loadporntubemovies

Coco'sVideos:AnEmpiricalInvestigationofVideo-PlayerDesignFeaturesandChildren'sMediaUseAlexisHiniker1,2,SharonS.
Heung2,Sungsoo(Ray)Hong2,JulieA.
Kientz21.
InformationSchool,UniversityofWashington,Seattle,WA,USA2.
HumanCenteredDesignandEngineering,UniversityofWashington,Seattle,WA,USA{alexisr,heungs,rayhong,jkientz}@uw.
eduABSTRACTInthisstudy,wepresentCoco'sVideos,avideo-viewingplatformforpreschoolersdesignedtosupporttheminlearn-ingtoself-managetheirmediaconsumption.
Wereportre-sultsfromathree-weekexperimentaldeploymentin24homesinwhichpreschoolersusedthreedifferentversionsoftheplatform:onethatisneutraltothelimitstheyset,onethatenforcesthelimitstheyset,andonethatattemptstoerodethelimitstheysetbyautomaticallyplayingadditionalcon-tentaftertheplannedcontentisfinished("post-play").
Wefoundthatpost-playsignificantlyreducedchildren'sauton-omyandlikelihoodofself-regulation,extendedvideo-view-ingtime,andledtoincreasesinparentintervention.
Wefoundthatthelock-outmechanismdidnotreducevideo-viewingtimeorthelikelihoodofparentintervention.
To-gether,ourresultssuggestthatavoidingplatformsthatworktounderminetheuser'sintentionsismorelikelytohelpchil-drenself-regulatetheirmediausethanrigidparentalcon-trols.
AuthorKeywordsChildren,mediause,interactiondesign&children,parentalmediation,preschoolers,child-computerinteraction,paren-talcontrols,auto-play,post-play.
ACMClassificationKeywordsH.
5.
m.
Informationinterfacesandpresentation(e.
g.
,HCI):Miscellaneous;INTRODUCTIONEntertainmentmediaplaysacentralroleinthelivesofyoungchildren,withtheaveragepreschoolerwatchingmorethanthreehoursoftelevision,film,andothervideoprogrammingeachday[10,31].
Children'stelevisionandvideoscanpro-videproductivelearningopportunitiesforkids[15]anduse-fulrespitesfromcaregivingforadults[24],inadditiontoservingasanenjoyablepartofdailylife.
However,childde-velopmentresearchalsosuggestsfamiliessetlimitsontheamountoftimepreschoolersspendconsumingpassivevideocontent,asheavyviewinghasbeenlinkedtoincreasedriskofobesity[26],reductionsinimaginativeplay[8],andsleepdisruption[14].
Inaddition,thelowcognitivedemandandhigh-rewardexperienceofviewingvideosmakeiteasyforviewerstoengageinextendedconsumption[11].
Inthisstudy,weexaminetheroleofdesignasitrelatestochildren'stransitionstoandfromvideoviewingexperiences.
Aspartofthis,weinvestigatedtwodifferentexistingdesignparadigmswiththepotentialtoinfluencechildren'stransi-tionbehaviors.
First,avarietyofcommercialproductsknownas"parentalcontrols"offertosupportparentsinset-tingandenforcinglimitsonchildren'suseoftechnology,in-cludingcontrolsspecificallytargetingvideoviewing[41].
Second,indirectcontrasttothelimit-enforcinggoalsofpa-rentalcontrols,manypopularplatformsservingvideosin-cludepost-playfeaturesandnext-videosuggestionsthatseektopromotecontinuedviewingandminimizenaturalstoppingpointsbyautomaticallyplayingnewcontentwhenthese-lectedvideoends.
Priorworkhasshownthatparentsfindpost-playandrelatedfeaturesfrustratingandbelievetheseimpedetheirfamily'sabilitytosetboundaries[24].
Theextenttowhicheitherlimit-enforcing(parentalcontrols)orlimit-eroding(post-play)designsinfluencefamilies'be-haviorsinpracticeisnotrobustlyunderstood.
PriorworksuggeststhattheauthoritariandesignandrigidityofparentalFigure1:Theplaylist-buildingscreen:Coco'sVideossup-portsplanningoutaplaylistofvideostowatch.
Here,playlistitemsaccumulateontheleftastheyareselected.
Ontheright,theusercanchoosebetweenoneofthreetabs.
Here,the"Recommended"tabisselected,andrecommendedchannelsareshown.
Theusercandrilldownintoanychanneltoselectindividualvideos.
Permissiontomakedigitalorhardcopiesofallorpartofthisworkforpersonalorclassroomuseisgrantedwithoutfeeprovidedthatcopiesarenotmadeordistributedforprofitorcommercialadvantageandthatcopiesbearthisnoticeandthefullcitationonthefirstpage.
Copyrightsforcom-ponentsofthisworkownedbyothersthantheauthor(s)mustbehonored.
Abstractingwithcreditispermitted.
Tocopyotherwise,orrepublish,topostonserversortoredistributetolists,requirespriorspecificpermissionand/orafee.
RequestpermissionsfromPermissions@acm.
org.
CHI2018,April21–26,2018,Montreal,QC,Canada2018Copyrightisheldbytheowner/author(s).
Publicationrightsli-censedtoACM.
ACM978-1-4503-5620-6/18/04…$15.
00https://doi.
org/10.
1145/3173574.
3173828controlsareunlikelytobestservefamilies'needs[30,32,41],whileotherworkhasshownthattheseinterfacesareoftenpoorlyunderstoodanddifficulttouse[21,29].
Tothebestofourknowledge,nopriorworkhasexaminedtheeffectpost-playinthecontextoflimit-setting,despitethefactthatitisacommonfeatureofInternetvideo-on-demandplatformssuchasNetflixandYouTube.
Weundertookthecurrentprojectinpursuitoftwospecificgoals.
Thefirstwastobuildonourpastdesignwork[22]tocreateavideo-viewingplatformtosupportpreschoolersinplanningtheirmediaconsumptionwithintention.
Thesec-ondwastoconductanexperimentalstudytounderstandhowdesignsintendedtoeitherenforceorerodefamilies'limitsinfluenceparents'andchildren'sexperienceswiththissys-teminareal-worldcontext.
Todothis,wecreated"Coco'sVideos,"whichwedeployedinthehomesof24familieswithpreschoolersforthreeweeks.
Weconductedawithin-sub-jectscomparisonoffamilies'responsestothreedifferentver-sionsofthesystem.
Intheseconditions,theplatformalterna-tively:1)remainedagnostictofamilies'limits,2)activelyattemptedtoenforcefamilies'limitswithalock-outmecha-nism,or3)activelyattemptedtounderminefamilies'limitswithapost-playmechanism.
Wefound,first,thatchildrenengagedwiththecoreexperi-encewithintentionanddisplayedautonomousdecision-makingastheyplannedtheirvideoviewing.
Inaddition,childrentookownershipoftheirtransitionexperienceastheyendedtheirviewingandmovedontootheractivities.
Sec-ond,wefoundthatpost-playextendedchildren'sviewingtime,ledtoincreasedinterventionfromparents,reducedev-idenceofchildren'sautonomy,andwasperceivednegativelybyparentsrelativetotheotherversionsofthesystem.
Third,wefoundthatthelock-outmechanismdidnotappeartore-ducechildren'sautonomy,althoughitalsodidnotreduceviewingtimeorincreasetheneedforparentstointervene.
Asmediacorporationsincreasinglyseektoengageandmon-etizetheattentionoftheirpreschoolaudience[7],itisusefulforthedesigncommunitytounderstandhowspecificdesignchoicesinfluencechildren'spatternsofengagementanddis-engagement.
Whilechildren'smediaplaysapositiveroleindailylifeformany,families'usagepatternswillalwaysnec-essarilyincludebothdisconnectingandconnecting.
Thecon-tributionofthisworkistosupportdesignersinunderstand-inghowtheirchoicesmayinfluencechildren'sabilitytoau-tonomouslyself-regulatetheiruseofmediaandparents'self-efficacyinsupportingtheirchildren'smediahabits.
RELATEDWORKPreschoolers,MediaUse,andMediaEffectsMediauseinearlychildhoodhasbeenthesubjectofexten-sivescholarship.
Priorworkhasshownthatmostchildrenusedevicesasinfantsandtoddlers,including97%ofchil-drenundertheageoffour[27].
Formanychildren,mediauseisaroutinepartofdailylife,withtheaveragepreschoolerspendingapproximatelythreehourswithtechnologyeveryday[10,31,47].
Althoughmoderntechnologyoffersawidevarietyofdigitalactivitiesrangingfromvideochattovirtualreality,watchingvideosispreschoolers'mostcommondig-italpastime[24,47].
Alargebodyofworkinchilddevelopmenthasexaminedtheeffectsofmediauseonchildren'swellbeing,linkingexten-sivefast-pacedvideocontenttoattentiondisorders[12,42]andviolentmediaexposuretoincreasesinhostilityandtheperceptionofhostilityinothers[17,40].
