#OTLip18 Abstracts

Programme

9:00-9:30Coffee and pastries
9:30-10:20Keynote: Angelo Cangelosi (University of Plymouth) - Developmental Robotics for Language Learning, Trust and Theory of Mind: An Interdisciplinary Journey
10:20 Emmanuel Senft (University of Plymouth) - Safely Teaching Robots to Interact with Vulnerable Populations
10:40 Ilaria Torre (Trinity College Dublin) - Playing with Trust
11:00-11:15 Refreshments
11:15 Thomas Colin (CogNovo, University of Plymouth) - Intention and Insight
11:35 Frank Loesche (CogNovo, University of Plymouth) - Eureka as a Model for 2222 Years of Interdisciplinary Research
11:55 Tara Zaksaite (CogNovo, University of Plymouth) - Effectiveness of Financial Education: Evidence from an Interdisciplinary Research-Project
12:15 Sarah Levinsky, Adam Russel (University of Falmouth) - Tools that Propel: on (not) Knowing in the Body and Time
12:45 - 13:30 Lunch & Posters session
13:30-14:20Keynote: Sophia Lycouris (University of Edinburgh) - Interdisciplinarity and Innovation: Learning from the Arts
14:20 Jacob Downs (University of Sheffield) - Headphones and the Affordances of Technological Intimacy
14:40 Aska Sakuta (University of Chichester) - Embodied flow states and its role in movement performance
15:00-15:15 Refreshments
15:15-16:05Keynote: Joanna Rączaszek-Leonardi (University of Warsaw) - Interdisciplinarity in Research on Basic Problems in Cognition: The Case of Relation of Symbols to Dynamics
16:05 Klara Łucznik (CogNovo, University of Plymouth) - Sync To The Others Not To The Movement - The Investigation Into Shared Physiological Dynamics In Dance Improvisation
16:25 Kathryn B. Francis (University of Reading; CogNovo), Ilaria Torre (Trinity College Dublin; CogNovo), Klara Łucznik (University of Plymouth; CogNovo) - Discussion: Endeavouring to be Interdisciplinary
17:15 END

Location: Jill Craigie Cinema, Roland Levinsky Building, University of Plymouth, Drake Circus, Plymouth, PL4 8AA, UK

Keynotes

Developmental Robotics for Language Learning, Trust and Theory of Mind: An Interdisciplinary Journey

Angelo Cangelosi19 June 2018 9:30am

Growing theoretical and experimental research on action and language processing and on number learning and gestures clearly demonstrates the role of embodiment in cognition and language processing. In psychology and neuroscience, this evidence constitutes the basis of embodied cognition, also known as grounded cognition (Pezzulo et al. 2012; Borghi & Cangelosi 2014). In robotics, these studies have important implications for the design of linguistic capabilities in cognitive agents and robots for human-robot communication, and have led to the new interdisciplinary approach of Developmental Robotics (Cangelosi & Schlesinger 2015). During the talk we will present examples of developmental robotics models and experimental results from iCub experiments on the embodiment biases in early word acquisition and grammar learning (Morse et al. 2015; Morse & Cangelosi 2017) and experiments on pointing gestures and finger counting for number learning (De La Cruz et al. 2014). We will then present a novel developmental robotics model, and experiments, on Theory of Mind and its use for autonomous trust behavior in robots. The implications for the use of such embodied approaches for embodied cognition in AI and cognitive sciences, and for robot companion applications will also be discussed.


