Computers in Human Behavior 25 (2009) 599–608

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier .com/locate /comphumbeh

An interactive tool to promote musical creativity in people with dementia

Philippa Riley *, Norman Alm, Alan NewellSchool of Computing, University of Dundee, Perth Road, Dundee, DD1 4HN Scotland, UK

a r t i c l e i n f o

Article history:Available online 1 October 2008

Keywords:MultimediaHCIDementiaAlzheimer’s DiseaseMusicCreativity

0747-5632/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.chb.2008.08.014

* Corresponding author. Tel.: +44 (0)1382 386537;E-mail address: philippariley@computing.dundee.a

a b s t r a c t

Dementia is a growing problem that affects the lives of those diagnosed and caregivers, with symptomshaving an effect on memory, communication, the ability to learn new skills and problems with behaviour,such as aggression, agitation and depression. Participation in activities can improve quality of life for peo-ple with dementia, reducing behavioural problems and aiding relaxation. Research has established thatpeople with dementia can be both artistically and musically creative, and have an appreciation of musiceven in the latter stages of the disease. The symptoms of the disease, however, mean that supportingmusic making activities with this group of people is challenging. This paper describes a prototype systemdesigned to enable people with dementia to create music, using a touch screen interface to control a sys-tem which utilises chords to create pleasant-sounding music regardless of any prior musical knowledge.Results of usability studies suggest the system is easy to use, and that pleasant-sounding music can becreated with it. Participants, including people with dementia, appeared actively engaged during use ofthe system, many reporting they enjoyed the experience. Future testing will establish the degree to whichpeople with dementia could be musically creative using such a system.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Dementia currently affects over 27.5 million people worldwide(Wimo, Jonsson, & Winblad, 2006), and with the aging populationsteadily increasing, the number of people with dementia is ex-pected to rise. Ferri et al. (2005) estimate that the number willdouble every 20 years and will exceed 81 million in the year2040. Given the problems associated with the condition, such anincrease in the number of people with dementia is likely to havefar-reaching consequences on families and available communityresources. Dementia, a condition that more commonly affects olderpeople, encompasses a range of brain disorders that progressivelydamage extensive areas of the brain, eventually leading to cerebralfailure. The main symptoms of dementia are memory impairment,a decline in cognitive function, and problems with executive func-tioning (affecting the ability to use imagination, organise and for-ward plan). Normal everyday functioning is affected because ofthis damage, and an individual’s condition will deteriorate overtime (LoGiudice, 2002).

Although finding a cure for such a debilitating condition isimportant, researchers also recognise the need to provide a betterquality of care for people with dementia with facilities that en-hance day-to-day living. There are various ways in which thesafety and care of people with dementia can be improved, for in-stance by training and educating caregivers (Jacques & Jackson,

ll rights reserved.

fax: +44 (0)1382 385509.c.uk (P. Riley).

2000; Kitwood, 2005) and using assistive technology (Astell,2005). Another successful strategy is to provide meaningful activ-ities that people with dementia can participate in on a daily basis.These activities should reflect both what individuals would like todo and what they are able to do. Volicer and Bloom-Charette (1999,p. 8) believe the provision of meaningful activities for people withdementia ‘is the most important factor, but the hardest to achieve’.Their investigations, involving people with dementia who also hada diagnosis of depression, found a 90% improvement in mood whenparticipants were involved in meaningful activities.

Although there are benefits associated with participation inactivities, the symptoms of the condition make it difficult to deviseactivities that are accessible. This is caused by the person’s problemswith short-term memory, communication difficulties, decreasedconcentration levels and their inability to learn new skills. Researchis therefore being carried out to develop novel and interesting activ-ities for this group that are nevertheless failure-free – that is, activ-ities for which there is always a successful outcome. If effectivelyimplemented, such activities could enhance the lives of people withdementia and improve their social interaction and day-to-day living(Alm et al., 2005). Research has established that people with demen-tia have the capacity to continue to participate in a variety of activ-ities and tasks. It has been suggested that they can retain existingand develop new creative skills, both artistically and musically, de-spite the dementia (Miller, Boone, Cummings, Read, & Mishkin,2000; Miller, Yener, & Akdal, 2005).

It has been thought for some time that music can be experi-enced and enjoyed by people with dementia, and recent research

600 P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608

suggests this is evident even in the latter stages of the condition(Clair, Mathews, & Kosloski, 2005; Cuddy & Duffin, 2004). How-ever, without a specialist music therapist present, musical activi-ties provided by caregivers tend to be passive, i.e. listening tomusic or to others singing. Where music therapists provide activemusic making sessions, they often opt for simple percussioninstruments for basic rhythm making (Vink, 2002). This is becausethe use of traditional musical instruments for this group would beimpractical, not only because the symptoms of the conditionwould make it difficult to learn to play such an instrument, but alsobecause playing musical instruments successfully usually requiresprior training. Thus, although there are recognised benefits associ-ated with the provision of musical activities, caregivers lack the re-sources and possibly the skill to support active music makingsessions. There is thus a need to develop a novel instrument thatcould make creative music making accessible to people withdementia.

This research explored the development of a system that couldenable people with dementia to be creative through active musicmaking, whether they had pre-existing skills or not. The systemwas designed to be easy for caregivers to support, providing anengaging and enjoyable activity for those participating, with theadded potential to increase social interactions if used in pairs orgroups. It was hoped that the quality of life for people with demen-tia and caregivers could be improved through this creative musicalactivity.

2. Literature review

2.1. Dementia

Dementia is caused by widespread disorders of the brain. Thereare over seventy conditions that cause dementia, with each onehaving a ‘particular pattern of cognitive change that may beaccompanied by physical signs’ (LoGiudice, 2002). Alzheimer’s Dis-ease (AD) is the most prevalent of these causes, accounting for 60%of cases, with Lewy body variant (of Alzheimer’s Disease) account-ing for 20% of cases, frontotemporal dementia (FTD) (includingPick’s) 5%, vascular dementia (VaD) 5% and the remaining causes5% (Terry, 2006).

Dementia is progressive and leads to multiple dysfunctions inthe brain causing problems with memory, communication andjudgement (Jacques & Jackson, 2000). Various characteristics havebeen found to exist in all types of dementia (Aldridge, 2005; LoG-iudice, 2002), including short and long-term memory impairment;multiple cognitive dysfunctions, including memory impairment,language disorders and problems with abstract thinking and judge-ment (aphasia), an inability to comprehend what is said, felt andheard (agnosia), difficulty finding names of words (anomia) andproblems carrying out motor activities (apraxia); and frontal exec-utive dysfunction, affecting the ability to plan, to use one’s imagi-nation (abstraction) and the capacity to organise.

Memory loss is often seen to be the essence of the condition(Huub, 2005), this generalisation possibly arising because themajority of people with dementia suffer from AD where memoryloss is a prominent feature. For those with AD, short-term memoryloss (working memory) becomes more severely affected as thedementia progresses to the extent where it is virtually lost (Almet al., 2007). However, people suffering from FTD and VaD aremuch less likely to have problems with memory impairment. InFTD, memory is preserved to some extent (Mendez et al., 2006),and in VaD, memory impairment may be mild, or not present atall (Roman, 2003). In contrast to Huub (op cit), Roman (2003) be-lieves frontal executive dysfunction and cognitive decline are thecore symptoms of dementia, and not memory impairment.

