An Interactive Gravitational-Wave Detector Model for Museums and Fairs
S. J. Cooper, A. C. Green, H. R. Middleton, C. P. L. Berry, R. Buscicchio, E. Butler, C. J. Collins, C. Gettings, D. Hoyland, A. W. Jones, J. H. Lindon, I. Romero-Shaw, S. P. Stevenson, E. P. Takeva, S. Vinciguerra, A. Vecchio, C. M. Mow-Lowry, A. Freise
AAn Interactive Gravitational-Wave Detector Model for Museums and Fairs
S. J. Cooper, A. C. Green,
1, 2
H. R. Middleton,
1, 3, 4 and C. P. L. Berry
1, 5
R. Buscicchio, E. Butler, C. J. Collins, C. Gettings, D. Hoyland, A. W. Jones, J. H. Lindon,
1, 6
I. Romero-Shaw,
1, 7, 4
S. P. Stevenson,
1, 8, 4
E. P. Takeva,
1, 9
S. Vinciguerra,
1, 10, 11
A. Vecchio, C. M. Mow-Lowry, and A. Freise Institute for Gravitational Wave Astronomy and School of Physics and Astronomy,University of Birmingham, Birmingham, B15 2TT, United Kingdom Department of Physics, University of Florida, Gainesville, FL 32611, USA School of Physics, University of Melbourne, Parkville, Vic, 3010, Australia OzGrav, Australian Research Council Centre of Excellence for Gravitational Wave Discovery Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA),Department of Physics and Astronomy, Northwestern University,1800 Sherman Avenue, Evanston, IL 60201, USA Elementary Particle Physics Group, School of Physics and Astronomy,University of Birmingham, Birmingham, B15 2TT, United Kingdom Monash Centre for Astrophysics, School of Physics and Astronomy, Monash University, VIC 3800, Australia Centre for Astrophysics and Supercomputing,Swinburne University of Technology, Hawthorn, 3122, Victoria, Australia Department of Physics and Astronomy, University of Edinburgh, James Clerk Maxwell Building,Peter Guthrie Tait Road, Edinburgh, EH9 3FD, United Kingdom Max Planck Institute for Gravitational Physics (AlbertEinstein Institute) Callinstraße 38, 30167 Hannover, Germany Leibniz Universit¨at Hannover Welfengarten 1, 30167 Hannover, Germany
In 2015 the first observation of gravitational waves marked a breakthrough in astrophysics, andin technological research and development. The discovery of a gravitational-wave signal from thecollision of two black holes, a billion light-years away, received considerable interest from the me-dia and public. We describe the development of a purpose-built exhibit explaining this new areaof research to a general audience. The core element of the exhibit is a working Michelson inter-ferometer: a scaled-down version of the key technology used in gravitational-wave detectors. TheMichelson interferometer is integrated into a hands-on exhibit, which allows for user interaction andsimulated gravitational-wave observations. An interactive display provides a self-guided explana-tion of gravitational-wave related topics through video, animation, images and text. We detail thehardware and software used to create the exhibit, and discuss two installation variants: an indepen-dent learning experience in a museum setting (the Thinktank Birmingham Science Museum), and ascience-festival with the presence of expert guides (the 2017 Royal Society Summer Science Exhibi-tion). We assess audience reception in these two settings, describe the improvements we have madegiven this information, and discuss future public-engagement projects resulting from this work. Theexhibit is found to be effective in communicating the new and unfamiliar field of gravitational-waveresearch to general audiences. An accompanying website provides parts lists and information forothers to build their own version of this exhibit.
I. INTRODUCTION AND OVERVIEW
Gravitational waves are ripples in space and timefirst predicted as a consequence of the general theoryof relativity by physicist Albert Einstein in 1916 .A century later and after decades of technologicaldevelopment, the first observation of gravitationalwaves was on the 14 September 2015 . The sig-nal came from two black holes orbiting each othera billion light-years from Earth . The black holesmerged together, creating a new bigger black hole.The gravitational waves produced by this event spread out across the Universe, eventually reachingthe Earth, where their miniscule effect was detectedby the Laser Interferometer Gravitational-Wave Ob-servatory (LIGO) . LIGO has since been joined inobserving gravitational waves by Virgo and futuredetectors are planned with KAGRA and LIGO In-dia . The current global gravitational-wave detectornetwork has made many new observations ; thebeginning of a new kind of astronomy.In anticipation of increased media coverage andpublic interest in gravitational-wave astronomybrought about by the first detections, the Educa- a r X i v : . [ phy s i c s . e d - ph ] A p r ion and Public Outreach (EPO) group of the LIGOScientific Collaboration has worked to develop re-sources and activities aimed at informing and inspir-ing the general public, prospective students, and thewider scientific community about our work . Ourgroup at the University of Birmingham has a stronghistory of involvement with public engagement withresearch . Here, we describe our work develop-ing an interactive model gravitational-wave detectordesigned to demonstrate the key technologies thathave enabled gravitational-wave astronomy, and in-troduce the public to this new field of astronomy.Museum and science-fair exhibits are an effectiveway of increasing interest in science and rais-ing awareness of scientific concepts . Visits toscience museums have been shown to improve long-term science knowledge and adult memories ofschool field trips can often recall something learntduring their childhood experience . We have cre-ated an interactive exhibit that can be used bothwhen an expert is present to explain it and as astand-alone,non-facilitated piece which a member ofthe public can use to learn independently. The ex-hibit teaches the public about gravitational waves,how they have been detected, and the kinds of astro-physical events which can be observed using them.The resulting piece is a long-term installation at theThinktank Birmingham Science Museum (Think-tank) and was featured at the 2017 Royal SocietySummer Science Exhibition (RSSE).In this article we provide a detailed description ofthe design and implementation of our exhibit. Weprovide an overview of the distinguishing featuresof our exhibit in Section II, and cover the techni-calities of the hardware and software in Section III,with links to the detailed design for others to use .In Section IV and Section V we describe two usecases of the exhibit in a museum and science fairsetting respectively. Finally in Section VI, we dis-cuss the impact of our exhibit, measured throughsurveys as well as anecdotal examples of the publicreception, and look to the future of these activitiesin Section VII. II. DESIGN CONSIDERATIONS
Gravitational-wave science involves wide extremesof scale in the Universe. The colliding black holesand neutron stars that we observe have masses manytimes the mass of the Sun and can be billions oflight-years from Earth. However, the resulting grav-itational waves arriving at Earth create minusculechanges in distance: a typical black hole collision moves the components of the LIGO instruments bya thousandth of the width of a proton. To detectsuch changes, high-precision instrumentation is re-quired, and on a large scale: each detector site isseveral square kilometres. It can therefore be chal-lenging to communicate the science of gravitationalwaves in a human-relatable way.Our exhibit is a model gravitational-wave detec-tor, demonstrating the core technology of current de-tectors like LIGO: the
