10 years innovative research - rvo.nl · 2 | iop men-machine-interaction | 10 years innovative...

23
IOP Men-Machine-Interaction 10 years innovative research

Upload: others

Post on 01-Aug-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

IOP Men-Machine-Interaction10 years innovative research

Page 2: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3

The MMI mission 4

Interview - Gerrit van der Veer 7

Interview - Mark Neerincx 10

Project description - Collaboration at a distance 12

Project description - SuperAssist 14

Interview - Berry Eggen 16

Interview - Jorrit Kuipers 18

Project description - Virtual Assistant 20

Interview - Martijn Vastenburg 22

Project description - Independent@Home 24

Interview - Vassilis Javed Khan 26

Interview - Lucas Noldus 28

Project description - Task-adaptive information distribution 31

Project description - Audio Navigation Assistance 32

Project description - SiteGuide 34

Project description - À Propos 36

Interview - Some other projects 38

Interview - Sander Leer and René Collier 40

Contents

Page 3: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

4 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 5

The MMI mission

Understanding and improving the ways in which people interact with ever more complex applications is important in all parts of society. Whether it concerns the design of a smart phone or a car navigation device, the set-up of a control room in a production plant or the development of a telesurgical device, it is crucial that we understand how man and machine can work together better. An equally relevant consideration is the insight into how machines can support human-to-human interaction.

In its first phase (1999-2004), the IOP-MMI programme focused on basic research in areas such as multimodal interaction, 3D navigation and user-centred design. In its second phase, which covered 2004 to the present, the emphasis has been on applications in which one or more users interact with an intelligent system. This research has resulted in improved interaction design principles that are applicable in industry as well as society. In recent years, additional attention has been paid to projects that address the MMI needs of small and medium- sized enterprises (SMEs).

Research challengesModern technology supports the interaction between the participants in professional or consumer applications. Technological advancement, however, creates its own problems: potentially useful or pleasant applications may become so complicated

that their benefits are overshadowed by how complex their operations are. One way to guarantee a satisfying user experience is to replace explicit user control by the adaptive behaviour of a ‘smart’ system. The user, too, is adaptive and can learn which way of interacting with the system leads to optimal performance and to the achievement of his intended goal.The IOP-MMI programme has therefore addressed the following key question:

Application domainsUser-system interaction is currently an issue for almost every professional and consumer application. Professionals may be willing to invest time and effort in learning to operate a new system, but ultimately the efficiency, effectiveness, safety and satisfaction of their work will improve if the application is easy to use. Consumers, too, will benefit from interacting with a ‘smart’ system that knows their likes and dislikes, their situation-dependent preferences and routines and can therefore anticipate their intentions.

End of an eraThis booklet was first presented at the event that marked the end of IOP Man-Machine Interaction. During the ten years that this programme was active, we focused on supporting research on new and groundbreaking insights within the subject, and have tried to gain broad support for the relation between man and machine within the industry. It is our hope that at the end of the programme, both companies and research institutes continue the initiative of the IOP in pushing the technology forward in such a way that the Netherlands remains an important player in the field of Man-Machine Interaction.

René CollierChairman of the Advisory Board

Joep van WijkProgramme Coordinator

What dynamic knowledge of one another should systems and humans acquire and apply in order to optimally achieve the user’s goals?

Page 4: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

6 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 7

In Van der Veer’s assessment, MMI is fairly well established in the Netherlands. “In essence, only Denmark and the Netherlands can compare themselves to Silicon Valley – which is considered the cradle of MMI. I think that the tradition in both countries of taking design into consideration plays a role in this, as well as – very importantly – the education systems, which offer ample room for thinking about IT and the human factor in that context.”

Science and the business communityThe technology companies in Silicon Valley learned to appreciate the value of MMI through trial and error, under fierce pressure from their competitors. Van der Veer: “There too, they had to discover that it takes more than just technology: when you develop product functions you only

truly add value if your reasoning is fully user-driven.” MP3 players are an easy way to illustrate the issue, he explains. In terms of technical features, they can do about the same thing as iPods – and yet there is a world of difference. “In the United States, MMI is purely an issue for the business community, and has been from the start,” he continues. “It is a completely different approach from the European tradition, in which we often look for a sleek scientific theory to examine a successful product which incorporates complex systems.”

Van der Veer is pleased that international cooperation in the field of MMI is increasingly gaining momentum.

Gerrit van der Veer is acknowledged worldwide as an authority in the field of Human-Computer Interaction. At the Association for Computer Machinery (ACM), he is the first non-American president of ACM SIGCHI, the international society of ACM members who are interested in human-technology and human-computer interaction. In that capacity and on the strength of his own expertise, Van der Veer is in a position to assess the progress and status of the field of human-machine interaction in the Netherlands and other parts of the world.

