Introduction On the surface, developing a digital therapeutic prototype seems like a straightforward task. Simply combining the relevant therapeutic techniques from psychology and implementing a digital interface to deliver them appears to achieve the goal. However, as the visual fidelity of the prototype begins to increase, knowledge from a wider variety of disciplines becomes necessary. Additionally, when the digital platform is an immersive medium such as virtual reality (VR), this complexity only increases further.
Immersive technologies such as virtual reality introduce levels of interactivity that traditional desktop or mobile devices are incapable of. Touch, movement, sense of scale, and even lighting are of utmost importance to creating an immersive experience and effective therapeutic. These new areas of attention represent a widening of knowledge required to develop a digital therapeutic prototype in VR.
Additionally, working with metaphor and the unconscious requires understanding sociocultural differences in metaphoric language use and associated imagery. Without a personalized or cultural understanding of metaphor, the techniques of metaphor therapy are far less effective. This once again represents another layer of complexity in developing a digital therapeutic.
All of these factors require extending beyond siloed thinking and incorporating an interdisciplinary approach. A wide array of disciplines interacting in a dynamic and ever evolving manner is ultimately req uired to develop a nontrivial, digital therapeutic prototype.
Interdisciplinary work
As the process of developing a digital therapeutic prototype began to expand, so did the number of questions and considerations surrounding development. Which therapeutic techniques can be translated into a virtual environment? What fidelity is “good enough” for the initial draft of a prototype? How should abstract concepts such as thoughts be represented visually? How can a digital therapeutic experience be customized to an individual? What methods of locomotion should be implemented? How should the environment and lighting be designed to evoke the affectual tone of depression?
These are but a few of the broader questions that arose specific to development. To assist in managing a project that seeks to find clarity around such a wide range of topics an interdisciplinary strategy was adopted. Prominent interdisciplinary scholars Matthew Miller and Veronica Boix-Mansilla define interdisciplinary work as, “a kind of integrated work, [which] builds on disciplinary perspectives ... and combines them to create a product, develop an explanation, or propose a solution that would have been unattainable through single disciplinary means.” [2]
The following work seeks to reflect on overview the interdisciplinary approaches involved in creating a digital therapeutic prototype focused on metaphor therapy. This work will include an overview of the primary disciplines and the role they played in development, followed by an explanation of the integrative strategies used, then an argument for the added value of interdisciplinarity, and finally limitations and suggestions for future development.
Disciplinary Perspectives
The primary disciplinary perspectives in this work can be derived from the degree plan to which they belong, that is, “development of digital therapeutics.” This phrase offers three distinct categories, development, digital and therapeutics. The first, development, includes the applied field discipline of entrepreneurship. The second, digital, implies the use of technology and as such the disciplines of Information Systems, Computer Science and Human Computer-Interaction are included. Finally, therapeutics, in this context refers to psychotherapy, and draws primarily from the discipline of psychology. These three broader categories also each have their own smaller “branches” that incorporate knowledge from adjacent or closely related disciplines. This section will explore some of the key theories by which this project was developed as well as some of the information from the supplementary perspectives included in these larger categorizations.
The discipline of entrepreneurship primarily serves the role of the organizing framework for prototype development and project management. The key source for this project was Eric Ries’ The Lean Startup. From this, the concept of a minimum viable product is utilized. The minimum viable product (MVP) is defined by Ries as something that, “helps entrepreneurs start the process of learning as quickly as possible...the goal of the MVP is to begin the process of learning, not end it.” [4]. Although there are some key sources included for project management and development strategies, this perspective also incorporates extracurricular experiences. For example, work at the Imaging Research Center in developing a virtual reality data visualization platform laid the foundation for the digital therapeutic prototype built in Unreal Engine. Without this applied invaluable experience, targeting virtual reality as a platform for development may have been impossible or never considered to begin with. Similarly, applied experiences, such as product development with startup Ortus Academy, that utilize an agile development schedule were essential for project planning and prototype prioritization. From these experiences, the scrum methodology’s product backlog was adapted to create an organizational “board'' for development requirements, goals and issues. This rapidly changing dynamic board served as a running log of daily development cycles and overall project progress at the individual level. Without such a board, and by extension the entire discipline of entrepreneurship, the project can stagnate and become uncontrollable.
