofthesearchisevolving,andtheywanttoknowhowtheiractionsarecontributingtothesearchresponseasawhole,sotheycanunderstandtheirimpactontheoperationandfeellessisolatedfromtheirteammates.Whilethisisthecase,weweretoldbySARmanagersthatfieldworkersaresupposedtobefocusedonthein-the-momentdemandsofthewildernessenvironmentandthesearchtasktheyweregiven.Thispresentsanimportanttension.Ateam’sfocusontheirassigneddutiesisbeneficialtotheoperationasawhole(andthusthesafetyandlivelihoodofthesubject),asitincreasesthelikelihoodthattheyfindthesubjectsooner.Ontheotherhand,remainingawareofone’srolewithinthebiggerpictureandinrelationwitheveryoneelsecouldboostateam’smorale,asitcouldserveasareminderthattheteam’sfocusedactionsaremeaningfultotheoperationasawhole.Afieldworker’sabilitytohearotherremoteteamsmayhelpboostthismorale,similartohowplayersofonlinevideogamesfeelahighersenseofteamcommitmentwhentheyareabletohearandcommunicatewiththeirteammembers[20].Especiallyinlargeanddifficultsearches,WSARworkersoftenwanttoknowiftheyaremakingprogressandwhatisgoingonelsewhere.Thisiswhytheyareinterestedintheradiochatter.Buttoomuchofthiscanbecomeadistraction.Insomecases,fieldteamshaveevenshutofftheradiobecausetheywerehearingtoomuchradiochatter,anditwashinderingtheirfocus.

“It depends on the task but sometimes [hearing radio traffic] is obnoxious, where you have to turn the radio down and hopefully remember to turn it back up again.” – P3 (Field Team Leader)

Whilethisisthecase,thereisalsoaninherentcuriosityaboutwhatothersaredoing,andwhethertheyhavefoundclues:

“I know that I'm always listening; like if I'm not busy, say we're just walking down a trail and looking for clues, I'm always interested in what other people are talking about.” – P3 (Field Team Leader)

Thiscuriosityispartnosiness,butalsoadesiretounderstandthebiggerscopeofthesearch.Thiscouldbebeneficialinthatitcouldreduceafieldteam’ssenseofisolation,boosttheirmorale,andencourageagreatersenseofbelongingwiththeentirerespondingorganization(i.e.,allpersonnelwhoarepartoftheresponse).Thisideaoftryingtofostera‘largeteam’approachisalsosomethingthatresonatedwithSARmanagers.

“I felt that I was always working hard to sort of foster an atmosphere of inclusivity and sharing information. So I'm naturally in favour of as many people knowing the bigger picture as possible.” – P10 (SAR Manager)

Participantstoldusthatthisawarenesscouldalsohaveapotentialutilitarianpurpose.Forexample,itcouldallowfieldteamstoknowifotherteamshavefoundorpassedalonginformationthatisrelevanttothemandtheirdutiesandallowthemtocoordinatewithothernearbyteamsifdeemednecessary.However,notallSARmanagersagreedwiththis:

“[The] right [approach] isn't necessarily to give the field teams more information about where they are in the context of other efforts, but the right level of detail.” – P13 (SAR Manager)

Workersunderstandthisfundamentaltension,asthefieldworkers’tasksalternatebetweenmomentsofboredom(walkingthroughtheforest)andtimeswhenextremefocus



isrequired(e.g.,navigatingagully).Essentially,increasedawarenessofotherfieldteamscouldbeusefulwhentheirworkbecomesmoretightlycoupled,whentheyneedtocollaborateonsomething,butlessusefulwhentheirworkislooselycoupled(whichismostofthetime),whentheyarefocusedontheirtaskassignment,eveniftheincreasedawarenessleadstoaboostinmorale.

5.6 Inter-Team Communica-on

WhilefieldteamscangenerallyoverhearalotofcommunicationbetweenCommandandotherteams,protocolprohibitsafieldteamfromcommunicatingdirectlywithanotherfieldteamwithoutgoingthroughCommandfirst,eventhoughfieldteamscangenerallyheareachotherontheradio.ThisisunlesstheyhaveexplicitpermissionfromCommandtocommunicatedirectly.Participantstoldusthattherearetworeasonsforthis:(1)Commandwantstohavecontrolandawarenessofallinformationpassingthroughtheradiochannel,and(2)Commandwantstopreventtheradiochannelfromhavingtoomuchtraffic.

“[Command doesn’t] want to miss anything, and if one team was talking to another team […] maybe that could happen.” – P6 (Field Worker)

SARmanagersalsotoldusthatpermissiontocommunicatedirectlywithanotherfieldteamisusuallygrantedforoneoftworeasons:(1)theteamneedstodirectlycoordinateresourcesoractionswithanothernearbyteam,or(2)theteamneedstoactasaradiorelaytoCommandforanotherfieldteamthatisoutofradiorangeofCommandbutwithinradiorangeofthefirstteam.

“[A field team] can ask Command for permission to talk directly to another team, and that would be because they need to share information, but Command is still monitoring and still aware of that information.” – P4 (SAR Manager)

Duringthesimulatedoperation,weobservedoneinstanceofateamrequestingpermissiontotalkdirectlywithanotherteam.Inthiscase,itwasduringtherescuestageoftheoperation,andoneoftheteamsneededtosplitintotwosmallerteamstoextractsubjectsinseparatenearbyspots.Giventhattheyweredeployedwithequipmentforoneteam,theyneededtocoordinatetheirsharedresourcesviatheradio.Onereasonthatitwasbeneficialfortwofieldteamstocommunicatedirectlywitheach

otherinthesetypesofsituationsistoavoidanylossofinformationfromhavingmessagesgothroughCommandfirst.

“If you're running into obstacles out in the field that are hard to get past. […] it's often useful to communicate directly rather than have a go-between that could distort the message.” – P6 (Field Worker)

Themoredirectcommunicationis,themorequick,efficient,andclearitis.Thisisespeciallyimportantwhenteamsarecoordinatingwitheachotherorpassingalonglower-leveldetails.Teamshavedonethiswhen,forexample,theyneededtoinformanearbyteamofhazards,orsharenavigationalinstructions.Inthesecases,itwasespeciallyhelpfultohavelocalknowledgeandguidancefromsomeonewhowasactually‘outthere’.