Michelson interferometer .This optical configuration is often used to measurechanges in distance; this is explained further in Sec-tion III. The exhibit highlights both the behaviourof gravitational waves—changing relative distanceson a small scale—and the technologies necessary tomeasure this behaviour. Such an interferometer isa common item in the tool-kit of gravitational-waveeducation and outreach and is often used in under-graduate laboratory experiments . The Univer-sity of Birmingham has expertise in designing andbuilding instruments such as interferometers. There-fore, it was possible for us to build our own exhibit,and in doing so showcase both the technical exper-tise and research of the University in this field.The exhibit design is driven by three main aims:(a) present gravitational-wave topics and conceptsso that they are accessible for a broad audience,(b) attract interest in the exhibit using an appealingand exciting design , and (c) be suitable for usein both a museum and a science-fair setting. In amuseum, an exhibit needs to work as a stand-alonepiece, whilst at fairs it is accompanied by experts toguide a visitor through the demonstration and an-swer any questions. Exhibiting at the Thinktank al-lows us to engage our local community, highlightingthe activities taking place in the city of Birmingham.The RSSE was a national science fair providing usan opportunity to work in collaboration with severalother universities.The audiences in both settings are typically non-scientists with a general interest in science. Weengage our audience by pitching the exhibit mate-rial to the right level, enabling them to build upontheir current understanding , and conveyingthe subject in an interactive , varied and funway .Our initial design considerations were the size andweight of the model, how the public would interactwith the exhibit in its stand-alone use case withinthe museum, and how demonstrators would interactwith the exhibit when explaining gravitational-wavescience to small groups at fairs . We wanted thegravitational-wave detector model to be as large asfeasibly possible: aesthetically we wanted something2hiny and interesting to the public , and practicallythe larger size allows the components to be more eas-ily viewed. At the same time, the model needed tobe small enough to be easily transported for eventsand to and from the museum. We settled on a circu-lar aluminium base with a diameter of 0 . .The gravitational-wave detector model can beused on its own, or in combination with screens andbuttons that visitors can use to interact with themodel and learn more about it . We have devel-oped custom exhibit software, which can be adaptedto suit a specific audience and the particular inter-active configuration in use, either in a museum orat a science fair (see Section IV and Section V fordetails on these audiences).In the museum, the software is set up so that userscan guide themselves through the exhibit with thehelp of multimedia material. The software and in-teraction with the physical hardware needed to bedurable to cope with high usage and to operate in-dependently without maintenance for extended pe-riods of time. It also needed to follow the museum’shealth and safety protocols to be suitable for unsu-pervised use by all ages. The information presentedneeded to be self-explanatory, suitable for a rangeof interaction times, and use a range of informationdelivery options (e.g., video, images, text).At the Thinktank, our exhibit is located in agallery containing several unrelated science exhibits.Each is housed in a large bay (approximately 2 m × . . III. TECHNICAL DESIGNA. Hardware
Gravitational-wave detection requires measure-ment of small changes in distances. A gravitational-wave detector can be thought of as a precise ruler.The key component of gravitational-wave detectorslike LIGO and Virgo which enables this precisemeasurements is the
Michelson interferometer .The model gravitational-wave detector is a workingMichelson interferometer. It compares the paths oftwo laser light beams to detect changes in distance.While it cannot detect gravitational waves, it canpick up vibrations in the room, even when there isno apparent disturbances to the exhibit. This pro-vides an intuitive means to illustrate that these in-struments are capable of sensing vibrations imper-ceptible to humans. Laser BeamsplitterMirror MirrorInterference Pattern
Lens PZTs
FIG. 1. Schematic optical layout of a Michelson Inter-ferometer. A beamsplitter is used to split laser lightequally into two perpendicular directions. Each beam re-flects off a mirror, and the two beams recombine again atthe beamsplitter. The interference pattern appearing atthe output of the interferometer depends on differencesbetween the two paths taken by the two beams. Piezo-electric transducers (PZTs) are used to precisely moveone of the mirrors, changing the interference pattern tosimulate an observation of the gravitational wave.
The core constituents of a Michelson interferom-eter are a laser, two mirrors and a beamsplitter asillustrated in Fig. 1. The laser beam first hits thebeamsplitter, where it is split in two. The beamstravel in two arms at 90 ◦ to each other to mirrorsat the ends of each arm. After reflecting from themirrors, the light from the two arms recombines atthe beamsplitter where the two beams interfere. Theresulting light hits a screen where it can be viewed3 ime FIG. 2. Illustration of the ring interference pattern pro-duced by our interferometer. When the interferometeris disturbed, the resulting concentric rings of the inter-ference pattern change over time (left to right in theillustration) from light to dark and back again. and it is this recombined light that produces thechanging interference pattern produced by the in-terferometer, i.e. the output of the detector changesfrom dark to light when a gravitational wave passesthrough the detector.The interference pattern in LIGO is a single spotof light changing over time. To demonstrate thechanges clearly to our audience, we used lenses tocreate an extended interference pattern of a seriesof concentric rings as illustrated in Fig. 2. Whenthere is no change in distance, these rings remainstill, but if there is some disturbance to the interfer-ometer the rings will either breath in and out (forsmall motions) or seem to be zooming in or out (forlarger motions). By viewing the interference pat-tern, the audience can build an understanding thatzooming in one direction or the other corresponds tothe detector sensing either an increase or decrease indistance. Fig. 2 gives an examples of how the rings’motion makes both the direction and the magnitudeof the motion visually apparent, while maintaininga LIGO-like operation.This basic setup of the interferometer can be builtwith all grades of components, from inexpensivecraft mirrors and laser pointers suitable for classesof students to lab-grade optics. The majority of theoptical components we use (beamsplitter, mirrors,mounts, etc.) are either lab-grade parts, or bespokeparts manufactured by the University of Birming-ham’s mechanical workshop (see the accompanyingwebsite for example parts). This ensures the long-term stability of the configuration, achieves a shinyaesthetic appeal, and allows the public to encounterequipment which is frequently used in research labo-ratories. Large mirrors and a large beamsplitter (all2-inch diameter) are used to increase their visibilityand emphasise their importance. The resulting ex-hibit Michelson is shown in Fig. 3 with the laser onthe left. The exhibit contains more components than FIG. 3. The Michelson interferometer at the core of ourexhibit, shown in situ at the Thinktank Birmingham Sci-ence Museum. shown in the Fig. 1 schematic: in addition we usea screen and a webcam to display the interferencepattern and record it.We use a helium–neon (He–Ne) laser with an ex-posed view of the glowing gas. Seeing the glow ofthe exposed laser both attracts audiences and em-phasises the light source of the setup. We use aclass 1 laser for safety reasons and ensure that allbeams are contained in the circular base plate. Theprimary hazard is the high voltage required for thelaser (1 kV when running). All active componentsare encased in a plastic box underneath the opticalbase so that these cannot be accessed. The laseris mounted inside an acrylic tube with acrylic end-caps, and grounded to the base plate. The laserand all optics are encased inside an acrylic dome,protecting the optical components from damage andmisalignment.This risk of laser induced eye damage is extremelyremote due to the use of a low-powered laser andinability to misalign the interferometer. The Com-puter Aided Design (CAD) for the full bespoke lasermounting is shown in Fig. 4.LIGO is fine-tuned and controlled so that the in-terference pattern is almost completely dark unlessa gravitational wave passes through the detector.Our interferometer reacts to any kind of shakingmotion, meaning that the interference pattern con-stantly flickers. To observe the interference patternwe tested a large variety of screen materials. Thebrightest, highest contrast pattern was observed us-ing red card. The use of two diverging lenses (focal4