The complexity of dementia, the way it affects individuals dif-ferently, and the potential to suffer from multiple causes of demen-tia at the same time, adds to the difficulty in fully understandingthe condition. Although some patterns are certain, such as the var-ious stages of deterioration those with AD will progress through, itis still impossible to predict effects precisely, i.e. determining howquickly individuals will progress from one stage to another. In thepast, the final stages of dementia were thought to leave individualswithout any mental powers, incapable of understanding, commu-nicating or being able to reason, needing everything done for them(Jacques & Jackson, 2000). However, it is now thought that peoplein the latter stages may still retain some mental abilities, such asthe capacity to recognise emotion in facial expressions (Luzzi, Pic-cirilli, & Provinciali, 2007), forward plan and set goals (Sabat, 2001)and to appreciate music (Sixsmith & Gibson, 2007). Dementia iscurrently one of the top five causes of death in the Western world(Huub, 2005), it being thought that as the functions of the braincontinue to deteriorate, its inability to adjust to changes in theenvironment mean life can no longer be supported.

2.2. Dementia care

Dementia has a huge impact on the person with the condition, aswell as those in the family unit and other caregivers (LoGiudice,2002). People with dementia may need continual care, dependingon the severity of their condition and this can make it extremely dif-ficult for carers to cope, whether family or professional. The treat-ment of people with dementia has been, and in some cases is stillless than ideal (BBC, 2007), and this is partly due to people not rec-ognising or appreciating the remaining capabilities of individualswith the condition (Sabat, 2001). Individuals are currently providedwith a variety of medications and care methods, including:

Medical; drugs continue to be developed to manage the symp-toms of dementia.

Person-centred care; in the past, poor results received in testsmeasuring cognitive function meant people in the latter stages ofdementia were considered to be in a vegetative state with noawareness. However, Sabat’s (2001) research suggested peoplewith dementia were able to retain the capacity to do and feelthings that such cognitive tests were not designed to measure. Sa-bat found that people with dementia retained the ability to expe-rience shame, embarrassment and pride; feel concern for others;maintain dignity; display and experience self-hood and find novelforms of communication to compensate for an inability to commu-nicate verbally.

Residential care homes and day centres; as well as housing peoplewith dementia and providing 24-h care, respite is also provided forpeople with dementia still living at home, allowing family mem-bers to have a break from care duties.

Provision of activities; shown to have a positive effect on peoplewith dementia, increasing relaxation, enjoyment, empowering theperson with dementia and ultimately improving quality of life.

Technology – assistive and leisure; assistive technologies are usedto help people with dementia retain some independence, or to helpothers to care for people with dementia, in that systems can bedeveloped as useful reminders as well as to alert carers when aperson is out of bed, or in a position of danger (e.g. near a kettle,or oven) (Astell, 2005). Leisure technologies aim to improve theday-to-day living of people with dementia by providing access toactivities that might otherwise be unavailable or too difficult todo (Alm et al., 2007).

2.3. Activities for people with dementia

There are various challenging behavioural symptoms associatedwith dementia, for example apathy, agitation, aggression and

P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608 601

shouting. These symptoms can be extremely stressful for the per-son with dementia, as well as for family members and professionalcarers. Volicer, Simard, Pupa, Medrek, and Riordan (2006) believesome behavioural symptoms occur because people with dementiaare unable to initiate meaningful activities for themselves. Theyfound the benefits of providing continuous activity programmesfor people with dementia resulted in a decrease in restlessnessand disruptive behaviour. They found individuals were more likelyto express happiness during activity sessions and experience in-creased undisturbed sleep. Improvements in behaviour and moodalso led to a reduction in the amount of psychotropic medicationsgiven to those participating in the study. Involvement in activitiesmay have additional positive effects on people with dementia byhelping them become more relaxed, and by reducing agitationand boredom (Jacques & Jackson, 2000).

There are various types of basic activities in which people withdementia can participate, including daily activities such as wash-ing-up, making sandwiches and clearing away items from the tableafter meals. The ability to carry out such tasks may remain possibleas they may continue to be familiar to people with dementia, andthey are therefore considered by Dowling (1995) to be the mostvaluable of all activity types. As the abilities of people with demen-tia gradually deteriorate, even basic tasks can become impossibleand so carrying out familiar activities can be extremely empower-ing, giving a sense of being useful to others. Similarly, participationin leisure activities can be beneficial as they enable people withdementia to actively involve themselves in hobbies and interests,rather than being non-participative. Leisure activities work wellboth for groups and individuals, for example groups participatingin art sessions at a day care centre or individuals listening to astory on a cassette or cd.

2.4. Creativity in people with dementia

The research of Gottlieb-Tanaka, Small, and Yassi (2003, p. 130)illustrates that people with mild/moderate dementia ‘are just asable to engage in creative expression as able seniors’, although theymay require more time to process ideas and then express them.Various researchers have investigated the potential for people withdementia to be creative, and it has been suggested that people withdementia may retain creative abilities (Crutch & Rossor, 2006;Cuddy & Duffin, 2004; Miller et al., 2000, 2005), and may even beable to develop new creative abilities because of the dementia(Miller & Hou, 2004; Miller et al., 2000, 2005). Miller et al. (2000)reported that people developed new musical and artistic skillsafter the onset of FTD. Pickles and Jones (2006) discuss a manwho, as his dementia became more severe, expressed himself moreeasily through his piano playing than he could verbally. Music hasthe ability to induce and portray emotions, and his wife noted thatthe music he played gave her an insight into his feelings at thattime.

Hannemann’s (2006, p. 60) research on creativity in people withdementia has led him to believe that ‘creative art expression mightplay a major role in expression of their unspeakable emotions’.Joining in with others and accomplishing different tasks can beempowering, and creative activities can provide new and novelways for people with dementia to communicate, which is particu-larly valuable for those people unable to communicate verbally.Also, feelings of isolation and depression are lessened when peoplewith dementia participate in creative activities together. Hanne-mann (2006, p. 64) claims ‘the act of creative expression and socialcontext are vital to the mental and physical wellbeing of the partic-ipants’. Activities such as music and painting can positively influ-ence the sense of wellbeing of people with dementia, with thoseparticipating in artwork and music showing improvements inself-esteem and mood (Hannemann, 2006) and decreased agitation

(Volicer & Bloom-Charette, 1999). Dowling (1995) claims the pro-cess of painting was extremely engaging for people with dementia.During creative sessions people became happier, with smiles oftenoccurring straight away. The people participating would begin toreminisce, and on occasions unprompted singing took place.

2.5. Technology for people with dementia

Although participation in activities has been shown to improvequality of life for people with dementia, the symptoms of dementiacan render some activities inaccessible. The inability to learn newskills, along with memory impairment and communication prob-lems can mean many activities are difficult or impossible for peo-ple with dementia to accomplish. Activities thus need to be chosenwith care and with individuals in mind, but this can be problematicfor family members and carers who struggle to find activities thatare accessible (Dowling, 1995). As well as the symptoms of thecondition preventing participation, people with dementia may alsobe aware of their own deterioration and so may be less willing tojoin in activities that could render them with a feeling of embar-rassment or failure. There is therefore a need to develop a varietyof leisure activities that have been tailored to meet the needs ofpeople with dementia, i.e. are failure-free, and that are easily sup-ported by carers.