Interdisciplinarity and Innovation: Learning from the Arts

Sophia Lycouris19 June 2018 1.30pm

Is there a technique to develop interdisciplinary methodologies? Despite the popularity of interdisciplinary research in the recent years, theoretical exploration about what this might entail is not extensive. Researchers tend to engage with interdisciplinary research for the purposes of developing solutions for specific everyday life problems which are too complex to deal with from within single disciplines. As a result of this, they tend to start with concrete problems, and proceed immediately with directly relevant practical steps, without investing time to explore or develop models of interdisciplinary research. Allen Repko, who is one of the very few scholars interested in developing a theory of interdisciplinarity, argues that successful integration across disciplinary approaches is a fundamental principle in successful interdisciplinary research.
But how can integration be achieved? Integration implies a fusion of heterogeneous elements, which generates a coherent and unified outcome, the constituent parts of which are not anymore identifiable as heterogeneous. The product of this fusion is something new that did not exist before and owes its coherence to the emergence of a new system that holds it together. But is this not what Blue Skies research is about? This is the point where the arts come in to explode the potential of future interdisciplinary research. Artists are the world’s specialists in generating objects (or projects) which did not exist before and (if successful) appear as perfectly coherent. Good artistic composition includes various techniques of putting together coherently objects and projects that did not exist before, as informed by historical, geographical and cultural contexts. The arts have perfected the "science" of composition through a multitude of material and conceptual explorations undertaken in the recent years, centuries and millennia.
This presentation will discuss how the arts can contribute to the development of uniquely creative approaches to interdisciplinary research, by referring to examples of projects in which artistic perspectives were in productive dialogue with non-art based approaches to research. The potential to develop techniques for interdisciplinary methodologies of this type will be also discussed by addressing the challenges involved in running such projects.


Interdisciplinarity in research on basic problems in cognition: The case of relation of symbols to dynamics

Joanna Rączaszek-Leonardi19 June 2018 3.15pm

The relation of “symbolic” cognition to the dynamics of a body acting in the environment remains one of the hardest problems in cognitive sciences, despite two decades of work on embodied cognition and underscoring its action-driven, situated, and distributed nature. The ways, in which our behaviour depends on seemingly abstract structures such as, for example, natural language, and the way such structures arise from the dynamics is still not clear. Over 60 years ago, shortly after cognitive sciences were born as a research domain, several researchers advocated a particular take on this relation (e.g., Polanyi, 1968, Pattee, 1969). Instead of the mapping relation (which still is the most prevalent take on ‘meaning’ of symbols), a more ‘cybernetic’ dependency of dynamical processes on symbolic structures was advocated: namely that of control.
This way of looking at symbol/dynamics relation changes the basics of theory formulation of multiple phenomena in multiple cognitive domains. The theoretical stance on this basic issue affects the way we think about language, problem solving, decision making and even movement control, both in the individual and in the collective aspect. Research in those multiple domains can thus bear out the benefits of the control approach. In my talk I will illustrate this in the area of language development and collective task performance in adults, with some hints to the possibility of creating artificial intelligent systems. This will serve a further goal of reflection on the particular way such research engages various disciplines. When we deal with theories about very basic assumptions about cognitive systems, research in various disciplines is called for not (or not only) because an interdisciplinary expertise is needed, but because many disciplines rely on this basic assumption, making it, in a way transdisciplinary. Strategies for doing research that verify claims of such theories are discussed.

Oral presentation

Safely Teaching Robots to Interact with Vulnerable Populations

Emmanuel Senft (University of Plymouth)19 June 2018 10:20am

Robots are expected to take a ubiquitous place in our future society, entering our schools, our care facilities, and our homes. As a result, robots will have to socially interact with vulnerable populations, e.g. when involved in medical therapies or education, where they should learn from domain specialists lacking technical expertise. As such, users should be provided with an efficient and safe way to teach robots to interact socially with humans. We propose and evaluate a new method to progressively teach robots to interact with humans online from in-situ demonstrations from a teacher. The robot rapidly gains autonomy while being supervised by a teacher in control of the robot's behaviour. Our method is evaluated in a child-tutoring scenario, a complex and non-deterministic environment. Results show that using demonstrations and supervision from a teacher, the robot can progressively acquire an efficient action policy, presenting a similar outcome on the teaching session as the teacher's policy. This generic principle of acquiring new behaviour has potential applications to a wide range of problems from optimising the behaviour of social robots to allowing non-technical users to use machine learning.