Otherworkhasex-ploredthepotentialoftelevisionandothermediatosupportpositivegrowth,includinglearning,creativity,andprosocialbehavior.
Thesestudieshaveshown,forexample,thatvideocanenablesocialandemotionallearning[34]aswellasskilldevelopmentandschool-readiness[15,36],especiallywhencoupledwithparentsupport.
Risk,Fear,andParentalControlsChildren'seageradoptionoftechnologyhasinspiredsomesocialconcernandanumberofpubliccampaignstolimitchildren'sexposuretodigitalmedia[6,9,33,43].
Althoughthelargeamountoftimechildrenspendwithtechnologycallsforexaminationsofwaysinwhichtechnologyusein-fluenceschildren'slong-termwellbeing[38],itisusefultonotethatthesequestionsareaskedwithinaculturalframeofreferencethatforegroundsrisktochildren[25],andindoingso,mayoverlookopportunitycostsandcreatenewburdensforfamilies.
Giventhiscontext,itisunsurprisingthatare-centreviewofexistingparentalcontrolsreportsthatthesetoolshavearisk-averseorientationandfocusexclusivelyonrestrictingandcontrollingchildren'sbehavior[41].
Asare-sult,otherworkinHCIhascalledforare-imaginingofthedesignofparentalcontrols[32]tomovebeyondrestrictingchildrenandtowardmentoringchildrenindevelopinghealthyandproductivemediahabits.
Inthatspirit,recentdesignresearchinHCIhasexploredin-terfacestosupportparentsandchildreninselectingandfil-teringcontentcollaboratively[20]andworkingtogethertoplanthewaysinwhichtheywilluseappsandgames[22].
Otherworkhasinvestigateddesigningforparentsandchil-drentousedigitalmediatogether[5,39].
Anethnographyoffamilies'technologypracticessuggeststhatthissupportiveapproachmaybemoreusefulforfamiliesthantraditionalpa-rentalcontrols.
Theauthorsfoundthatfamilies'boundariesarefluidandunlikelytobebest-servedbytoolsthatassumepre-definedanddeterministiclimits[30].
Webuiltonthispastworkbyfirstconductingdesignre-searchtounderstandfamilies'responsestotoolsforpromot-ingintentionalityandautonomyratherthancompliance.
Tohelpfurtherourunderstandingoftheeffectivenessofparen-talcontrols,wealsoexaminetheresultofaddingalock-outmechanismtothisexperience.
SupportingChildren'sAutonomyBythetimechildrenturnthree,theyhavetheemergentabil-itytoplan,setgoals,andchoosetheirownactionswithin-tention[13].
Evidence-basedpreschoolcurriculasupportchildrenindevelopingtheseskills[3,37],andexpertsined-ucationandchilddevelopmentrecommendthatpreschoolersregularlyhavetheopportunitytoplanhowtheywillspendtheirtimeandreceivefeedbackaboutthebehaviorstheyen-actrelativetotheirself-definedplans.
Bydoingso,childrenstrengthentheirabilitytomakepurposefulchoicesandtheirabilitytoself-regulatetheirbehavior[13].
Similarly,self-determinationtheoryexplainsthatchildren'sabilitytoself-regulaterequiresintrinsicmotivationtoenactspecificbehaviorsandinternalizationofnorms[19].
Under-standingsocialexpectationsandvaluesandthenhavingtheopportunitytoautonomouslychoosetoenactthesenormsen-ableschildrentoexerciseexecutivefunction,controlim-pulses,andadopttheexpectationsoftheircommunity.
Incontrast,controllingchildrenanddemandingcomplianceworkstounderminetheirsenseofautonomyandabilitytoself-regulate[28],suggestingthatparentalcontrolsmayinfactdiminishchildren'sself-controlwithrespecttomediause.
Webuildonthisexistingworkbydesigningasystemintendedtoforegroundchildren'schoice-makingandpromptthemtosetgoals,ratherthanrestrictingtheirbehavior.
SYSTEMDESIGNDesignProcessThisworkwasinformedbyaseriesofinterviewswith27parentsofpreschoolersandadiarystudywithanother28familiesexploringtheirexperiencesendingtheirtimewithentertainmentmediaandtransitioningbacktothephysicalworld.
Aspartoftheseinterviews,parentsreviewedasetofstoryboardsofdesignideasfortransitionsupport.
Story-boardsincludeddesignsinwhichthetechnologythechildwasusing"gottired"andtoldthechilditneededtogotosleep,apairedsystemthatallowedtheparenttomakechangestoachild'stabletsettingsthroughaseparatephone,timers,andotherfeatures.
Weevolvedthesestoryboardsasweconductedadditionalinterviewstoincorporatepartici-pants'feedback.
Wealsoconductedaparticipatorydesignworkshopwithsevenpreschoolersinourtargetagerange[23]andalabstudyofarelatedsystemtoexaminehowchildrenthinkabouttheirtabletuseandhowtheyrespondtonudgesfromthesystem[22].
Ourfinaldesignincorporatedinsightsfromallthreeofthesedesignprojects.
Weexpandonourpriorworkinthelab[22]bycreatinganewsystemthatfocusesonvideoconsumption(ratherthanappsorgames),deployingthissysteminthewild(ratherthanevaluatingitinthelab),and,exploringthesysteminrelationtopost-playandparen-talcontrols.
Coco'sVideosTheresultofthisdesignworkwasanappforAndroidde-vicesthatwecalled"Coco'sVideos.
"Coco'sVideosallowstheusertobuildandplayaplaylistofvideos,drawingcon-tentfromYouTubeviatheYouTubeAPI.
Uponopeningtheapp,theonlyinteractivecontentisa"play"buttontobegintheexperience.
Pressingthisbuttontakestheusertoascreenwithaclock(Figure2,left)whereacartooncharacternamedCocoasks,"Hithere!
Howmuchtimeshouldwespendwatchingvideos"A"next"buttonallowstheusertomovepastthistimelimitscreenanddisplaysasetofninepossible"nextactivities"fromwhichtheusercanselect(Figure2,middle).
Asthesearepresented,Cocosays,"Whenyou'redonewatchingvid-eos,whatwillyoudonext"Thecategoriesofactivities,suchas"Readabook,""Playoutside,"and"Leavethehouse"weregeneratedbyclusteringnext-activitiesthatchildrenen-gagedinafterusingtechnologyasreportedin381diaryen-trieswecollectedaspartofaseparateprojectonchildren'smediause.
Wealsoincluded"Somethingelse!
"asaflexiblecatch-allforactivitiesthatmightnotfitourcategorizations.
Theusercanselectanyonecategory,switchingasmanytimesasneeded.
Assoonasacategoryisselected,a"next"buttonbecomesvisibleinthebottomrightcornerofthescreen.
Pressing"next"fromthisscreentakestheusertoavideo-selectionscreen(Figure1)whereCocosays,"Ok!
Nowlet'spicksomevideos.
"Apaneontherightsideofthescreendis-playsthreedifferenttabs,eachofwhichcanbeusedtofindcontent:History:Thedefaulttabisthehistorytab,whichdisplaysthemost-recentlyviewedvideosasthumbnailswithtitlesanddurationslistedinreversechronologicalorderofviewing.
Iftheapphasneverbeenusedtowatchvideos,nocontentisdisplayed.
Recommended:UsingratingsfromCommonSenseMe-dia[44],weselectedtwelvedifferentYouTubechannelswithparticularlyhigh-qualitycontentdesignedforourtar-getagerange,includingCBeebies,DanielTiger'sNeigh-borhood,PBSKids,andCBCKids.
Ourrecommendationstabdisplayseachofthesechannelsinagridasathumbnailandtitle(Figure1).
Tappinganyoneofthesechannelsdis-playsagridofitsvideos.
Search:Thesearchtabdisplaysatextboxwithamicro-phoneiconandamagnifyingglass,givingtheuserabilitytosearchbytypingorbyspeaking(weenabledtextsearch-ingwiththeassumptionthatonlyparentswouldusethisfeatureandexpectedchildrenonlytousevoicesearch).
SearchingusesYouTube'sAPItoreturnthetop21most-relevantsearchitems,displayedasthumbnailsinagridwithtitlesanddurations.
Ausercanloadindefinitelymanyadditionalresults(inbatchesof21)asneeded.
Tappingavideothumbnailinanyofthesetabsaddsthevideotoapersistentplaylistontheleftsideofthescreen.
Onceadded,anyvideocanalsoberemoved.
Onlyvideoswhosedurationfitswithintheremainingbudgetofthetotaltimelimitcanbeadded,andasthetotaldurationoftheplaylistgrows,ineligiblevideosaregrayedoutanddisabled.
Ifausertapsadisabledthumbnail,Cocosays,"Wedon'thaveenoughtimeforthatone!
"Pressinga"play"buttonatthebottomoftheplaylisttransitionstoafull-screenvideoplayerFigure2:Screenshotsforcreatingandwatchingaplaylist.