Various systems have already been developed specifically forpeople with dementia. Much of this technology comes under thecategory of assistive technology, and this is generally used in twoways. Firstly to help carers by alerting them if/when a person withdementia may be in danger, and secondly, to help individuals withdementia to increase, maintain or improve operational ability intheir daily lives (Astell, 2005). Additionally, other systems devel-oped specifically for people with dementia can be better catego-rised as ‘leisure technology’, and are designed to allowparticipation in leisure pursuits that may be unavailable, or other-wise difficult or impossible to do. This type of technology can im-prove quality of life for people with dementia by providingactivities that the individual has never tried, or by offering newways to continue participating in a hobby or interest previouslyenjoyed. Current leisure technologies already developed for peoplewith dementia include:

2.5.1. Music memory lane and video memory laneThese systems enable people with dementia to listen to and

watch nostalgic music and video. Research showed these systemshad a positive impact in engaging people with dementia, with areduction in fidgeting also observed. The music memory lane sys-tem was found to be more effective than live music, and partici-pants chose to use the media memory lane systems more oftenthan other activities, remaining engaged for longer periods (Olsen,Hutchings, & Ehrenkrantz, 2000).

2.5.2. Picture gramophone (ENABLE project)This system was developed to provide a simple interface where

people could select music to play, watch the lyrics of the song ap-pear onscreen, and view photographs associated with the song. Themajority of participants were thought to benefit from using thesystem, choosing to sing and hum along to songs. A positive effectwas seen in the mood of most participants (Topo et al., 2004).

2.5.3. CIRCA reminiscence systemCIRCA enabled people with dementia to enjoy reminiscence in a

novel way using a hypermedia system. This technology facilitatedcommunication and reminiscence using generic photographs andnostalgic music via interaction through a touch screen. Users wereabsorbed by it, and it captured their imaginations, prompting them

602 P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608

(in some cases) to share stories that even close family membershad not heard before (Alm et al., 2007, 2005).

2.5.4. Living in the momentContinuing research by the developers of CIRCA has resulted in

the development of an interactive entertainment system, specifi-cally designed to establish if people with dementia can use, engagein and enjoy the system without the constant guidance and sup-port usually needed from carers. The system provides a series ofvirtual environments that the user can explore and interact with(Alm et al., 2006b).

2.5.5. The INDEPENDENT projectAs part of this project, researchers developed a simple music

player for people with dementia, and also a system that wouldstream images into the homes of people with dementia givingthem a ‘window on the world’ (Orpwood, Sixsmith, Torrington,Chadd, & Chalfont, 2007).

2.5.6. New technologies for creativityAlthough new technologies are being developed for people with

dementia, there is still little being designed to enable them to be cre-ative. When developing new technologies for creativity, it is impor-tant to consider the person with dementia in terms of their creativeabilities and how the symptoms of dementia affect them. This shouldhelp ensure the activity becomes more accessible to them. New tech-nologies must continually incorporate ways to bypass and managethe problems caused by the condition. It has been established thata simple, intuitive interface is best when designing technology forpeople with dementia (Alm et al., 2007). Making the technology sim-ple to use may enable a person with dementia to more easily partic-ipate in the activity, even though their ability to learn new skills isbecoming progressively more difficult or even impossible.

Given the potential for music to communicate emotion, musiccould be a way for people with dementia to express themselves cre-atively. It may be that active music making could enable people withdementia to communicate through their own musical compositions,which could be both an enjoyable and potentially empowering expe-rience. Music making sessions could be developed for both individ-uals and groups. In terms of musical creativity tools, it appearsthat nothing has been developed specifically for people with demen-tia. However, a system called Hyperscore that was developed for useby novice composers (Machover, 2004), was taken to a hospital andamongst the patients who used it were two people with dementia.Boulanger, a researcher at the MIT Media Lab, commented that‘‘one group of patients with dementia worked in pairs, with one indi-vidual managing the computer interface and the other making com-position suggestions. Their ability to create interesting pieces using acollaborative format was very promising, and almost shocking com-pared to findings . . . in traditional music therapy, where collabora-tion between individuals is difficult without the facilitator beingvery directive” (Boulanger, personal communication). However,Boulanger does not say at which stage of dementia the patients were,how long they played with the software, or whether they enjoyed itand were engaged, although he does suggest the patients used thesystem together without any help. Therefore, further evaluationsare required in order to establish whether Hyperscore can be usedby people with dementia to create meaningful music.

3. Research

3.1. Research question

This research concentrated on developing technology to sup-port people with dementia in their participation in musical activ-

ity. The aim of this research was to provide an activity currentlyunavailable – the creation of meaningful musical compositions.The intention was to develop a system that would enable peoplewith dementia to create music without the need for prior musicalor instrumental skill, via a user-friendly, intuitive, failure-free sys-tem. If this can be achieved using technology that is both intuitivefor the user and easy for carers to support, then the act of musicmaking may be more widely available to people with dementiawho do not have access to music therapists, or relatives and carerswho are musical.

3.2. Preliminary research

The initial phase of the research involved one of the researchers(Riley) volunteering at an Alzheimer’s day care centre in Dundee,Scotland for one morning a week, taking part in activities and mak-ing informal observations of clients and carers. She observed thegroup whilst participating in all activities, and notes were writtendirectly after leaving the centre. The researcher continues to volun-teer at Alzheimer Scotland to gain further insight into the world ofpeople with dementia attending day centres. On joining, the groupconsisted of 10 clients, 3 carers and 2 volunteers. The clients gen-erally appeared to have mild dementia, although two clients dis-played symptoms of moderate to severe dementia. The purposeof the observations was to recognise and appreciate the effect ofdementia, not only on the clients, but also on those caring for them.The researcher had 12 years experience of helping to care for afamily member with dementia, and was therefore aware of the ef-fect of the condition on the family unit. Initial observations high-lighted the difficulty carers had in finding varied activities thatthey (a) felt confident enough to take and (b) believed clients coulddo.

Activities observed over the first six weeks consisted of discus-sion groups and physical activities (such as bowling), and it wascommon for some clients to be unable or unwilling to participate.After six weeks the researcher asked clients and carers if theywould like musical activity sessions introduced. They agreed totry this and she began bringing in her keyboard, playing traditionalScottish songs and old music hall songs. This enabled her to ob-serve how the clients reacted to musical activities. The followingpositive changes were observed immediately:

� Some clients became noticeably more relaxed.� Some clients began to dance.� Many clients began to sing or hum.� Many clients would tap their feet in time to the music, and clap

their hands.

One client who often sat crying and picking her hands, or whowas otherwise aggressive and abusive, instantly became calmwhen the music began and would happily dance on her own, orwith a carer or volunteer. If sitting, she would make pronouncedmovements with her hands and arms, seemingly absorbed in themusic. Other clients who did not always participate in other activ-ities reacted well to the music, singing or humming and tappingtheir feet, etc. It appeared they were enjoying the activity, and atleast four clients continued to make verbal reference to the musicsessions on a weekly basis. Two of the clients often asked the re-searcher if she had brought her keyboard, although one would callit a ‘music box’ and the other, who had more severe communica-tion difficulties, would imitate playing a piano in order to get herquestion across. These sessions worked well, and so other musicalactivities were introduced in order to bring variety to the groupsessions. As well as introducing other songs (such as those sungby Frank Sinatra, Shirley Bassey and Elvis Presley) for group singingand dancing sessions, musical activities also included:

P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608 603

� Reminiscing about pieces of music.� Debating the meaning of songs after they had been played.� Guessing the instrument sound or sound effect.� Teaching the researcher old Scottish tunes by clients, carers and

the other volunteer singing a song line by line. The researcherwas able to learn the tune and eventually everyone sang thesong whilst the researcher played.