The second disciplinary category included in this project is the digital category. This category, similar to entrepreneurship, is applied in nature and draws from the disciplines of Information Systems, Computer Science and Human Computer-Interaction. Most applied is computer science and the programming optimizations from within this discipline. Computer science offers algorithmic optimizations and data structures that are necessary to produce an application as computationally demanding as a VR experience. Although the specifics of implementation are handled by commercially available game engines such as Unreal Engine, operating within a rendering engine requires baseline knowledge of programming, game physics and 3D graphics. Complementary to this is the design and usability aspects of the Information Systems and Human Computer-Interaction disciplines. Each of these offer principles and design cycles for developing interactions for a system. In this case, designing and iterating on different levels of prototype fidelity was the biggest contribution. Low fidelity prototyping, often hand drawn or described verbally, generates a basic shared understanding of what the system might look like without much functionality. This perspective similarly offers principles for interface design. Although interfaces are limited in the prototype, general design principles such as affordances and learnability help to provide metrics for usability of the high fidelity prototype design. For example, when designing a key and lock system in the prototype, one would expect to be able to turn and pull a key/lock system. Design in a virtual environment demands realism and simulation of good design.
Finally, the disciplinary categorization referred to as “therapeutics” includes the discipline of psychology. However, psychology alone is not an accurate representation of the depth of this category. Subdisciplines from within psychology such as clinical psychology and cognitive psychology were used extensively to form the foundational knowledge for the project. Supplementary disciplines such as anthropology, linguistics, and history are also included in this categorization as they seek to define and expand the concept of metaphor therapy. For example, understanding metaphor therapy inherently involves a rich understanding of metaphor itself. However, defining metaphor within the context of metaphor therapy is also complex and too requires the perspective of multiple disciplines. For example, Lakoff and Johnson’s conceptual metaphor theory offer a linguistic definition of metaphor as, “...understanding and experiencing one kind of thing in terms of another” [1]. Although this provides an inclusive definition of metaphor, to fully understand its use requires sociocultural contexts. Developing this requires knowledge from the disciplines of anthropology and history among others. History provides a consensus definition of metaphor in specific circumstances and the development of a term's use over time. For example, metaphors for depression including darkness evolved from historical medical terms relating to black bile in the body. Alongside this, cultural understandings of ritual and healing are similarly important to understanding universal healing metaphors and how they are put into action. Each of these supplementary perspectives build pieces of the psychological categorization. At a more direct level, clinical psychology is essential to understanding metaphor therapy. Dr Richard Kopp offers a clinician’s guideline for implemented metaphor therapy, offering a multi stepped approach for identifying, clarifying and transforming client generated metaphors. Similarly, cognitive therapeutic techniques such as thought diffusion explain some of the specific methods of action that metaphor therapy uses. These techniques are the fundamental source of cognitive change that are used within the prototype.
Integrating Perspectives
Handling such a vast base of knowledge and disciplinary perspectives demands an integrative approach. Furthermore, solving a pragmatic problem such as translating an established therapeutic technique into a digital medium lends itself towards an instrumental interdisciplinary approach. Allen Repko defines instrumental interdisciplinarity as:
“a pragmatic approach that focuses on research, borrowing from disciplines, and practical problem solving in response to the external demands of society. Borrowing alone, however, is not sufficient; it must be supplemented by integration.” [5].