Left:Thefirstscreenoftheexperienceallowsparentandchildtochoosethedurationoftheexperience.
Middle:Selectingtheactivitythatwillfollow,suchasreadingabook,goingoutsideortakingabath.
Right:Thevideoplayer.
Oneminutebeforetheplaylistends,awarningnotifiestheuseroftheupcomingtransition.
whichplaysthrougheachvideointheplaylistintheorderspecifiedbytheuser.
Ifthereismorethanonevideointheplaylist,atthestartofthelastvideo,thevideocontentpausesandanoverlayisaddedabovethecontentthatdisplaysanalarmclockandamessage.
Audioplays:"We'realmostdone!
Getreadytosaygoodbyewhenthisvideoends.
"Tappinganywhereonthescreendismissesthiswarningandresumesplayingthevideo,butthevideoneverrestartsunlessthescreenistapped.
Ifthisfinalvideoismorethanoneminutelong,thenoneminutebeforeitends,thevideopausesagainandanoverlaywithanalarmclockisagaindisplayedabovethecontent(Figure2,right).
Audioplays,"Oneminuteleft,thenit'stimetosaygoodbye.
"Onceagain,thevideodoesnotresumeuntiltheusertapsthescreen.
Whentheplaylistends,atransitionscreenappearsdisplay-ingtextandanimagethatcorrespondtothenext-activitythechildselectedwhenconstructingtheplaylist(e.
g.
,takingabath,eating,orplayingwithtoys).
Cocosays,"Nowit'stimeto[nextactivity].
Areyoureadyto[nextactivity]"Asde-scribedinthemethodssection,wecreatedthreevariantsofthistransitionscreentorepresentourthreestudyconditions.
ThedefaultversionofthetransitionscreenisshowninFigure3,left.
METHODSParticipantsTwenty-fourfamiliescompletedallproceduresandwerein-cludedinourfinalsample.
Theseparticipantswererecruitedthroughaninstitutionalparticipantpooldrawnfromregionalbirthrecords.
Allfamilieshadatleastonechildbetweentheagesof3and5yearsold(inclusive)whowasconsideredthetargetofthestudy.
Inadditiontothisagespecification,asec-ondinclusioncriterionwasthattargetchildrenusescreenmediaatleasttwiceperweekonaverage.
Ourdataover-sam-pledwhitechildrenandfamilieswithtwomarriedparents,andhouseholdincomeamongparticipantswashigherthantheregionalmedianof$80,349[1].
Participantde-mographicsareshowninTable1.
ProceduresWeconductedanexperimentalstudywitharandomizedcompleteblockdesign,whereeachparticipantwastreatedasablockandexperiencedeachofthreeexperimentalcondi-tions.
Parentsfromtheparticipantpoolwhorespondedtoaphonesolicitationweredirectedtoascreenersurvey,com-posedofquestionsaboutfamilydemographicsandthetargetchild'smediause.
FamiliesthenreceivedaGalaxyTabEAndroidtabletbymailwithacopyofCoco'sVideospre-installedandconfig-ured.
Familieswereaskedtogivethetargetchildtheoppor-tunitytouseCoco'sVideosatleastfivetimesperweek,foratleastfiveminuteseachtime,foraperiodofthreeweeks.
Theyweretoldthattheywerewelcometousetheappasmuchastheywishedaboveandbeyondthatlowerbound.
Asathank-youfortheirparticipation,familieskeptthetabletattheconclusionofthestudy.
Overthecourseofthethree-weekstudyperiod,theappro-tatedbetweeneachofthreedifferentconditions:neutral,controlled,andpost-play.
Theexperienceineachofthesethreeconditionswasidentical,exceptfortheelementsdis-playedonthetransitionscreen.
Thatis,inallthreeconditionschildrensetatimelimit,choseaplannednextactivity,andconstructedaplaylistintheexactsameway.
Thetransitionscreendifferedacrossthethreeconditionsinthefollowingways:Neutral:Intheneutralcondition,alarge"home"buttonwasalsodisplayedinthebottomrightcornerofthescreen.
ChildGenderMale(58%),Female(42%)ChildAgeMean(sd)=3.
6(0.
92)years;Age3(N=16),Age4(N=5),Age5(N=4)ChildRaceWhite(88%),Asian(4%),Mixed(8%)Householdincome(US$)25–50K(8%),50–75K(4%),75–100K(21%),100–125K(21%),125–150K(8%),150Kormore(38%)ParentEducationHighSchool(8%),TradeSchool(4%),AssociateDegree(4%),Bachelor'sDe-gree(42%),AdvancedDegree(42%)ParentMaritalStatusPartnership(8%),Married(92%)Table1:ParticipantDemographicsPressingthisbuttonbroughtthechildbacktothebegin-ningoftheexperienceandenabledthechildtocreateanewplaylist(Figure3,left).
Post-play:BorrowingterminologyfromtheNetflixfea-ture[45],wecreateda"post-play"condition.
Here,chil-drensawtheexactsamescreenasintheneutralcondition(includingthehomebutton),andtheyalsosawasmallem-beddedvideoplayerinthetoprightcorner.
Whenthetran-sitionscreenappeared,theappqueriedtheYouTubeAPIforrecommendationsbasedonthefinalvideointheplaylistthathadjustcompleted.
Theserecommendationswerethenplayedseriallyinthissmallembeddedplayer.
Thechildcouldmaketheplayerfull-screenorotherwiseengagewiththetoolbar(e.
g.
,topauseoradvancethecon-tent)(Figure3,middle).
Controlled:Inthecontrolledcondition,childrensawtheexactsamescreenasintheneutralcondition,exceptnohomebuttonwaspresent.
Asnoelementofthescreenwasinteractive,theuserwaslockedoutoftheapponcetheyreachedthispoint.
Afterathree-minutedelay,theappwouldresetandreturntothehomescreenautomatically(Figure3,right).
Allparticipantssawallthreeoftheseversionsoftheapp,eachforaperiodofoneweek.
Theorderofconditionswascounterbalancedacrossparticipants,andbetween3and5familieswereassignedtoeachpossibleordering.
Theorder-ingwashard-codedintotheappbyparticipantID,andtheappautomaticallydisplayedthecondition-specifictransitionscreenbasedontheIDandstartdate.
Thus,theexperienceautomaticallychangedslightlyeverysevendayswithoutanyupdatesornotifications.
Theappcollectedthreetypesofdataduringthedeployment:Appusage:Theapploggedalloftheuser'sinteractionsandinputs,includingthetimelimitstheyselected,thetermstheysearchedfor,thevideostheychose,theamountoftimetheyspentwatchingvideos,andthetimestampwhentheypressedeachbutton.
Systemusage:Theappalsotrackedusageacrossthede-viceforaperiodofthreeminutesafteraplaylistended,documentingwhichappswereusedandforhowlong.
Wealsologgedwhenthetabletwasturnedoff.
Audiocapture:Beginningoneminutebeforeeachplaylistended,thetabletmicrophonebeganrecordingambientau-diodata.
Thissamplestretchedfromoneminutebeforeuntiltwominutesafterthetransition.
AudiorecordingswerestoredinthecloudusingAmazonWebServicesS3platform.
Inadditiontodiscussingthisaudiosamplingwitheachfamilyduringtheconsentprocess,wealsodis-playedareddotonthescreenanytimetheappwasrecord-ingaudio(andinformedthemthattheycouldlookforthisindicator).
Wealsogavefamiliestheabilitytodeleteanyrecordingtheydidnotwantustoaccess,althoughnofam-ilieschosetodothiswithanyoftheirrecordings.
AudiocapturepersistedforthreeminutesaslongasCoco'sVid-eoswasrunning(e.
g.
,evenifthescreenwasturnedoffortheappwasmovedtothebackground).
Aftercompletingtheexperimentalprocedures,allpartici-pantswereinvitedtocompleteafollow-upsurveyabouttheirexperienceusingCoco'sVideos.
Theseweresentoutinbatches,andbecauseofparticipants'staggeredstartdates,parentsreceivedthesurveybetween2weeksand2monthsaftertheirstudyperiodended.
Inthesurvey,weaskedpar-entstoreportwhattheylikedanddislikedabouttheapp,whattheirchildlikedanddisliked,howtheyfeltaboutthelock-outmechanism,howtheyfeltaboutpost-play,andwhethertheycontinuedusingtheappafterthestudyended,amongotherquestions.
Twentyparentscompletedthesurveyandoneparentcompletedthefirsthalf.
DataAnalysisQuantitativeanalysis:Weusedourapp-usageandsystem-usagelogstoconstructadatasetofdistinctplaylistswatchedbyeachchild.
Foreachplaylist,wefirstnotedwhetherornotthechildactuallysawthetransitionscreen(e.
g.
,ifthechildexitedearly,heorshedidnothaveengageinthedesignedtransitionexperience).