� Name that tune, where many clients would recognise the tunevery quickly. It was interesting to observe clients choosing toguess the song after nine or 10 notes, but who were often sing-ing the song after only two or three.

� With the researcher being English, much fun was had with theclients transposing Scottish words into English, and then singingthe song in an English accent.

The decision to concentrate on musical creativity was based onthe belief that other creative activities involving art (painting, col-lage work and crafting) appeared to be more readily available forpeople with dementia than musical creativity sessions. This is pos-sibly due to a number of reasons, including:

� Music making activities are more difficult to support than artsessions in terms of the skills required by the person takingthe session, and so there may be a lack of confidence on the partof non-musical carers to take active music making sessions.

� The materials required to participate in art sessions (paper,paint, pencils, etc.) may be cheaper and easier to obtain thanmusical instruments.

� There may be a lack of music therapy specialists available totake sessions.

� There are no musical instruments currently available that wouldbe simple enough for a person with dementia to play withoutprior training, and the symptoms of dementia may prevent evenpreviously trained/accomplished musicians from continuing toplay.

From her observations, the researcher believed a new musicalinstrument that was both intuitive and simple to use would pro-vide a musical activity easy for carers to support, where the spec-ialised knowledge of music therapists would not be necessary.

3.3. The system

It was proposed a system be developed that addressed the fol-lowing requirements.

� Is easy and intuitive to use.� Plays music instantly, i.e. providing one-touch feedback.� Allows users to express themselves creatively.� Can be used by individuals, in pairs or in groups.� Does not require a prior knowledge of music.

Although providing an easy and intuitive design for people withdementia is by no means a trivial task, the CIRCA reminiscence sys-tem has already successfully been developed for this group of peo-ple. CIRCA (Computer Interactive Reminiscence and ConversationAid) uses a touch screen that people with dementia can press in or-der to make choices to view photographs and listen to pieces ofmusic from an era appropriate to them. This system has been sub-ject to detailed evaluation and has proved to be easy to use by peo-ple with dementia. CIRCA encouraged people with dementia totake a more equal role during reminiscence than happened withtraditional reminiscence, where carers were under more pressureto direct the sessions and promote discussions. People withdementia were able to decide which material they wanted to view,

and this was empowering for them as they were given more con-trol over the session (Alm et al., 2006a).

CIRCA was developed by a multidisciplinary team and theircombined expertise in the areas of psychology, design and soft-ware engineering enabled them to develop technology that had aprofessional, aesthetically pleasing and intuitive design. Almet al. (2007) established that when developing for people with cog-nitive dysfunction, design considerations should incorporate thefollowing features:

Use of touch screens; found to be used successfully by peoplewith dementia. There is a physical connection between the userand the display, with users receiving immediate feedback whenthey touch the screen – something that cannot be imitatedwhen watching another person use a mouse or keyboard.Controls positioned at the bottom of the screen; reducing fatiguewhen users hold their arms out to press the screen.Large format screen and large font sizes; invites participation andprovides easy to read prompts for those with visualimpairment.Minimal use of text; although people with dementia may retainthe ability to read and understand text, their limited short-termmemory can make it difficult to remember long passages oftext.Use of a hypermedia structure with limited options for selection;enabling the user to navigate easily between multimedia, andin the case of CIRCA, ensuring the user was never in the wrongplace or lost, because there is no wrong place to be. The usercould navigate to all sections easily, and the design encouragedthem to explore and hopefully enjoy the material provided, ulti-mately providing a failure-free experience.Attractive design; Alm et al. (2007) believe the professional andinviting design of their system is, in part, responsible for engag-ing users. They relate this to the ‘aesthetic-usability effect’,where aesthetic designs are thought to be easier to use thandesigns that are less aesthetically pleasing (Lidwell, Holden, &Butler, 2003).

The proposed system for musical creativity follows on from thedesign methods used for CIRCA, and provides the user with a sim-ple and uncluttered interface. As with CIRCA, the system was givena name – ‘‘ExPress Play” – and a professional look with a polishedvisual display. This name encapsulated the whole purpose of thesystem – to ‘express’ oneself through musical play, and to ‘press’to play. Minimal text with a large font size was used. Navigationwas kept to a minimum, ensuring that users could not ‘get lost’,providing a failure-free environment. Buttons were generally posi-tioned at the bottom of the screen to reduce movement. UnlikeCIRCA where the screen is positioned vertically, it is intended toposition the touch screen horizontally at a slight incline on a tablein front of the person. This is a familiar position generally adoptedby many people when doing activities such as writing letters anddrawing. This provides a more comfortable position for extendedplay sessions than a vertical touch screen.

Music making is a pastime generally limited to people who havemastered playing an instrument. However, the researcher’s ownexperience of music gave her insights into the possibility of usingchords as a means of being able to play pleasant-sounding musicregardless of prior musical knowledge. A chord is a set of threeor more notes that are played simultaneously and are usually thefirst, third and fifth notes taken from a musical scale (Levitin,2006). It has been established that any chord can follow anotherand always sound harmonic (Jourdain, 2002), and so chords shouldbe able to be used successfully in a music making system. Thesechords can be further grouped to enhance harmony, as in Western

1 For interpretation of the references to colour in this figure text, the reader isreferred to the web version of this article.

Fig. 1. Welcome screen.

604 P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608

culture chords are split into two categories – major and minor. Thisis an important way to give emotional meaning to western music –major chords are generally thought to portray a happy, activemood and minor chords, a melancholy or sad mood. This portrayalof emotion is recognisable within Western culture, regardless ofmusical training (Levitin, 2006). Given the above, it would seemplausible that the system could not only be developed for peoplewith dementia to do creative music making, but also as a way forthem to express emotion.

It was therefore decided to develop a system that would allowpeople with dementia to play music specific to three moods (hap-py, sad and angry). The design would have an uncomplicated nav-igation system and a simple interface. Selections providedthroughout the system’s interface would be kept to a minimumin order that users would not become confused. Visual feedbackwould be provided instantly during music play to help individualswith severe short-term memory loss to remain engaged whileusing the system, as well as adding to the aesthetics of the design.As dementia can affect a person’s ability to read and comprehendtext, it was important to provide both visual and audio promptsthroughout their interaction with the system. Those able to readand comprehend words would be given the option of a ‘text’ but-ton displaying textual prompts, while those with some communi-cation difficulties would be given the option of a ‘picture’ buttondisplaying facial emotion (e.g. a happy face).

3.4. Prototypes

The system was designed based on knowledge already gainedfrom others designing technology for people with dementia (Almet al., 2007), and from the researcher’s own experience (seeabove) and her knowledge of how dementia affects motor skills,communication and memory. Small-scale prototypes were cre-ated using PowerPoint to establish how the system might lookand work, and to explore how the system could express emotionthrough the use of chords. Hawthorne (2007) has commentedthat the use of paper and simplified prototypes is less effectivewith older people. Where this is the case, such an effect wouldbe exacerbated by the symptoms associated with dementia. Thusit was decided to develop a functional prototype at an earlystage in the development of the system. It was important thatthe system should operate correctly at all times, as inappropriateactions occurring during evaluations could cause participants tobecome unnecessarily anxious or confused. Prototypes wereshown to colleagues and peers, and to experts in the field ofdementia in order to gain initial reactions to design, usabilityand potential. Responses were positive, especially with regardto the novelty of the design.