IG. 4. Computer aided design used to develop bespokeparts for the exhibit. The exposed He–Ne laser tube washoused inside an insulating acrylic tube and mounted onadapted lab-grade posts. This image was rendered usingBlender . length −
50 mm) allows us to create a large beam-spot. These, combined with a small difference in theinterferometer’s arm lengths, produce the ringed in-terference pattern that changes depending on therelative length of the arms.The Michelson interferometer described here canbe used as a stand-alone piece. In this configura-tion, it is well suited for use with small groups un-der the guidance of a trained demonstrator, enablingmore direct engagement . For example, the outerdome can be removed and properties of the interfer-ometer explored, such as demonstrating alignmentof optics, or pushing on the base to bend it slightly,changing the relative arm lengths.Without a trained demonstrator present, it is diffi-cult for the public to interact with the interferometerin a meaningful and safe way due to the requirementsof safety and robustness. To solve this problem, cus-tom exhibit software was developed to allow the userto learn more about the exhibit (Section IV), and in-teract directly with the interferometer without theneed to physically touch it. This reduces the risk ofdamage to the equipment or misalignment of the in-terferometer, and ensures that the interactions withthe exhibit are meaningful and repeatable. This in-teraction mechanism uses three piezoelectric trans-ducers (PZTs) mounted behind one of the end mir-rors in an equilateral triangle to minimise the tilt ofthe mirror. The PZTs convert electrical voltage intomotion, moving the mirror by up to 2 µ m. They aredriven using a combination of a Raspberry Pi andan Arduino Uno. Through the PZTs, the user can send a gravitational-wave signal , a set of predefinedcustom signals, to the gravitational-wave detector model. These simulated signals are amplified andsimplified versions of the kinds of waveforms thatLIGO and Virgo are searching for. There are twochirp signals similar to those observed so far frommerging black holes and neutron stars , and two asyet unseen signals: a continuous-wave signal which isexpected from rotating neutron stars , and a burstsignal which could come from events like supernovaexplosions . The simulated signals, while not exactreplicas of the real events, produce visually differ-ent interference patterns, allowing members of thepublic to see the connection between an astrophys-ical object and the interference pattern observed ina LIGO-like detector.The user can push one of four buttons to select agravitational-wave signal. We use four arcade-stylebuttons chosen for their colourful and rugged nature;the bold design makes them easy to identify, attrac-tive to children and suitable for prolonged use .The chosen gravitational-wave signal is sent to theRaspberry Pi and then to the Arduino Uno, whichtransfers the signal to the PZTs, resulting in a mov-ing interference pattern. Although the actual mo-tion of the mirror is undetectable by eye (a fewhundred nanometres), the change in the interferencepattern is apparent, giving an indication of the preci-sion measurements possible with real gravitational-wave detectors. The average maximum frequency foreach of the detections during the first and second ob-serving runs of LIGO and Virgo is several hundredhertz , whereas the human eye struggles to noticeflickering over 25 Hz. The signals are therefore de-signed to fit this frequency limit and are amplifiedby twelve orders of magnitude when compared withthe real signals to make their effect clear.The ability to interact with the exhibit stimulatesmembers of the public to ask more in-depth ques-tions on the nature of the exhibit and also the ofthe real gravitational-wave detectors . At sciencefairs, jumping or walking near the exhibit provokedquestions about how to remove seismic noise fromthe detector output. This creates an opportunity fordemonstrators to talk about different noise sourcesin the detector. As the output signal is not perfect,it also encourages questions about the data analysisinvolved to identify and characterise astrophysicalsources. B. Software
Museum exhibits need to be self-explanatory :typically a specialist will not be present to guidethe user or answer any questions. To achieve a self-5xplanatory exhibit, we use interactive software topresent a mixture of video, images, and text, pro-viding a varied range of learning materials . Thespecific configuration of software and hardware usedin the Thinktank and the RSSE are detailed in Sec-tion IV and Section V, respectively.There is existing software to create museum ex-hibits, such as Open Exhibits or Intuiface . Adrawback of these is that they are either written inolder programming languages such as Actionscript3, as is the case with Open Exhibits, or are costlyto design and run, like Intuiface. Existing solutionstherefore lacked the flexibility required for our ex-hibit.We have developed our own exhibit software basedon a number of Javascript libraries including re-veal.js, socket.io and johnny-five . The advan-tages of using these libraries are that they; are mod-ern; have a relatively low barrier to entry; can bevisualised easily through the use of HTML and CSS,which are key web technologies; are easy to develop;and are widely supported and will be supported foryears to come. This allowed more time and effortto be spent on content creation, rather than thefunctionality of the exhibit. Wide support for thissoftware makes installation in a variety of locations,with different sets of requirements, easier to man-age. The packages used in this project are also open-source, reducing the overall cost.The software is flexible in terms of the availablefeatures and customisation. It can be used with oneor two display screens and is touch-screen compat-ible. A set of navigable pages can be created inorder to guide a user through the materials. Thetop-level menu provides a selection of topics, leadingto sub-pages with further information. This allowsusers to direct their own learning , and poten-tially build upon the understanding achieved at aprevious visit . Within the sub-pages, a combina-tion of animations, images and text can be displayed.The display can also be timed to return to the homescreen after a set idle time, which is desirable whenthe exhibit software has been left on a sub-page sothat it is ready for the next user. Selecting a topiccan also trigger a pre-recorded video to begin play-ing, as well as simultaneously displaying live datafrom the exhibit.In the gravitational-wave detector model, a web-cam is used to show an enlarged view of the currentinterference pattern on the screen, and a photodi-ode embedded in the centre of the screen takes alight intensity reading, which is live-plotted usingcustom-written graphing software. This helps largergroups of people to clearly see the interference pat- tern from a distance and the changing intensity read-ing in response to different signals. A schematic viewof all the signal paths used in our exhibit is shownin Fig. 5. As described in Section III A, the user candirectly interact with the exhibit via a selection ofbuttons. Pressing one of these selects a simulatedgravitational-wave signal to send to the interferom-eter. The buttons can also be lit up in patterns viathe software, making them attractive to push. WebcamPC Display (Webcam, Graphing,Exhibit Content)
Raspberry PI Arduino (Buttons)
ButtonsArduino (PZT)
Interferometer PZTPD Arduino (PD)
30 m Gap
FIG. 5. Signal flow chart for the exhibit. A Raspberry PIruns the exhibit software and interfaces with the piezo-electric transducers (PZTs) and the photodiode (PD).When used at the Thinktank Birmingham Science Mu-seum, the PC is separated by 30 m from the rest of theexhibit, so communication must be be done via ethernetrather than a simpler, faster USB connection (USB islimited to 5 m).