5.7 Use of Other Remote Communica-on Modali-es

OuranalysisrevealedthatCommandneedstofullyunderstandtheinformationthatfieldworkerspasstotheminordertomakegooddecisionsbasedonit.Theyalsoneedtounderstandthisinformationasquicklyaspossible.Wheneverysecondcounts,delaysand



misunderstandingscanleadtoareducedlikelihoodthatthesubjectisfoundalive.Thereisapotentialforothercommunicationmodalitiesandinformationstreamssuchaspicturesandvideostosupportquickerunderstandingofinformation.However,someSARmanagersareconcernedthatbringinginmoremodalitiescouldintroducetheriskofinformationoverload,anditcouldbecometoomuchinformationforCommandtomanageandcontrol.Thereisopportunitytopresentinformationthatisrichinvisualdetailthroughmeans

suchasphotosandvideos.Justlikeinfirefighting[28],itcanbeeasytomakemistakesinlisteningandcommunicatingovertheradioinWSAR,andthesemistakescanhaveseriousconsequences.Inparticular,WSARworkerscaneasilymisskeydetailsortakealongtimetodescribeorunderstandinformationthatisrichindetail.Thisbecomesmorelikelygiventhestressfuldemandsoftheirwork.Forexample,ifafieldworkerisfacingalotofstressduetoenvironmentaldemandsorthedemandsoftheirassignedtask,theymaynotbeabletodescribesomeimportantpieceofinformationinacoherentway.Similarly,iftheradiooperatoratCommandisunabletolistentodetailsthroughtheradioaseasilyduetodistractionsinthecommandpost,theymayeasilymissacrucialdetailgivenbyafieldteam.Examplesofinformationrichindetailsthatcanbehardtodescribeorunderstandviathe

radioincludethevisualpropertiesofclues(e.g.,footprints,objects),thegeographicallayoutofaspot(e.g.,wheretrees,rocks,hills,andbodiesofwaterarefound),andfirst-aidinformation(e.g.,themedicalstateofthesubject).Sendingimagesorvideosofthistypeofinformationcouldbebeneficial.

“I could imagine using [pictures or videos] in situations where, if you're in the bush and you're looking for an example helicopter landing zone, if you're looking for a route out, or […] you can send pictures of the condition people are in.” – P9 (Field Worker)

Inrecentyears,somefieldworkershavebeguntosendimagesofcluestoSARmanagersviaSMS/MMSmessaging.

“We'll obviously be taking pictures with our phones and in the last five or six years we, text messaging is so good I'll just text my SAR manager a picture and he'll be like, ‘Yep, that's their footprint’ or ‘Nope, that's not their foot.’” – P3 (Field Team Leader)

WhilepicturesandvideoscouldaidfieldworkersindescribingtheconditionsofascenetoCommandorhelpthemconveyamessage,verbaldescriptionsarestillusefulinsomecircumstances;eitherontheirownorcoupledwithimagesorvideos.

“Terrain's usually better described by whoever's out there. It's hard to look at a picture and know what the real situation is.” – P8 (SAR Manager)

WhilesomeSARmanagersandagenciesembracenewercommunicationchannels,theycanstillleadtopotentialchallenges.Forone,addingmorechannelscouldmakeitincreasinglychallengingtomaintainamentalmodeloftheoperation,astheworkersatCommandmayhavetopayattentiontomultipleinformationstreams.

“I know from being a manager I receive information via voice, text, email and radio already, and synthesizing an operating picture out of those different streams can be very challenging. There are many things competing for my attention during a search.” – P13 (SAR Manager)

Inadditiontothis,addingmorestylesofcommunicationaffectsworkprotocols,requirestrainingintheiruse,andpotentiallyintroducesmorepointsoffailureintheoperation.

“The strength of the current system is that it is durable and fault tolerant, where most digital systems introduce more single points of failure and fragility. A functioning radio, a white board, some pens, and some paper is the basics of a SAR management system. Adding more tools and



technology introduces opportunity for failure in the devices, protocols and training of the individuals.” – P13 (SAR Manager)

Therearetwothingsthatthissuggests:(1)aggregatingexistinginformationstreamsbeforeintroducingnewonesmightbemorebeneficial,and(2)carefulconsiderationshouldbetakenbeforeintroducinganewinformationstream,todeterminewhetherornotitisactuallynecessaryorhelpfultotheoperation.

6 DISCUSSION AND CONCLUSIONS

WenowdiscussourresultsandtheirimplicationsforthedesignofremotecollaborationtechnologiesforWSAR.OurgoalistohelpreadersunderstandthespecificcommunicationcontextsandchallengesofWSARworkers,sotheycanunderstandwhatdesignsolutionsandapproachescanbegintoaddressthosechallenges.WealsohopetohelpdirectotherresearchersinformingresearchquestionsforfurtherstudyingtechnologyuseinWSARdistributedcollaboration.Whilemaintainingthesharedmentalmodeloftherespondingagencyasawholeis

importantforaresponse,therearemanychallengesuniquetoWSARthatmakethischallenging.Someoftheseareduetotechnologicalfactors(i.e.,communicatingovertheradio).Othersrelatetoenvironmentalfactors,andsomerelatetocurrentcommunicationandworkprotocols.Webelievethatitisworthexploringpotentialdesignsolutionsnotonlytothechallengesthatariseduetotechnologicalfactors,butalsotothosethatareduetoenvironmental,social,andprotocolfactors.Next,weexploreanddiscussthesechallengesanddiscussourrecommendationsforfutureresearchanddesignworkinthisspace.Ourrecommendationsarebasedonourfindings,butanynewsolutionswarrantfurthervalidation,especiallyinsettingsthatareasclosetoreal-worldaspossible.