In terms of developing the system to express emotion throughmusic, there are various attributes that can be applied in combina-tion to portray specific moods. The greater the number of attri-butes used, the more reliably the music portrays the intendedemotion. Attributes include tempo (the speed or pace of the mu-sic), pitch (how high or low a note is), intervals (the relationshipbetween two pitches), timbre (tone quality), attack (the way asound begins) and articulation (techniques that affect the way sin-gle, or multiple notes are performed). Juslin (2005) suggests com-mon attributes used for different emotions include:

Happiness; fast tempo, major key, ascending pitch, perfect 4th/5th intervals, staccato articulation, smooth and fluent rhythm,bright timbre, fast tone attacks and sharp contrasts between‘long’ and ‘short’ notes.Sadness; slow tempo, minor key, narrow pitch range, descend-ing pitch, small intervals (e.g. minor 2nd), legato articulation,dull timbre, slow tone attacks and slow vibrato.

Anger; fast tempo, minor mode, dissonance (discord), major7th/augmented 4th intervals, staccato articulation, complexrhythm, sudden rhythm changes (e.g. syncopations), sharp tim-bre and fast tone attacks/decays.

The chords used to portray the happy emotion incorporatedmany appropriate attributes. The rhythm used – four quavers,the last tied to a minim – had a contrast of short and long notesand a ‘smooth and fluent’ rhythm with a fast tempo. The soundused was a clarinet with a bright timbre, and the chords used wereall from major keys. The attributes of the sad emotion includedproviding users with the ability to play sustained chords. Thesound used was cellos with a slow vibrato and dull timbre. Inter-vals were small (minor 2nd), thus giving a narrow pitch range overthe 11 chords provided. With regards to a slow tempo and changesin timing, this would be dependent upon how the user played mu-sic on the system. The chords used for the angry emotion portrayedthe mood by incorporating the following attributes. The soundused imitates that made when slamming a hand down hard on apiano over multiple adjacent keys. Dissonance was thereforeachieved through there being only a single interval between notes(giving a ‘clashing’ effect). The initial sound was accented andattacking before finally falling away. However, as with the sadmood, rhythm and tempo would be controlled by the user as theyplayed.

3.5. Layout and design

The interface was designed to be uncluttered, providing onlynecessary options to users. Colours used matched general precon-ceptions of the emotions being portrayed, with the happy musicscreen having a yellow background, the sad music screen havinga blue background and the angry music screen, a red background.1

Other screens were given a neutral green background, and all back-ground colours were muted rather than garish, as it was thought thiswould be more aesthetically pleasing for the intended user agegroup (50+).

The interface provided a welcome screen with a visual andaudio prompt asking the user to ‘‘press anywhere to start” (seeFig. 1). When pressed, the user navigates to a screen that has theoption to select either ‘‘words” or ‘‘pictures”. Depending on the

P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608 605

choice made the user navigates to a screen with word or pictureprompts, the only difference being that buttons either have texton them or pictures displaying facial expressions. Similarly, thesubsequent prompt either asks the user to ‘‘press on a word belowto select a mood”, or ‘‘press on a face below to select a mood”.

Once the user selects a mood (happy, sad or angry), they navi-gate to one of six screens where they play happy, sad or angry mu-sic. The user is prompted to ‘‘press in this box to start playing”.When the screen is touched, two types of instant feedback are gi-ven simultaneously – audio and visual – and so users hear a chordplaying whilst also seeing circular shapes appear on the screen (seeFig. 2). These circular shapes not only provide instant and constantfeedback, but also add interest to the screen. As the user’s fingermoves, so they begin to draw on the screen leaving a trail of circlesbehind. This visual trail provides a continuous prompt that some-thing happens when the screen is touched. This kind of promptingis particularly important for those with severe short-term memoryloss, and may increase the length of time a user plays with thesystem.

As the user moves over the screen, different sized circles willappear. The size of the circle is dependent upon the speed thatthe user moves their finger across the screen. The faster they move,the smaller the circles that appear. Different sized circles were in-cluded as being more aesthetically pleasing than circles of thesame size. The circle colour was chosen to match the mood se-lected, with the happy screen having yellow circles, the sad screenhaving blue circles and the angry screen having red circles. The cir-cle colour was made darker than the background and the circleshape made 3-dimensional, rather than 2-dimensional, in orderthat the circles would stand out in the foreground. The end ofthe trail of circles is set to fade as the user moves around thescreen, and this ensures both the background and the trail beingcreated are easily visible to the user at all times.

When different areas of the screen are touched, chords of differ-ent pitch and volume play. Moving up the screen causes the pitchof the chord to become higher, whereas moving down the screencauses the pitch to become lower. This provided an intuitive inter-face where a ‘‘higher” note was ‘‘higher” up the screen. A furthervisual clue to the change in pitch was provided by the inclusionof horizontal lines on the screen. As the user moves over a line,so the chord sound changes pitch. Volume is dependent on wherethe user touches the screen on the horizontal axis. As the usermoves to the right the sound will become louder, and as they moveto the left the sound will become quieter, being in keeping with the

Fig. 2. Showing angry music screen including trail produced during play.

normal stereotype of a volume control. Being able to change vol-ume may enable users to add more feeling to their music, as ithas been suggested that dynamics are used in musical composi-tions to change the emotional effect of melody and harmony (Jour-dain, 2002). As the intended user may have a hearing impairment,however, volume has been set so that it does not fall below a cer-tain value.

During play, if a finger is dragged from the top of the screen tothe bottom of the screen, a passage of chords will be played in se-quence (starting with a high pitched chord that gradually gets low-er). The music produced should sound both harmonious andmusical. By touching different parts of the screen, users playchords of different pitch and volume and may be able to developa melody, whilst also playing music that portrays the mood ini-tially selected.

4. Evaluation

Three pilot studies were carried out to test the usability of thesystem using a functional prototype. The tests provided qualitativedata through observations and video recordings, and a question-naire in the second Pilot Study provided quantitative data. Dueto ethical issues in research involving people with dementia, initialevaluations were performed with users who did not have demen-tia. Once it was ascertained that the system was working suffi-ciently well to be introduced to people with dementia, a furthertrial was done with this group.

4.1. Method

4.1.1. Pilot Study 1The first study examined reactions to the usability and func-

tionality of the primary design, and involved four participantswho did not have dementia. Participants were two females (aged37 and 61) and two males (aged 37 and 43), of which two wereacademics, one a retired secretary and one a teacher. The re-searcher observed the participants while they navigated theinterface and played with the system. Participants were advisedto play as long as they wanted, and to comment or ask for helpif required.

4.1.2. Pilot Study 2The second study tested the usability and effectiveness of the

system design with participants who did not have dementia, butwho would be comparable in age to the intended users of the sys-tem. These tests were intended to highlight usability and design is-sues that people with dementia might find problematic, but thatcould be too difficult for them to communicate verbally. In termsof design, it was thought that the likes and dislikes of these partic-ipants would be comparable to people with dementia of a similarage range. Each participant was provided with an informationsheet and a consent form. Five females and six males participated– three participants being aged 50–60, three 61–70, two 71–80 andone aged 80+. Two did not give their age, but were within the over-all age range of the other participants.