Interview

Gerrit van der Veer President of ACM SIGCHI

“ Nothing wrong with focus”

Page 5: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

8 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 9

After all, the approaches used by the business community and science should complement each other: “We need each other very much, and we have found each other.” For the time being, this intense level of cooperation is limited to Europe and the United States. “In developing countries such as China and India, MMI is still in its infancy. However, those countries’ interest in the topic is increasing rapidly, now that they no longer focus exclusively on efficient production, but also want to develop complex products and systems.”

UnacceptableIn terms of how much businesses value MMI, there is a major gap between the group of front-runners, including IBM, Xerox, Microsoft, Apple and Philips on the one hand and most other technology companies on the other. Van der Veer: “MMI has established itself among these front-runners. At that level there is an awareness that the MMI knowledge used in products and services makes an undeniable contribution to business capital. The biggest problems are encountered in medium sized enterprises launching specialised technical inventions for professional users. In hospital equipment, for instance, you often see that the user aspects have been completely ignored. The professionals who work with these devices are primarily interested in how well the device functions and are therefore less inclined to complain about user-friendliness. Frequently, no alternatives exist for such equipment: they save lives,

so any faults are overlooked, despite the fact that we know that it can be done so much better, making the work safer, more economic, and more pleasant. That’s simply not acceptable.”

Smart contributionIn the opinion of Van der Veer, the IOP MMI programme has made smart contributions in the past 10 years to increasing the level of knowledge available in the field of MMI and to transferring that knowledge to the business community, particularly SMEs. “It’s hard enough to find funding for research in the Netherlands as it is, but for multidisciplinary research it’s even harder. MMI is largely an interaction between psychology, information science, and design. Most knowledge institutes house those disciplines in different faculties, and it is still difficult to find funding for joint research. With IOP MMI, however, joint funding was possible.” The cooperation that IOP MMI fostered between knowledge institutes and the business community also worked well, in Van der Veer’s opinion. “When it comes to time, businesses and scientists have a completely different outlook. That complicates cooperation, but it does produce results. The private sector keeps the pressure on and is quicker to distinguish between what can be classified as the pursuit of a hobby and what really matters in terms of accelerating developments. That provides focus, and there is nothing wrong with that.”

Page 6: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

10 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 11

“When it comes to the provision of information, there are two factors that make it hard for you in a crisis situation,” Neerincx explains. “The first is that you receive a huge amount of information from a great many different sources, which makes it easy to overlook crucial information. The second is that you don’t really have time for system interaction in a crisis situation.” Systems that demand attention for complex operating procedures are less suited for use in crisis situations. During the course of the research carried out in the framework of IOP MMI, a great deal has changed. Neerincx: “The internet developed incredibly quickly, the smart phone has broken through to the main-stream and we now have large, interactive screens and tables. In order to make effective use of these new technological options, research increasingly needs to focus on the human component, on the context in which people use the resources at their disposal: how do we design applications that truly offer intelligent support? Systems that help you set priorities, that understand something about the context, that ensure that information is distributed optimally, that ‘sense’ the pressure you’re under at a given moment, so that they don’t overwhelm you with important information when you are unable to process it.” The projects run by Neerincx primarily focused on public order and safety. “But it is very easy to imagine a broader application for what we are discovering here about man-machine interaction,” he says. “Consider a smart,

environment-based support system for people who are chronically ill, a system that offers advice about medication or food in specific situations.”

The IOP MMI programme facilitates constructive collaboration between researchers and companies. Neerincx: “The great thing about that is that you guarantee scientific quality, while still maintaining the focus on developing concrete applications for the relatively short term. It accelerates the process of translating knowledge into applications. The end result is that we were able to produce both publications and demonstrators.”

Buildings have collapsed and dozens of emergency response workers are doing their utmost to bring the injured to safety and rescue the survivors from beneath the rubble. People are running in all directions. Reports come in quick-fire by walkie-talkie, telephone, text messages and email. But in this mishmash of information, what is urgent and lifesaving, and what can wait a bit? Mark Neerincx, Professor of Man-Machine Interaction, focused on smart support for Urban Search & Rescue teams in chaotic situations, working in the framework of IOP MMI.

Interview

Mark Neerincx Researcher with TNO and professor of MMI at Delft University of Technology

“ Support in chaotic situations”

Page 7: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

12 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 13

Urban search and rescue (USAR) missions are complicated and time-critical operations. The time window for finding survivors after a large-scale disaster (earth quake, tunnel accident,...) is usually only a few days. In such situations, USAR teams can use all the help they can get to make their collaborative performance efficient and effective. The members of such a team are part of a complex system of humans, dogs, equipment and information and communication technology. The project develops support systems that improve collaboration at a distance between

all the units of such an organisation, taking the particular phase of the mission into account. The functionalities are being tested iteratively in field exercises and in real missions. The team-centred approach is a prime example of how a complex system can be engineered with a focus on its functionality rather than on the mechanics of the technology involved.