6.1 Implicit Communica-on and Awareness

Thereareopportunitiesfortechnologydesigntofosterandsupportteamawareness,cohesion,andstayingintheloop.Inparticular,thereisanopportunitytosupportimplicitcommunication,coordination,andawarenessremotelyinadditiontoexplicitcommunication.Implicitcommunicationandawarenessarealmostnon-existentbetweenfieldteamsandbetweenthefieldandCommand.WSARworkersfeelthatthebackgroundinformationprovidedfromoverhearingradiochatterissometimesbeneficialpotentiallybecauseitmaybehelpingtofillagapleftbymissingimplicitawarenessandcommunication.WSARworkersareseekingeverybitofhigher-levelawarenessthattheycanget.Fillingthisgapcouldprovideameaningfulimpact.Whilewedonotyetknowwhatthisimpactwouldbe,werecommendfuturedesignworktoexploresolutionstoprovidingadditionalawareness,andstudytheimpactsofthesesolutions.Toillustratesomesimpleexamples,researchcouldexploretechnologiesthatallowfieldworkerstoseetheareastheyhavecovered,theareastheircolleagueshavecovered,acollectionofcluestheircolleagueshavefound,andmessagestheyhavesentovertime.Awarenessofthesethingscouldcontributetotheconsistencyofthesharedmentalmodel[10]amongsttheworkers,whichcouldthereforebenefitcollaborationandteamwork.AsmaintainingasharedmentalmodelisoneofthekeygoalsofWSARremotecommunication,werecommendexploringandtestingdesignsolutionsthataimtogiveWSARworkersasmuchawarenessaspossible.WSAR,however,bringsaboutuniquechallengesindoingthis.Forone,whenafieldteam

isoutofradiocontactandisolatedfromeveryoneelse,theysimplywillnotbeabletoreceiveupdatesfromotherteams.Sometechnologicalsolutions,likeradiorepeatersandmesh-networkingtechnologies(e.g.,[72]),helpminimizethisdisconnectedness.However,



morecouldstillbedonetoprovideWSARworkerswithrelevantinformationandawarenesswhiledisconnected.Forexample,itmaybeworthwhiletoexploretechnologiesthatpresentfieldworkerswithrelevant‘offline’information;i.e.,informationthatisalreadythere,andcanbepresentedtotheuserattherelevanttimewhileoutofradiocontact,or‘offline’.Forexample,itcouldbebeneficialtoshowafieldteamhowmuchoftheirassignedareastheyhavecovered,orshowCommandaprediction(throughprobabilitymodelsorartificialintelligence)ofwhereout-of-contactteamsmaybelocatedandhowmuchprogresstheyarelikelytohavemadeatthecurrenttime,basedontheirgivenassignmentsandotherfactorssuchasweather.Whiletheinformationmaynotbeperfectlyaccurate(e.g.,itmaybeoutofdateor‘stale’),itcouldstillprovideCommandworkerswithmoretoworkwiththanjustreceivingnothing.Thesystemcouldalsoexplicitlytelltheuserthattheinformationisjustaprediction,orevengiveanindicationofhowlikelyitistobeaccurate(i.e.,a‘confidence’rating).Asasimilarexampleonthefieldside,whenfieldworkersare‘offline’,technologycouldgivethemrelevantinformationsuchasexpectedweatherchangesthatareimminent,whereotherfieldteamsarepredictedtobe,andpredictionsofwhenCommandmightwantanupdatefromthem.Thesearesimple‘offline’solutionstomaintainingasharedmentalmodelthatmaybeworthexploringinfuturework.Onethingtonoteisthat‘offline’inthiscaseisnotabinarystate.Ateam1canbe

‘completelyonline’(i.e.,abletocontactallotherteams),‘partiallyonline/offline’(i.e.,abletocontactsomebutnotallotherteams),or‘completelyoffline’(i.e.,notabletocontactanyotherteam).Furthermore,the‘severity’of‘online/offline’statedependsnotonlyonhowmanyotherpersonnelateamisabletocontact,butwhomtheyareabletocontact.Forexample,beingunabletocontactCommandmaybeworsethanbeingunabletocontactanotherfieldteamsearchingtheothersideofthesearcharea.Evenwiththeimportanceofasharedmentalmodel,ourfindingsalsorevealedan

importantdesigntension:whilefieldworkersoftenwantmoreimplicitawareness,theyneedtobefocusedontheirduties.Thus,fieldteamsshouldnotreceivetoomuchinformationthatisirrelevanttothemandtheirduties.Ifthishappens,theycouldquicklybecomedistractedorexperiencementaloverload.Whenthishappens,theycouldeasilybecomeoverwhelmedand/orstarttomissorignoreimportantdetailsgiventothem,andthusstarttoperformtheirdutieswrong.Onlythemostimportantdetails,suchasinformationrelevanttotheteam’sduties,basicbigger-picturedetails(suchasiseveryoneokay,hasthesubjectbeenfound,etc.),andthefieldteams’contributionstothebiggerpictureofthesearchshouldbepresentedtotheteam.Thisisconsistentwiththeideafrompreviousworkthatcompletesharedmentalmodelsarenotalwaysnecessaryforsuccessfulcollaboration[12],aslongastheinformationsharedisup-to-date,consistent,andleadstoworkerscarryingoutindividualdutiesinsuchawaythatithelpstheorganizationasawholepushclosertotheircollectivegoal.Ideally,ateamshouldbegivenenoughinformationtocarryouttheirdutiesproperly,keeptheirspiritsup,andunderstandtheircontributionstothesearchresponseasawhole.Ofcourse,itcanbehardtoknowjusthowmuchinformationistherightamountforfieldteams.Systemscouldbedesignedtoeasilyprovideadjustableamountsofinformation.Forexample,WSARmanagerscouldselectandfilterwhatinformationshouldbesharedwitheachfieldteaminsoftware.Thiscouldeasilybeupdatedatvariouspointsintime,dependingonthesituation.Fieldteamscouldsimilarlyhavesystemsthatprovidethemwithinformationwheretheycanselecttoseemoreorless,dependingontheircontext.

1TeamsinthiscasereferstobothfieldteamsandCommand.



Additionally,whiletherecouldbesomebenefittointroducingsomeimplicitcommunicationandawarenessbetweenfieldteams(thoughthisshouldbeexploredfurther),webelievethatmorepotentialbenefitcouldcomefromintroducingincreasedimplicitawarenessbetweenfieldteamsandCommand.Tosomeextentthisalreadyexists,asCommandcanobservetheGPSlocationsofteams,lookatformsanddocumentation,andeavesdroponradioconversations.Thoughevenwithallofthis,Commandstillneedstoputalotofeffortintocommunicatingexplicitlywiththefieldteamstogetanupdatedpictureoftheirstatuses.Muchofthisinformationstilldoesnotcomeautomaticallyorimplicitly.Gapsinradiocoverageworsenthisproblem.Byallowingformorestatusinformationfromfieldteamstocomeinautomatically,thiscouldsavetimeonCommand’spartandallowthemtoputmoreattentiontowardotheractivities.Asasimpleexample,itcouldbeworthittoexplore360°cameraswornbyfieldworkersthatautomaticallytakeandsendgeotaggedphotosoftheirsurroundingstoCommand,wheretheyarethendisplayedwithindigitalmapsofthesearchterrain.Photoscouldbesharedeitherperiodically(e.g.,every10minutes)orduringkeyeventssuchaswhentheyhavereachedacertainlocationorwhentheyarestoppedforalongtime.SoftwarecouldallowWSARmanagersto‘scrollthrough’timeandseehowthecontentandlocationofthephotoschangeovertime.