IV. LONG-TERM INSTALLATION AT THETHINKTANK BIRMINGHAM SCIENCEMUSEUM
The Thinktank is part of Birmingham MuseumsTrust , a registered charity responsible for man-agement of museum sites and collections owned byBirmingham City Council. There are eight museumsites managed by Birmingham Museums Trust, in-cluding the Birmingham Museum and Art Galleryand Sarehole Mill as well as the Thinktank Birm-ingham Science Museum . The Thinktank receives230 ,
000 visitors per year, including 45 ,
000 fromschools, 10 ,
000 from other school-aged groups, and152 ,
000 general visitors, 95% of whom visit withchildren . It houses a wide variety of objects andexhibits, ranging from natural sciences, includingfossils and wildlife specimens, to science and indus-try, including a planetarium and a large collectionof steam engines.6ne of our main considerations when translatingour experience with hands-on demonstrations to alongstanding museum piece was developing a robustdesign. Accessibility of information is also impor-tant. It is impractical for a museum exhibit to re-quire an expert person present to explain it at alltimes; therefore, we needed to find alternative waysto convey our enthusiasm for the subject.The exhibit, initially installed in June 2016 ,is housed in the Futures Gallery. The gallery con-sists of a series of bays, each focused around a centralprop, as depicted for our exhibit in Fig. 6. Props aremounted to a narrow post connected to a raised falsefloor which curves up to form a low (approximately0 . . It is 1 . ◦ forward, and located in front of the rear projectedscreen, which is predominantly used to display videocontent.The rear projected screen constantly shows a largelive feed of the interference pattern, allowing visitors to clearly see a zoomed in view of how the patternchanges due to floor vibrations as they move around,or as they simulate a signal via the arcade buttons.A simultaneous view of the actual pattern on thescreen inside the interferometer dome, along withreal time graphing of the photodiode output, showsthat this video is indeed a live feed from the instru-ment they see in front of themIn a museum setting, visitors will often spend lim-ited time interacting with any particular exhibit. Itis important to make the scientific information quickand easy to access, whilst also providing depth andvariety for longer interactions . The software de-scribed in Section III B was developed specificallyto achieve this, and to enable direct interaction withthe exhibit. The content was developed in collabora-tion with the Thinktank to ensure the language wassuitable and accessible for both children and adults.The colour scheme was checked to be colour-blindfriendly using Color Oracle and the fonts chosenwere sans-serif as there are indications that thesefonts may be more dyslexia-friendly .There are two points of user interaction in themuseum installation of the exhibit. First, the touch-screen computer in front of the exhibit, and secondlythe four buttons used to simulate exaggerated ex-amples of gravitational-wave signals. The buttonscan be pressed at any time during interaction withthe exhibit. The touch-screen provides a selection oftopics and a quiz.Typically, selecting a topic will trigger a short (un-der 90 s) video to play on the rear screen, duringwhich visitors are free to click through several short( <
80 words per page) text and graphic pagesthat expand on the content of the video. The videoduration is displayed on the screen, so that a vieweris aware that the video is short . An example of thetouch-panel display is shown in Fig 7. Gravitational-wave astronomy is an area of current rapid progress;therefore, the software has been configured to al-low for periodic expansions as new research resultsemerge, and to provide new content for the repeatvisitor .The videos themselves cover four core topics: (a)an explanation of the gravitational-wave detectormodel on display and what the interference patternmeans; (b) the nature of gravity and how gravita-tional waves are produced; (c) how interferometersare used to search for gravitational waves; (d) thedetection of gravitational waves, including the firstdetection in 2015 and other observations since. Eachvideo is subtitled, and features enthusiastic membersof our group as well as video clips, graphics and an-imations created by others in the LIGO Scientific7 a) Photograph (b) Sketch FIG. 6. Our exhibit as configured for use in the Thinktank Futures Gallery. Visitors stand next to a touchscreenpanel which they can use to select from a range of topics and either read, watch videos, or take a quiz. Pushing thearcade-style buttons sends a signal to the driven mirror, emulating the effect of different types of gravitational wavesignals. The interferometer model itself is mounted at an angle on a narrow post behind a low barrier approximately1 m from the viewer. All cables to and from the model, touch-screen, and buttons are concealed inside the post andunder the raised false floor. Approximately 2 m from the viewer, at the back of the exhibit, a second projected screendisplays a live feed of the interference pattern alongside the currently selected video. The static text panel to the leftcontains some overview information about the exhibit.