Further information: Mark NeerincxE [email protected]

Project discription

Collaboration at a distance

Page 8: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

14 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 15

The SuperAssist project developed personal assistants that filter, provide and enrich information. They take part in local and/or remote inquiry, diagnosis and recovery actions. Each assistant improves the supervision of its individual user,

but co-operation with other assistants and their users provides additional profits, such as adequate and efficient medical and technical information transfer from patient to specialist.

The project addressed the following questions: How can a virtual assistant provide a chronically ill person with information about her health in an effective and pleasant way? How can the patient be motivated to adopt a healthy lifestyle? Which dialogue features facilitate the transfer of factual and emotional information?The resulting models and methods of telecare were implemented and evaluated in working prototypes.

Further information: Mark NeerincxE [email protected]

In the transmural health care domain, patients and specialists (medical and technical) are involved in the usage, supervision and maintenance of electronic medical dossiers and health monitoring equipment. Information technology can support these tasks. For instance, a rule-based system can inform a patient and his medical specialists about risky changes in the patient’s state of health.

Project discription

SuperAssist

Page 9: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

16 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 17

“ Focus on experience”

The booming development of the internet and web technology has been one of the drivers behind that shift, according to Eggen. Another significant driver is the fact that computer technology is increasingly incorporated into its environment: “Every-thing is becoming smaller, more powerful and wireless at a breathtaking rate.” The technology makes it possible to develop intelligent environments. As a result, researchers, developers and designers of business processes all face entirely new challenges. Eggen: “These environments ‘know’ their users and are aware of the physical, social and cultural context in which they are used. They adapt to the user.

This adaptation needs to be meaningful to the user and take place in an appropriate way. Only then will people consider these systems ‘intelligent’.” This has repercussions for all the disciplines involved in developing man-machine interactions for this new

generation of products and services, according to Eggen. ‘Intelligent environments’ are penetrating deep into users’ lives. The concept of ‘usability’ takes on an entirely different dimension as a result. Eggen: “How do you take an effective approach to the preferences and norms and values of people and groups of people when, for example, designing systems that support social interaction? Acceptance and actual use in the longer term are the marks of success. In order to develop a sound scientific basis for it, we are conducting more and more research in the real world, outside the lab.”To design intelligent environments that offer users genuine added value, technology and social sciences need to cooperate integrally with the design and business disciplines, Eggen believes. A national MMI research agenda, such as the one that IOP MMI has helped set for the past 10 years, can contribute to this. Eggen: “It should be noted that cooperation can only lead to ground-breaking excellence when the supporting disciplines are world-class. Fortunately there are a number of leading groups in the Netherlands that meet this standard.”

Berry Eggen has been involved in the IOP Man-Machine Interaction programme from the outset. Eggen headed various research projects that were supported by the programme and saw the focus of the IOP MMI research projects shift. What started as ease-of-use for simple systems ended as far more complex combined systems, products and services, which together form intelligent environments for their users. Eggen: “There is a much stronger focus on the experience now.”

Interview

Berry Eggen Head of the User Centered Engenering Capacity Group, Department of Industrial Design, Eindhoven University of Technology

Page 10: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

18 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 19

Jorrit Kuipers, director of Green Dino: “We have managed to offer about one-quarter of the total teaching programme for gaining a driving license through simulation. In that respect, there is still room for improvement. The validated results so far are really good. The success rate for students who learned the lesson material using this method is slightly higher than students who learned it from a ‘real’ instructor.” One of the major advantages of the simulator is that all of the students’ behaviour and responses can be recorded and analysed. Kuipers: “Everyone involved can learn a great deal from it.

For example, we see patterns in the behavioural differences between men and women and between different types

of people in the data from the simulators. That could, for instance, make it possible to adapt the lessons directly to personality traits, improving the results even more.”

“ Still a huge gap between science and the market”

Green Dino is the market leader in the Netherlands in the field of car driving simulators. In cooperation with Delft University of Technology and with support from IOP MMI, a simulator was developed that may lead to a revolutionary change in how we learn to drive: not only does it simulate the car and the traffic, it also simulates the driving instructor. Research has shown that the virtual driving instructor certainly does just as well as his flesh-and-blood competitor, and in fact even performs a little better. You might think that the market would be cheering enthusiastically, but you would be forgetting human nature. No matter how crystal clear the results may look, the idea that a machine can be a better teacher than a person is simply not an outcome that driving instructors and their clients can accept yet. What’s more, the ‘fun’ aspect of driving in a real car is never quite the same in a simulator.