6.2 Communica-on Modali-es and Informa-on Streams

Basedonourfindings,webelievethatWSARworkers,teams,andagenciescouldbenefitfromhavingmultiplemodalitiesofcommunicationandinformationsharingattheirdisposal,eachusefulforcertainsituations(e.g.,photosmaybeusefulfordescribingclues,mapsmaybeusefulfordescribinglocations,andtextmaybeusefulforquickstatusupdates).Currently,however,alotof(thoughnotall)remotecommunicationhappensviatheradio,whichdoesnotdowellatpresentingdifferenttypesofinformationinusefulways,atusefultimes,orgivingitthenecessaryprioritization.InformationPresentation,Prioritization,andClassification.Oneofthebiggestchallengeswefoundisthatthereisalotofinformationcominginfromthefield,andSARmanagershavetobringallofthistogetherandmakesenseofit.Whiletherecouldbeapotentialbenefittointroducingotherchannelssuchasvideo,bringinginmoreinformationstreamsintroducesagreaterriskofmentaloverload.Beforefocusingonintroducingmorechannels,designersshouldfirstfocusonaggregatingtheexistingchannelstogetherandpresentingtheinformationinasimplifiedwaytothenecessarypeople.Suchasystemcouldwork,forexample,asasharedcommunicationworkspace.Inaddition,partofthetaskofsortingthroughandpresentinginformationisrelativelymundane,andcouldlikelybeautomated,thussavingtimeandallowingworkerstoattendtomoreimportantduties.Asystemthataggregatesinformationstreamscouldbedesignedtosupportmultiple

communicationmodalities,allowworkerstoseethestatusesofotherworkersandteams,allowCommandtoreferbacktopreviously-collectedinformation,andallowtheagencytokeeparecordofeverythingthathappens,forliabilitypurposes.WhilesuchasystemcouldbehelpfulinWSAR,itmustbedesignedsothatitiseasytosearchandsortthroughthisinformation.Iftheworkspaceisdesignedlikeaflat-prioritysystem,itmaybecometooeasyforimportantinformationtogetlostinaseaofnoise,asitoftendoesincollaborationplatformslikeSlack[9,68].Basedonourfindings,werecommendpresentinginformationindifferentways(e.g.,asalocationonamaporaneventonatimeline)andwithdifferentlevelsofdetail,dependingonwhoisviewingitandinwhatcontexttheyareviewingit.Asanexample,ifamanageratCommandqueriesataskassignmentnumber,theymaybeinterestedinseeingtheroughsearchpathtakenfortheassignmentandtheareacoveredbyit.Ifafieldteamleaderqueriesthesameassignmentnumber,theymaybeinterestedin



seeingdetailsonthesearchtechniquestocarryout,thelandmarksinthefieldtowatchoutfor,andtheequipmentneeded.Furthermore,afieldworkermaybeinterestedinseeingasearchpathinrelationtotheirownfirst-personviewoftheenvironment,whereasamanageratCommandmightbeinterestedinseeingitoverlaidonanoverheadmap.AsynchronousCommunicationandInformationAggregation.Ourfindingsindicatethat,asaresultofradioreceptiongaps,informationcanpropagateslowly.AlotoftheinformationexchangebetweenfieldteamsandCommandhappensbeforeandaftertaskassignments,duetoboththeunreliablereceptioninthefieldandthefactthatfieldteamsareheavilyfocusedontheirtaskswhileinthefield.Whileithasitsdownsides,lackofrealtimecommunicationscanalsosometimesbebeneficialforfieldteams,asitallowsthemtofocusonthetaskofsearchingandcollectinginformation.Fromthis,weinferthereispotentialopportunityforasynchronouscommunicationandinformationsharingthatshouldnotbeignoredinfutureexplorationsofWSARremotecollaboration.Asanexample,afieldteammaywanttosharesomepieceofinformationwithCommand,

buttheydonotneedthemtovieworrespondtoitimmediately.AteamoutsideofradiocoveragecouldqueueupamessagetoCommandthatwouldsendassoonastheyregaincontactwithCommand.Additionally,ateamwouldnothavetowaitforCommandtobecomeavailablebeforesendingamessagetothem.Ateamcouldsendaless-urgentmessagetoCommandquickly,thencarryonwiththeirduties.WSARworkershavecitedthisasareasonforsendingandreceivingtextmessagesandphotos.FocusandDistraction-FreeCommunicationintheField.Giventhementalandphysicaldemandsthatfieldworkersface,communicationsshouldbeassimple,minimal,quick,anddistraction-freeaspossibleforthem.Technologyshouldprovideminimaldistractionsfromfieldworkers’immediatesurroundings,allowthemtocommunicatehands-freeifpossible,minimizetheamountoftimetheyneedtospendsendingandreceivingmessages,allowthemtoanswerless-importantmessageswhentheyarelessbusy,andallowthemtofocusonlisteningandbeingonthelookoutforthesubject.Asynchronouscommunication,asdescribedabove,helpswiththis,allowingfieldworkerstofocusontheirdutiesandrespondwhentheyarebetterableto.Intermsofsynchronouscollaboration,hands-freetechnologiessuchaswearablecameras(e.g.,[45,63]),wearableaugmentedreality(e.g.,[44,61]),anddronesforvideocommunication(e.g.,[40])showpotentialpromise.