OnlyplayliststhatendedwithoneofFigure3:Transitionscreenexperiences.
Left:Neutralexperiencewherechildhastheoptiontopresstopressthehomebutton(bottom)right.
Middle:Post-playexperiencewherechildhastheoptiontopressthehomebuttonandcreateanewplaylist(bottomright)andvideosrelatedtothemostrecentlyviewedvideoplayautomatically(topright).
Right:Controlledexperiencewherenohomebuttonisdisplayedandnocontentonthescreenisresponsive.
Inallthreeconditions,theactivitythechildselectedashisorhernextactivitywhenconstructingtheplaylistisreflectedonscreen(here,reading).
Inallthreeconditions,whenthescreenisdisplayed,Cocosays,"Nowit'stimetoread.
Areyoureadytoread"thethreetypesoftransitionexperiences(post-play,neutral,andcontrolled)wereincludedinouranalysis(N=411).
Qualitativeanalysis:Thesystemcaptured3-minuteaudiore-cordingsforthetransitionperiodsurrounding332playlists.
In33oftheseinstances,thechildendedthesessionafterthewarningbutbeforethetransitionscreenappeared,andin7instances,theaudiowasnotclearenoughtoanalyze.
Thus,ouraudio-clipdatasetwascomposedoftheremaining292.
Toanalyzeaudioclips,tworesearchersfirstlistenedtoeachclipindividuallyandtranscribedeachone.
Wethendividedtheclipsacrosstheresearchteamandperformedanindivid-ualopencodingtoidentifypotentialthemes,whichweredis-cussedasagroupandevolvedintoaxialcodesbasedonsim-ilaritiesacrossresearchers.
ExamplesoffinalcodesincludedwhetherthechildrespondedtotheCococharacter,whethertheparentintervenedtostopthechildfromwatchingaddi-tionalcontent,whetherthechildandparentagreedtochangetheplannedtransition,andwhetherthechildmadestate-mentsabouthisorherautonomywhentransitioning.
Onere-searchercodedall292audioclipsbasedonthiscodingscheme;asecondresearchercodedarandomlyselected10%toassessinterraterreliability.
Cohen'sκwas.
783.
RESULTSChildren'sUsagePatternsOverthecourseofthreeweeks,participantscollectivelycre-ated597differentplaylistsusingCoco'sVideos,anaverageof24.
88playlistsperchild(SD=9.
25).
Theaverageplaylistlasted18.
44minutes(SD=21.
00)andcontained8.
2videos(SD=9.
60).
Childrenchosetoendtheplaylistearly31%ofthetime,oftentoadjustandrestarttheactivelistofvideos.
Theamountoftimechildrenspentwatchingasingleplaylistrangedfromlessthan1minutetoexactly2hours.
Childrenfrequentlyre-watchedvideostheyhadseenbefore.
Collectively,childrenincluded2,452videosintheirplaylists,and53%ofthesewereselectedfromthe"History"tabthatdisplayedtheir21most-recentlywatchedvideosinreversechronologicalorder.
Thistabwasalsotheonethatwasdisplayedbydefault.
Whenchildrenaddednewcontent,theyweremorelikelytochoosefromour"Recommenda-tions"library(Figure1)thantheyweretosearchforcontentontheirown.
Ofthe1,149videoschildrenselectedthatwerenotalreadyintheirrecenthistory,69%camefromourrec-ommendationslibrary,while31%wereaddedbysearching,suggestingthatpresentingdefaultcontentselectionscouldbeaneffectivemeansoffunnelingchildrentowardspecificex-periences.
TransitionTimingUsingloggeddatafromthedevice,wereconstructedthewayinwhichthetabletwasusedwhenaplaylistended.
Asonemeasureofachild'sresponsetoatransition,wecalculatedtheamountoftimebetweenwhentheplaylistendedandwhentheactivesessionended,whichincludedtimespentwatchingpost-playvideos,timespentcreatingandwatchingnewplaylists,andtimespentusingotherappsonthetablet.
Wethencomparedthisadditionalpost-transitionsessiontimeacrossthethreeconditions.
Becausedatawerenon-nor-mallydistributed,wefirstperformedalogtransformationonpost-transitionsessionduration.
WethenusedablockANOVAtoaccountfornon-independenceofsamples,withcondition(neutral,post-play,controlled)astheindependentvariableandparticipantIDastheblockingfactor.
Thisre-vealedasignificantdifferencebetweengroupsinpost-tran-sitionsessionduration(F(23)=4.
204,p<.
001,η2=.
233).
Posthocanalysisrevealedthatpost-playsessionshadsignif-icantlymorespill-overtimethanneutralsessions(meandif-ferenceinlogs=.
189,95%CI[.
002,.
377],p=.
047)andsignificantlymorespill-overtimethancontrolledsessions(.
322,95%CI[.
119,.
525],p<.
001).
Thedifferencebetweentheneutralandcontrolledconditionswasnotsignificant(.
1328,95%CI[-.
066,.
331]).
ABonferronicorrectionwasappliedtoallcomparisons.
EngagingwithCocoThroughouraudiorecordings,wedocumentedthatchildrenwerehighlyresponsivetoCoco'spresenceandfrequentlyspokedirectlytoCocoand/orrepeatedCoco'swords.
Acrossallconditions,thetransitionscreen(Figure2)wasalwaysac-companiedbyaudioaskingthechildabouthernextactivity.
Forexample,ifachildindicatedwhencreatingtheplaylistthatshewouldreadabookwhenitended,attheendoftheexperienceCocowouldsay,"Nowit'stimetoread.
Areyoureadytoread"Cocomadethisstatementregardlessofcon-dition.
Childrenroutinelyrepliedtothisquestion,speakingdirectlytoCoco.
Forexample,whenCocoaskedonechild,"Areyoureadytoeat"thechildresponded,"Mm-hmm,I'mhungry!
"OtherchildrenrepliedtoCoco'stransitionquestionsayingthingslike,"I'mreadytoplayoutside,""Yes!
""I'mgoingtogetasnack,""Ofcourse[I'mready],""I'mreadytodosomethingelse,"and"Goodbye!
"Inaminorityofinstances,childreninteractedwithCocobypushingbackagainstthenormsshepresented.
Occasionally,childrenrepliedtoCoco'squestionbysayingthingslike,"No,Idon't[wantto],""I'mnot[ready],"or"No!
Nopenopenopenope,no.
"Ouraudiorecordingsalsosuggestthatchildrencametoex-pectthisinteractionwithCocoasaroutinepartoftransition-ing.
Asoneplaylistapproacheditsend,thechildanticipatedCoco'sscriptandaskedhismother,"Isshegonnasay,'Timetosaygoodbye'"Anotherchildresistedputtingthetabletawaybeforethetransitionaudioplayed,saying,"Iwanttoseewhatshe[Coco]says.
"Cocothensaidthatitwastimeforbedandaskedthechild,"Areyoureadyforbed"Thechildresponded,"Yeah,"andthenbeganpreparingforbed.
Athirdchildturnedoffthescreenbeforetheaudiocouldfin-ishplaying,saying,"I'mreadyforlunch.
"Hethenquietlystatedthewordsthatwouldhaveplayedifhehadletthem:"Areyoureadytoeat"suggestingthatthisaudiowasapre-dictablepartofhisroutine.
AnotherchildrespondedtoCoco'squestionbysayingthatshecouldnoteatdinnerbe-causeherfatherhadnotyetcomehomefromwork.
Sheex-plainedtohermother,"Iwanttohear,'It'stimetoeat'again…whenDaddygetshome,"thepointwhenthemealcouldbegin.
Intheseandmanyotherinstances,childrendemonstratedthatthisdialogwithCocobecameameaning-fulandexpectedpartoftransitioning.
ThoughchildrenengagedwithCocoandrespondedtothecharacterdirectlyinallthreeconditions,theywerelesslikelytodosowhenpost-playvideosweredisplayed.
Duringthetransition,childrenspokedirectlytoCoco16%ofthetimewhenpost-playvideosweredisplayedand40%ofthetimewhentheywerenot.
Achi-squaretestcomparingthepres-enceofthisdirectresponse(Y,N)acrossconditions(con-trolled,neutral,post-play)revealedthatthisdifferencewassignificantχ2(2)=16.
75,p<.
001.
Post-hocanalysisrevealedthatchildrenweresignificantlylesslikelytospeakdirectlytoCocointhepost-playconditionthanineithertheneutral(Z=2.
81,p=.
005)orthecontrolled(Z=4.
07,p<.
001)condition.
Therewasnodifferencebetweentheneutralandcontrolledconditionswithrespecttochildren'slikelihoodofrespondingdirectlytoCoco.
Theseresultssuggestthatthecompetingcontentofpost-playmadeitlesslikelythatchil-drenattendedtoandparticipatedinthedesignedtransitionexperience.
PromotingCoco'sTransitionNormsWesawthatchildrenfrequentlyrespondedtothetransitionpromptbyinternalizingCoco'sstatementsasnormsandau-tonomouslyenacting—andevenenforcing—transitionex-pectations.