Interview

Jorrit Kuipers Green Dino BV

The simulated driving instructor might be a success for researchers, but the product is far from market-ready, Kuiper notes. “To begin with, I must say that the support from IOP MMI was excellent. Thanks to the programme, we were able to make huge strides. At the same time, there is still a large gap between science and the market. In the Netherlands, we tend to think in

terms of commercial opportunities fairly early on, but we forget that the road to the market is long and difficult. If you have a prototype, you’re far from done. It also involves less tangible factors such as acceptance, complex regulatory frame-works, the availability of capital and the market conditions. For this product it is no different.”

Page 11: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

20 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 21

In one subproject, sophisticated driving assessment methods were developed in order for the VA to ‘understand’ the student’s behaviour in the simulator. The assessment software reveals the

student’s strengths and weaknesses, keeps track of her progress, and gives predictions and recommendations based on current and past performance.

In the second subproject, student-adaptive feedback and instructions were elaborated on the basis of the driving assessment as well as on individual characteristics, such as preferred learning styles. Special consideration was given to the nature and the timing of the feedback.

The most promising research results were implemented in the simulator produced by Green Dino Virtual Realities. Performance data are stored on these commercial simulators, which yields information on thousands of driving students. In turn, this is particularly useful to continuously improve and evaluate how different versions of the VA affect driving and learning performance. Further information: Peter WieringaE [email protected]

Using driving simulators for training purposes provides important benefits such as reduced costs and guaranteed safety while driving in controlled virtual environments. In order to extend the benefits of such training, a virtual assistant (VA) is needed who is able to autonomously and effectively train student drivers.

Project discription

Virtual Assistant

Page 12: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

22 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 23

The Independent@home field studies produced intriguing results. “Firstly,” says Vastenburg, “we gained a better idea of what motivates the various stakeholders to be enthusiastic about intelligent environments for the elderly. Both home care organisations and informal caregivers see technology as a chance to improve coordination between care providers. Home care organisations primarily see opportunities for working more efficiently. The lack of human and material resources in that sector could perhaps be alleviated by technology. Among the informal caregivers (family, friends), you mainly see that they benefit from the worries being removed: did dad go to the doctor? Did he take his medicine? By using technology to answer those types of questions, not only

do you take away the worry, caregivers and the elderly also have more time left, literally, for better quality contact with each other.”

Vastenburg believes that with support from the IOP MMI programme, Independent@home has demonstrated its value. “For this project, we developed a platform and a series of new solutions that can help the elderly and their caregivers to stay in their own homes for as long as possible. But we are still at the beginning of researching how we can support product designers in developing applications that can offer genuine added value in practice. Which solutions have potential, and which don’t? What is the best way to implement technology for this? To answer those questions, we have to head out into the field, see what works and what doesn’t in the real world and – not to be forgotten – what the elderly, caregivers and home care organisations can actually manage in practice. This type of research is expensive and does not immediately yield market-ready products and services. That means that the time it takes to earn back the investment is too long for the business sector, and the return is uncertain in any case. Having said that, this type of support for the elderly has great significance for society. The elderly would like to continue living at home for as long as possible. A programme like IOP MMI makes cooperation between business and science possible, bringing the realisation of that desire within our grasp.”

Interview

Martijn Vastenburg Delft University of Technology and project leader of Independent@home

“It’s not until the practical application of something that you really see whether it works or not”

The Independent@home project supported by IOP MMI is researching the options for enabling the elderly to live independently for longer, assisted by an intelligent environment. The task here is not to achieve the most technically advanced solution, but rather the most desirable from the perspective of the user. Independent@home project leader Martijn Vastenburg: “A great deal is technically possible when it comes to supporting the elderly and chronically ill with ICT-style solutions. But the gap between that and actual practice in the day-to-day lives of the elderly and chronically ill is huge. Product designers face the challenge of bridging this gap. With Independent@home, we’ve taken large strides to support the designers.”

Page 13: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

24 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 25

The Independent@Home project explores new design directions to support independent living. A variety of services has been developed including SnowGlobe (a presence display for seniors), a virtual physical activity coach for seniors, an awareness display for professional caregivers and an iPhone awareness app for family caregivers. In order to offer better support to service designers, a service platform was developed that integrates sensors and services in the home, and enables a uniform user-friendly inter - action style. The sensors are used to detect the living routines of the inhabitants. The system then activates services that are relevant to these routines.

The use of awareness displays, a key project research area, has been studied in several exploratory studies on design. Using awareness systems, caregivers might better be able to consider the actual and changing needs of individual clients, and better be prepared for the home visits. At the same time, awareness displays have been found to support social wellbeing. In the case studies, the researchers considered functional requirements as well as the context of use and the human values. Both the platform and a series of prototype services have been evaluated in longitudinal field tests in the homes of older volunteers

and of their professional and family caregivers. The findings from the field studies show that awareness displays contribute both to the efficiency of the care process and to the peace of mind of caregivers and seniors.

A new spin-off company, ConnectedCare.nl, has been established in order to further develop the concept of using sensor systems to support care networks.

Further information: Dr Martijn VastenburgE [email protected]

Elderly people indicate that they prefer to live in their own homes as long as possible. Independent living contributes to wellbeing, and at the same time lowers the pressure on professional care. Technology could play a central role in supporting independent living, through home automation, by providing information to caregivers, and also by encouraging seniors to stay active in a physical and social sense. Industry indicates that end-users have problems using the present stand-alone products, which tend to be complex to use. Moreover, present products do not make optimal use of the sensor information.

Project discription

Independent@Home

Page 14: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

26 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 27

The research supported by IOP MMI which earned Vassilis Javed Khan his doctorate in November 2009, caused an outcry in the press. One experiment from his research was interpreted as the introduction of a ‘Big Brother’-style system for the constant monitoring of children’s movements. The source of the report was a short press release from the ANP press bureau.

Interview

Vassilis Javed Khan Eindhoven University of Technology

According to Javed Khan, “In many cases, the media published the report without checking the background. What we want to know is how we can offer busy families better support in their need to communicate. More specifically: what needs do families have for information that go beyond the many channels of communication that are already available?The experiment referred to by the news-papers involved a total of five children and their parents.”

The experiment itself, and primarily the various studies that preceded it, showed that parents who have children in primary school feel a particularly strong need to know what their child does in the classroom. Javed Khan explains, “Not in such a way that they can check up on them, but rather so that they can communicate more meaningfully with them. The children themselves offer few specific details about what happens. If you do have that information, as a parent, you are able to ask much more specific questions, and the quality of communication increases.”

The type of research that Javed Khan is working on is still in its infancy. “There is still a great deal that we do not know about communication and communication needs in modern families and a great deal of research still needs to be done before tangible products can be developed that actually meet a need. Despite that, practical applications are on the horizon. That is the strength of a programme such as IOP MMI, as Javed Khan explains. “The programme promotes cooperation between science

and SMEs in particular, which become more innovative as a result. That is good news for the economy and the future of the Netherlands.” Javed Khan is practising what he preached: after completing his PhD, he launched a spin-off company called Kidzframe, for parents whose children attend a day-care centre. The group leaders at the day-care centre can easily use the service to send parents photos and short messages. He is also concluding a pilot study at a psychiatric clinic for patients with obsessive-compulsive disorders. Javed Khan: “It helps patients to return to their own home environment more quickly. Contact with and monitoring of the patient remains in place and therapy can be continued for some time by a ‘digital coach’.”

“The programme promotes cooperation between science and SMEs in particular, which become more innovative as a result.”

Page 15: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

28 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 29

Lucas Noldus, who started out as a biologist, is now CEO of Noldus Information Technology, a mid-sized enterprise that develops software for the support and analysis of behavioural studies. Founded in 1989, the company serves universities and laboratories all over the world. Noldus: “My interest in MMI developed from this. Our first tools worked perfectly from an IT perspective, but we quickly discovered that they had problems in terms of user-friendliness. Since then, we have been following the developments in MMI closely and conducting our own research, for example in our experience lab.”

Interview

Lucas Noldus member of the iop mmi advisory board, CEO of Noldus IT

Noldus has been a board member for IOP MMI for almost the full ten years. “I think that we have achieved quite a lot in terms of promoting MMI in the Netherlands. Many designers working on interactive systems have fully taken on board that the user needs to be central. The IOP MMI projects have certainly contributed to the fact that the user has become the central point around which all developments have to take place.”

Another important contribution made by IOP MMI is the improved cooperation between all kinds of stakeholders in the MMI field. “Ten years ago, providers and consumers of MMI knowledge could hardly find each other, if at all. That has definitely changed. The establishment of the Dutch branch of the professional association

for CHI (Computer-Human Interaction) also made an important contribution to that. CHI Nederland has over 500 members and after that of the United States, is now the biggest professional association in this field in the world.”

In terms of intensifying MMI in business, the IOP MMI programme achieved mixed success, according to Noldus. “The focus on MMI shifts in response to economic developments. Even now, hardly anywhere is the topic is considered as one of the core activities; it’s more part of the ‘packaging’.” Still, Noldus sees the future for MMI as far from bleak, even after the MMI programme comes to an end. “A programme such as this needs to be finite, otherwise it will lose its innovative drive. There will definitely be other programmes that give

“The focus on MMI shifts in response to economic developments”

MMI research the support it needs. MMI is a recurring theme everywhere and therefore has a guaranteed future, but it is evident that it is difficult to keep the topic on the agenda. For example, there is now a very strong focus on themes such as healthcare

and transport. But these are areas where ICT is everywhere and where you will see ample opportunities for contribution by MMI researchers.”

