AIL-Talks: Digital Emotions
Artificial Intelligence (AI) currently are experiencing a powerful renaissance of scientific and public attention. We believe
that the advance of these technologies will have an enormous impact on our ways of living, working, learning, communicating,
travelling, caressing, curing, and even dying.
However, there still remain two big issues to
be solved in order to allow for a fruitful human-machine-interaction: In order to become trustworthy to humans, how can machines
develop emotional behavior, and how do they have to look like?
In the lectures series four international
experts from AI, robotics, and biology discuss the state of the art in their respective research fields.
concept & moderation: Katharina Gsöllpointner
Dr. Katharina Gsöllpointner is a media and art theoretician with
a habilitiation in media studies. Her focus is on the crossover of media aesthetics, digital technologies, and the cybernetics
of art, and she has a passion for inter-, trans- and crossdisciplinarity. Beside working as a university lecturer and researcher
at the Department Digital Art at the University of Applied Arts Vienna, since 2017 she also is a staff member of the Department
Cross-Disciplinary Strategies where she teaches Artistic Strategies from a cross-disciplinary perspective.
Univ. Prof. Dr. Robert Trappl
Head of the Austrian Research Institute for
Artificial Intelligence Vienna; Professor Emeritus of Medical Cybernetics and Artificial Intelligence at the Center for Brain
Research, Medical University of Vienna
Why do humans have emotions? What functions do they have in mental processes and
in interactions with other humans? What is their anatomical and neurophysiological basis? Which different emotions do
we know? Are their facial expressions different in different cultures, i.e. learned or are they universal? How can robots
recognize emotions? Are there rules / diagrams which emotions robots should express as a result of different situations, in
order to make a "human-like“ impression? Do robots already have emotions and, if not, how are the chances that they will ever
have them? We will try to answer some of these questions to have a material for a vivid discussion.
Dr. Adam Miklosi
Director of the Family Dog Project, Head of Ethology, Eötvös Loránd University (Hungary)
can robots learn from Dogs? In recent years, robotics started to chase a dream: It wants to produce a companion for humans.
The main aim of social robotics is to build agents which can move into our homes, schools or hospitals, and which are regarded
as ’real’ companions. Some people imagine such robot as „real” people while others think it should be something radically
But why leave out those people from this scientific endeavour who know the most about companionship, behaviour
and social interaction: the ethologists?
Based on our experience in social robotics and ethology we came to the
conclusion that the human-dog relationship offers a perfect model to study the possibilities and constrains of robot design.
We consider the dog to be man’s first ’biorobot’. Thus we suggest that robots of the future should be by no means similar
to man but represent a "new species”.
The detailed study of the human-dog relationship led us to make proposals
for the behavioural capacities of companion robots that include complex social skills, like attachment, faithfulness, emotional
responsiveness, social monitoring etc. It will be the question of the not-so-distant future how our insights can be implemented
in real robotic agents.
Prof. Dr. Eva Hudlicka
at Psychometrix Associates & Visiting Faculty at the College of Information and Computer Sciences, University of Massachusetts
Amherst, Psychotherapist in private practice at therapy21st
Social robots and virtual affective agents are increasingly
becoming components of our digital landscape. This talk will discuss how computational models of emotions can enhance
their affective realism, and how this impacts the associated human-agent relationships. The talk will address
the nature of these relationships, their roles in human-agent interactions, as well as the emerging ethical considerations.
Univ.Prof. Dr. Mag. Helmut Leder
Professor of Cognitive Psychology and Head of Department of Basic Psychological
Research and Research Methods at the University of Vienna, Head of the Research Focus “Perceptual Aesthetics”, Department
of Psychology/University of Vienna
Psychology of the Arts: Move me… astonish me? Emotions in Aesthetic Exeriences:
150 years after its foundation as a psychological discipline, Empirical Aesthetics has arrived at a state in which descriptive
models open new, exciting pathways for empirical, experimental, and neuro-scientific progress towards a Psychology of the
Arts. Our theoretical model (Leder et al., 2004) has guided a large number of empirical studies, e.g. regarding style processing
(Leder & Nadal, 2014; Leder, Bär & Topolinski, 2012), the cognitive classification changes due to expertise (Belke
et al., 2010), effects of context and museum environments (e.g. Brieber et al.,2015), and art-related emotion (Leder et al.,
2013). Ten years after the model was proposed, based on our own and others’ developments, we have now provided an even more
comprehensive model theory (Pelowski, Markey, Forster, Gerger, & Leder, 2017) – comprising states of being strongly moved,
negative emotions and even transformative states.
Foto: Pinar Yoldas The Kitty AI. Artificial Intelligence
for Governance (2016)