Collaboration . This is intended to help the publicmake a human connection to the science we discuss,and to share in our excitement for the subject .The four presenters (two female and two male) wereall PhD students from the University of BirminghamInstitute for Gravitational Wave Astronomy. Stu-dent presenters were chosen as we considered themto be more relatable and intentionally avoided theold-professor stereotype of a physicist . Thevideos were produced in collaboration with the com-munications department of the College of Engineer-ing and Physical Sciences at the University of Birm-ingham.Finally, the quiz option facilitates a more activeinteraction where the user can test their knowledgeand receive confirmation of their understanding .The quiz is intentionally short (four questions) anddisplays the final score in a chalk-board image nextto Einstein.In summary, the software, interferometer andstatic display together provide the capability forvarying levels of self-guided interaction within a mu-seum setting. A short engagement allows for a littleinformation about gravitational waves to be gained;however, much more detailed information is avail-able if a visitor chooses a prolonged interaction. V. THE ROYAL SOCIETY SUMMEREXHIBITION
The RSSE is an annual week-long festival hostedat The Royal Society in London, celebrating cutting-edge science and technology in the UK. Each year,around 20 teams from university research groups andindustry are selected to develop a science-fair styleexhibit, staffed full-time by members of each group.The week caters to a range of audiences, includ-ing days for school groups, a Twilight Science openevening, press and media sessions, evening soir´ees forselected guests including politicians and celebrities,and is also open to the general public. The layout ofthe RSSE is such that visitors move from one exhibitstand to the next.In 2017, UK members of the LIGO EPO group were selected to jointly host a stand at the RSSEnamed Listening to Einstein’s Universe . Thespace for our stand was a 4 m × IG. 7. A screenshot of the custom exhibit software asused in the Thinktank. The screenshot shows one of thetopic pages—from the main menu users select a topicthey are interested in (gravity in the case shown here).They can then navigate through these information pagesusing the menu shown on the left-hand side, or by usingthe previous and next buttons at the bottom. They canthen return to the main menu via the
Back button in thebottom-left. The timer in the bottom-center indicatesthe duration of any videos currently playing in seconds. have been developed by the Birmingham group overthe last 10 years , and (b) our gravitational-wave detector model, on loan from the Thinktank.The hardware and software configuration used atthe RSSE is depicted in Fig. 8. In comparison to thatused in the Thinktank (see Section IV and Fig. 6),the core components—the interferometer and a dis-play screen—were identical; however, the interactiveself-guided content was not required because expertswould always be on hand to explain the model inperson. Instead, a smaller single television screenwas used to show the live video feed of the interfer-ence pattern alongside a graph plotting the inten-sity of light at the centre of the interference pat-tern over time. The simplified display allowed fortailored explanations to suit each visitor: a shortoverview at times of high footfall (such as for schoolgroups), or an expanded explanation when time andvisitor interest allowed. The additional electronicsrequired for the RSSE setup were concealed on alower shelf behind a tablecloth. The arcade buttonswere housed in a separate box to prevent the physicalbutton press shaking the exhibit and drowning outthe injected signal. The graph plotter and button (a) Photograph (b) Sketch
FIG. 8. Our exhibit as configured for use at the RSSE2017. The gravitational-wave detector model is mountedon three feet and sits on a round table 0 . Listening to Einstein’sUniverse stand (on the upper-right), and displays livefeeds of both the interference pattern, and a graph ofthe resulting measured signal. box were first developed for the RSSE and subse-quently implemented at the Thinktank.Over the course of the week, approximately 30volunteers staffed our stand in rotation, with vary-ing levels of expertise in interferometry, hardware,and software engineering. We produced a manualfor general maintenance of the model, and one ofthe Birmingham team was available if any furtherquestions arose. The robustness and safety of thedesign required for the Thinktank meant that trans-porting the model was relatively simple, and thatthe installation time and maintenance of the exhibitwere minimal.The model was located towards the back of the
Listening to Einstein’s Universe stand; therefore,the aesthetic choices made to attract visitors atthe Thinktank, such as the use of bright colouredlights and shiny components, were advantageoushere too . The dome on the interferometer meantthat we could safely invite visitors to take an up-close look at the optics, which was not only usefulfor explaining the model but also a necessity in thecompact space.The combination of aesthetic and practical designchoices made for a museum setting also resulted in9 model that is robust and exciting to use at sciencefairs such as the 2017 RSSE. By adapting the mul-timedia content provided, the configuration of ourexhibit can be tailored to suit a wide range of con-texts. VI. IMPACT
Our exhibit was installed at the Thinktank in June2016. Since then, it has spent over two years housedat the Thinktank Birmingham Science Museum, aswell as being included as part of the
Listening toEintein’s Universe stand at RSSE. An early proto-type was featured on the UK’s Channel 4 news dur-ing the media campaigns surrounding the announce-ments of the first gravitational wave and first binaryneutron star detections. Each stage of the exhibit’slife has enabled us to explore different aspects of userengagement and reception, as well as the broader im-pact of the project. In this section, we describe ourfeedback collection at the RSSE and the improve-ments we have implemented to the exhibit based onexperiences gained in a science-fair setting. We alsodescribe how observations in the museum setting ledto further modifications. Finally, we consider thewider impact of the project for our group and thecollaborations and opportunities resulting from it.
A. Feedback from the Royal Society SummerScience Exhibition 2017
As described in Section V, the RSSE is an annual,week-long festival celebrating innovation and ad-vancements in science, technology, engineering andmathematics (STEM) research by British researchgroups and companies. It is open to all and free. Theaudience ranges from school groups and the generalpublic to politicians and celebrities.In 2017, 10 ,
123 members of the public, 2002 stu-dents and 262 teachers visited the RSSE . Schoolgroups (age 14+) are required to register in ad-vance to attend the dedicated sessions, bringing upto 25 students per group . This means that stu-dents at the RSSE are more likely to be from schoolswith proactive class teachers who were motivated topursue out-of-classroom activities, and the studentsthemselves are a sub-selection from their year groupor class (e.g., most interested, most likely to benefitfrom attending, etc.), biasing the sample comparedto the entire population of schoolchildren. The audi-ence at the RSSE is therefore not ideal for engaginghard-to-reach demographics with research. Survey respondents Total (Paper | Electronic)
Total responses 171 (63 | | | | | The Exhibition was an opportunity to showcasethe LIGO Scientific Collaboration’s work to boththe public and to high-profile individuals. The highfoot-fall and many in-person interactions providedan occasion to gather feedback and measure the im-pact of our exhibit.We created two types of survey that were usedthroughout the week. The first used an establishedelectronic survey platform, completed via a pair oftablets at the exhibit stand. This targeted individ-uals or small groups, typically older teenagers oradults. Questions typically asked the user to ratetheir opinion on a Likert scale from 1 to 5. The sec-ond survey format was on paper, aimed at youngerattendees by using a range of graphical questionformats rather than the traditional multiple choicequestionnaire. This format was also better suitedto large groups since the paper forms could be dis-tributed quickly to an entire class. The surveys werecreated to be short, with eight and ten questions forthe paper and electronic versions respectively. Bothversions took no more than a few minutes to com-plete.The questions within each of the two survey typeswere not identical (due to the different survey for-mats); however, both aimed to assess change in theindividual’s interest in physics and gravitational-wave research, as well as their interest in specificparts of the
Listening to Eintein’s Universe stand.The demographics of those surveyed through bothformats are summarised in Table I. A conscious ef-fort was made to maintain a gender balance acrosseach session of those surveyed over the course of theweek.A summary of responses to key survey questionsis given in Table II. While the surveys generally dis-cussed the exhibit as a whole, many people spent asignificant fraction of their time at the interferome-10 ow much has your knowledge of gravitational waves changed?