6.4 Communica-on ‘Dead Zones’

Lastly,evenwiththeimplementationofmultipledesignsolutionstoWSARcommunicationchallenges,radio-reception‘deadzones’arestillanissueinWSAR.Oneissuethatworkersfacerelatedtothesedeadzonesisthattheygenerallydonotknowifandwhentheyareinone.WSARworkersdonotknowwhentheyhaveenteredadeadzone,norwhentheyhaveexitedit.Inaddition,neitherCommandnorotherteamsgenerallyknowifaspecificteamisinadeadzone.Onepossiblesolution,stemmingfromtheprincipleofseamfuldesign[14,15]istoallowfieldteamstosee(e.g.,asasimplestatusicon)whetherornottheyareabletocontactCommand.Furthermore,ifcommunicationwithotherfieldteamsispermitted,itcouldnotifytheteamwhichotherteamstheycancommunicatewith.Remotecollaborationtoolscouldrevealonamapwhereradiodeadzonesarelocated(similarto[7])andnotifywhenfieldteamsareinoneandwhentheyareabouttoenterone.Thiscouldallowafieldteamtoplanforwhentheywillenteradeadzoneandadjusttheirworkactivitiesbasedaroundthis.ItcouldalsoallowCommandtoknowwhenateamisinadeadzoneorwhentheyareabouttoenterone,sotheycouldprepareinadvanceofthis.Ratherthanjustbeingseenasachallenge,deadzonescouldalsobeseenasopportunitiesforfieldteamstofocus



ontheirsearchtaskandcollectinformationwithoutdistraction.Whenateamleavesadeadzone,asystemcouldremindthemtogetintouchwithCommandagain.Furthermore,whilenotdirectlyrelatedtousability,mesh-networkingtechnologiesand

communicationplatforms(suchas[72])areseeingincreasingadoptioninemergencydomains.InthecontextofWSAR,newcommunicationplatformscouldsupportautomaticinformationcollectionandtransmissionbetweenfieldteams’radioantennae,ultimatelyreachingCommand.Suchtechnologiescouldseamfullyreveal,forexample,whoateamcanreachcontactwithinameshnetwork,ifCommandhasreceivedorreadtheirmessage,orthelastconfirmedtimethatCommandwasabletoobservetheteam’sstatusorlocation.Weseeopportunityforexplorationofthisinfuturework.

7 LIMITATIONS AND FUTURE WORK

Overall,ourstudyopenedupthedesignspaceforWSARdistributedcollaboration,outliningtheinformation-sharingactivitiesandchallengesofWSARworkersandprovidinginitialthoughtsontheopportunitiesfornewandemergingtechnologiesandcommunicationmodalitiestobegintoaddressthesechallenges.ThiswillbeparticularlybeneficialforWSARworkersaroundtheworld,manyofwhomarevolunteers,aswellasforoutdoorenthusiastsandanyonewhoworksinwildernessareaswhomayeverneedWSARservices.Beyondthedesignconsiderationswementionedabove,wealsorecommendthatfuture

worklookintootherWSARcontextsbeyondjustthoseofWesternCanada,asdifferencesincontextscouldpotentiallyrevealnewerinsights.Mostofourparticipantswerefromagenciesnearsmallmountaintowns.Somewerefrommoreremoteregionsthatseefewertourists,andafewwerefromagenciesnearalargemetropolitancityorneartownsthatseemanytouristsengaginginoutdooractivities.Moreover,whileWSARprotocolsvaryslightlyacrossWesterncountries,theymaydifferquitewidelyinnon-Westerncontexts.Thedemographicsofourstudywerelargelylimitedtooldermen(averageageofabout

50years),giventhatthisisthedemographicthatisprevalentinWSARinCanada.StudyingmorewomenoryoungerWSARmemberswhoaremoretechnologyliteratemayrevealdifferentcommunicationandtechnology-usepatterns.Similarly,giventhatmostoftheSARworkersinourstudyweretrainedinthesameway(underthesamesetofguidelines),theirmindsetsabouthowthingsshouldbedoneinWSARmayhavebeenlimitedasaresultofthis.Thus,werecommendfutureworkincludeabroadersetofparticipants,andalsotakeintoaccountthosewhocouldparticipateinWSARinthefuture(ratherthanjustthosewhoparticipateinitnow).Lastly,ourobservationwascentredaroundasinglemock-searchexercisebasedona

scenarioinvolvingasearchformultiplemissingsubjects.Whilethisscenariowasdesignedtobecomplex,totrainWSARworkersinmultipleaspectsoftheirworkandtoexposeworkerstoavarietyofthingstheycouldencounterinarealoperation,werecognizethatthisdoesnotcoverthecompletescopeofpossibleWSARincidents.ThoughwecomplementourobservationfindingswithinterviewresponsesfromWSARworkersplayingavarietyofroleswithvariedexperienceandskilllevels,werecommendfuturestudiesinthisspace,includingvalidationsofdesignsolutions,exploreothertypesofWSARscenarios;e.g.,usageindifferenttypesofenvironments,rescue-onlyoperations,andsearchesforsinglesubjects.

ACKNOWLEDGEMENTS

WethanktheNaturalSciencesandEngineeringResearchCouncilofCanada(NSERC),AlbertaInnovatesTechnologyFutures(AITF),andtheUniversityofCalgaryforfundingthisresearch.WealsothanktheBritishColumbia(BC)SearchandRescueAssociation(BCSARA),



SARAlberta,andthenumerousSARagenciesandmembersfromBCandAlberta,Canadawhomwehavebeenincontactwithfortheirideasandsupportofthiswork.

REFERENCES [1] Sultan A. Alharthi, William A. Hamilton, Igor Dolgov, and Zachary O. Toups. 2018. Mapping in the Wild:

Toward Designing to Train Search & Rescue Planning. In Companion of the 2018 ACM Conference onComputer Supported Cooperative Work and Social Computing (CSCW ’18), 137–140.DOI:https://doi.org/10.1145/3272973.3274039

[2] Sultan A. Alharthi, Nicolas LaLone, Ahmed S. Khalaf, Ruth C. Torres, Lennart E. Nacke, Igor Dolgov, andZacharyO.Toups.2018.PracticalInsightsintotheDesignofFutureDisasterResponseTrainingSimulations.In Proceedings of the 15th International Conference on Information Systems for Crisis Response AndManagement.