Page 16: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

30 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 31

Two main factors determine how information should be distributed: relevance for the task that the user is performing and the information load in his current and future tasks. Irrelevant information should be suppressed if overload threatens the distribution, and the workflow should possibly be adjusted. In addition, the experience and skill level of the operators can be taken into account. The processing load must be distributed over the people performing the task in such a way that no single person becomes overloaded.

The application domain chosen for the TAID project was crisis and disaster management, where suboptimal communication between individuals and organisations is a well-known problem. The approach adopted Machine Learning (ML) techniques in order to adaptively select and distribute task-relevant information. Real-life communication logs from emergency response operations were used

to auto matically learn (a) to decide on the relevance of information from dialogues and (b) to optimally distribute this information to members of the organisation. Since relevance depends on the current status and the task that an actor is performing, an adaptive workflow model of the actors’ tasks in crisis and disaster management was made. This was done by using Brahms, the multi-agent modelling tool developed by NASA for simulating space missions. The workflow model was used to evaluate the effect of the information distribution system. The TAID project was done in close collaboration between the University of Amsterdam (Niels Netten and Maarten van Someren) and the University of Twente (Guido Bruinsma and Robert de Hoog).

Further information: Maarten van Someren [email protected]

The task-adaptive information distribution project (TAID) addresses the problem of selecting and distributing task-relevant information among people collaborating in highly dynamic work situations. Under these circumstances, managing the information flows becomes a difficult task.

Project discription

Task-adaptive information distribution

Page 17: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

32 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 33

IT solutions have recently been developed to provide cognitive support for the elderly, especially those in the early stages of Alzheimer’s, by reminding them of important activities such as taking their medication. Another type of application is a GPS-based device that alerts a caregiver if its owner leaves a predefined area. Beyond reminders and caregiver alerts, a mobile assistant could be designed that proactively helps an Alzheimer patient to safely reach a programmed destination in his vicinity (supermarket, post office, church,...). This project created a prototype of a navigation tool based on position sensing with GPS or RFID. It gives spoken instructions to its user on his way

to a predefined location. The emphasis was on the requirements for interaction: What sorts of verbal instructions are helpful (e.g. using landmarks instead of street names)? What level of physical detail is needed in location sensing and in verbal commands? How can a caregiver easily program the destination? How should an emergency be signalled?

Further information: Ben Kröse E [email protected]

Paul de GreefE [email protected]

Project discription

Audio Navigation Assistance

Page 18: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

34 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 35

Site optimisation can be done in one of two ways. One is to evaluate a site by comparing it to a collection of sites that are intended for the same users, and then detect the differences. These can be used to improve the site under discussion. The second approach makes use of logs that show how the site is being used. For instance, it is possible to infer from the behaviour of the visitors to a site whether if there are problems with navigating the site or whether the workflow is less than optimal. This diagnosis can lead to improvements in the design. A system was built that finds corresponding pages of a set of websites, finds characteristic elements of these sets such as words in the text and file names, and also the links to other sets of pages. This is used as a model for designing a new page or analysing an existing site. Both methods have been implemented and evaluated.A new idea that came up in the course of the project is the use of visual properties of webpages such as colours and lengths and

directions of lines to assess the impression that a page makes on the reader. This makes it possible to also evaluate not only the content, but also the ‘look and feel’ of a webpage. The SiteGuide project has resulted in several tools and methods for the analysis and assessment of web sites and web pages. SiteGuide was developed by Viktor de Boer and Vera Hollink. The result is being developed further in cooperation with Euro IT&C, an SME specialized in web design.

Further information: Maarten van [email protected]

One of the challenging application areas for interaction design is finding a good way to structure and optimise the content of a new website. The SiteGuide system can take a set of websites as input, extract the content and the structure of the pages and links, and present it to the user as a first draft. This draft is imported to a sketching tool that allows the user to modify the design freely.

Project discription

SiteGuide

Page 19: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

36 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 37

To be acceptable to the end user, this information should be very precise, since redundant or irrelevant information causes irritation, on top of an increased cognitive load due to the interruption. To maximise the precision of the information, the system’s user interface should be personalized to a high degree; this can be achieved by biasing the system to documents on the user’s computers and documents that the user downloads and reads. Given that the underlying technology is precise enough, and the confidence in its own precision is trustworthy, adaptive information provision by means of a message that pops up on the desktop should not disrupt the writing process, and thus lead to an end product of lower quality. Therefore, the project was two-pronged: one subproject focused on high-precision trainable information retrieval and recommendation techniques, the other dealt with how and when to suggest it to the user. In essence, the project resulted in a prototypical recommender system that presents information to professional writers in a manner that is perceived as timely, non-intrusive, and trustworthy. Investigating the question of how and when to suggest information to the user produced the unexpected scientific insight that computer-based writing in the internet age, where the internet increasingly serves as long-term memory storage, does not fit the ruling cognitive theories of the writing process very well; these theories require an upgrade.

