Introducing the COMPANIONS project - CiteSeerX

5 downloads 0 Views 120KB Size Report
relationship…with mutual caring and trust” [Gleitman, 2000, p467]. What is ... The medical benefits of pet ownership are well documented [Garity et al, 1989].
Introducing the COMPANIONS project: Intelligent, persistent, personalised multimodal interfaces to the internet. Oli Mival & David Benyon, Centre for Interaction Design Napier University Edinburgh [email protected] - [email protected] Abstract The paper introduces the COMPANIONS project, a 4 year, EU funded Framework Programme 6 project involving a consortium of 16 partners across 8 countries. It’s aim is to develop a personalised conversational interface, one that knows and understands its owner, and can act as an alternative access point to resources on the Internet, all the while nuturing an emotional involvement from it’s owner/user to invoke the shift from interaction to relationship. On a technical level it intends to push the state of the art in machine based natural language understanding, knowledge structures, speech recognition and text to speech. With these technical developments will come advanced interaction design elements, some of which were initiated on the SHEFC funded project, UTOPIA (Usable Technology for Older People: Inclusive & Appropriate), examining the potential for developing artificial companions for older people. The UTOPIA view of companionship and the Elderly Companionship is a concept that is familiar to all, yet defies simple explanation. Psychology considers it a central need, yet balks at a concise definition of what constitutes a companion beyond “a relationship…with mutual caring and trust” [Gleitman, 2000, p467]. What is clear, is the importance of companions to emotional well being. Indeed the loss of companions is considered a primary cause of depression among older people [Sluzki, 2000]. It is therefore important to consider that the loss of human companions is a natural consequence of growing old. There is a diminishing of the supportive ties of family members, of friends and of other relationships from previous, concurrent, and following generations through death or distancing by migration or relocation. Furthermore, social roles and ties are lost through retirement and any parental function is reduced as children grow up and become independent. This substantial erosion of social networks inevitably leads to the loss of companions and is often accompanied by an experience of emotional impoverishment, not infrequently experienced by the elderly as a pervasive depression “without a reason” [Gory and Fitzpatrick, 1992]. With consideration of this natural decline in human companionship, the potential value of developing artificial companionship become distinctly apparent. On a simple level, older people have relationships with companions, be they pets, friends or care assistants. But what constitutes the difference between an interaction and a relationship? To form a relationship, the user needs to care about the interaction, to invest emotion in it. The artificial companions evoke the emotional investment through replicating recognizable real world behaviour. The movement of AIBO’s head when he is stroked is remarkably realistic and is as endearing as the similar movement of an animal. Thus the user invokes affection and, crucially, attributes personality in much the same way as with a real pet. The importance of behaviour in the attribution of personality can be simply highlighted through the example of cats and dogs, the most common household pets. There is a strong cultural belief that cats have a higher intelligence than dogs, and that dogs are excitable and gullible compared to the cool, sophisticated elegance of cats. These personality attributes are derived from the behaviour and relationships humans have with each animal. In reality, cats have a much lower cerebral development than dogs and a relatively much smaller brain size. Yet intelligence is attributed through behaviour and human interpretation of that behaviour. Of course some of the products are more successful at evoking this personification behaviour within its user than others, though what factors are important to this process remains somewhat unclear. Realism of behaviour seems a likely contender as AIBO’s movement and people’s reaction to it suggests. However if technology as simple as the Tamagotchi can provoke such intense emotional responses as depression at its death, then the psychological impact must be as important as simple engineering issues. From this it may be suggested that the difference between a tool and a companion is a set of

characteristics, a personality, which transforms an interaction into a relationship and evokes an emotional investment. Products which achieve this we call personification technologies (Mival, 2004). To understand more clearly the potential factors at work in these relationships it is useful to examine the relationship between older people and their most basic companions, pets. The Useful Uselessness of Pets The medical benefits of pet ownership are well documented [Garity et al, 1989]. Pet ownership can lead to an enhanced emotional status and provides significant support in reducing emotional trauma following bereavement. Indeed not only emotional health but also physiological health is enhanced through contact with animals, particularly in the elderly. Furthermore, studies have shown that when animals enter the lives of older patients afflicted with Alzheimer’s disease or arteriosclerosis, the patients will laugh and smile more, are more socially communicative and less hostile to their care workers [Beck & Katcher, 1996]. However, some older people live in accommodation which does not allow pets or may suffer from psychological or physiological deterioration that make pet ownership problematic and potentially unsafe for the animal. In situations such as these, the use of artificial pets may be an alternative. It is interesting to note that pets are at the non-specific purpose end of the companionship spectrum. A cat serves no other function than to be a cat. Yet as discussed above, by simply being a cat it can affect the health and well-being of its owner. People take delight in its activities, and it is purely from its behaviour that the benefits are derived. Cats cannot read email to you, struggle as a webcam and do not react to guidance from a computer. They are autonomous objects driven by their own goals. Kaplan suggests that this autonomy, this non-functionality is an important design consideration when developing artificial pets, he suggests they should “be designed as free ‘not functional’ creatures” [Kaplan, 2001]. It is the intention to use these insights to drive the interaction design elements of the major new FW6 EU project COMPANIONS. Introduction to the COMPANIONS project The COMPANIONS project, a 4 year, EU funded FW6 project involving a consortium of 16 partners across 8 countries. The project’s vision is that of a personalised conversational, multimodal interface, one that knows its owner, is implemented on a range of platforms, indoor and nomadic, and based on integrated high-quality research in multimodal human-computer interfaces, intelligent agents, and human language technology. This project is an ECA (Embodied Conversational Agent) which differs from the ECA state of the art by having large-scale speech and language capacity; it also differs significantly from the standard “big engineering” approach to this area, by offering relatively simple architectures with substantial tested performance, based on extensive application of powerful machine learning methods. Large groups of EU citizens will need new forms of interface to the Internet if they are to get benefit from it as it grows more complex: these include huge numbers of mainstream citizens, since the Internet simply does not serve the average non-technical person as well as it does the academics who invented it. Beyond a few simple purchases such as holidays half the EU population make no effective use of the Web at all, and a recent survey (Light, 2005) shows one third of the UK population actively hostile to it. Use of the Internet to access and organise information is limited to current interaction mechanisms such as browsing. Such new forms of interface should be available across a wide range of platforms, from PCs and TV screens to mobile devices, including phones. The deluge of information on the Internet will increasingly be about individuals, and will include their own digital repositories (texts, videos, images) as well as information held about them. Most citizens will have little control over this, their own digital life, without some new form of assistive interface to the Internet. There is already an established need for individuals to organise their own life material, and to give a narrative structure to their lives, particularly when old, which now means shuffling old photographs on paper.