[3] H. Artman and Y. Wærn. 1999. Distributed Cognition in an Emergency Co-ordination Center. Cognition,Technology&Work1,4(December1999),237–246.DOI:https://doi.org/10.1007/s101110050020

[4] NiteshBharosa,JinKyuLee,andMarijnJanssen.2010.Challengesandobstaclesinsharingandcoordinatinginformationduringmulti-agencydisaster response: Propositions from field exercises. InfSystFront 12, 1(March2010),49–65.DOI:https://doi.org/10.1007/s10796-009-9174-z

[5] R.Bierhals,I.Schuster,P.Kohler,andP.Badke-Schaub.2007.Sharedmentalmodels—linkingteamcognitionandperformance.CoDesign3,1(March2007),75–94.DOI:https://doi.org/10.1080/15710880601170891

[6] GregoryA.BigleyandKarleneH.Roberts.2001.TheIncidentCommandSystem:High-ReliabilityOrganizingfor Complex and Volatile Task Environments. AMJ 44, 6 (December 2001), 1281–1299.DOI:https://doi.org/10.5465/3069401

[7] GregorBroll and SteveBenford. 2005. SeamfulDesign for Location-BasedMobile Games. InEntertainmentComputing-ICEC2005(LectureNotesinComputerScience),155–166.

[8] JenniferL.Burke,RobinR.Murphy,MichaelD.Coovert,andDawnL.Riddle.2004.MoonlightinMiami:FieldStudy of Human-Robot Interaction in the Context of an Urban Search and Rescue Disaster ResponseTraining Exercise. Human–Computer Interaction 19, 1–2 (June 2004), 85–116.DOI:https://doi.org/10.1080/07370024.2004.9667341

[9] Ann Frances Cameron and Jane Webster. 2005. Unintended consequences of emerging communicationtechnologies:InstantMessagingintheworkplace.ComputersinHumanBehavior21,1(January2005),85–103.DOI:https://doi.org/10.1016/j.chb.2003.12.001

[10] JanisA.Cannon-Bowers,EduardoSalas,andSharolynConverse.1993.Sharedmentalmodelsinexpertteamdecisionmaking.InIndividualandgroupdecisionmaking:Currentissues.LawrenceErlbaumAssociates,Inc,Hillsdale,NJ,US,221–246.

[11] MichaelD.CardwellandPatrickT.Cooney.2000.Nationwideapplicationof the incidentcommandsystem:Standardizationisthekey.FBIL.EnforcementBull.69,(2000),10.

[12] JohnM.Carroll,MaryBethRosson,GregorioConvertino,andCraigH.Ganoe.2006.Awarenessandteamworkin computer-supported collaborations. Interact Comput 18, 1 (January 2006), 21–46.DOI:https://doi.org/10.1016/j.intcom.2005.05.005

[13] J. Casper and R. R. Murphy. 2003. Human-robot interactions during the robot-assisted urban search andrescue response at the World Trade Center. IEEETransactions on Systems,Man, and Cybernetics, Part B(Cybernetics)33,3(June2003),367–385.DOI:https://doi.org/10.1109/TSMCB.2003.811794

[14] M. Chalmers, I. MacColl, and M. Bell. 2003. Seamful design: showing the seams in wearable computing.(January2003),11–16.DOI:https://doi.org/10.1049/ic:20030140

[15] MatthewChalmersandAretiGalani.2004.SeamfulInterweaving:HeterogeneityintheTheoryandDesignofInteractive Systems. In Proceedings of the 5th Conference on Designing Interactive Systems: Processes,Practices,Methods,andTechniques(DIS’04),243–252.DOI:https://doi.org/10.1145/1013115.1013149

[16] Victor Cheung, Nader Cheaib, and Stacey D. Scott. 2011. Interactive Surface Technology for a MobileCommand Centre. In CHI ’11 Extended Abstracts on Human Factors in Computing Systems (CHI EA ’11),1771–1776.DOI:https://doi.org/10.1145/1979742.1979843

[17] ApoorveChokshi,TeddySeyed,FranciscoMarinhoRodrigues,andFrankMaurer.2014.ePlanMulti-Surface:AMulti-Surface Environment for Emergency Response Planning Exercises. InProceedingsoftheNinthACMInternational Conference on Interactive Tabletops and Surfaces (ITS ’14), 219–228.DOI:https://doi.org/10.1145/2669485.2669520

[18] HerbertH.Clark.1996.UsingLanguage.CambridgeUniversityPress.[19] JosephL.CooperandMichaelA.Goodrich.2006. Integrating critical interfaceelements for intuitive single-

display aviation control of UAVs. In Enhanced and Synthetic Vision 2006, 62260B.DOI:https://doi.org/10.1117/12.666341

[20] LauraDabbish,RobertKraut,andJordanPatton.2012.CommunicationandCommitmentinanOnlineGameTeam. InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems (CHI ’12),879–888.DOI:https://doi.org/10.1145/2207676.2208529



[21] AudreyDesjardins,SaulGreenberg,RonWakkary,and JeffHambelton.2016.AvalancheBeaconParks:SkillDevelopment andTeamCoordination in aTechnologicalTrainingGround. InProceedingsofthe19thACMConference on Computer-Supported Cooperative Work & Social Computing (CSCW ’16), 872–886.DOI:https://doi.org/10.1145/2818048.2835200

[22] AudreyDesjardins,CarmanNeustaedter,SaulGreenberg,andRonWakkary.2014.CollaborationSurroundingBeaconUseDuringCompanionAvalancheRescue. InProceedingsofthe17thACMConferenceonComputerSupported Cooperative Work & Social Computing (CSCW ’14), 877–887.DOI:https://doi.org/10.1145/2531602.2531684

[23] PatrickDohertyandPiotrRudol.2007.AUAVSearchandRescueScenariowithHumanBodyDetectionandGeolocalization.InAI2007:AdvancesinArtificialIntelligence(LectureNotesinComputerScience),1–13.

[24] PaulDourishandVictoriaBellotti.1992.AwarenessandCoordinationinSharedWorkspaces.InProceedingsof the 1992 ACM Conference on Computer-supported Cooperative Work (CSCW ’92), 107–114.DOI:https://doi.org/10.1145/143457.143468

[25] Mica R. Endsley. 1995. Toward a Theory of Situation Awareness in Dynamic Systems.HumFactors 37, 1(March1995),32–64.DOI:https://doi.org/10.1518/001872095779049543

[26] MicaR.Endsley.2011.DesigningforSituationAwareness:AnApproachtoUser-CenteredDesign,SecondEdition(2nded.).CRCPress,Inc.,BocaRaton,FL,USA.