Playing with Trust

Ilaria Torre (CogNovo, Trinity College Dublin)19 June 2018 10:40am

Computer avatars, digital personal assistants, social robots… they are among us! And they will continue to be. As we build machines that are able to do increasingly more difficult or unwanted jobs, several questions for interdisciplinary researchers open up: how do we make people co-exist with machines in society? How do we make people trust them? How machine-like should they be? These questions are popping up at conferences on Human-Machine Interaction all around the world. The answer of most experimentalists seems to be to build a new machine feature and send a questionnaire round their friends or students, asking how trustworthy that machine looks/sounds/feels like. However this approach only collects data on someone's explicit attitudes towards an object, besides being filled with cognitive biases from the experimenter and participant alike. While immersing people in a realistic environment where we can observe how they interact with a certain machine is not always feasible, perhaps a good compromise is to create a game where we can observe how much people cooperate and trust with a certain machine. I have been trying this approach for a few years, and it seems to at least give some results. I would like to use this opening statement for an informal discussion on methods in Human-Machine Interaction, or Human-X Interaction more in general. What are the pros and cons of different approaches? Is it ethical to use games to elicit particular behaviours? How can we integrate participants' opinions better?


Intention and Insight

Thomas R. Colin (CogNovo, University of Plymouth)19 June 2018 11:15am

Insights, also called "aha!"-moments, are crucial psychological events during which a representation is suddenly shaken off and profitably replaced by another, such that a solution to a conundrum is apprehended "in a flash". How are such sudden bursts of creativity possible? I present a model of insight according to which the crucial component is intention. I argue that insight relies on a flexible repertoire of intentions, enabling agents to investigate different strategies "on a whim" by switching between intentions. Thereby, they can suddenly modify entire strategies for tackling a situation as well as the associated representations. I will discuss the approach from three disciplinary perspectives: (1) Machine Learning and more specifically Reinforcement Learning, which provide the mathematical formalism and building blocks for the model, (2) cognitive psychology and neuroscience, which provide supporting empirical evidence, and (3) the philosophy of intention and action, which relate the model to existing conceptions of action, intention, and reasoning. Results in the different disciplines will be related to one another, highlighting correspondences that have gone unnoticed due to disciplinary barriers. Indeed, besides presenting and defending a theory of insight and intention, I will also briefly discuss the difficulties associated with bridging together fields related by their subject matter, but separated by their vocabulary and methodology.


Eureka as a Model for 2222 Years of Interdisciplinary Research

Frank Loesche (CogNovo, University of Plymouth)19 June 2018 11:35am

Since Eureka’s first mention in an architectural text 2000 years ago and the further exploration by eminent scientists from different disciplines 200 years ago, placing the research in a single discipline has primarily generated confusing and opposing definitions in the past 20 years. Based on my research in the past 2 years, I highlight these conflicts, suggest terminology that helps to situate the research between Cognitive Science, Design, and Education, and provide a case study of how taking a step back assists in advancing the understanding of insight. In interviews with creative practitioners, I show how their education influences the creative process and leads to conflicts between what they learn and what they experience in their practice. Supported by the results from a controlled laboratory study, I suggest a theoretical model resolving these conflicts. I conclude that insight is an example for a research topic that is more fruitfully discussed across disciplinary boundaries.


Effectiveness of Financial Education: Evidence from an Interdisciplinary Research-Project

Tara Zaksaite (CogNovo, University of Plymouth), William Brambley, Shaheen Akter & Olga Siemers (The Open University, Milton Keynes)19 June 2018 11:55am

Increases in personal debt and low levels of saving are growing societal problems (e. g., Bricker et al., 2017; Doty, Collins, Rustgi & Kriss, 2008). Given the wide-reaching importance of this topic for academics, educators, policy-makers, and the general public, it was ideally placed to be considered from an interdisciplinary perspective. Naturally, this work already intersects Economics and Psychology, therefore we united these disciplines within an interdisciplinary project, which explored the effectiveness of a short, behaviour-focused online course on financial management. We collaborated with several non- academic organisations, including credit unions, a community-volunteer initiative, and a magazine. Our team consisted of two Economists, a Psychologist, and a Sociologist. This talk will describe this project, with particular focus on its interdisciplinary aspects, and will outline the preliminary findings.


Tools that propel: on (not) knowing in the body

Sarah Levinsky (University of Falmouth)19 June 2018 12:15pm

Tools that Propel (TTP), an interactive video installation that catalyses new choreographic thinking, emerged from discussions between myself and Adam Russell around the intersections between our separate, but interrelated practice-based research – mine choreography, and Adam's interactive digital media. Confronting its subjects with a life-size projection of themselves and other bodies, it blends live 'mirror-like' video and recorded fragments from the recent past that resemble their current movement. The dancer improvises with 'ghosts' of themselves and others tracked by the sensor before them; this entanglement encourages breaking of habits and mining of memories, exploring subtle variations.
This presentation examines getting lost in order to discover something new as key to our collaborative venture and the development of new embodied knowledge whilst using TTP. The work's logic grew through and across disciplinary terrains as we navigated 'without a compass […] so that together we [could] build that which we [did] not know what might be' (André Lepecki). Mirroring this, when using TTP, dancers are propelled to new physical discoveries and embodied understanding as it draws on the decisions and movements of their corporeal selves but disturbs them, feeding back the incidental and gestures which start midway through the trajectory they previously knew.


Tools that propel: on (not) knowing in time

Adam Russell (University of Falmouth)19 June 2018 12:15pm

This is a companion proposal to Sarah Levinsky's submission, both reflecting on our interactive video installationTools that Propel. Within the theme of Off the Lip 2018 we propose reflecting on the ways in which two very different research perspectives both gave rise to and are informed by this work.Tools that Propel is most obviously concerned with creative processes within full-body movement improvisation driven by temporal juxtapositions of past and present motion. But our collaboration itself can be considered a juxtaposition of different movements.
My own research concerns technology not as a tool that satisfies purposes known in advance by its makers and users, but as 'tools for not knowing'. More specifically this research is informed by Bergson's Matter and Memory (1896), particularly Martha Blassnigg's remarks on the book at Off the Lip 2015. Here, matter that remembers is matter that can do somethingother than a determinate habitual action-reaction circuit, by bringing different past memories into the present duration. I will suggest that we consider collaborations not just between agents separated in space (e.g. Sarah and myself), but between agents separated in time, including between our own past and present intentions.


Headphones and the Affordances of Technological Intimacy

Jacob Downs (University of Sheffield)19 June 2018 2:20pm

The spatial affordances of headphone technologies are intimately, if implicitly, known to all who use them. Individuals can choose to enter into relational acoustic realities, auditorily distancing themselves from extraneous sound and directing perceptual attention towards an interiorized sonic ecology. Here, the interstitial spaces between the tympanic membranes of ears and headphones produce a semi- permeable binaural envelope, broadly overwhelming the auditory faculties. How do listeners make use of these perceptual technologies, and in what ways can we make sense of such experiences as interdisciplinary researchers? Within this short paper, I outline some of the central issues that pervade discussions of headphones and technological intimacy to begin to answer these questions. I use methods from the burgeoning philosophical field of postphenomenology to analyse empirical data collected from headphone listeners about their experiences of sonic intimacy. Engaging not only with 'positive' experiences of these technologies, I consider cases of some contemporary torture practices involving the forcible, non-volitional appendage of headphone technologies to the bodies of detainees and the channelling of intensely loud sound into the ears. I conclude with reflections upon the context-specific multistabilities of such perceptual technologies and consider avenues for future research.


Embodied flow states and its role in movement performance

Aska Sakuta (University of Chichester)19 June 2018 2:40pm

My research project uilises an interdisciplinary methodology to explore the connections between our cognitive states and movement performances, through the idea of embodied Flow states (Nakamura & Csikszentmihalyi, 2014). The research begins with the idea of meditation through/as movement, wherein a mover enters a peak state of experience and performance (Sellers-Young, 1993; Yuasa et al., 1993). This state is often associated with the idea of “Flow” (Fraleigh, 2000; Krein & Ilundain, 2014), a state of sharpened intuition, enhanced focus, and full immersion (Privette, 1983). Here, the research asks the question: Is there a relationship between the phenomenon of "Flow" and the quality of the movement performance?
This presentation will first describe the theoretical and methodological structure of this research, then present its provisionary results. This includes discussions on how some scientific accounts suggest that the state of “Flow” can be neurologically represented by a deactivation in some of the executive functions in the brain, thereby potentially giving way to more primitive functions including intuitive motor control (Austin, 2010; Dietrich, 2004). Following this discussion on the theoretical connections between Flow, embodiment, and meditative states of consciousness, the presentation will describe the interdisciplinary methodology of this research, which examines the relationship between embodied Flow states and movement performance through multiple-trial case studies.


Sync To The Others Not To The Movement – The Investigation Into Shared Physiological Dynamics In Dance Improvisation

Klara Łucznik, Jon May & Emma Redding (CogNovo, University of Plymouth)19 June 2018 4.05pm

A growing number of studies suggest that there is a crucial role for shared physiological dynamics in social coordination, rapport, empathy and even team performance. This paper investigates the presence and temporal development of shared interpersonal physiological dynamics of heart rate and breathing rate during dance improvisation, a free, creative movement practice. Further, it examines whether coordination of physiological state is an underlying component of the group flow experience. Group flow appears in a successful, effortless collaboration where coordination of actions proceeds smoothly in an empathic way. Therefore, it was hypothesised that higher levels of group flow will be related to a higher level of dancers' coordination on the physiological.
As expected, group improvisation score led to higher coordination on a physiological level in the group, measured by heart and breathing rate, in comparison to the solo score. However, the comparison of the coordination of activity level showed that dancers were not in the higher movement (activity level) synchrony in the group tasks than in solo task. These findings suggest that 'empathic projection', the alignment of physiological states by aligning the emotional reaction to the situation, facilitates the shared heart-rate and breathing-rate dynamics in the group dance improvisation. The investigation of group flow experience and shared physiological dynamics did not show any significant links.


Discussion: Endeavouring to be Interdisciplinary

Kathryn B. Francis (University of Reading; CogNovo), Ilaria Torre (Trinity College Dublin; CogNovo), Klara Łucznik (University of Plymouth; CogNovo)19 June 2018 4:25pm

This panel discussion will focus on the challenges of promoting interdisciplinarity. Talks within the panel will explore case studies to form a discussion around the following questions:

  1. How can we foster and nurture open attitudes towards collaboration?
  2. How do we lay the foundations (spatial & temporal) for interdisciplinary collaboration?
  3. What does interdisciplinary success look like?
In addressing question one, the panel discussion will draw on recent work exploring the current state of interdisciplinarity within research divisions, pressures facing researchers who are looking to utilise multiple disciplinary approaches, and how openness might play a role in the transfer and sharing of disciplinary knowledge (Torre et al., forthcoming).
When discussing how we might create spaces (and time) for such endeavours, speakers will discuss platforms that have been designed for interdisciplinary collaboration. Within this discussion, we will explore the value of innovative summer schools, multiple disciplinary seminars, and mixed research groups, in providing positive multidisciplinary experiences for early-career researchers.
Across the discussion, speakers will endeavour to evaluate the short-term and long-term impact of interdisciplinary initiatives and how we might assess the success of these projects within research contexts that often use assessment criteria developed in single-disciplinary settings.