Inmanycases,childrenpromptlyturnedoffthetabletwithoutparentoversightinresponsetothetransitionscreen.
Childrenoftenexplainedaloudthattheyweremov-ingontothefollow-upactivitytheyhadplannedforthem-selves,andinouraudioclipstheycanbeheardsayingthingslike,"Timetosleep,""It'stimetoeat,andI'meating,"or,"I'mreadytogooutside…Iclicked'goingoutside'aftermyvideos.
"Inoneinstance,achildrespondedtoCoco'sassertionthatitwastimetodo"somethingelse"bysaying,"NowI'mnotgo-ingtowatchanymore…I'mgoingtodosomethingelse.
"Hismotherreplied,"Ok,timetodosomethingelse.
Whatareyougoingtodo"Thechildrepliedthathewantedtodanceandinstructedhismotherto"turnonmusic.
"Hecouldthenbeheardsinginganddancinginthebackground.
Similarly,onemotherjokedwithherchild,saying,"WhatwillyoudonextCleanthelivingroomIsthatwhatyouwanttodonext"towhichthechildrepliedflatly,"No,Ipickedplayoutside,"highlightingheragencyinselectinghernextactivityandref-erencingherin-appselectionastheplanofrecord.
Acrosstheseandmanyotherexamples,wesawthatchildrentookownershipofthetransitionandconnecteditwiththeirownintentionsandtheirself-definedplan.
Inotherinstances,childrenwentbeyondadoptingthesenormsforthemselvesandworkedtosetthesesameexpecta-tionsforothers.
InresponsetoCoco'squestion,"Areyoureadyforbed"onechildannouncedtohisfamily,"Every-body!
It'stimeforbed!
"AnotherchildturnedtohisbrotherandrepeatedCoco'smessage,sayingtohim:"It'stimetoeat.
Areyoureadytoeat"Athirdchildwatchedtheplaylisttogetherwithherfather,andwhentheone-minutewarningwasdisplayed,explainedtohim,"wehaveoneminuteleft,"implyingthattheexpectationsappliedtobothofthem.
Chil-dreninformedtheirparentsoftransitionexpectations,sayingthingslike,"Dad,it'stimetoread,"and"Seewhatthetabletsaid"Anotherchildexplainedfirsttohismotherandthenhisfather,"Wehavetogonow,"imposingCoco'sstatementthatitwastimetoleavethehouseoneachofhisparents.
Wecodedeachaudioclipforthepresenceoftheseautono-mousstatementsoftransitionownership.
Weobservedthatchildrenwerelesslikelytotransitionautonomouslyorstateandpromotethesenormswhenpost-playvideosweredis-played.
Achi-squaretestcomparingthefrequencywithwhichchildrentransitionedautonomously(Y/N)acrosscon-ditions(neutral,post-play,controlled)revealedthatthisdif-ferencewassignificantχ2(2)=14.
655,p=.
001.
Posthocanalysisrevealedthatchildrenwerelesslikelytodisplayau-tonomywhenpost-playvideoswereshownthanineithertheneutral(Z=-2.
50,p=.
012)orthecontrolled(Z=-3.
85,p<.
001)condition.
Therewasnodifferencebetweentheneutralandcontrolledconditionsinchildren'slikelihoodofdisplay-ingautonomy.
Finally,wesawthataschildrenembracedthesenorms,theyoccasionallydidsomaladaptively.
Onechildexplainedtohismotherthatafteryouselectyournextactivity,Cocowill"makeyoudoit;"hismotherthenexplainedthatthenext-activitysuggestionswerenotaboutpolicingbehaviorandwerethereto"giveyouanideaofwhat'snext.
"Anotherchildexplainedafterfinishingaplaylistthathehadtogobackanddotheentireexperienceoveragainfromthebegin-ning,becausehehadpickedthewrongfollow-upactivityandnowcouldnotgotobedbecauseCocohadnotsaid,"Nowit'stimeforbed.
"Inthesecasesandafewothers,childrenfocusedonrigidlyadheringtoCoco'sstatedexpectations.
Thesebehaviorssuggestthatwhilechildreninthisagegroupmayeagerlyadoptnormspresentedbythesystem,theymayalsohaveastrictinterpretationthatismorecloselyalignedwiththetool'sliteralpresentationthanitsbroadergoals.
UnderminingLimitsWhilechildrenoftenendedtheirsessionasplanned,atothertimestheychosetounderminethelimitstheyhadsetforthemselvesandcontinuedwatchingvideos.
Inresponse,par-entssometimesenforcedtheplannedtransition,sometimesignored(orperhapswereunawareof)thechild'sdecisiontocontinuewatching,andsometimesexpressedagreementthatcreatinganewplaylistfittheirfamily'sscheduleandneedsandsupportedtheirchildinunderminingthepre-existingplan.
ThefrequencyofeachofthesescenariosisshowninFigure4.
Whenchildrenchosetoignoretheirplannedtransition,par-ents'mostcommonresponsewastointervene.
Inthesein-stances,childrenwatchedpost-playvideos,returnedtothehomescreenandcreatedanewplaylist,orevenkilledtheappandrestartedittoworkaroundthelock-outmechanism.
Inouraudioclips,parentscanthenbeheardsteppingintostoptheexperience,andsayingthingslike,"IsitoverGetyourshoes,""Alright,that'sit,""Whenthetimecomesup,you'resupposedtohitthebuttonandstopwatching,ok""It'stimetoturnitoff,"and"C'mon,youweregonnaeat.
WannapushthisIthinkyoupushthistoturnitoff.
"However,inanon-trivialminorityofcases,whenchildrenchosetoignoreaplannedtransition,parentssupportedthisdecisionandcollaborativelyrevisedtheplanwiththechild,consistentwithpriorworkonhowfamiliesexperiencelimit-setting[30].
Forexample,inoneinstance,Cocotoldthechild,"Nowit'stimetoeat,"andthechild'smotherrepliedthatitwasnotactuallytimetoeatyet,becausedinnerwouldnotbereadyforanotherthirtyminutes.
Themotherandchildthenagreedthatthechildshouldconstructanotherplaylist.
Inanotherinstance,Cocoannouncedthatitwastimetoread,andthechildignoredthispromptandbeganmakinganewplaylist.
Hisfatherthensaid,"Youwantsomehelp"andthefatherandchildthencollaborativelybuiltanewplaylistto-gether.
Anothertime,onemotherinvitedherchildtoignoretheplannedtransition,saying,"Ifyouwanttowatchmorevideos,youcanhitthatbutton[home].
"Thoughthiswasnotparents'mostcommonresponse,itoccurredroutinelyenoughtosuggestthatthinkingofsupportsinthisspaceastoolsforenforcingstrictandwell-definedlimitsmaynotbethebestreflectionoffamilies'needsandpractices.
ResponsestoPost-PlayParentsexpressedfrustrationwithpost-playbothin-the-mo-mentastheirchildusedthesystemandinretrospectastheycompletedourpost-studyfollow-upsurvey.
Whenparentsobservedpost-playvideosduringthestudy,theycouldbeheardinaudioclipssayingthingslike,"It'stimetoplay[withtoys],butthenthey'replayinganothervideo,""Idon'tknowwhyitgivesusthisextravideoattheveryend,"and"What's*that*videointhecornerThat'sweird!
"Parentsexpressednegativefeelingsaboutourpost-playex-perienceinthepost-studyfollow-upsurveyaswell.
Weaskedifparentspreferredwhenthesevideoswereshown,preferredwhentheywerenotshown,orhadnopreference;71%saidthattheypreferredwhenthesevideoswerenotshown,and29%saidthattheyhadnopreference.
Noparentsreportedthattheypreferredthepresencepost-playvideos,suggestingthatthiscommonfeatureofcommerciallyavail-ablevideoplatformsisoutofsyncwithsomefamilies'pref-erences.
Weaskedparentstoexplainthischoice,andtheytoldusthatpost-playmakestransitioningmoredifficultbyholdingthechild'sattentionwithcontentheorshehadnotplannedtoengagewith.
Parentscalledpost-play"distracting,"ex-plainedthatit"underminedthe'time'sup,'"and"madeithardtotellherthatthevideosweredone.
"Theyfeltthatthepost-playexperienceledtotheirchild"beingsuckedin,""drasticallyincrease[d]hisresistancetostopping,"andmadeit"morelikelythatmydaughterwouldasktokeepwatching.
"Inadditiontofeelingthatiterodedtransitions,parentshadanegativereactiontothefactthatitgavetheplatformcontrolofcontentchoices,sayingthat,"Idisliketheideaofrandomvideosbeingshowntomychild,"and"mydaughterlikestobeincontrolofwhichvideosshewatches.
"ResponsestotheLock-OutMechanismInourfollow-upsurvey,parentsweredividedintheirreac-tiontothelock-outmechanismwhichpreventedchildrenfromcreatinganewplaylistoncethetransitionscreenap-peared.