[27] StephenM.FioreandEduardoSalas.2004.Whyweneedteamcognition. InTeamcognition:Understandingthe factors that drive process andperformance. American Psychological Association, Washington, DC, US,235–248.DOI:https://doi.org/10.1037/10690-011

[28] ElenaGabor.2015.Wordsmatter:radiomisunderstandingsinwildlandfirefighting.Int.J.WildlandFire24,4(June2015),580–588.DOI:https://doi.org/10.1071/WF13120

[29] WilliamW.Gaver.1991.SoundSupportForCollaboration. InProceedingsoftheSecondEuropeanConferenceon Computer-Supported Cooperative Work ECSCW ’91, Liam Bannon, Mike Robinson and Kjeld Schmidt(eds.).SpringerNetherlands,Dordrecht,293–308.DOI:https://doi.org/10.1007/978-94-011-3506-1_22

[30] MichaelA.Goodrich,BryanS.Morse,DamonGerhardt,JosephL.Cooper,MorganQuigley,JulieA.Adams,andCurtis Humphrey. 2008. Supporting wilderness search and rescue using a camera-equipped mini UAV.JournalofFieldRobotics25,1–2(2008),89–110.DOI:https://doi.org/10.1002/rob.20226

[31] Michael A. Goodrich and Alan C. Schultz. 2007. Human-robot Interaction: A Survey. Found. TrendsHum.-Comput.Interact.1,3(January2007),203–275.DOI:https://doi.org/10.1561/1100000005

[32] CarlGutwinandSaulGreenberg. 1998.Design for individuals, design for groups: tradeoffsbetweenpowerandworkspaceawareness.(1998).

[33] Carl Gutwin and Saul Greenberg. 2001. The importance of awareness for team cognition in distributedcollaboration.(2001).

[34] Carl Gutwin and Saul Greenberg. 2002. A Descriptive Framework ofWorkspace Awareness for Real-TimeGroupware. Computer Supported Cooperative Work (CSCW) 11, 3–4 (September 2002), 411–446.DOI:https://doi.org/10.1023/A:1021271517844

[35] Stephen E. Hannestad. 2005. Incident command system: A developing national standard of incidentmanagementintheUS.InProcofISCRAMConference.

[36] Christian Heath, Marina Jirotka, Paul Luff, and Jon Hindmarsh. 1994. Unpacking collaboration: theinteractionalorganisationoftradinginacitydealingroom.ComputSupportedCoopWork3,2(June1994),147–165.DOI:https://doi.org/10.1007/BF00773445

[37] Christian Heath and Paul Luff. 1992. Collaboration and control: Crisis management and multimediatechnologyinLondonUndergroundLineControlRooms.ComputSupportedCoopWork1,1–2(March1992),69–94.DOI:https://doi.org/10.1007/BF00752451

[38] JamesHollan,EdwinHutchins,andDavidKirsh.2000.DistributedCognition:TowardaNewFoundationforHuman-computer Interaction Research. ACM Trans. Comput.-Hum. Interact. 7, 2 (June 2000), 174–196.DOI:https://doi.org/10.1145/353485.353487

[39] EdwinHutchins.1995.CognitionintheWild.MITPress.[40] Brennan Jones, Kody Dillman, Richard Tang, Anthony Tang, Ehud Sharlin, Lora Oehlberg, Carman

Neustaedter,andScottBateman.2016.ElevatingCommunication,Collaboration,andSharedExperiencesinMobileVideoThroughDrones. InProceedingsofthe2016ACMConferenceonDesigningInteractiveSystems(DIS’16),1123–1135.DOI:https://doi.org/10.1145/2901790.2901847

[41] Brennan Jones, Anthony Tang, and Carman Neustaedter. 2019. Drones for Remote Collaboration inWildernessSearchandRescue.In1stInternationalWorkshoponHuman-DroneInteraction.RetrievedJune7,2019fromhttps://hal.archives-ouvertes.fr/hal-02128391

[42] Justice Institute of British Columbia. 1999. Ground Search and Rescue (GSAR) Manual (2nd ed.). JusticeInstitute of British Columbia. Retrieved fromhttp://www.jibc.ca/sites/default/files/emd/pdf/SAR100%20GSAR%20Participant%20Manual.pdf

[43] Justice Institute of British Columbia. 2015. Search and Rescue Management Level 1 Participant Manual(Selected Pre-Read Material). Retrieved fromhttp://www.jibc.ca/sites/default/files/emd/pdf/EMRG_1783_PreRead_Chapters_for_Web_20150624.pdf



[44] Shunichi Kasahara and Jun Rekimoto. 2014. JackIn: Integrating First-person Viewwith Out-of-body VisionGeneration for Human-human Augmentation. In Proceedings of the 5th Augmented Human InternationalConference(AH’14),46:1–46:8.DOI:https://doi.org/10.1145/2582051.2582097

[45] Shunichi Kasahara and Jun Rekimoto. 2015. JackIn Head: Immersive Visual Telepresence System withOmnidirectionalWearableCameraforRemoteCollaboration. InProceedingsofthe21stACMSymposiumonVirtual Reality Software and Technology (VRST ’15), 217–225.DOI:https://doi.org/10.1145/2821592.2821608

[46] Md.NafizHasanKhanandCarmanNeustaedter.2019.AnExploratoryStudyoftheUseofDronesforAssistingFirefighters During Emergency Situations. InProceedingsofthe2019CHIConferenceonHumanFactorsinComputingSystems(CHI’19),272:1–272:14.DOI:https://doi.org/10.1145/3290605.3300502

[47] H.Kitano,S.Tadokoro,I.Noda,H.Matsubara,T.Takahashi,A.Shinjou,andS.Shimada.1999.RoboCupRescue:search and rescue in large-scale disasters as a domain for autonomous agents research. In IEEESMC’99Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat.No.99CH37028),739–743vol.6.DOI:https://doi.org/10.1109/ICSMC.1999.816643

[48] RobertJ.Koester.2008.LostPersonBehavior:ASearchandRescue.dbsProductionsLLC.[49] Bernard Osgood Koopman. 1980. Search and screening: general principles with historical applications.

PergamonPressNewYork.[50] Wendy E.MacKay. 1999. Is Paper Safer? TheRole of Paper Flight Strips in Air Traffic Control.ACMTrans.