Participants were given a brief explanation of the purpose of thesystem, and told that the study would help identify potential prob-lems with the usability and design of the system. Two camcordersrecorded participants while they used the system, and the re-searcher also observed all the experiments. Although a strict timelimit could have been implemented it was more appropriate to rig-orously test the usability of the system, and also allow users to ex-press themselves creatively for as long as they wanted. At the endof the session participants were asked to fill out a questionnaire.Questions regarding age and musical experience were included,

Table 1Selected questions asked to participants in Pilot Study 2

Question Strongly/agree

Unsure Strongly/disagree

ExPress Play is an easy way to play music 100 0 0It did not take me long to learn how to use it 82 0 18I liked how ExPress Play looked 100 0 0I liked the colours used in ExPress Play 91 0 0I found the look of ExPress Play patronising 0 9 82ExPress Play enabled me to play music that

sounded pleasant91 0 0

I was able to make up and play my own musicaltunes

55 45 0

I liked the music that I composed 55 45 0I enjoyed using ExPress Play 100 0 0I would have another go if I had the opportunity 100 0 0I enjoyed being able to play music 100 0 0I thought the happy music accurately portrayed a

happy mood91 9 0

I thought the sad music accurately portrayed asad mood

91 9 0

I thought the angry music accurately portrayed aangry mood

91 9 0

606 P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608

as well as 27 statements representing usability, learnability, de-sign, enjoyment and musicality of the system. Positive and nega-tive questions were included, for example:

� I found ExPress Play easy to use.� ExPress Play is an easy way to play music.� I did not understand how to use it at first.� I found the look of ExPress Play patronising.� ExPress Play enabled me to play music that sounded pleasant.� I thought the sad music accurately portrayed a sad mood.

Participants were asked to select their level of agreement foreach statement using a five-point Likert scale (strongly agree,agree, unsure, disagree and strongly disagree).

4.1.3. Pilot Study 3The third study was carried out with a group of people with

dementia and consisted of observing ten participants with mildto moderate dementia using the system. In some cases the partic-ipant used the system with only the researcher present, but in fourcases it was appropriate for both the researcher and a carer to bepresent.

4.2. Results

4.2.1. Pilot Study 1: Four participants who did not have dementiaParticipants used the system for between 5 and 15 min. All par-

ticipants said they enjoyed using the system to play music. Oneparticipant also said she thought her late mother (who had demen-tia) would have enjoyed using the system. Participants did notcomment negatively on the usability of the system, although theprompt to play music obviously caused problems. The prompt ask-ing the user to ‘‘press in this box to start playing” led two partici-pants to press down on the screen and pull their finger upimmediately. This caused a chord to play for the duration of thepress (less than a second) and for one circle to appear. A demon-stration was required to illustrate that sound would play whenthe user touched the screen (whether pressing down or draggingaround) and would stop when they removed it. One participantwondered if a clicking sound heard when moving between adjoin-ing chord sounds would be annoying for users. Another participantsuggested that intervals used between chord sounds should beexamined.

4.2.2. Pilot Study 2: 11 older participants who did not have dementiaChanges made to the design of the system on completion of Pi-

lot Study 1 included changing the prompt for playing from ‘‘pressin this box to start playing”, to ‘‘drag your finger around the screento play”. Participants played for as long as they wished. Sessionsvaried in time from 3 to 21.5 min. Six people previously playedan instrument – one for 0–1 years, two for 2–10 years, two for20+ years, one did not specify. No participants currently playedan instrument.

Table 1 shows the results of selected questions asked to partic-ipants relating to the usability, learnability, design, enjoyment andmusicality of the system. Most participants agreed it did not takethem long to learn how to play, with all participants finding thesystem an easy way to make music. This was supported by obser-vations recorded of participants navigating through screens with-out any real difficulties and positive comments being made, suchas ‘‘it was easy to use and fun to play”. All participants stated thatthey enjoyed using the system, and this was further supported bycomments made such as, ‘‘great fun”, ‘‘I’d be quite happy to playwith that” and ‘‘I enjoyed that”.

Results show participants believed they were able to playpleasant-sounding music using the system, and this is in line with

participants being observed tapping their feet, nodding their headand moving their body rhythmically to the music they were creat-ing. Although over half of participants answered that they couldmake up their own musical tunes, 45% were unsure and this corre-lates with the same number of participants being unsure whetherthey liked the music they composed.

Participants liked the design of the system and the colours used,and the majority did not think that the design of the system waspatronising. All participants who answered agreed that the happy,sad and angry music accurately portrayed the appropriate mood.One participant commented, ‘‘that’s very expressive”, and someparticipants went on to comment that they thought people withdementia would be able to use the system to express how theywere feeling through the mood they selected.

Observational data showed participants navigating the systemin different ways when wanting to select another sound. Somepressed the ‘Finish’ button to start again, whereas others pressedone of the other face/text buttons to change mood. Participantswho pressed the ‘Finish’ button did not appear to be negatively af-fected by going back to the start each time. No participants usedthe ‘Start Again’ button. The most common comment from partic-ipants was to ask why the circle shapes changed size.

Participants appeared to enjoy the whole experience, withcomments made such as ‘‘you could really get a lot of peoplein to that. You can hear it, you can see it”, ‘‘fascinating designs”,‘‘I really like that” and ‘‘the music’s fascinating”. All participantsagreed they would have another go on the system if given theopportunity.

4.2.3. Pilot Study 3: 10 participants with mild to moderate dementiaThe initial screen asking users to ‘press anywhere to start’

caused difficulties for some participants who appeared apprehen-sive about where to press given the choice of the whole screen.The ‘Start Again’ button, which appeared at the bottom of thescreen when the user began playing music, distracted some partic-ipants causing them to say things like ‘‘start again . . . , why?”.Pressing this button then confused them as it directed them backto the mood selection screen. Additionally, some participants acci-dentally selected buttons when – perhaps either due to mobilityproblems or their misunderstanding of the consequences – theheel of their hand rested on the bottom of the screen. Again, thisproved to be distracting and confusing, especially when playingmusic at the time.

P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608 607

Problems occurred with the prompt on the play screen, ‘‘dragyour finger around the screen to play music”. The word ‘around’caused at least one participant to move in a circular motion withinthe play screen. Another participant traced his finger ‘around’ theedge of the screen, though whether this is attributed to the promptwording is not clear. The words ‘play music’ in this prompt alsocaused some participants to comment that they were ‘not musical’every time they saw/heard the prompt, and they then had to beencouraged to continue using the system with comments that itwas designed to be used by anyone, regardless of musical experi-ence. It became evident very quickly that additional prompting bythe researcher or carer was necessary for many participants. It alsoappeared that selection of mood was commonly decided by howthe participant was feeling at that moment. In some cases partici-pants did not want to select the sad or angry moods at all, comment-ing that they did not feel that way or did not like other people to feelthat way. Participants had to be encouraged to select these othermoods in order for them to experience the sad and angry options.

All participants appeared to enjoy using the system. One partic-ipant laughed animatedly every time she played/heard a sound onthe system. Another participant, when she initially began makingmusic, commented, ‘‘it’s beautiful. . . I’ve never seen anything likethat before”. Other comments included ‘‘this is marvellous”, ‘‘I can-nae believe that”, ‘‘I like this” and ‘‘brilliant. . . I like the combina-tion of the music”. Some participants commented that they didnot like the ‘angry’ music, although interestingly most continuedto play after making the comments. Various comments were madedescribing what the trail of circles looked like, from a ‘caterpillar’to a ‘necklace’, and some participants seemed to be trying to drawpatterns or pictures. When both the researcher and a carer werepresent, the playing of the system became a more social activity.Communication was promoted when the participant was playing,with them giving immediate feedback about what was happeningwhen they played. Additionally, if the carer played (whether todemonstrate to the participant, or for their own entertainment),conversation continued, with participants commenting on whatthe carer had played. All laughed when a smiling participant saidto the carer who was playing, ‘‘that’s not bad..!”.