Morethanhalfofparentsappreciatedthissupportsaying,"thiswasgreatasitmadesurescreentimeendedifwewerenotrighttheretotakeitaway,"and"itmadecuttingoffwatchingtimeeasierformeandhim.
"However,35%ofparentsdisagreed,explainingthatthiswastoorestrictiveandmadeitimpossibleforthemtochangetheirplans.
Thesepar-entssaidthingslike,"ifwebothagreedthatshecouldwatchmore,itwouldhavebeennicetoallowit,"and"sometimesIallowedhimtocontinuedependingonwhatIwasdoingandhowwellhebehavedafterhisfirstsession.
"Forthesepar-ents,apre-definedcontractdidnotmatchtheirneeds,astheyvaluetheabilitytoreviseboundaries.
However,severalparentswhoreportedvaluinghard-and-fastboundarieswerealsodissatisfiedwiththelock-outmechanism,becausetheyfeltitwasnotstrictenough.
Theseparentsexplainedthatanapp-levelcontrolwasinsufficientsaying,"hewouldexittheappandopenYouTubedirectly"andthat,"IwouldratherusethetimeronmyiPadwhichshutsthetabletoff.
"Anotherparentexplainedthatafeaturetolockchildrenoutoftechnologyis,"tryingtosolveaprob-lemitcan'tsolve,"andexplainedthatforherfamily,"it'snotausefulfeaturewithoutagloballock,whichhasitsownsetofproblems.
"Parents'PerspectivesonCoco'sVideosParentshadahighlyfavorableresponsetothestructureandgoalsofCoco'sVideos,saying,"IlovedthattheapptoldherFigure4:Thewayinwhichasessionended,byfrequency0102030405060ChildchoosestotransitiononscheduleChildandparentdecidetogetherthatchildshouldwatchmore…Childcontinueswatchingvideos;noresponsefromparentChildcontinueswatchingvideos;parentintervenesNumberofAudioClipsControlledNeutralPost-Playitwastimetostopandnotme.
Itmadetransitioningeasier,"callingita"greatexperience,"andreportingthat,"IbecamemoreawareofwhatandhowIwasdoingthings.
"However,eightparentsreportedthattheappwouldnotfittheirlifestyleinthelongrun,astheplatformdoesnotprovidethecontenttheirchildwatches,andasaresult,electednottouseitafterthestudyended.
Parentsreportedthattheyappreciatedhavingbuilt-inre-mindersfortheirchildthattheplaylistwasending,sayingthingslike,"makinghimhitthebuttontocontinueafterthelastvideowarningseemedtoreallyhelpasheforsureknewitwasthelastone,""thewarningsthatthescreentimewascomingtoanendwasalsoagoodfeaturethatworkedwellforher,"and"[I]likedthatIdidn'thavetodoanynaggingorremindersabouthowmanyminutesofscreentimewereleft.
"Parentsalsorepeatedlysaidthattheirchildrenappreci-atedbeingabletochoosetheirnextactivity.
Thoughwedidnotexplicitlyaskparentswhetherchildrendisplayedownershipovertransitionsorinternalizednorms,thiswasaconsistentthemeinresponsestoopen-endedques-tionsaboutchildren'sreactions.
Parentsspontaneouslyre-portedthat,"Ithinkitmadehimfeellikeabigboy,""Mychildlovedmakingchoices,"andexplainedthatnowthey"decidetogether…howlongshe'llwatchbeforeshestarts.
"Nineofthe20parentswhocompletedtheentiresurveyspon-taneouslyreportedthatamongthethingstheyandtheirchildlikedbestwerewaysinwhichtheappgavethechildchoice,control,andownership.
Eightparentsreportedthatsincetheconclusionofthestudy,theyhadincorporatedtheappintotheirmediaroutineandcontinuedtouseitregularly.
Otherssaidthattheywouldliketodoso,butthatthecontentortheformfactordidnotfittheirviewinghabits,sayingthingslike,"ifitintegratedwithourregularappsand[could]bethelauncherlikeamazonkidsunlimitedwedefinitelywoulduseit.
"Parentshadava-rietyofsuggestionsforimprovementsthatwereunrelatedtothetransitionstructure,suchasaddingsupportforcastingtoaTV,integratingwithNetflix,NickJr.
,andAmazonVideo,changingthevoiceusedforaudio,andincreasingtheamountofcontentintherecommendationslibrary.
Nineteenofthe20parentswhocompletedtheentiresurveyexpressedinter-estinadoptingtheaspectsofthesystemthatsupportedplan-ningvideouseandtransitioningattheendoftheexperience.
DISCUSSIONCoco'sVideosandChildren'sAutonomyAschildrenusedCoco'sVideos,theyembracedtheconceptofplanningandexpressedownershipovertheactofselect-ingtheirnextactivity.
Parentsreportedthatthiswasoneoftheirchild'sfavoritepartsoftheappexperience.
Thisthreadofownershipcarriedforwardtochildren'stransitionsawayfromthescreen,andchildrenoftencommentedontheirself-definednext-activityastheyturnedoffthetablet.
Thesedemonstrationsofintentionalityareimportant,becausepriorworkhasshownthatplanningandpurposefuldecision-mak-ingleadtogainsinchildren'sabilitytoself-regulatetheirbehavior[37].
Wesawthatchildrennotonlyenforcedthistransitionforthemselves,theyalsoregularlyextendedtransitionexpecta-tionstothosearoundthem.
Regulatingothersisanimportantpartoftheprocessofinternalizingnormsandlearningtoreg-ulateone'sownbehavior[16],andourparticipants'actsofother-regulationareconsistentwiththeprocessofadoptingnewgoalsforthemselves.
However,wealsoobservedthataspartofitsinfluenceonchildren'sbehavior,theplatformshowedevidenceofper-suadingsomechildrentorigidlyadheretonormsandtreattheappasanauthoritarianfigure.
Children'soccasionalin-sistencethattheywererequiredtocomplywiththeapp'scommands,evenwhenthesenolongermatchedtheirowndesires,wasinconsistentwithourgoalofhelpingchildrenidentifyandfollowthroughontheirownintentions.
Weplantoiterateonourdesignchoicestoinvestigatehowwemightevolvethetransitionscreentoconsistentlysurfacethechild'sownauthority.
Futuredesignsthatincorporatevoicerecog-nitionandenablethecharactertorespondtothechildwouldallowustoexaminehowaninteractiveexperienceinfluenceschildren'sattitudesorbehaviors.
TheRoleofPost-PlayWhenCoco'sVideosautomaticallyplayedadditional,re-latedcontentaftertheplaylistended,childrenwerelesslikelytoverbalizestatementsofautonomyandlesslikelytoengageintheirplannedtransition,suggestingthatthisfea-tureunderminesthechild'slikelihoodofself-regulating.
Theirtabletsessionsweresignificantlylonger,andparentsweresignificantlymorelikelytostepinandintervene,sug-gestingthatthechildandparentaremorelikelytobeoutofsyncinthepresenceofthisfeature.
Wesawthatparentswerefrustratedbythepresenceofpost-playandfeltthatitdistractedtheirchild,madeexpectationsconfusing,andmadetransitionsmorepainful.
Despitethisconsistentlynegativereaction,post-playistodayastandard,defaultfeatureofmanyvideo-viewingplatforms,includingYouTube,Vimeo,Netflix,Hulu,AmazonVideo,andevenYouTubeKids.
Whilethisfeatureissometimespresentedasaconveniencefortheuserwiththeintentionofmakingiteasiertofindenjoyablecontent(e.
g.
,[45,46]),ourresultsshowthatinpracticeitworkstoerodefamilies'plannedbe-haviorsandstandsindirectcontrasttousers'desires.
Aspriorworkhasshownthatuserssometimesabandonsystemstheyfeelaretooaddictiveortooaggressiveindemandingengagement[2],itisworthexploringwhetherfeatureslikepost-playmightdiscourageadoptioninthelongrun.
TheRoleoftheLock-outMechanismOurchildparticipantswereequallyautonomousandequallylikelytodisengagefromvideoviewingwithorwithoutalock-outmechanism.
Ourresultsshowthatforourpartici-pants,aneutralexperiencethatremindschildrenoftheirownintentionsisaseffectiveasstrictcontrolsthatconstraintheirbehavior.
Drawingonself-determinationtheory[35],weex-pectedthatthelock-outmechanismwouldimpedechildren'ssenseofautonomyandintrinsicmotivationtoadheretotran-sitionnorms.
However,childparticipantswereequallylikelytomakeautonomousstatementsintheneutralandcontrolledconditions.
Thissuggeststhatevenifsuchcontrolsarenotnecessary,itispossiblethatforyoungchildrentheyalsoarenotproblematic.
Parentshadmixedreactionstothelock-outmechanism,withsomevaluingenforcementandothersdislikingthefactthattheapptookcontrol.