Making sense of this pragmatic problem solving is aided by a useful organizational metaphor. In this case the interdisciplinary strategy can be described loosely as the process of making pottery with a pottery wheel. The right mix of clay to support pottery workproper pottery work is analogous to the foundational knowledge provided by psychology, history, anthropology, and linguistics. In order to spin the clay into pottery, a wheel and sculpting tools are required, the supportive tools for working clay are the technical disciplines of computer science and information systems. Properly utilizing and integrating the mixed clay and careful use of the pottery wheel is the outcome of interdisciplinarity, a shaped pot or when things go wrong, it is squashed back down to a ball and changed to a teacup. The changing form of the piece, determined by the potter themselves, is the same as the iterative cycles of development of an entrepreneur. Together these perspectives come together to form a dynamic and creative process that can result in useful and often beautiful final pieces. No one piece is sufficient for developing this process. With just clay (psychology) one only has the raw material (knowledge) to create, with just a wheel (technology) one only has the tools to create. When the individual can combine these things together they engage in a dynamic process for creation. Such is the model of developing an interdisciplinary digital therapeutic prototype.
Organizing Perspectives - Concept Map
Interdisciplinary Strategies and Value
The first task involved in developing a digital therapeutic prototype is to develop an understanding of the traditional therapeutic context sought for translation to a digital medium. As previously described, this is metaphorically “creating the right clay mix” in the process of pottery. This requires a complex interweaving of the explanatory disciplines of psychology, linguistics, history and anthropology. In an effort to achieve this, Miller and Boix-Mansilla’s strategy of building complex-multi causal explanations was used. This strategy is defined as:
“borrowing concepts and findings from a variety of disciplines to construct complex explanations of a phenomenon under study...findings from one discipline raise questions to be answered through the tools of another discipline, which in turn [may] reveal causes that had not been previously considered”[2].
Without such a strategy, deep understanding of metaphor is not possible and universal use of metaphoric symbolism from the discipline of anthropology would never be incorporated. By blending many theories together, the platform from which to build a prototype becomes increasingly valuable. Simply taking a psychotherapeutic technique and creating a 1:1 digital module does not enrich the potential of digital therapeutics. Creating a multi-faceted explanation of the power of metaphor and metaphor therapy paves the way for a potentially more beneficial and widely applicable therapeutic product.
Secondly, as metaphor is so central to the concept of metaphor therapy, it is logical to include attention to analogies and metaphoric language use from the component disciplines. The strategy of reasoning through analogies is a natural fit for this project. Miller and Boix-Mansilla define this strategy as:
“Analogy involves mapping the properties and relations from one domain onto those of another domain … Analogical thinking allows us to apply the theories and concepts of one discipline to those of another domain.”[2].
This strategy was used when transforming metaphoric descriptions of depression and abstract therapeutic techniques into tangible objects in VR. For example, the concept of depression, described historically as dark or heavy, was mapped into a real time rendering engine through dynamic lighting and environment building. Similarly, the therapeutic process of thought diffusion is described as, like creating distance between individuals and their thoughts. This process was also translated into VR through design of an interaction with an object to literally pull a thought from an avatar into space and away from the user. These descriptions and theories, allow for a clear mapping into the digital disciplines and implementation.
As the process of creating a digital therapeutic prototype requires deep explanation of therapeutic technique and translation into a usable digital interface, the strategy of bridging the explanation-action gap was used. This is defined as:
“a strategy for integrating disciplinary contributions that is common in applied fields, such as the professions … one or more domains may be used to define the nature of a problem or explain dimensions of a problem, while knowledge from other domains may be used to guide interventions, implementations, or solutions. This strategy involves assuming that certain disciplinary perspectives may provide explanatory leverage while other perspectives may offer pragmatic or action-oriented leverage” [2].
In this case the split is somewhat well explained already;, psychology provides the explanatory knowledge and the technical disciplines provide the means for putting the knowledge into action and implementing a prototype. Guiding this process overall is the use of entrepreneurship and agile development to manage the complexity. Here the added value of interdisciplinarity is perhaps at its greatest. With only a rich explanation of the value of metaphor therapy, there is no progress towards aiding those suffering from depression in isolation or clinicians overwhelmed with work. Immersive digital therapeutics offer a virtual space to engage with therapeutic themes and practices outside of a traditional clinical setting. Consequently, a client is now afforded the freedom to participate in therapy in the comfort of their homes at a time they deem convenient and appropriate. Among other benefits, this freedom of engagement promotes more therapeutic “action” outside of scheduled, therapist led sessions.