No change Large increase (Electronic) 1% 80%How much has your knowledge of LIGO changed?
No change Large increase (Electronic) 7% 66%Rate your interest in physics before & after visiting the RSSE ∗ ∗∗ (Paper) 49% 13%TABLE II. Selected survey results from the RSSE 2017. The left column shows the question where the labelsElectronic and Paper are as described in Table I. The right columns show some key results. * 83% of those specifying no change were already Highly or Very highly interested in physics before attending the RSSE. Most peopledescribed their interest before the RSSE as OK (32%) or High (32%), and after the RSSE as
Very highly (51%). Percentage increasein interest across genders: female 75%, male 48%, other/undisclosed 75%.** All of those specifying a 2-step increase were aged 11–16 (22% of 14–16 year-olds and 8% of 11–13 year-olds) ter model, and, therefore, the results are consideredreflective of the model gravitational-wave detector.According to questions from the electronic survey,we found that
Listening to Eintein’s Universe wassuccessful in both its core goal of increasing peo-ple’s awareness of gravitational waves (80%
Largeincrease ), and in informing them about LIGO (66%
Large increase ), which was just one of several re-search projects mentioned at the exhibit. Similarly,the paper survey indicated that 62% of those askedwere more interested in physics than they were previ-ously, while a large majority of the rest were already
Highly interested in physics before they arrived. Fur-ther analysis was possible using the data from thepaper forms. The exhibit was particularly effectiveat engaging some of the 11–16 age range: all of thoseindicating a large increase in interest in physics (8 ofthe 63 people surveyed) were in this age bracket. Italso effectively engaged girls, 75% of whom indicatedincreased interest. It is possible that this is a resultof our efforts to deliberately include both male andfemale volunteers in every session of the exhibition.These results indicate that we had a positive im-pact on those who attended, but cannot tell us more,such as long-term impact on the attendees, how wemight improve our engagement with the public us-ing the interferometer in the future, or how to reachaudiences who have less prior interest in, or accessto, STEM subjects. This is an area of active explo-ration for the group in the future.
B. Feedback from the Thinktank FuturesGallery Installation
We now consider our exhibit installation in theThinktank, with a focus on the audience reachedand the changes we made over the course of the in- stallation. With a long-term installation, the exhibitcan come into contact with more people over a pro-longed duration.In order to gauge the impact of our exhibit in thissetting, we monitored visitor interactions with theexhibit and performed a short survey in the museum.The number of survey participants was small, andthus robust conclusions could not be drawn basedon response statistics. Despite the small participa-tion, we found some useful information and severalareas of improvement were clear from observationsof museum visitor behaviour and interaction.In its first iteration, the exhibit was placed to-wards the back of the Futures Gallery in a rela-tively dark corner. Opposite this position was oneof the museum highlights:
RoboThespian , a talking,singing robot. Many visitors were observed to headdirectly for the robot, skipping the back corner ofthe gallery entirely . On the day of the survey,only 13 of 200 people who entered the gallery inter-acted with our exhibit. In light of this behaviour,we have worked with the museum to place our ex-hibit closer to the gallery entrance, providing both amore prominent position, where visitors are liable tospend more time , and also offering better lightingto attract the visitor.Of 13 people who interacted with the exhibit, ninetook part in the survey. They were asked about theirprior knowledge of gravitational waves. We foundthat three had not heard of gravitational waves be-fore seeing the exhibit and none had any awarenessof the University of Birmingham’s involvement inthe discovery of gravitational waves. All found theexhibit at least Quite informative and
Fairly easy touse .Our observations at the Thinktank, and our expe-riences at the RSSE, led us to make modifications toimprove upon the museum installation design. The11nitial installation did not have interaction buttons,or a navigable touch screen. The only means of in-teraction with the exhibit was through a roller-ballmouse and a single click button. In the RSSE, wefound that the interaction worked well through thearcade-style buttons. The subsequent addition of atouch screen has brought the exhibit more up-to-date with modern technology and, therefore, morefamiliar to the visitor .Our future work will included continued monitor-ing of visitor interaction with the exhibit to assessthe success of the modifications and new position,as well as further areas of improvement. C. Wider Impact
We also consider the wider impact of this work.The experience gained in designing, implementingand evaluating the exhibit, as well as exposure toscience communication professionals, has enabled usto explore new means of sharing gravitational-wavescience to non-expert audiences in engaging and ac-cessible ways. It has also led to further work beyondthe exhibit itself.One such project is an interdisciplinary collabo-ration with audio–visual digital artist Leon Trim-ble . The project, Gravity Synth , is a musical in-strument combining a Michelson interferometer witha modular synthesiser . The interference patternfrom this interferometer is converted into sound viaa photodiode, and processed through a modular syn-thesiser, exploring the relationship between gravi-tational waves, vibrations and sound. The Grav-ity Synth has been performed at a variety of eventsranging from arts and music festivals such as LunarFestival (2018), Future Everything (2018) and TheSuperposition (2017), to science orientated eventsincluding Cheltenham Science Festival (2019), Pintof Science (2017), Interact Engagement Symposium(2017) and was featured on the BBC’s Digital Planet18th birthday show (2019) .We have also formed collaborations with otheruniversity departments who are keen to include aMichelson interferometer as part of their own publicengagement schemes. As a result of this work, ourgroup has built a similar interactive interferometerfor the University of East Anglia as well as additionalsmaller, more portable variations for our own use.Our website details component lists and instruc-tions for use by others to build their own interfer-ometer. Alongside this, we are investigating makinglow-cost interferometer kits with novelty elementssuch as building blocks as a more affordable and fun means for schools to create their own Michelson in-terferometers similar to existing examples from theLIGO EPO group which use magnets and glue . VII. CONCLUSIONS AND FUTUREACTIVITIES