Parentsexplainedthatthisfeaturemadeitharderfortheirfamilytonimblyadjusttheirplannedmediause,consistentwiththemanyinstancesinwhichweob-servedfamiliesunderminingandrevisingtheirownlimits.
Further,aspriorworkhasshownthatparentsfeelguiltyaboutusingtechnologytooccupytheirchild[9,24,30],inter-facechoicesthatpromotestrictboundaries,especiallyboundariesthatparentsmayinfactdecidetochange,havethepotentialtounintentionallyshameusersfortheirchoiceswhilesimultaneouslyfailingtoprovidethemwiththeexpe-riencethatbestfitstheirneeds.
DesignImplicationsWeseeseveralwaysthatsystemsforpreschoolersmightconsiderthevaluesandbehaviorsweobservedinthisstudy:Designtosupportautonomy:Wesawthatchildrenspokedi-rectlytoCoco,developedaparasocialrelationshipwiththecharacter,andansweredCoco'squestionseventhoughthecharacterwasneverinteractive.
Framingthetransitionasaquestionledchildrentoreplyaloud,announcingtheirinten-tionsandreinforcingtheirautonomy.
Ourresultssuggestthatdesignsthat:1)provideopportunitiesforplanningandmak-ingchoices,2)remindchildrenoftheirintentions,and3)askquestionsofthechild,havethepotentialtopromoteauton-omyandscaffoldmediaself-regulation.
Supportflexiblelimitsandallowrevision:Consistentwithpriorwork[30],wesawthatfamiliesdonotalwayswanttointegratemediaintodailylifebysettingpre-definedandcon-sistentlimits.
Whilefamiliesvaluedthecleartransitionpointwepresented,theyhadamixedexperiencewithbeinglockedintothisplan.
Controlsthatenforcelimitsnotonlyfailtosupportthecommonuserscenarioofrevisingplans,theyalsodenychildrentheopportunitytopracticetheimportantactofself-regulating.
Respectfamilies'limits:Whilefor-profitmediacorporationswillalwayshaveanincentivetoseekuserengagement,dis-engagementwill,bynecessity,alsobeapartofmediause.
Designingforpositivedisengagementexperienceshasthepotentialtoimproveoverallusersatisfactioninthelongrun.
Families'transitionexperiencesweremoreprolongedandinvolvedmorefrictionwhenpost-playwasturnedon.
Ourresultssuggestthatavoidingpost-playismorehelpfulinlim-itingchildren'smediaconsumptionthanadoptingalock-outmechanism.
LimitationsandFutureWorkThereareanumberoflimitationstotheclaimswecanmakefromthisdata.
Weconductedthisworkwithasmallsamplethatwasover-representativeofmiddle-classandwhitefam-ilies.
Thisisakeylimitation,aspriorworkhasshownthatcontrollingand"nononsense"parentingstrategiesareusefuladaptationsinlow-resource,high-stressenvironmentsandcanfosterchildren'sself-regulationinthesecontexts[4].
Priorworkalsoshowsthatcontrollingandauthoritarianpar-entingstylesoperatedifferentlyindifferentpopulations[18].
Weplantoconductfutureworkwithmorediversefamiliestodevelopamorecompleteunderstandingoftherelation-shipsamongthedesignfeatureswestudied,parentingprac-tices,demographics,children'sself-regulation,andusersat-isfaction.
Wealsodidnotdirectlysolicitfeedbackfromchildrenabouttheirexperiences,aperspectivethatwouldbeavaluablecomplementtothedatawepresenthere.
Andalthoughourthree-weekdeploymentgaveusaccesstomanyusagein-stancesforeachfamily,itwastooshortforustodeveloparobustunderstandingofparticipants'long-termhabitsinre-sponsetothesefeatures.
Finally,givenourdesignchoices,wewerelimitedtothecontentavailableonYouTube,whichdidnotcomprehensivelyincludeallofthecontentourpar-ticipantstypicallyengagewith.
Futureworkremainstoin-vestigatethesedesignchoicesinthecontextofchildren'sex-istingmediaroutines.
CONCLUSIONToday'spreschoolerswillcomeofageinaworldofcon-stantlyconnecteddevicesthatwillprovidethemwiththeop-portunitytopluginatalmostanymoment.
Forthesechil-dren,defininghowtheywanttoengagewiththismediaeco-systemandmanagingtheirongoingmediaconsumptionislikelytoforeverbeapartofdailylife.
Inthisstudy,wepre-sentonepossiblealternativetotoday'sparentalcontrols,inwhichwemoveawayfromthetraditionallyauthoritariande-signsofthisspace.
Ourresultsshowthatthroughdesign,wecansupportchildreninbecomingautonomoususers.
Or,throughfeatureslikepost-play,wecanunderminetheirabil-itytoself-regulate.
Childparticipantsconsistentlyshowedevidenceofwantingtotakeownershipoftheirtransitions,andtheydemonstratedagencyinbothengaginganddisengagingwithtechnology.
Weconcludethatthedesigncommunitydoesnotneedtocreatetoolstopoliceexcessivemediause—wesimplyneedtostopdesigningexperiencestoencourageit.
ACKNOWLEDGMENTSThisworkwasfundedbytheUniversityofWashingtonRoy-altyResearchFund.
REFERENCES1.
GuyBalk.
2016.
$80,000median:IncomegaininSeattlefaroutpacesothercities.
SeattleTimes.
2.
EricP.
S.
Baumer,PhilAdams,VeraD.
Khovanskaya,TonyC.
Liao,MadelineE.
Smith,VictoriaSchwandaSosik,andKaitonWilliams.
2013.
Limiting,leaving,and(re)lapsing:AnExploringofFacebookNon-usePracticesandExperiences.
InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems.
3257-3266.
https://doi.
org/10.
1145/2470654.
24664463.
ElenaBodrovaandDeborahJ.
Leong.
2007.
ToolsoftheMind:TheVygotskianApproachtoEarlyChildhoodEducation.
PearsonEducation,UpperSaddleRiver,NJ.
4.
GeneH.
BrodyandDouglasL.
Flor.
1998.
MaternalResources,ParentingPractices,andChildCompetenceinRural,Single-ParentAfricanAmericanFamilies.
ChildDevelopment69,3:803–816.
https://doi.
org/10.
1111/j.
1467-8624.
1998.
tb06244.
x5.
MindyBrooks,AshleyFenwick-Naditch,andEricaBranch-Ridley.
2011.
ElectricRacer:Anintergenerationalgamingexperiencedesignedtopromoteliteracy.
TheNewCoviewing:DesigningforLearningthroughJointMedia:30–33.
6.
AriBrown.
2011.
Mediausebychildrenyoungerthan2years.
Pediatrics128,5:1040–5.
https://doi.
org/10.
1542/peds.
2011-17537.
BenjaminBurroughs.
2017.
YouTubeKids:TheAppEconomyandMobileParenting.
SocialMedia+Society3,2:205630511770718.
https://doi.
org/10.
1177/20563051177071898.
SandraL.
Calvert,Calvert,andSandraL.
2015.
ChildrenandDigitalMedia.
InHandbookofChildPsychologyandDevelopmentalScience.
JohnWiley&Sons,Inc.
,Hoboken,NJ,USA,1–41.
https://doi.
org/10.
1002/9781118963418.
childpsy4109.
ValerieCarson,MarianneClark,TanyaBerry,NicholasLHolt,andAmyELatimer-Cheung.
2014.
AQualitativeExaminationofthePerceptionsofParentsontheCanadianSedentaryBehaviourGuidelinesfortheEarlyYears.
InternationalJournalofBehavioralNutritionandPhysicalActivity11,1:65.
https://doi.
org/10.
1186/1479-5868-11-6510.
Childwise.
2017.
TheMonitorPreschoolReport.
11.
DimitriAChristakisandFrederickJZimmerman.
2006.
Theelephantinthelivingroom:maketelevisionworkforyourkids.
Rodale.
12.
DimitriAChristakis,FrederickJZimmerman,DavidLDiGiuseppe,andCarolynAMcCarty.
2004.
Earlytelevisionexposureandsubsequentattentionalproblemsinchildren.
Pediatrics113,4:708–713.
https://doi.
org/10.
1542/peds.
113.
4.
917-a13.
AnnS.
Epstein.
2003.
Howplanningandreflectiondevelopyoungchildren'sthinkingskills.
YoungChildren58,5:28–36.
14.
JenniferFalbe,KirstenDavison,RebeccaFranckle,ClaudiaGanter,StevenGortmaker,LaurenSmith,ThomasLand,andElsieTaveras.
2015.
SleepDuration,Restfulness,andScreensintheSleepEnvironment.
Pediatrics132,2:e365-e375.
https://doi.
org/10.
1542/peds.
2014-230615.
ShalomMFisch,RosemarieTTruglio,andCharlotteFCole.
1999.
TheimpactofSesameStreetonPreschoolChildren:AReviewandSynthesisof30years'Research.