5. Discussion

Pilot Study 1 suggested minor changes be made to the interface,and these were implemented. Pilot Study 2 provided data fromparticipants who were the same age as the target user group. Expe-rience of playing a musical instrument did not appear to affect theamount of time participants in Pilot Study 2 used the system, withthe six previous players using it for an average of 9 min 50 s, andthe five non-players, an average of 6 min 33 s. Quantitative results,together with analysis of video recordings, showed the system waseasy to use, enjoyable to play and provided a way to produce pleas-ant-sounding music. Participants navigated through the screensquickly, and were able to begin playing music with minimal addi-tional prompting. No changes were made to the design of the de-vice as a result of this study.

On completing Pilot Study 2, the system appeared to be at anappropriate stage to introduce to people with dementia. PilotStudy 3 established a variety of usability issues not apparent instudies 1 and 2, affirming that people with dementia have prob-lems with simplified tasks. The design of the welcome screenneeded revisiting, as a lot of additional prompting was requiredto help users navigate. The screen asking users to select ‘words’or ‘pictures’ seemed an unnecessary complication to the design –with prompting again required to help users navigate this screen– and navigation could be simplified by removing this screen.The wording of the prompt ‘‘drag your finger around to play music”

caused various problems, and so this could be changed to a moresimplified statement to help users understand more easily whatthey are being asked to do.

Some participants preferred not to select either the sad or angrymoods, seemingly because of what they believed they would becommunicating. Although this supports the hypothesis that peoplewith dementia could use music as a viable means of communica-tion, this device has primarily been developed for musical creativ-ity. In this version of the system the use of faces to illustrate moodmight have been a factor in users not wanting to select two of thesounds provided. It is thus proposed that two versions of the sys-tem be created for evaluation – one that has sound selection bymood (using facial icons), and one that has sound selection byinstrument (using pictures of instruments, e.g. clarinet, cello andpiano). This should establish if selections made are affected whenselection of sound is by mood, and may also identify whetherinstrument selection is a more effective way of engaging users inthis creative activity.

It was noticeable that some participants accidentally made but-ton selections with the heel of their hand when they were intend-ing to play. Placing buttons at the bottom of the screen had beendone to aid comfort when pressing (Alm et al., 2007). Repositioningof buttons to the top of the screen should prevent accidental selec-tion, and as the screen is positioned horizontally rather than verti-cally, moving the buttons to the top of the screen should not causemobility problems for users.

Participants in all studies became engaged in the activity, butthis was more meaningful in Pilot Study 3. The users with demen-tia all appeared to be engaged in the activity, with one particularlypoignant moment being when a participant whose verbal commu-nication was limited and whose mobility was particularly affected,put both his hands on the screen. He focussed wholly on the screenand appeared totally absorbed in the sounds he was creating. Thelocation of his fingers and thumbs on the screen, along with a con-stant change in the pressure of his fingers on the screen, caused thepitch of the chords to change. However, it is not clear whether theparticipant was changing the pressure purposely or not. This was aspecial moment as the participant stayed like this for some time(2 min 55 s), and when prompted again, played in the same wayagain for 1 min 22 s. He communicated at times, but it was notpossible to understand what he was saying. However, his body lan-guage and the intense manner in which he focussed on the screenwhilst keeping his hands there suggested he was enjoying thewhole experience.

Although no additional data was gained in terms of usabilityfrom Pilot Study 2 – the healthy older adults did not find naviga-tion or use of the system difficult – parallels can be drawn betweenhow both groups reacted to the design of the system and theirengagement and behaviour when playing. The results in Table 1show participants from Pilot Study 2 liked the design of the system,while observations of participants in Pilot Study 3 saw one partic-ipant leave because she thought it was ‘‘for bairns” – that is, chil-dren. Participants from both groups had similar behaviours,smiling on first hearing the music, making comments of enjoymentand moving their body or tapping their feet whilst playing.

6. Conclusions

The study with people with dementia established that theywere able to use the system to create pleasant-sounding music,and that with minor adjustment, the system could provide an easyway to bring active music making to people with dementia,whether for use at home, in residential housing or in day care cen-tres. Using the system to play music to two other people provided apositive experience for participants, and, if the prompting system

608 P. Riley et al. / Computers in Human Behavior 25 (2009) 599–608

is made more effective, the system could provide an activity forindividual use, as well as for use in pairs or groups. As Holmes,Knights, Dean, Hodkinson, and Hopkins (2006) showed, live musicis received particularly well by people with dementia, and this sys-tem could give a sense of empowerment to users if playing for oth-ers, as well as the potential benefits for those listening. This systemhas been shown to be easily supported by carers, which confirmsits potential as a regular activity for people with dementia.

These studies suggest that carrying out evaluations withhealthy older adults may be a useful way to gather informationwhen designing for people with disabilities of the same age. Obser-vations may also highlight similarities in behaviour between bothgroups, and although this does not help designers establish if theirsystems for creativity and leisure activities are usable, they may beable to use this information to gain an insight into how their targetgroup might react.

Further evaluations will be used to establish whether peoplewith dementia can use the system to be musically creative. If thisis shown to be the case, further selections could be added to futureversions. This could include providing options to select music styleas well as mood, for example sounds that portray classical, countryor jazz genres. A variety of instrument sounds, screen colours andshapes could be provided, enabling the system to be fully customi-sable for individuals. Additional information gained from furtherevaluations will focus on the potential of people with dementiato communicate emotion through music, leading to future researchto develop a tool that could enable people with dementia who havesevere communication difficulties, to express their feelings.

These investigations have suggested that active music makingcould be used as a means for people with dementia to be creative,a way for them to be empowered by sharing their musical cre-ations with others, and potentially as a means for those with lan-guage problems to communicate emotion to others.

Acknowledgements

Special thanks go to all the participants taking part in the pilotstudies, and especially to the clients and carers at Alzheimer Scot-land in Dundee, Scotland. This research was funded by the Engi-neering and Physical Sciences Research Council.

References

Aldridge, D. (2005). Dialogic-degenerative diseases and health as a performedaesthetic. In D. Aldridge (Ed.), Music therapy and neurological rehabilitation:performing health (pp. 39–60). London: Jessica Kingsley Publishers.

Alm, N., Dye, R., Astell, A., Ellis, M., Gowans, G., & Campbell, J. (2007). Makingsoftware accessible for users with dementia. In J. U. Lazar (Ed.), Niversalusability: Designing computer interfaces for diverse users (pp. 299–316). NewYork: Wiley.

Alm, N., Astell, A., Ellis, M., Dye, R., Gowans, G., & Campbell, J. (2006). A cognitiveprosthesis to support communication by people with dementia. In Internationalworkshop on cognitive prostheses and assisted communication (CPAC2006) at theintelligent user interfaces conference (IUI2006) (pp. 30–36). Sydney, Australia.