At this exciting time in gravitational-wave re-search and discovery, our aim for this project wasto bring this research to a wider community in anaccessible way. We have designed and built a physi-cal exhibit and custom-made exhibit software whichare able to explain what gravitational waves are, howthey are detected, and the recent discoveries. By in-stalling the exhibit in the Thinktank BirminghamScience Museum, we have a long-term means of in-creasing the community awareness of the researchtaking place at a local university.Looking further afield, the exhibit has also beenshown at the 2017 RSSE. Attending a national sci-ence festival like this allows us to have a geographi-cally wider reaching impact in the shorter term.We have monitored the reception to our exhibitand taken action in response to what we have learnt.Upon re-installation to the Thinktank, the exhibithas upgraded custom software, greater user interac-tion, and is now in a more prominant museum posi-tion. We will continue to monitor and improve uponthe installation via collaboration with the Thinktankstaff and surveying the musuem visitors.This project has led to further work ingravitational-wave public engagement, including col-laborations with artists bringing this research to apotentially new audience at arts and music festivals.In the long term, the project will have a lasting roleon an international scale with online instructionsand parts lists to enable others, including schoolgroups, to build their own versions of this exhibit. ACKNOWLEDGEMENTS
This project was funded by Science and Technol-ogy Facilities Council Small Award (project numberST/N005767/1) and also by the Royal Astronomi-cal Society Public Engagement Grant (2015). Theauthors are grateful to both the Thinktank Birm-ingham Science Museum and The Royal Society fortheir support and advice for the exhibit develope-ment and installation. The authors express theirdeep gratitude to all those who have helped withthis project since its inception, in particular: Steve12rookes and all those in the University of Birming-ham mechanical workshop; John Bryant and DavidStops for additional technical support; AlejandroVigna G´omez for his contributions to the exhibitvideos; Siyuan Chen and Carl-Johan Haster for use-ful discussions for the exhibit development; and Ju-lia Dancu and Luke Scantlebury-Smead for conduct-ing the Thinktank survey. The authors thank LauraTrouille and Joey Shapiro Key for their commentson this manuscript; the LIGO Education and PublicOutreach working group for their support and for use of multimedia resources; and Sarah Cole, Kris VogtVeggeberg and Jayatri Das for useful recommenda-tions for literature on exhibit design. This workwas also supported by STFC Grant ST/N000633/1and the University of Birmingham. CPLB is sup-ported by the CIERA Board of Visitors ResearchProfessorship, and NSF Award PHY 1912648. HMis also supported by the Australian Research CouncilCentre of Excellence for Gravitational Wave Discov-ery (OzGrav) (project number CE170100004). Thiswork has been assigned LIGO document numberP2000036. Albert Einstein. Approximative Integration of theField Equations of Gravitation.
Sitzungsber. Preuss.Akad. Wiss. Berlin (Math. Phys.) , 1916:688–696,1916. BP Abbott, R Abbott, TD Abbott, et al. Observa-tion of gravitational waves from a binary black holemerger.
Physical review letters , 116(6):061102, 2016. B. P. Abbott et al. Properties of the BinaryBlack Hole Merger GW150914.
Phys. Rev. Lett. ,116(24):241102, 2016. B. P. Abbott, R. Abbott, T. D. Abbott, et al. GWTC-1: A Gravitational-Wave Transient Catalog of Com-pact Binary Mergers Observed by LIGO and Virgoduring the First and Second Observing Runs.
Physi-cal Review X , 9(3):031040, Jul 2019. J Aasi, BP Abbott, R Abbott, et al. Advanced LIGO.
Classical and Quantum Gravity , 32(7):074001, 2015. F. Acernese et al. Advanced Virgo: a second-generation interferometric gravitational wave detec-tor.
Class. Quant. Grav. , 32(2):024001, 2015. Yoichi Aso, Yuta Michimura, Kentaro Somiya, et al.Interferometer design of the KAGRA gravitationalwave detector.
Phys. Rev. D , 88(4):043007, August2013. Bala Iyer, Tarun Souradeep, C.S. Unnikrishnan,et al. LIGO-India. Technical report M1100296-v2dcc.ligo.org/LIGO-M1100296/public, Nov 2011. B. P. Abbott et al. GW150914: First results fromthe search for binary black hole coalescence with Ad-vanced LIGO.
Phys. Rev. , D93(12):122003, 2016. BP Abbott, R Abbott, TD Abbott, et al. GW151226:Observation of gravitational waves from a 22-solar-mass binary black hole coalescence.
Physical ReviewLetters , 116(24):241103, 2016. BP Abbott, R Abbott, TD Abbott, et al. GW170104:Observation of a 50-solar-mass binary black hole co-alescence at redshift 0.2.
Physical Review Letters ,118(22):221101, 2017. B.. P.. Abbott et al. GW170608: Observation of a 19-solar-mass Binary Black Hole Coalescence.
Astrophys.J. , 851(2):L35, 2017. BP Abbott, R Abbott, TD Abbott, et al. GW170814:A three-detector observation of gravitational wavesfrom a binary black hole coalescence.
Physical ReviewLetters , 119(14):141101, 2017. BP Abbott, R Abbott, TD Abbott, et al. GW170817:observation of gravitational waves from a binaryneutron star inspiral.
Physical Review Letters ,119(16):161101, 2017. B. P. Abbott et al. GW190425: Observation ofa Compact Binary Coalescence with Total Mass ∼ . M (cid:12) . 2020. Virgo Collaboration educational resources.public.virgo-gw.eu/educational-resources/. Larry E. Suter. Visiting science museums during mid-dle and high school: A longitudinal analysis of studentperformance in science.
Science Education , 98(5):815–839, 2014. Neta Shaby, Orit Ben-Zvi Assaraf, and Tali Tal. ‘Iknow how it works!’ student engagement with ex-hibits in a science museum.
International Journal ofScience Education, Part B , 9(3):233–252, 2019. Katherine P. Dabney, Robert H. Tai, John T. Almar-ode, et al. Out-of-school time science activities andtheir association with career interest in stem.
Interna-tional Journal of Science Education, Part B , 2(1):63–79, 2012. Paichi P. Shein, John H. Falk, and Yuh-Yuh Li. Therole of science identity in science center visits andeffects.
Science Education , 103(6):1478–1492, 2019. Anthony Lelliott. Understanding gravity: The roleof a school visit to a science centre.
InternationalJournal of Science Education, Part B , 4(4):305–322,2014. J. Falk and L. Dierking.
Learing From Museums .Rowman & Littlefield, Lanham, MD, 2000. John H. Falk and Mark D. Needham. Measuring the mpact of a science center on its community. Journalof Research in Science Teaching , 48(1):1–12, 2011. John H. Falk and Lynn D. Dierking. School field trips:Assessing their long-term impact.
Curator: The Mu-seum Journal , 40(3):211–218, 1997. Charlotte Bond, Daniel Brown, Andreas Freise,and Kenneth Strain. Interferometer Techniques forGravitational-Wave Detection.