MediaPsychology1,2:165–190.
https://doi.
org/10.
1207/s1532785xmep0102_516.
EmilyFoxandMichelleRiconscente.
2008.
MetacognitionandSelf-RegulationinJames,Piaget,andVygotsky.
EducationalPsychologyReview20,4:373–389.
https://doi.
org/10.
1007/s10648-008-9079-217.
DouglasA.
Gentile,SarahCoyne,andDavidA.
Walsh.
2011.
MediaViolence,PhysicalAggression,andRelationalAggressioninSchoolAgeChildren:AShort-termLongitudinalStudy.
AggressiveBehavior37,2:193–206.
https://doi.
org/10.
1002/ab.
2038018.
LeilaniGreening,LauraStoppelbein,andAaronLuebbe.
2010.
TheModeratingEffectsofParentingStylesonAfrican-AmericanandCaucasianChildren'sSuicidalBehaviors.
JournalofYouthandAdolescence39,4:357–369.
https://doi.
org/10.
1007/s10964-009-9459-z19.
WendyS.
Grolnick,EdwardL.
Deci,andRichardM.
Ryan.
1997.
InternalizationwithintheFamily:TheSelf-determinationTheoryPerspective.
JohnWiley&SonsInc.
20.
YasmeenHashish,AndreaBunt,andJamesEYoung.
2014.
InvolvingChildreninContentControl:ACollaborativeandEducation-orientedContentFilteringapproach.
InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems.
1797–1806.
https://doi.
org/10.
1145/2556288.
255712821.
HHendershot.
2002.
GreatExpectations:TheRiseandFalloftheV-Chip.
TelevisionQuarterly.
RetrievedSeptember25,2015fromhttps://scholar.
google.
com/scholarhl=en&q=great+expectations:+the+rise+and+fall+of+the+v-chip&btnG=&as_sdt=1,48&as_sdtp=#022.
AlexisHiniker,BongshinLee,KileySobel,andEunKyoungChoe.
2017.
Plan&Play:SupportingIntentionalMediaUseinEarlyChildhood.
InProceedingsofthe2017ConferenceonInteractionDesignandChildren.
85–95.
https://doi.
org/10.
1145/3078072.
307975223.
AlexisHiniker,KileySobel,andBongshinLee.
2017.
Co-DesigningwithPreschoolersUsingFictionalInquiryandComicboarding.
InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems.
5767–5772.
https://doi.
org/10.
1145/3025453.
302558824.
AlexisHiniker,HyewonSuh,SabinaCao,andJulieA.
Kientz.
2016.
ScreenTimeTantrums:HowFamiliesManageScreenMediaExperiencesforYoungChildren.
InProceedingsoftheSIGCHIConferenceonHumanFactorsinComputingSystems.
648–660.
https://doi.
org/10.
1145/2858036.
285827825.
SteviJacksonandSueScott.
1999.
RiskAnxietyandtheSocialConstructionofChildhood.
Riskandsocioculturaltheory:Newdirectionsandperspectives:86–107.
26.
RussellJago,TomBaranowski,JaniceCBaranowski,DeborahThompson,andKAGreaves.
2005.
BMIfrom3-6yofAgeisPredictedbyTVViewingandPhysicalActivity,notDiet.
InternationalJournalofObesity29,6:557.
https://doi.
org/10.
1038/sj.
ijo.
080296927.
HildaKKabali,MatildeMIrigoyen,RosemaryNunez-Davis,JenniferGBudacki,SwetaHMohanty,KristinPLeister,andRobertLBonner.
2015.
ExposureandUseofMobileMediaDevicesbyYoungChildren.
Pediatrics:peds.
2015-2151.
https://doi.
org/10.
1542/peds.
2015-215128.
RichardKoestner,RichardM.
Ryan,FrankBernieri,andKathleenHolt.
1984.
SettingLimitsonChildren'sBehavior:TheDifferentialEffectsofControllingvs.
InformationalStylesonIntrinsicMotivationandCreativity.
JournalofPersonality52,3:233–248.
https://doi.
org/10.
1111/j.
1467-6494.
1984.
tb00879.
x29.
DaleKunkel,WendyJoMaynardFarinola,KirstieFarrar,EdwardDonnerstein,EricaBiely,andLaraZwarun.
2002.
DecipheringtheV-Chip:AnExaminationoftheTelevisionIndustry'sProgramRatingJudgments.
JournalofCommunication52,1:112–138.
https://doi.
org/10.
1111/j.
1460-2466.
2002.
tb02535.
x30.
MelissaMazmanianandSimoneLanette.
2017.
Okay,OneMoreEpisode:AnEthnographyofParentingintheDigitalAge.
InProceedingsoftheACMConferenceonComputerSupportedCooperativeWorkandSocialComputing.
2273–2286.
https://doi.
org/10.
1145/2998181.
299821831.
PatriciaMcDonough.
2009.
TVViewingAmongKidsatanEight-YearHigh.
TheNielsenCompany.
RetrievedSeptember14,2017fromhttp://www.
nielsen.
com/us/en/insights/news/2009/tv-viewing-among-kids-at-an-eight-year-high.
html32.
MarijeNouwen,NassimJafarinaimi,andBiekeZaman.
2017.
ParentalControls:ReimaginingTechnologiesforParent-childInteraction.
RetrievedSeptember13,2017fromhttps://dl.
eusset.
eu/handle/20.
500.
12015/292833.
JennyS.
RadeskyandDChristakis.
2016.
MediaandYoungMinds.
Pediatrics138,5:e2–e6.
34.
EricERasmussen,AutumnShafer,MalindaJ.
Colwell,ShawnaWhite,NarissraPunyanunt-Carter,RebeccaL.
Densley,andHollyWright.
2016.
Relationbetweenactivemediation,exposuretoDanielTiger'sNeighborhood,andUSpreschoolers'socialandemotionaldevelopment.
JournalofChildrenandMedia.
10,4:443–461.
https://doi.
org/10.
1080/17482798.
2016.
120380635.
RichardM.
RyanandEdwardL.
Deci.
2000.
Self-DeterminationTheoryandtheFacilitationofIntrinsicMotivation,SocialDevelopment,andWell-being.
TheAmericanpsychologist55,1:68–78.
36.
GavrielSalomon.
1977.
EffectsofEncouragingIsraeliMotherstoCo-observeSesameStreetWithTheirFive-Year-Olds.
ChildDevelopment48,3:1146.
https://doi.
org/10.
2307/112837837.
LawrenceJ.
SchweinhartandDavidP.
Weikart.
1997.
TheHigh/ScopePreschoolCurriculumComparisonStudyThroughAge23.
EarlyChildhoodResearchQuarterly.
12,2:117–143.
https://doi.
org/10.
1016/S0885-2006(97)90009-038.
AricSigman.
2012.
TimeforaViewonScreenTime.
Archivesofdiseaseinchildhood97,11:935–42.
https://doi.
org/10.
1136/archdischild-2012-30219639.
LoriTakeuchi,ReedStevens,andothers.
2011.
TheNewCoviewing:DesigningforLearningthroughJointMediaEngagement.
InNewYork,NY:TheJoanGanzCooneyCenteratSesameWorkshop.
40.
BarbaraJ.
Wilson.
2008.
MediaandChildren'sAggression,Fear,andAltruism.
FutureofChildren.
18,1:87–118.
https://doi.
org/10.
1353/foc.
0.
000541.
BiekeZamanandMarijeNouwen.
2016.
ParentalControls:AdviceforParents,ResearchersandIndustry.
EUKidsOnline.
Retrievedfromhttp://eprints.
lse.
ac.
uk/65388/42.
FrederickJZimmermanandDimitriAChristakis.
2007.
AssociationsbetweenContentTypesofEarlyMediaExposureandSubsequentAttentionalProblems.
Pediatrics120,5:986–992.
https://doi.
org/10.
1542/peds.
2006-332243.
CampaignforaCommercialFreeChildhood.
RetrievedSeptember18,2017fromhttp://www.
commercialfreechildhood.
org/44.
CommonSenseMedia:Reviews&AgeRatings-BestMovies,Books,Apps,GamesforKids.
RetrievedSeptember18,2017fromhttps://www.
commonsensemedia.
org/45.
HowcanIPreventtheNextEpisodeofaShowFromPlayingAutomaticallyNetflixHepCenter.
RetrievedSeptember18,2017fromhttps://help.
netflix.
com/en/node/210246.
Autoplayvideos-Android-YouTubeHelp.
GoogleSupport.
RetrievedSeptember17,2017fromhttps://support.
google.
com/youtube/answer/6327615co=GENIE.
Platform=Android&hl=en47.
2013.
Zerotoeight:Children'smediauseinAmerica2013.
ACommonSenseMediaresearchstudy:1–31.
Retrievedfromhttps://www.
commonsensemedia.
org/research/zero-to-eight-childrens-media-use-in-america-2013