The development of the recommender system was greatly boosted by use cases provided by Tilburg University’s Information Services department, which co-operated in building an expert recommendation system (to be used for external media and PR purposes), by Trouw, the Amsterdam-based newspaper, which acted as the testing ground for a related news article recommendation system, and by CiteULike, a social bookmarking website for scientific articles, which offered its data for performing a part of the experiments described in the PhD thesis “Recommender Systems for Social Bookmarking” by PhD student Toine Bogers. In 2009, CiteULike implemented a new recommender system according to Bogers’ model.

The results of À Propos could be used as improvements in existing business applications, both online such as CiteULike’s social bookmarking system, and client-based such as in the IntelliGent tool of the project’s industrial project partner, Search Expertise Centrum BV.

Further information: Antal van den Bosch E [email protected] BovesE [email protected]

Project discription

À Propos

The purpose of the project was to develop a system that supports professional text writers by proactively offering them information that may be relevant for the document they are producing.

Page 20: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

38 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 39

Interview

SOME OTHER PROJECTS

TelepresenceThe project intended to increase the degree of involvement and social presence in two-person teleconferencing by providing direct eye-contact even during physical movement of the participants. Second, assessment methodology was developed for the evaluation of social presence in telecommunication situations, including subjective measures, but especially objective behavioural measures that are sensitive to different types of interaction.

Further information: Lydia MeestersE [email protected]

Boosting DomoticaThe project created a platform for SME’s in home automation that allows them to collaborate on innovative applications. It also developed and transferred a methodology to discover at an early stage the wished of the consumers.

Further information: Liek VoorbijE [email protected]

FeetBackThe focus of the project is on continuously monitoring and influencing the movement pattern of diabetes patients, since inactivity may lead to ulcers on their feet. Patients receive feedback on the risks and advice on how to change their behaviour. The content and the style of the guidance are tailored to the individual preferences, so as to maximize the persuasive impact.

Further information: Lou BovesE [email protected]

‘Human-in-the-loop’ user interfaces in car interiors

Drivers tend to keep their workload fairly constant. Since many primary tasks of the driver have been automated in modern cars, the number of desirable secondary tasks increases (navigation, in-car info-tainment,…) and their control needs to receive more attention. This situation requires a novel view and implementation of the user interface, based for instance on haptic control elements.

Further information: Alex SerrarensE [email protected]

Videophone support in healthcareThe aim of this telecare project was to facilitate the efficicient and natural interaction between a client and his care provider. Guidelines and recommendations were developed for the use of video- phony to support the timely intake of medication so that, for this application, a home visit can be replaced by coaching from a distance.

Further information: Lydia MeestersE [email protected]

Haptic feedback in telesurgeryThe project focuses on the design of controllers that enable haptic feedback in robotics for medical applications. This feedback regulates the interaction between the ‘master’, i.e. the control panel operated by the surgeon, and the ‘slave’, i.e. the robot performing the intervention. Haptic feedback augments the visual information on which the surgeon relies so far. It can adapt to the preferences and the nature of the task.

Further information: Maarten SteinbuchE [email protected]

Design patterns for starters and DIY’s

Re-usable knowledge was compiled for “visual design” by software engineers with no expertise in the HCI domain. A pattern library or toolbox was composed of existing and internationally accepted pattern collections, rewritten and illustrated for the target users. The toolbox was completed with a “pattern wizard” that has been put on-line in the public domain.

Further information: Gerrit van der VeerE [email protected]

Creative design of interactive systems

The project has formulated basic principles and guidelines for “experience design”. They have been tested in case studies and put into practice in in a “visual design” and a “virtual worlds” course. They have also been applied in the redesign of the website of a major broadcasting company.

Further information: Gerrit van der VeerE [email protected]

Page 21: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

40 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 41

“A tough legacy,” is how Sander Leer, the brand-new chairman of CHI-NL, refers to it. “But it’s a legacy that we look forward to taking on. It is of course in the interest

of the profession to see MMI continue developing strongly in the Netherlands. But it also benefits the innovative capacity of the business community, the continued, often multidisciplinary development of scientific MMI research, and education. It is inevitable that in order to achieve this the various players will need to cooperate and exchange information. CHI-NL is eager to act as a facilitator in that context.”