Alm, N., Astell, A., Gowans, G., Vaughan, P., Ellis, M., & Dye, R. (2006). Living in themoment: An interactive entertainment system for people with dementia. InInternational workshop on cognitive prostheses and assisted communication(CPAC2006) at the intelligent user interfaces conference (IUI2006) (pp. 16–19).Sydney, Australia.

Alm, N., Dye, R., Astell, A., Ellis, M., Gowans, G., & Campbell, J. (2005). Makingsoftware accessible to people with severe memory deficits. In Proceedings ofaccessible design in the digital world. Dundee, Scotland.

Astell, A. (2005). Developing technology for people with dementia. Psychiatric Times,22(13), 900.

Astell, A., Ellis, M., Alm, N., Dye, R., Gowans, G., & Campbell, J. (2005). Usinghypermedia to support communication in Alzheimer’s disease: the CIRCAproject. In Proceedings of the 11th international conference on human–computerinteraction, Las Vegas, USA.

BBC. (2007). Dementia patients ‘are ignored’. Available from: <http://news.bbc.co.uk/1/hi/health/7113112.stm> Retrieved 27.11.07.

Clair, A. A., Mathews, R. M., & Kosloski, K. (2005). Assessment of active musicparticipation as an indication of subsequent music making engagement for

persons with midstage dementia. American Journal of Alzheimer’s Disease andOther Dementias, 20(1), 37–40.

Crutch, S. J., & Rossor, M. N. (2006). Artistic changes in Alzheimer’s disease.International Review of Neurobiology, 74, 147–161.

Cuddy, L. L., & Duffin, J. (2004). Music, memory, and Alzheimer’s disease: is musicrecognition spared in dementia, and how can it be assessed? Medical Hypothesis,64, 229–235.

Dowling, J. R. (1995). Keeping busy: A handbook of activities for persons with dementia.Baltimore, USA: The John Hopkins University Press.

Ferri, C. P., Prince, M., Brayne, C., Brodaty, H., Fratiglioni, L., Ganguli, M., et al. (2005).Global prevalence of dementia: A Delphi consensus study. Lancet, 366(9503),2112–2117.

Gottlieb-Tanaka, D., Small, J., & Yassi, A. (2003). A programme of creative expressionactivities for seniors with dementia. Dementia, 2(1), 125–135.

Hannemann, B. T. (2006). Creativity with dementia patients: Can creativity and artstimulate dementia patients positively? Gerontology, 52, 59–65.

Hawthorne, D. (2007). Interface design and engagement with older people.Behaviour and Information Technology, 26(4), 333–341.

Holmes, C., Knights, A., Dean, C., Hodkinson, S., & Hopkins, V. (2006). Keep musiclive: Music and the alleviation of apathy in dementia subjects. InternationalPsychogeriatrics, 18, 623–630.

Huub, B. (2005). What is dementia? In B. Huub (Ed.), The simplicity of dementia: Aguide for family and carers (pp. 13–20). London: Jessica Kingsley Publishers.

Jacques, A., & Jackson, G. A. (2000). Understanding dementia. London: ChurchillLivingstone.

Jourdain, R. (2002). Music, the brain, and ecstasy: How music captures our imagination.New York: Harper Collins Publishers.

Juslin, P. N. (2005). From mimesis to catharsis: expression, perception and inductionof emotion in music. In D. Miell, R. MacDonald, & D. J. Hargreaves (Eds.), Musicalcommunication (pp. 85–115). Oxford: Oxford University Press.

Kitwood, T. (2005). Dementia reconsidered: The person comes first. Maidenhead: OpenUniversity Press.

Levitin, D. J. (2006). This is your brain on music: The science of a human obsession. NewYork: Dutton Books, Penguin Group.

Lidwell, W., Holden, K., & Butler, J. (2003). Universal principles of design. Gloucester,MA: Rockport Publishers.

LoGiudice, D. (2002). Dementia: an update to refresh your memory. InternalMedicine Journal, 32, 535–540.

Luzzi, S., Piccirilli, M., & Provinciali, L. (2007). Perception of emotions on happy/sadchimeric faces in Alzheimer disease: Relationship with cognitive functions.Alzheimer Disease and Associated Disorders, 21(2), 130–135.

Machover, T. (2004). Hyperscore: A graphical sketchpad for novice composers. BTTechnology Journal, 22(4), 171–179.

Mendez, M. F., McMurtray, A., Chen, A. K., Shapira, J. S., Mishkin, F., & Miller, B. L.(2006). Functional neuroimaging and presenting psychiatric features inFrontotemporal dementia. Journal of Neurology, Neurosurgery and Psychiatry,77(1), 4–7.

Miller, B. L., Boone, K., Cummings, J. L., Read, S. L., & Mishkin, F. (2000). Functionalcorrelates of musical and visual ability in Frontotemporal dementia. BritishJournal of Psychology, 176, 458–463.

Miller, B. L., & Hou, C. E. (2004). Portraits of artists: Emergence of visual creativity indementia. Archives of Neurology, 61(6), 842–844.

Miller, B. L., Yener, G., & Akdal, G. (2005). Artistic patterns in dementia. Journal ofNeurological Sciences, 22(3), 245–249.

Olsen, R. V., Hutchings, B. L., & Ehrenkrantz, E. (2000). ‘‘Media memory lane”:Interventions in an Alzheimer’s day care center. American Journal of Alzheimer’sDisease, 15(3), 163–175.

Orpwood, R., Sixsmith, A., Torrington, J., Chadd, J., & Chalfont, G. (2007). Designingtechnology to support quality of life of people with dementia. Technology andDisability, 19(2–3), 103–112.

Pickles, V., & Jones, R. A. (2006). The person still comes first: the continuing musicalself in dementia. Journal of Consciousness Studies, 13(3), 73–93.

Roman, G. C. (2003). Vascular dementia: distinguishing characteristics, treatmentand prevention. Journal of the American Geriatrics Society, 51(5 Suppl. II),S296–S304.

Sabat, S. R. (2001). The experience of Alzheimer’s Disease: Life through a tangled veil.Oxford: Blackwell Publishers Ltd.

Sixsmith, A., & Gibson, G. (2007). Music and the wellbeing of people with dementia.Ageing and Society, 27, 127–145.

Terry, R. D. (2006). Alzheimer’s disease and the aging brain. Journal of GeriatricPsychiatry and Neurology, 19(3), 125–128.

Topo, P., Maki, O., Saarikalle, K., Clarke, N., Begley, E., Cahill, S., et al. (2004).Assessment of a music-based multimedia program for people with dementia.Dementia, 3(3), 331–350.

Vink, A. (2002). A survey of music therapy practice with elderly people in theNetherlands. In D. Aldridge (Ed.), Music therapy in dementia care (pp. 119–138).London: Jessica Kingsley Publishers.

Volicer, L., Simard, J., Pupa, J. H., Medrek, R., & Riordan, M. E. (2006). Effects ofcontinuous activity programming on behavioural symptoms of dementia.Journal of the American Medical Directors Association, 7(7), 426–431.

Volicer, L., & Bloom-Charette, L. (1999). Assessment of quality of life in advanceddementia. In L. Volicer & L. Bloom-Charette (Eds.), Enhancing the quality of life inadvanced dementia (pp. 3–20). London: Brunner-Mazel Inc.

Wimo, A., Jonsson, L., & Winblad, B. (2006). An estimate of the worldwideprevalence and direct costs of dementia in 2003. Dementia and GeriatricCognitive Disorders, 21(3), 175–181.