Finally, to conceptualize the desired outcome as a step towards a much larger goal and refined product, the technique of extension redefinition is used to create a common metric for all the various perspectives by extending the concept of a minimum viable product. Extension Redefinition is described by Repko as:
“The technique of extension addresses conflict between disciplinary concepts or assumptions by extending the meaning of an idea beyond the domain of one discipline into the domain of another discipline.” [5]
In this project, disciplines measure progress by various metrics when viewed in silos creating conflict around the idea of progress. Psychology measures in improvement of depressive symptoms, computer science in algorithmic efficiency, and entrepreneurship in product-market fit profit. ExtendingRedefining the concept of a minimum-viable product from entrepreneurship into the field of digital therapeutics as a “minimum-viable therapeutic” reveals a common goal. No longer is progress measure by profit or business potential, but rather therapeutic potential. Rigid psychometrics can be implemented after a proof of concept is developed. Pursuit of algorithmic perfection can be lessened to improve usability and aesthetic design. The competing goals are now aligned towards a common task, to make measurable progress towards developing a novel therapeutic intervention. Progress can be measured in a variety of ethical ways that protect user safety while evaluating an experimental digital therapeutic prototype. In future work, affective response can be self-reported and indications of changes in emotional valence (good/bad) can be used as an indication of progress. Similarly, open ended user feedback can be collected in semi-structured interviews to gain a general sense of the prototype. Evaluations such as these can be used to determine initial effectiveness of a novel digital therapeutic with limited risk to participants.
Limitations
There are a number of limitations in this interdisciplinary work previously described. As prototype development and iteration continued, more perspectives became clear and of greater importance. For example, anthropology and the role of ritual, while discussed early, was only fully connected to the project when the prototype was developed. Having seen the action of thought diffusion in VR, ritual became a central theme. This limited time to investigate and implement the knowledge of ritual, symbol, and anthropological understandings of healing.
Similarly, as the prototype became more developed, the role of visual arts and animation became clear. Designing for a VR environment demands high quality assets, environments and lighting to promote realism and immersion. A significant amount of time, or collaboration could be spent on optimizing the visual components of the prototype. Other modalities could also be incorporated here as well, such as sound design and visual effects. Once the prototype reached a functional state, the breadth of refining touches became clear, without much time or base knowledge required for implementation.
Other limitations include the incorporation of advanced technologies, such as machine learning and affective computing, to create a generative individualized experience. Cutting edge systems such as Open AI’s DALL-E model can generate custom imagery for a user based on complex text entries [2]. Generative machine learning algorithms would allow for an experience with truly individualized imagery and meaning.
Perhaps the biggest limitation was the inability to adequately run user testing for the prototype. Given the time constraints of developing a VR prototype and the social distancing requirements of COVID-19, human testing of a head mounted device is nearly impossible. User testing is the next stage of development and iteration for the prototype. Affectual response measured by biometrics or even basic self reporting could serve as a useful quantification of results and impact of the experience on someone self identified with depression or depressive symptoms.
References
Lakoff, G., & Johnson, M. (2008). Metaphors we live by. University of Chicago press.
Miller M. and Boix Mansilla V. (2004). “Thinking Across Perspectives and Disciplines.” Interdisciplinary Studies Project, Project Zero, Harvard Graduate School of Education. GoodWork Paper 16. Cambridge, MA.
Ramesh, A., Pavlov, M., Goh, G., Gray, S., Voss, C., Radford, A., ... & Sutskever, I. (2021). Zero-shot text-to-image generation. arXiv preprint arXiv:2102.12092.
Reis, E. (2011). The lean startup. New York: Crown Business, 27.
Repko, Allen (2012). Interdisciplinary Research, Process and Theory. Second Edition. SAGE Publications.