Professional MMI service deskRené Collier is chairman of the IOP MMI advisory committee. He is satisfied that IOP MMI can make a financial contribution to the launch of a professional MMI service desk by CHI-NL. Collier: “We have been working with CHI-NL for years now and bump into each other at conferences and seminars. The fact that we have faith in each other is apparent from the fact that we are organising the closing conference of IOP MMI together. But of course CHI-NL is an association that relies entirely on the voluntary contributions and activities of its members. Setting up a solid service desk on that basis is asking too much. That’s why we have provided funding to allow the service desk to start professionally in the coming year.”

IOP MMI has made a major contribution to putting the field of man-machine interaction on the map in the Netherlands in the past 10 years. One of the results of the programme is better cooperation on MMI among the universities, research sector and business community in the Netherlands. Stakeholders have become better acquainted with each other and learned to appreciate each other, thus meeting one of the key aims of the programme. The association of MMI experts, CHI Netherlands, has been designated the ‘heir’ to the IOP MMI programme. CHI-NL needs to help ensure that MMI does not drop off the priority list when IOP MMI ends and that there continues to be a platform where businesses, researchers and educators can find each other.

Interview

Sander Leer and René Collier Chairman of CHI-NL Chairman of the IOP MMI advisory board

“A tough legacy”

Facilitating knowledge transfer“The trick is to share the MMI knowledge that is already available with a wider audience,” Leer says. “Within the profession, but also with entrepreneurs. With support from IOP MMI, we will be setting up CHI-Business, which is intended to become the point of convergence for supply and demand in the MMI area of expertise. Our current estimates show that it should

be possible for a service desk like this to wwbe financially self-supporting within the foreseeable future. It should be noted that CHI-NL is and will continue to be an association, which will not, for instance, initiate or fund research. Our aim with CHI-Business is to bring together parties that could benefit each other, thus facilitating the transfer of scientific knowledge to the business community.”

Page 22: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

42 | IOP Men-Machine-Interaction | 10 years innovative research

The user at the heart of it allMMI research and practical MMI applications could still do with a helping hand, both Collier and Leer believe. Collier, who was MMI group leader and innovation manager at Philips before he retired, says: “MMI still needs to carve out a niche for itself. Its significance is acknowledged fairly well in very big companies, where there is more of an awareness that it makes quite a difference whether you have to send your operators to Florida for two weeks to learn to use a professional device, or whether a mid-week trip to Eindhoven will suffice. But medium sized enterprises in particular are lagging behind. One of the complicating factors is that it is not always easy to

demonstrate the return on investment (ROI). Still, effective interaction design that puts the user at the heart of it all will always earn itself back. It should be noted here that effective interaction design enables you to make appealing products and motivate people to actually use them, and keep using them. The user experience has to be positive for that to happen; that aspect is often underestimated. Many companies have the most fantastic inventions lying around, with applications that only stay interesting for a few weeks once they’re tested on ‘real people’, and then quickly fall into disuse. Not because the technology isn’t sound, but because the development process did not put enough emphasis on the user.”

FlywheelSander Leer confirms that impression: “I come from the IT world, where you see that technology has developed so rapidly that the role of the user is lagging behind a bit. For MMI, catching up is absolutely vital to get the most out of the technology. This requires a dialogue between companies that develop products and services, know - ledge institutes, and practitioners who can integrate MMI knowledge into products. IOP MMI set that flywheel of knowledge development in motion, and over the past 10 years CHI-NL made it tangible in the form of conferences and user groups. We greatly look forward to ensuring that this momentum continues. It is fully in line with our aims as an association, which focus on networking and facilitating easier access to knowledge institutes.”

Page 23: 10 years innovative research - RVO.nl · 2 | IOP Men-Machine-Interaction | 10 years innovative research IOP Men-Machine-Interaction | 10 years innovative research | 3 The MMI mission

The aim of on IOP programme is to intensify the

interaction between companies in the Netherlands

and the knowledge infrastructure within a specific

discipline, by providing an extra impulse to research

within a theme and by actively involving industry in

the execution of this research. Combining the forces

of companies and knowledge infrastructure leads to

the formation of new networks and the

reinforcement of existing ones.

All information in this publication are written by:

Poco Más Communicatie

Published by:

NL Agency

NL Innovation

Juliana van Stolberglaan 3

PO Box 93144

2500 AC The Hague | The Netherlands

P +31 (0) 88 602 54 96

E [email protected]

www.agentschapnl.nl/iopmensmachineinteractie

© NL Agency | juni 2010

Publication: 3IMMI1001

NL Agency is an agency of the Dutch ministry of Economic Affairs responsible for the implementation of

sustainability, innovation and economic development programmes for various governmental bodies. NL Agency

is a department of the Dutch Ministry of Economic Affairs that implements government policy for sustainability,

innovation, and international business and cooperation. It is the contact point for businesses, educational

institutions and government bodies for information and advice, financing, networking and regulatory matters.

The division NL Innovation helps Dutch businesses to innovate, by providing finance, advice and contacts.