Living Rev. Rel. , 19:3,2016. Thor Labs. Michelson interferometer educational kit. Particle Toys Nikhef. Nikhef interferometer. Dennis Ugolini, Hanna Rafferty, Max Winter, CarstenRockstuhl, and Antje Bergmann. Ligo analogy lab—aset of undergraduate lab experiments to demon-strate some principles of gravitational wave detection.
American Journal of Physics , 87(1):44–56, 2019. P. J. Fox, R. E. Scholten, M. R. Walkiewicz, andR. E. Drullinger. A reliable, compact, and low-costMichelson wavemeter for laser wavelength measure-ment.
American Journal of Physics , 67(7):624–630,Jul 1999. Bob Peart. Impact of exhibit type on knowledge gain,attitudes, and behavior.
Curator: The Museum Jour-nal , 27(3):220–237, 1984. Dorothy Lozowski Boisvert and Brenda Jochums Slez.The relationship between exhibit characteristics andlearning-associated behaviors in a science museumdiscovery space.
Science Education , 79(5):503–518,1995. John H. Falk and Leslie M. Adelman. Investigatingthe impact of prior knowledge and interest on aquar-ium visitor learning.
Journal of Research in ScienceTeaching , 40(2):163–176, 2003. John Falk. The director’s cut: Toward an improvedunderstanding of learning from museums.
Science Ed-ucation , 88(S1):S83–S96, 2004. John H. Falk, Carol Scott, Lynn Dierking, LeonieRennie, and Mika Cohen Jones. Interactives andvisitor learning.
Curator: The Museum Journal ,47(2):171–198, 2004. Jennifer A. Fredricks, Tara Hofkens, Ming-Te Wang,Elizabeth Mortenson, and Paul Scott. Supportinggirls’ and boys’ engagement in math and science learn-ing: A mixed methods study.
Journal of Research inScience Teaching , 55(2):271–298, 2018. John J. Koran Jr., Mary Lou Koran, and Sarah J.Longino. The relationship of age, sex, attention, andholding power with two types of science exhibits.
Cu-rator: The Museum Journal , 29(3):227–235, 1986. Toni Dancstep (n´ee Dancu) and Lisa Sindorf. Exhibitdesigns for girls’ engagement (edge).
Curator: TheMuseum Journal , 61(3):485–506, 2018. Cody Sandifer. Technological novelty and open-endedness: Two characteristics of interactive exhibitsthat contribute to the holding of visitor attention ina science museum.
Journal of Research in Science Teaching , 40(2):121–137, 2003. J. Falk and L. Dierking.
The Museum ExperienceRevisited . Routledge, New York, 2013. Robin Meisner, Dirk vom Lehn, Christian Heath,et al. Exhibiting performance: Co-participation inscience centres and museums.
International Journalof Science Education , 29(12):1531–1555, 2007. Constanze Hampp and Stephan Schwan. The role ofauthentic objects in museums of the history of sci-ence and technology: Findings from a visitor study.
International Journal of Science Education, Part B ,5(2):161–181, 2015. Blender Online Community.
Blender - a 3D modellingand rendering package . Blender Foundation, Sticht-ing Blender Foundation, Amsterdam the Netherlands,2019. Luisa Massarani, Lara Mucci Poenaru, Jessica Nor-berto Rocha, Shawn Rowe, and Sigrid Falla. Adoles-cents learning with exhibits and explainers: the caseof maloka.
International Journal of Science Educa-tion, Part B , 9(3):253–267, 2019. Keith Riles. Recent searches for continuousgravitational waves.
Modern Physics Letters A ,32(39):1730035–685, Dec 2017. B. P. Abbott, R. Abbott, T. D. Abbott, et al. Firsttargeted search for gravitational-wave bursts fromcore-collapse supernovae in data of first-generationlaser interferometer detectors.
Phys. Rev. D ,94(10):102001, Nov 2016. Sue Allen and Joshua Gutwill. Designing with multi-ple interactives: Five common pitfalls.
Curator: TheMuseum Journal , 47(2):199–212, 2004. Bruce Wyman, Scott Smith, Daniel Meyers, andMichael Godfrey. Digital storytelling in museums:Observations and best practices.
Curator: The Mu-seum Journal , 54(4):461–468, 2011. Minda Borun and Jennifer Dritsas. Developingfamily-friendly exhibits.
Curator: The Museum Jour-nal , 40(3):178–196, 1997. Toni Dancstep (n´ee Dancu) and Lisa Sindorf. Cre-ating a female-responsive design framework for stemexhibits.
Curator: The Museum Journal , 61(3):469–484, 2018. John H. Falk, John J. Koran R., and Lynn D. Dierk-ing. The things of science: Assessing the learningpotential of science museums.
Science Education ,70(5):503–508, 1986. Ideum. Open Exhibits. openexhibits.org. reveal.js – The HTML Presentation Framework. re-vealjs.com. Socket IO. socket.io. Johnny-Five: The JavaScript Robotics & IoT Plat-form. johnny-five.io. Private communication. Lauren Deere, 2018. Anna Green. Gravitational waves in the Birminghamscience museum.
LIGO Magazine , 9:20–21, Sep 2016. Thinktank Birmingham Science Museum. Exhibitiondesign and construction guidelines. internal documentfor exhibitors, 2011. Color Oracle 2018. colororacle.org. British Dyslexia Association – Typefaces for dyslexia2018. bdatech.org/what-technology/typefaces-for-dyslexia/. Louis S. Nadelson. Who is watching and who is play-ing: Parental engagement with children at a hands-onscience center.
The Journal of Educational Research ,106(6):478–484, 2013. Jill Hohenstein and Lynn Uyen Tran. Use of questionsin exhibit labels to generate explanatory conversationamong science museum visitors.
International Journalof Science Education , 29(12):1557–1580, 2007. Royal Society Summer Exhibition2017. royalsociety.org/science-events-and-lectures/2017/summer-science-exhibition. LIGO Scienitifc Collaboration universities and insti-tutions at RSSE: University of Birmingham, Impe-rial College London, University of Glasgow, CardiffUniversity, University of the West of Scotland, Uni-versity of Southampton, University of Sheffield, AEIHannover, AEI Potsdam–Golm, Milde Marketing. RSSE ‘Listen to Einstein’s Uni-verse’ page. royalsociety.org/science-events-and-lectures/2017/summer-science-exhibition/exhibits/listening-to-einsteins-universe. L. Carbone, C. Bond, D. Brown, et al. Computer-games for gravitational wave science outreach: Blackhole pong and space time quest.