These needs have not been well met by current interfaces based on browsing, profiling and adaptation, and there is good evidence that a more directly personal interface will be more acceptable. A technological solution to this need, in part at least, is a persistent, personalised, companion agent, one that will “know” its owner, chat to the elderly to relieve their boredom, and become the multi-modal interface agent to the Internet for that owner, whatever their age or technical competence. The project calls these agents COMPANIONS. COMPANIONS will learn about their owners: their habits, their needs and their life memories. This will allow them to assist with carrying out specific Internet tasks, which will be facilitated by having complex models of their owners, by which we mean whole-life-memories, or coherent autobiographies, built from texts, conversations, images and videos. Some of this will already be in digital form, but some will be information gleaned from conversations with the COMPANION, information relatives and friends will want later, after the owner’s death, but might never have been able to ask, such as “where did you and your husband first meet?” The barrier to COMPANIONS so far, beyond very primitive forms, has been lack of progress in the adaptability of speech and language technology. The objectives of the proposal are to develop autonomous, persistent, affective and personal interfaces, or COMPANIONS, embedded in the Internet environment, with intelligent response in terms of speech and language, integrated with the manipulation of visual images and their content. We intend to give a higher level of performance in speaker independent speech recognition via robust dialogue management capacity, one that also moves beyond treating the content of communication from all modalities interlingually, i.e. without regard to the mode in which they were originally expressed. Machine learning will be at the core of our approach (especially via the learning of language input structure and of dialogue frames). COMPANIONS must be believable, intuitive, and above all humane conversational interfaces, and must be proved acceptable to our sample target social groups; moreover those groups will be consulted before the platforms are integrated. COMPANIONS will be autonomous and have original aspects of persistent human personality to establish loyalty and trust between users and such agents. They will be sensitive to limited emotion in speech and to the content of images, and will be themselves capable of demonstrating emotional/affective behaviour through speech and visual appearance (e.g. an avatar on a PC screen or mobile). COMPANIONS will also, by communicating with each other enable and enhance communication between the human users, rather than only between humans and these machine artifacts. An early implementation of a COMPANION is PhotoPal. PhotoPal allows people to view their photos and talk about them with their COMPANION. Photos are automatically tagged with the relevant dialogue allowing PhotoPal to build up a rich representation of the person’s activities and relationships. This allows PhotoPal to sort, style and send photos and for people to reminisce with their COMPANION. A second COMPANION is being developed in the domain of personal heath and fitness. CONCLUSIONS A significant aspect of the research in COMPANIONS is concerned with the form of embodiment of the COMPANION. It is not just that the embodiment might be in the form of a domestic animal. It could be something wholly new, but that demonstrates animal-like characteristics; particularly dependability, trust and affection. Companionship is certainly a characteristic that will arise from the interactions and relationship that results. ACKNOWLEDGEMENTS This work is funded by the European Commission under contract IST 034434.

REFERENCES Garity, T., Stallones, L., Marx, M. & Johnson, P. (1989) Pet Ownership and Attachment as Supportive Factors in the Health of the Elderly. Anthrozoos. Vol. 3, No.1 1989, pps 35-44 Gleitman, H. (2000) Psychology. Oxford Press Gory, K.L. & Fitzpatrick, K. (1992) The effects of environmental contexts on elderly depression. Journal of Aging and Health, 4(4):459-479 Kaplan, F. (2001) Free creatures: The role of uselessness in the design of artificial pets. In Proceedings of the CELE-Twente workshop on interacting agents 2001. Light, A. (2005) “EU Survey on Public Website Usage shows Potential" Usablility News, 9 February 2005 Mival, O., Cringean, S., and Benyon D. (2004) Personification Technologies: Developing Artificial Companions for Older People, ACM Press, 1--8. Sluzki, C. (2000) The extinction of the galaxy: Social networks in Family Process

the elderly patient. New York: