Cohort 2 (2020-2024)

Visions of the workplace-of-the-future include applications of machine learning and artificial intelligence embedded in nearly every aspect (Brynjolfsson & Mitchell, 2017). This “digital transformation” holds promise to broadly increase effectiveness and efficiency. A challenge to realising this transformation is that the workplace is substantially a human social environment and machines are not intrinsically social. Imbuing machines with social intelligence holds promise to help build human-AI teams and current approaches to teaming one human and one machine appear reasonably straightforward to design. However, if there are more than one human and more than one system that are working together we can see that the complexity of social interactions increases and we need to understand the society of human-AI teams. This research proposes to take a first step in this direction to consider the interaction of triads containing humans and machines.

Our proposed testbed will be concerned with automatic image classification and we choose this since identity and location recognition is a primary work context of our industrial partner Qumodo. Moreover, there are many image classification systems that have recently shown the ability to approach or exceed human performance. There are two scenarios we would like to examine involving human-AI triads and we term them the sharing problem and the consensus problem:

In the sharing problem we examine two humans teamed with the same AI and examine how the human-AI team is influenced by the learning style of the AI, which after initial training can either learn from a single trainer or from multiple trainers. We will examine how trust in the classifier evolves depending upon the presence/absence of another trainer and the accuracy of the other trainer(s). To obtain precise control the “other” trainer(s) could either be actual operators or simulations obtained by parametrically modifying accuracy based on ground truth. Of interest are the questions of when human-AI teams benefit from pooling of human judgment and if pooling can lead to reduced trust.

In the consensus problem we use the scenario of a human manager who must reach a consensus view based on input from a pair of judgments (human-human, human-AI). This consensus will be reached either with or without “explanation” from the two judgments. To make the experiment tractable we will consider the case of a binary decision (e.g. two facial images are of the same person or a different person). Aspects of the design will be taken from a recent paper examining recognition of identity from facial images (Phillips, et al., 2018).

In addition to these experimental studies we also wish to conduct qualitative studies involving  surveys or structured interviews in the workplace to ascertain whether the experimental results are consistent or not with people’s attitudes towards the scenarios depicted in the experiments.

As industry moves further towards AI automation, this research will have substantial impact on future practices within the workplace. Even as AI performance increases, in most scenarios a human is still required to be in the loop. There has been very little research into what such a human-AI integration/interaction should look like. Therefore this research is of pressing importance across a myriad of different sectors moving towards automation.

References
[BRY17] Brynjolfsson, E., & Mitchell, T. (2017). What can machine learning do? Workforce implications. Science, 358(6370), 1530-1534.

[PHI18] Phillips, P. J., Yates, A. N., Hu, Y., Hahn, C. A., Noyes, E., Jackson, K., … & Chen, J. C. (2018). Face recognition accuracy of forensic examiners, superrecognizers, and face recognition algorithms. Proceedings of the National Academy of Sciences, 115(24), 6171-6176.

The aim of the project is to investigate the combination of Human Computer Interaction and social/huggable robots for care, the reduction of stress and anxiety, and emotional support. Existing projects, such as Paro and the Huggable, focus on very simple interactions. The goal of this PhD project will be to create more complex feedback and sensing to enable a richer interaction between the human and the robot.

The plan would be to study two different aspects of touch: thermal feedback and squeeze input/output. These are key aspects of human-human interaction but have not been studied in human-robot settings where robots and humans come into physical contact.

Thermal feedback has strong associations with emotion and social cues [Wil17]. We use terms like ‘warm and loving’ or ‘cold and distant’ in everyday language. By investigating different uses of warm and cool feedback we can facilitate different emotional relationships with a robot. (This could be used alongside more familiar vibration feedback, such as purring). A series of studies will be undertaken looking at how we can use warming/cooling, rate of change and amount of change in temperature to change responses to robots. We will study responses in terms of, for example, valence and arousal.

We will also look at squeeze interaction from the device. Squeezing in real life offers comfort and support. One half of this task will look at squeeze input, with the human squeezing the robot. This can be done with simple pressure sensors on the robot. The second half will investigate the robot squeezing the arm of the human. For this we will need to build some simple hardware. The studies will look at human responses to squeezing, the social acceptability of these more intimate interactions, and emotional responses to them.

The output of this work will be a series of design prototypes and UI guidelines to help robot designers use new interaction modalities in their robots. The impact of this work will be enable robots have a richer and more natural interaction with the humans they touch. This has many practical applications for the acceptability of robots for care and emotional support.

References
[Wil17] Wilson, G., and Brewster, S.: Multi-moji: Combining Thermal, Vibrotactile & Visual Stimuli to Expand the Affective Range of Feedback. In Proceedings of the 35th Conference on Human Factors in Computing Systems – CHI ’17, ACM Press, 2017.

Designing and implementing autonomous drones to assist and interact with people in social contexts, so-called “social drones”, is an emerging area of robotics. Human-drone Interaction (HDI) applications range from supporting users in exercising and recreation [1], to providing navigation cues [2] and serving as flying interfaces [3]. To truly make drones “social”, we must understand how humans perceive them and behave around them. Thus, researchers have traditionally run experiments that allow observing users in direct contact with drones. However, such studies can be difficult, expensive and inflexible. For example, it can be difficult, infeasible, or even dangerous to conduct an experiment in the real world to study the impact of different drone sizes or flying altitudes on the user’s behavior. On the other hand, if valid experiments can be conducted in immersive virtual reality (VR), researchers can reach out to a higher number and potentially more diverse participants, and control the environment variables to a greater degree. For example, changing the size of a drone in VR involves merely changing a variable’s value. Similarly, drones in VR are not bound by the physical limitations of the word. But if VR-based studies of human-drone interactions are to be used as a springboard for informing our understanding of HDI in the real world, it is imperative to understand the extent to which findings generalize to in situ HDI.

This project aims to explore the use of immersive virtual reality (VR) as a test bed for studying human behavior around social drones. The main objectives are to understand whether results from studies conducted in VR would match results from corresponding real-world settings, and use VR to inform embodied/in-person HDI studies. As prior work suggests [5], it is expected that some behaviors will be similar across VR and the real world, thereby allowing researchers to use VR as an alternative to real world studies in some contexts. However, understanding the limitiations of VR for developing social drones will be vital as well.

To this end, the project will involve studying and comparing human proxemic behavior around drones both in VR and in the real world. While proxemic behavior has been investigated for human-robot Interaction [4], it has never been studied when interacting with drones. It is expected that attributes of drones, such as their flying altitude or their size, impact how people distance themselves from drones. As has also been studied in HRI proxemics, the extent to which people’s preferred distance to social drones differs based on first and third person viewpoints is also of interest to examine. The results from the real world and VR studies will be compared to allow an assessment of the opportunities, challenges, and limitations of using virtual reality to conduct experiments on introducing drones in close quarters to people to serve social purposes, and provide guidelines can help researchers decide whether to employ VR in their experiments and what factors to account for if doing so.

References
[1] Florian “Floyd” Mueller and Matthew Muirhead. 2015. Jogging with a Quadcopter. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (CHI ’15). Association for Computing Machinery, New York, NY, USA, 2023–2032.

[2] Pascal Knierim, Steffen Maurer, Katrin Wolf, and Markus Funk. 2018. Quadcopter-Projected In-Situ Navigation Cues for Improved Location Awareness. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI ’18). Association for Computing Machinery, New York, NY, USA, Paper 433, 1–6.

[3] Markus Funk. 2018. Human-drone interaction: let’s get ready for flying user interfaces! interactions 25, 3 (May-June 2018), 78–81.

[4] Jonathan Mumm and Bilge Mutlu, “Human-robot proxemics: Physical and psychological distancing in human-robot interaction,” 2011 6th ACM/IEEE International Conference on Human-Robot Interaction (HRI), Lausanne, 2011, pp. 331-338.

[5] Ville Mäkelä, Rivu Radiah, Saleh Alsherif, Mohamed Khamis, Chong Xiao, Lisa Borchert, Albrecht Schmidt, and Florian Alt. 2020. Virtual Field Studies: Conducting Studies on Public Displays in Virtual Reality. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (CHI ’20). Association for Computing Machinery.

Trustworthiness is a property of an agent or organisation that engenders trust in others. Humans rely on trust in their day-to-day social interactions, be they in the context of personal relationships, commercial negotiation, or organisational consultation  (with healthcare providers or employers for example). Social success therefore relies on the evaluation of the trustworthiness of others, and our own ability to present ourselves as trustworthy. If autonomous agents are to be used in a social environment, it is vital that we understand the concept of trustworthiness in this context [DEV18].

Some formal models of trust for autonomous systems have been proposed (e.g. [BAS16]), but these models are geared specifically towards autonomous vehicles. Any proposed model must be evaluated by testing. In many cases this would involve deploying complex hardware in sufficiently realistic scenarios in which trust would be a consideration. However, it is also possible to investigate trust in other scenarios. For example, it has been shown that different interfaces to an automatic image classifier change the calibration of human trust towards the classifier [ING20]. Relevant to social processing, in [GAL19] trust was examined via the use of videos. Here, the responses of human participants to videos involving an autonomous robot in a range of scenarios were used to investigate different aspects of trust.

Another way to generate user data to test formal models is using mobile games.  In a recent paper [KAV19], a model of the way that users play games was used to investigate a concept known as game balance. A software tool known as a probabilistic model checker [KWI17] was used to predict user behaviour under the assumptions of the model. Subsequently the game has been released to generate user data in order to evaluate the credibility of the model used.

In this PhD project you will use a similar technique to evaluate trust for autonomous systems. The crucial aspects are the formal models of trust and the question of how to design a suitable game so that the way users respond to different scenarios reflect how much they trust (autonomous robot or animated) characters in the game. You will:

1. Develop and evaluate models of trust for autonomous robots
2. Devise a mobile game for which players will respond according to their trust in autonomous robot or animated characters
3. Use an automatic technique such as model checking or simulation to determine player behaviour under the assumptions of your trust models
4. Analyse how well player behaviour matches that predicted using model checking

References
[DEV18] Trustworthiness of autonomous systems– K. Devitt, Foundations of Trusted Autonomy, Studies in Systems, Decision and Control, 2018

[BAS16] Trust dynamics in human autonomous vehicle interaction: a review of trust models – C. Basu et al. AAAI 2016.

[ING20] Calibrating trust towards an autonomous image classifier: a comparison of four interfaces – M. Ingram et al., submitted.

[GAL19] Trusting Robocop: Gender-Based Effects on Trust of an Autonomous Robot – D. Gallimore et al. Frontiers in Psychology 2019

[KAV19] Balancing turn-based games with chained strategy generation – W. Kavanagh, A. Miller et. al. IEEE Transactions on Games 2019.

[KWI17] Probabilistic model checking: advances and applications – M. Kwiatkowska et al. Formal System Verification 2017.

Research on tactile skin or e-skin has attracted significant interest recently as it is the key underpinning technology for safe physical interaction between humans and machines such as robots. Thus far, eSkin research has focussed on imitating some of the features of human touch sensing. However, skin is not just designed for feeling the real world, it is also a medium to express feeling through gestures. For example, the skin on the face, which can fold and wrinkle into specific patterns, allows us to express emotions such as varying degrees of happiness, sadness or anger. Yet, this important role of skin has not received any attention so far. Here, for the first time, this project will explore the emotion signal generation capacities of skin by developing programmable soft e-skin patches with embedded micro actuators that will emulate real skin movements. Building on the flexible and soft electronics research in the James Watt School of Engineering and the social robotics research in Institute of Neuroscience & Psychology, this project aims to achieve the following scientific and technological goals:

1. Identify suitable actuation methods to generate simple emotive features such as wrinkles on the forehead
2. Develop a soft eSkin patch with embedded microactutors
3. Use dynamic facial expression models for specific movement patterns in the soft eSkin patch
4. Develop an AI approach to program and control the actuators

Virtual reality (VR) is a powerful entertainment tool allowing highly immersive and richly contextual experiences. At the same time, it can be used to flexibly manipulate the 3D (virtual) environment allowing to tailor behavioural experiments systematically. VR is particularly useful for social interaction research, because the experimenter can manipulate rich and realistic social environments, and have participants behave naturally within them [RB18].

While immersed in VR, a participant builds an inner map of the virtual space and stores multiple expectations about the environment mechanisms i.e., where objects or rooms are and their interaction with them, but also about physical and emotional properties of virtual agents (e.g. theory of mind). Using this innovative and powerful technology, it is possible to manipulate both the virtual space and virtual agents within the virtual world, to test internal participants’ expectations and register their reactions to predictable and unpredictable scenarios.

The phenomenon of “change blindness” demonstrates the surprising difficulty observers have in noticing unpredictable changes to visual scenes[SR05]. When presented with two almost identical images, people can fail to notice small changes (e.g. in object colour) and even large changes (e.g. object disappearance). This process arises because the brain cannot attend to the entire wealth of environmental signals presented to our visual systems at any given moment, and instead use attentional networks to selectively process the most relevant features whilst ignoring others. Testing which environmental attributes drive the detection of changes can give useful insights on how humans use predictive processing in social contexts.

In this PhD the student will run behavioural and brain imaging experiments in which they will use VR to investigate how contextual information drives predictive expectations in relation to changes to the environment and agents within it. They will investigate if change detection is due to visual attention or to a social cognitive mechanism such as empathy. This will involve testing word recognition whilst taking the visuospatial perspective of the agents previously seen in the VR (e.g. [FKS18]). The student will examine if social contextual information originating in higher brain areas modulates the processing of visual information.  In brain imaging literature, an effective method to study contextual feedback information is the occlusion paradigm [MPM19]. Cortical layer specific fMRI is possible with 7T brain imaging; the student will test how top-down signals during social cognition activate specific layers of cortex. This data would contribute to redefining current theories explaining the predictive nature of the human brain.

The student will also develop quantitative models in order to assess developed theories. In recent work [PMT19], model checking was proposed as a simple technology to test and develop brain models. Model checking [CHVB18] involves building a simple, finite state model, and defining temporal properties which specify behaviour of interest.  These properties can then be automatically checked using exhaustive search. Model checking can replace the need to perform thousands of simulations to measure the effect of an intervention, or of a modification to the model.

References
[MPM19] Morgan, A. T., Petro, L. S., & Muckli, L. (2019). Scene representations conveyed by cortical feedback to early visual cortex can be described by line drawings. Journal of Neuroscience, 39(47), 9410-9423.

[SR05] Simons, D. J., & Rensink, R. A. (2005). Change blindness: Past, present, and future. Trends in cognitive sciences, 9(1), 16-20.

[RB18] de la Rosa, S., & Breidt, M. (2018). Virtual reality: A new track in psychological research. British Journal of Psychology, 109(3), 427-430.

[FKS18] Freundlieb, M., Kovács, Á. M., & Sebanz, N. (2018). Reading Your Mind While You Are Reading—Evidence for Spontaneous Visuospatial Perspective Taking During a Semantic Categorization Task. Psychological science, 29(4), 614-622.

[PMT19] Porr, B., Miller, A., & Trew, A. (2019). An investigation into serotonergic and environmental interventions against depression in a simulated delayed reward paradigm. Adaptive behaviour, (online version available).

[CHVB-8] Clarke, E. M., Henzinger, T. and Veith, H. & Bloem, R (2018). Handbook of model checking. Springer.

Virtual and Mixed Reality systems are socio-technical applications in which users experience different configurations of digital media and computation that give different senses of how a “virtual environment” relates to their local physical environment. In Human-Computer Interaction (HCI), we recently developed computational models capable of representing physical and virtual space, solving the problems of how to recognise  virtual spatial regions starting from the detected physical position of the users (Benford et al., 2016). The models are bigraphs [MIL09] derived from the universal computational model introduced by Turing Award Laureate Robin Milner. Bigraphs encapsulate both dynamic and spatial behaviour of agents that interact and move among each other, or within each other. We used the models to investigate cognitive dissonance, namely the inability or difficulty to interact with the virtual environment.

How the brain represents physical versus virtual environments is also an issue very much debated within Psychology and Neuroscience with some researchers arguing that the brain makes little distinction between the two [BOZ12]. Yet more in line with Sevegnani’s work, Harvey and colleagues have shown that different brain areas represent these different environments and that they are further processed in different time scales HAR12; ROS09]. Moreover, special populations struggle more with virtual over real environments [ROS11].

The overarching goal of this PhD project is, therefore, to adapt the computational models developed in HCI and apply them to psychological scenarios, to test whether the environmental processing within the brain is different as proposed. This information will then refine the HCI model and ideally allow a refined application to special populations.

References
[BEN16] Benford, S., Calder, M., Rodden, T., & Sevegnani, M., On lions, impala, and bigraphs: Modelling interactions in physical/virtual spaces. ACM Transactions on Computer-Human Interaction (TOCHI), 23(2), 9, 2016.

[BOZ12] Bozzacchi., C., Giusti, M.A., Pitzalis, S., Spinelli, D., & Di Russo, F., Similar Cerebral Motor Plans for Real and Virtual Actions. PLOS One (7910), e47783, 2012.

[HAR12] Harvey, M. and Rossit, S., Visuospatial neglect in action. Neuropsychologia, 50, 1018-1028, 2012.

[MIL09] Milner, R.,  The space and motion of communicating agents. Cambridge University Press, 2009.

[ROS11] Rossit, S., Malhotra, P., Muir, K., Reeves, I., Duncan G. and Harvey, M., The role of right temporal lobe structures in off-line action: evidence from lesion-behaviour mapping in stroke patients. Cerebral Cortex, 21 (12), 2751-2761, 2011.

[ROS09] Rossit, S., Malhorta, P., Muir, K., Reeves, I., Duncan, G., Livingstone, K., Jackson H., Hogg, C., Castle P., Learmonth G. and Harvey, M., No neglect- specific deficits in reaching tasks. Cerebral Cortex, 19, 2616-2624, 2009.

The aim of this project is to investigate, in the context of social interactions,the interaction between a driver and an autonomous vehicle. Autonomous cars are sophisticated agents that can handle many driving tasks. However, they may have to hand control to the human driver in different circumstances, for example if sensors fail or weather conditions are bad [MCA16, BUR19]. This is potentially difficult for the driver as they may have not been driving the car for a long period and have to quickly take control [POL15]. This is an important issue for car companies as they want to add more automation to vehicles in a safe manner.Key to this problem is whether this interface would benefit from conceptualizing the exchange between human and car as a social interaction.

This project will study how best to handle handovers, from the car indicating to the driver that it is time to take over, the takeover event, and then the return to automated driving. They key factors to investigate are: situational awareness (the driver needs to know what the problem is and what must be done when they take over), responsibility (who’s task is it to drive at which point),  the in-car context (what is the driver doing: are they asleep, talking to another passenger), and driver skills (is the driver competent to drive or are they under the influence).

We will conduct a range of experiments in our driving simulator to test different types of handover situations and different types of multimodal interactions involving social cuesto support the 4 factors outlined above.

The output will be experimental results and guidelines that can help automotive designers know how best to communicate and deal with handover situations between car and driver. We currently work with companies such as Jaguar Landrover and Bosch and our results will have direct application in their products.

References
[MCA16] Mcall, R., McGee, F., Meschtscherjakov, A. and Engel, T., Towards A Taxonomy of Autonomous Vehicle Handover Situations, Publication: Proceedings of the International Conference on Automotive User Interfaces and Interactive Vehicular Applications, pp. 193–200, 2016.

[BUR19] Burnett, G., Large, D. R. & Salanitri, D., How will drivers interact with vehicles of the future? Royal Automobile Club Foundation for Motoring Report, 2019.

[POL15] Politis, I, Pollick, F and Brewster S. Language-based multimodal displays for the handover of control in autonomous cars, Publication, Proceedings of the International Conference on Automotive User Interfaces and Interactive Vehicular Applications, pp. 3–10, 2015.

Aims
Increasing research suggests that intelligent virtual agents are most effective and accepted when they adapt themselves to individual users. One way virtual agents can adapt to different individuals is by developing an effective model of a user’s traits and using it to anticipate dynamically varying states of these traits as situational conditions vary. The primary aims of the current work are to develop: (1) empirical methods for collecting data to build user models, (2) computational procedures for building models from these data, (3) computational procedures for adapting these models to current situations. Although the project’s primary goal is to develop a general framework for building user models, we will also explore preliminary implementations in digital interfaces.

Novel Elements
One standard approach to building a model of a user’s traits—Classical Test Theory—uses a coherent inventory of measurement items to assess a specific trait of interest (e.g., stress, conscientiousness, neuroticism). Typically, these items measure a trait explicitly via a self-report instrument or passively via a digital device. Well-known limitations of this approach include its inability to assess the generalizability of a model across situations and occasions, and its failure incorporate specific situations into model development. In this project, we expand upon the classic approach by incorporating two new perspectives: (1) Generalizability Theory, (2) The Situated Assessment Method. Generalizability Theory will establish a general user model that varies across multiple facets, including individuals, measurement items, situations, and occasions. The Situated Assessment Method replaces standard unsituated assessment items with situations, fundamentally changing the character of assessment.

Approach
We will develop a general framework for collecting empirical data that enables building user models across many potential domains, including stress, personality, social connectedness, wellbeing, mindfulness, eating, daily habits, etc. The data collected—both explicit self-report and passive digital—will assess traits (and states) relevant to a domain across facets for individuals, measurement items, situations, and occasions. These data will be to Generalizability Theory and the Situated Assessment Method to build user models and establish their variance profiles. Of particular interest will be how well user models generalize across facets, the magnitude of individual differences, and clusters of individuals sharing similar models. Situated and unsituated models will both be assessed to establish their relative strengths, weaknesses, and external validity. Once models are built, their ability to predict a user’s states on particular occasions will be assessed, using procedures from Generalizability Theory, the Situated Assessment Method, and autoregression. Prediction error will be assessed to establish optimal model building methods. App prototypes will be developed and explored.

Outputs and Impact
Generally, this work will increase our ability to construct and understand user models that virtual agents can employ. Specifically, we will develop novel methods that: (1) collect data for building user models, (2) assess the generalizability of models; (3) generate state-level inferences in specific situations. Besides being relevant for the development of intelligent social agents, this work will contribute to deeper understanding of classic assessment instruments and to alternative situated measurement approaches across multiple scientific domains. More practicially, the framework, methods, and app prototypes we develop are of potential use to clinicians and individuals interested in understanding both functional and dysfunctional health behaviours.

References
[BLO12] Bloch, R., & Norman, G. (2012). Generalizability theory for the perplexed: A practical introduction and guide: AMEE Guide No. 68. Medical Teacher, 34, 960–992.

[PED20] Pedersen, C.H., & Scheepers, C. (2020). An exploratory meta-analysis of the state-trait anxiety inventory through use of generalizability theory. Manuscript in prepration.

[DUT19] Dutriaux, L., Clark, N., Papies, E. K., Scheepers, C., & Barsalou, L. W. (2019). Using the Situated Assessment Method (SAM2) to assess individual differences in common habits. Manuscript under review.

[LEB16] Lebois, L. A. M., Hertzog, C., Slavich, G. M., Barrett, L. F., & Barsalou, L. W. (2016). Establishing the situated features associated with perceived stress. Acta Psychologica, 169,119–132.

[MAR14] Stacy Marsella and Jonathan Gratch. Computationally Modeling Human Emotion. Communications of the ACM, December, 2014.

[MIL11] Lynn C. Miller, Stacy Marsella, Teresa Dey, Paul Robert Appleby, John L. Christensen, Jennifer Klatt and Stephen J. Read. Socially Optimized Learning in Virtual Environments (SOLVE). The Fourth International Conference on Interactive Digital Storytelling (ICIDS), Vancouver, Canada, Nov. 2011.

According to Emmanuel Schegloff, one of the most important linguists of the 20^thCentury, conversation is the “primordial site of human sociality”, the setting that has shaped human communicative skills from neural processes to expressive abilities [TUR16]. This project focuses on these latter and, in particular, on the use of nonverbal behavioural cues such as laughter, pauses, fillers and interruptions during dyadic interactions. In particular, the project targets the following main goals:

• To develop approaches for the automatic detection of laughter, pauses, fillers, overlapping speech and back-channel events in speech signals;
• To analyse the interplay between the cues above and social-psychological phenomena such as emotions, agreement/disagreement, negotiation, personality, etc.

The experiments will be performed over two existing corpora. One includes roughly 12 hours of spontaneous conversations involving 120 persons [VIN15] that have been fully annotated in terms of the cues and the phenomena above. The other is the Russian Acted Multimodal Affective Set (RAMAS) − the first multimodal corpus in Russian language, including approximately 7 h of high-quality close-up video recordings of faces, speech, motion-capture data and such physiological signals as electro-dermal activity and photoplethysmogram [PER18].

The main motivation behind the focus on nonvebal behavioural cues is that these tend to be used differently in different cultural contexts, but they can still be detected independently of the language being used. In this respect, an approach based on nonverbal communication promises to be more robust to the application over data collected in different countries and linguistic areas. In addition, while the importance of nonverbal communication is widely recognised in social psychology, the way certain cues interplay with social and psychological phenomena still requires full investigation [VIN19].

From a methodological point of view, the project involves the following main aspects:

• Development of corpus analysis methodologies (observational statistics) for the investigation of the relationships between nonverbal behaviour and social phenomena;
• Development of signal processing methodologies for the conversion of speech signals into measurements suitable for computer processing;
• Development of Artificial Intelligence techniques (mainly based on deep networks) for the inference of information from raw speech signals.

From a scientific point of view, the impact of the project will be mainly in Affective Computing and Social Signal Processing [VIN09] while, from an industrial point of view, the impact will be mainly in the areas of Conversational Interfaces (e.g., Alexa and Siri), multimedia content analysis and, in more general terms, Social AI, the application domain encompassing all attempts of making machines capable to interact with people like people do with one another. For this reason, the project is based on the collaboration between the University of Glasgow and Neurodata Lab, one of the top companies in Social and Emotion AI.

References
[PER18] Perepelkina O., Kazimirova E., Konstantinova M. RAMAS: Russian Multimodal Corpus of Dyadic Interaction for Affective Computing. In: Karpov A., Jokisch O., Potapova R. (eds) Speech and Computer. Lecture Notes in Computer Science, vol 11096. Springer, 2018.

[TUR16] S.Turkle, “Reclaiming conversation: The power of talk in a digital age”, Penguin, 2016.

[VIN19] M.Tayarani, A.Esposito and A.Vinciarelli, “What an `Ehm’ Leaks About You: Mapping Fillers into Personality Traits with Quantum Evolutionary Feature Selection Algorithms“, accepted for publication by IEEE Transactions on Affective Computing, to appear, 2019.

[VIN15] A.Vinciarelli, E.Chatziioannou and A.Esposito, “When the Words are not Everything: The Use of Laughter, Fillers, Back-Channel, Silence and Overlapping Speech in Phone Calls“, Frontiers in Information and Communication Technology, 2:4, 2015.

[VIN09] A.Vinciarelli, M.Pantic, and H.Bourlard, “Social Signal Processing: Survey of an Emerging Domain“, Image and Vision Computing Journal, Vol. 27, no. 12, pp. 1743-1759, 2009.

Face signals play a critical role in social interactions because humans make a wide range of inferences about others from their facial appearance, including emotional, mental and physiological states, culture, ethnicity, age, sex, social class, and personality traits (e.g., see Jack & Schyns, 2017). These judgments in turn impact how people interact with others, oftentimes with significant consequences such as who is hated or loved, hired or fired (e.g., Eberhardt et al., 2006). However, identifying what face features drive these social judgment is challenging because the human face is highly complex, comprising a high number of different facial expressions, textures, complexions, and 3D shapes. Consequently, no formal model of social face signalling currently exists, which in turn has limited the design of artificial agents’ faces to primarily ad hoc approaches that neglect the importance of facial dynamics (e.g., Chen et al., 2019). This project aims to address this knowledge gap by delivering a formal model of face signalling for use in socially interactive artificial agents.

Specifically, this project will a) model the space of 3D dynamic face signals that drive social judgments during social interactions, b) incorporate this model into artificial agents and c) evaluate the model in different human-artificial agent interactions. The result promises to provide a powerful improvement in the design of artificial agents’ face signalling and social interaction capabilities with broad potential for applications in wider society (e.g., social skills training; challenging stereotyping/prejudice).

Modelling the face signals will be derived using methods from human psychophysical perception studies (e.g., see Jack & Schyns, 2017) that extends the work of Dr Jack to include a wider range of social signals used in social interactions (e.g., empathy, agreeableness, skepticism). Face signals that go beyond natural boundaries such as hyper-realistic or super stimuli will also be explored. The resulting model will be incorporated into artificial agents using the public domain SmartBody (Thiebaux et al., 2018) animation platform with possible extension to other platforms. Finally, the model will be evaluated in human-agent interaction using the SmartBody platform with possible combination with other modalities including head and eye movements, hand/arm gestures, transient facial changes such as blushing, pallor, or sweating (e.g., Marsella et al., 2013).

Although there is not a current industrial partner, we expect the work to be very relevant to companies interested in the use of virtual agents for social skills training, such as Medical CyberWorld, and companies working on realistic humanoids robots, such as Furhat and Hanson Robotics. Jack and Marsella have pre-existing relations with these companies.

References
Jack, R. E., & Schyns, P. G. (2017). Toward a social psychophysics of face communication. Annual review of psychology, 68, 269-297.

Eberhardt, J. L., Davies, P. G., Purdie-Vaughns, V. J., & Johnson, S. L. (2006). Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing outcomes. Psychological science, 17(5), 383-386.

Chen, C., Hensel, L. B., Duan, Y., Ince, R., Garrod, O. G., Beskow, J., Jack, R. E. & Schyns, P. G. (2019). Equipping Social Robots with Culturally-Sensitive Facial Expressions of Emotion Using Data-Driven Methods. In: 14th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2019), Lille, France, 14-18 May 2019, (Accepted for Publication).

Stacy Marsella, Yuyu Xu, Margaux Lhommet, Andrew Feng, Stefan Scherer, and Ari Shapiro, “Virtual Character Performance From Speech”, in Symposium on Computer Animation , July 2013.

Stacy Marsella and Jonathan Gratch, “Computationally modeling human emotion”, Communications of the ACM , vol. 57, Dec. 2014, pp. 56-67.

Marcus Thiebaux, Andrew Marshall, Stacy Marsella, and Marcelo Kallmann, “SmartBody Behavior Realization for Embodied Conversational Agents”, in Proceedings of Autonomous Agents and Multi-Agent Systems (AAMAS), 2008.

Increasingly, location-based social networks, such as Foursquare, Facebook or Yelp are replacing traditional static travel guidebooks. Indeed, personalize venue recommendation is an important task for location-based social networks. This task aims to suggest interesting venues that a user may visit, personalized to their tastes and current context, as might be detected form their current location, recent venue visits and historical venue visits. The recent development of models for venue recommendation have encompassed deep learning techniques, able to make effective personalized recommendations.

Venue recommendation is typically deployed such that the user interacts with a mobile phone application. To the best of our knowledge, voice-based venue recommendation has seen considerably less research, but is a rich area for potential improvement. In particular, a venue recommendation agent may be able to elicit further preferences, ask if they prefer one venue or another, or ask for clarification in the type of venue, or distance to be travelled to the next venue.

This project aims to:

• Develop and evaluate models for making venue recommendation using chatbot interfaces, that can be adapted to voice through integration of text-to-speech technology, building upon recent neural network architectures for venue recommendation.
• Integrate additional factors about personality of the user, or other voice-based context signals (stress, urgency, group interactions) that can inform the venue recommendation agent.

Venue recommendation is an information access scenario for citizens within a “smart city” – indeed, smart city sensors can be used to augment venue recommendation with information about which areas of the city are busy.

References
[Man18] Contextual Attention Recurrent Architecture for Context-aware Venue Recommendation. Jarana Manotumruksa, Craig Macdonald and Iadh Ounis. In Proceedings of SIGIR 2018.

[Man17] A Deep Recurrent Collaborative Filtering Framework for Venue Recommendation. Jarana Manotumruksa, Craig Macdonald and Iadh Ounis. In Proceedings of CIKM 2017.

[Dev15] Experiments with a Venue-Centric Model for Personalised and Time-Aware Venue Suggestion. Romain Deveaud, Dyaa Albakour, Craig Macdonald, Iadh Ounis. In Proceedings of CIKM 2015.