Comput.-Hum.Interact.6,4(December1999),311–340.DOI:https://doi.org/10.1145/331490.331491[51] JohnE.Mathieu,ToniaS.Heffner,GeraldF.Goodwin,EduardoSalas,andJanisA.Cannon-Bowers.2000.The

influenceof sharedmentalmodels on teamprocess andperformance. JournalofAppliedPsychology 85, 2(2000),273–283.DOI:https://doi.org/10.1037/0021-9010.85.2.273

[52] SvenMayer,LarsLischke,andPawelW.Woźniak.2019.Drones forSearchandRescue. In1stInternationalWorkshoponHuman-DroneInteraction. Retrieved June 7, 2019 from https://hal.archives-ouvertes.fr/hal-02128385

[53] SusanMohammed,RichardKlimoski,andJoanR.Rentsch.2000.TheMeasurementofTeamMentalModels:We Have No Shared Schema. Organizational Research Methods 3, 2 (April 2000), 123–165.DOI:https://doi.org/10.1177/109442810032001

[54] R. R. Murphy. 2004. Human-robot interaction in rescue robotics. IEEE Transactions on Systems,Man, andCybernetics, Part C (Applications and Reviews) 34, 2 (May 2004), 138–153.DOI:https://doi.org/10.1109/TSMCC.2004.826267

[55] EmilyS.Patterson, JenniferWatts-Perotti*,andDavidD.Woods.1999.VoiceLoopsasCoordinationAids inSpaceShuttleMissionControl.ComputerSupportedCooperativeWork(CSCW)8,4 (December1999),353–371.DOI:https://doi.org/10.1023/A:1008722214282

[56] Christian Reuter, Thomas Ludwig, and Patrick Mischur. 2018. RescueGlass: Collaborative ApplicationsinvolvingHead-MountedDisplays forRedCrossRescueDogUnits.ComputSupportedCoopWork (October2018).DOI:https://doi.org/10.1007/s10606-018-9339-8

[57] YvonneRogers.1997.Abriefintroductiontodistributedcognition.RetrievedJuly24,(1997),1997.[58] Gavriel Salomon. 1997. Distributed Cognitions: Psychological and Educational Considerations. Cambridge

UniversityPress.[59] TonySalvador, JeanScholtz,andJamesLarson.1996.TheDenverModelforGroupwareDesign.SIGCHIBull.

28,1(January1996),52–58.DOI:https://doi.org/10.1145/249170.249185[60] J. Scholtz, J. Young, J.L. Drury, and H.A. Yanco. 2004. Evaluation of human-robot interaction awareness in

search and rescue. In 2004 IEEE InternationalConferenceonRoboticsandAutomation,2004.Proceedings.ICRA’04,2327-2332Vol.3.DOI:https://doi.org/10.1109/ROBOT.2004.1307409

[61] Aaron Stafford, Wayne Piekarski, and Bruce Thomas. 2006. Implementation of God-like InteractionTechniquesforSupportingCollaborationBetweenOutdoorARandIndoorTabletopUsers.InProceedingsofthe 5th IEEE and ACM International Symposium onMixed and Augmented Reality (ISMAR ’06), 165–172.DOI:https://doi.org/10.1109/ISMAR.2006.297809

[62] KateStarbirdandLeysiaPalen.2013.WorkingandSustainingtheVirtual“DisasterDesk.” InProceedingsofthe 2013 Conference on Computer Supported Cooperative Work (CSCW ’13), 491–502.DOI:https://doi.org/10.1145/2441776.2441832

[63] Anthony Tang, Omid Fakourfar, Carman Neustaedter, and Scott Bateman. 2017. Collaboration with 360°Videochat: Challenges and Opportunities. In Proceedings of the 2017 Conference onDesigning InteractiveSystems(DIS’17),1327–1339.DOI:https://doi.org/10.1145/3064663.3064707

[64] Zachary O. Toups and Andruid Kerne. 2007. Implicit Coordination in Firefighting Practice: DesignImplications forTeaching Fire EmergencyResponders. InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems(CHI’07),707–716.DOI:https://doi.org/10.1145/1240624.1240734

[65] Murray Turoff, Michael Chumer, Bartel Van de Walle, and Xiang Yao. 2004. The design of a dynamicemergencyresponsemanagementinformationsystem(DERMIS).JournalofInformationTechnologyTheoryandApplication(JITTA)5,4(2004),3.

[66] USDepartmentofHomelandSecurity.2008.NationalIncidentManagementSystem.



[67] Himanshu Zade, Kushal Shah, Vaibhavi Rangarajan, Priyanka Kshirsagar, Muhammad Imran, and KateStarbird.2018.FromSituationalAwarenesstoActionability:TowardsImprovingtheUtilityofSocialMediaData for Crisis Response. Proc. ACM Hum.-Comput. Interact. 2, CSCW (November 2018), 195:1–195:18.DOI:https://doi.org/10.1145/3274464

[68] AmyX. Zhangand JustinCranshaw.2018.MakingSenseofGroupChatThroughCollaborativeTaggingandSummarization. Proc. ACM Hum.-Comput. Interact. 2, CSCW (November 2018), 196:1–196:27.DOI:https://doi.org/10.1145/3274465

[69] FHWAOfficeofOperations-Glossary:SimplifiedGuidetotheIncidentCommandSystemforTransportationProfessionals.RetrievedApril2,2019fromhttps://ops.fhwa.dot.gov/publications/ics_guide/glossary.htm

[70] Small,UnmannedAircraftSearchForSurvivors InKatrinaWreckage.ScienceDaily.Retrieved June11,2019fromhttps://www.sciencedaily.com/releases/2005/09/050915003020.htm

[71] Lost Person Behavior – Apps on Google Play. Retrieved June 13, 2019 fromhttps://play.google.com/store/apps/details?id=com.azimuth_1.lpb&hl=en_CA

[72] goTenna Pro - Lightweight, Low-cost Tactical Mesh-Networking Comms. goTennaPro. Retrieved June 20,2019fromhttps://gotennapro.com/

ReceivedJune2019;revisedOctober2019;acceptedNovember2019.

Documents

Remote Communica-on in Wilderness Search and Rescue ...hcitang.org/papers/2020-group2020-wsar.pdf · Making information available in multiple formats (e.g., audio, pictures, videos)