KogWis 2016 Symposia

Symposium S1: Formal and Cognitive Reasoning
Christoph Beierle, Gabriele Kern-Isberner, Marco Ragni, Frieder Stolzenburg

Reasoning about (spatial) information is usually pervaded by uncertainty and subject to change. This is not unique to human reasonig, but it also applies to cognitive systems. Thus there is an increasing demand both from psychology and computer science for non-classical reasoning approaches. So far, many advanced and sophisticated approaches of knowledge representation and reasoning have not yet been made accessible for cognitive approaches, and insights gained from cognition are only rarely reflected in formal approaches. Generally, people employ both inductive and deductive reasoning to arrive at beliefs; but the same argument that is inductively strong or powerful may be deductively invalid. Therefore, a wide range of reasoning mechanisms has to be considered. The field of knowledge representation and reasoning oers a rich palette of methods for uncertain reasoning both to describe human reasoning and to model AI approaches. Beyond computational aspects, these methods aim to reflect the rich variety of human reasoning in uncertain and dynamic environments. The aim of this symposium is to address recent challenges and to present novel approaches to uncertain reasoning in its broad sense, including new insights from cognitive psychology, neuroscience, cognitive computing, and human computation, combining psychological models, uncertain (spatial) reasoning, and computer science. Reflecting this focus, the symposium “Formal and Cognitive Reasoning” at KogWis 2016 is organized jointly by the GI special interest groups Wissensrepräsentation und Schließen and Kognition.

Speakers:
Markus Knauff
, Justus-Liebig-Universität Gießen: New frameworks of rationality  
Christian Freksa, Universität Bremen: Affordance and constraints as complementary notions in problem solving 
Francois Bry, LMU Munich: Human computation: combining the computational power of machine with cognitive skills of humans

 

Symposium S2: Spatial Representation and Processing - What Information Do We Need?
Tobias Meilinger, Ramona Grzeschik

Space is a multi-faceted dimension and the awareness and the comprehension of spatial information is critical for the interaction with our environment. Different approaches have been performed to gain insight into the underlying mechanisms of spatial information process. This symposium will bring together a selection of experimental and theoretical approaches originating from Cognitive and Computer Science, Psychology, Linguistics, Geography, and Architecture to investigate the representation and processing of spatial information.
In this symposium, we will tackle the question how saliency and uniqueness of landmarks influence route learning, particularly in elderly wayfinders, and will discuss the relation between perceptual information, spatial representations and how they are integrated into spatial language. A computational model will be presented that explains perspective taking in terms of a joint influence of interference on the cognitive and the motor level. Further, we will look into the prediction of navigation ability which apparently cannot be reduced to a core processing capability. We will address the questions about the minimal room representations formed in different learning situations and how this representational selection does balance costs from integration and retrieval. Finally, a prototypical wayfinding aid will be presented which was developed with the focus on exploring and developing new means of providing navigational instructions. 
Bringing together researchers and approaches from different areas will afford different perspectives on the question of which information is essential when comprehending and acting in a spatial context, and to what extend information requirements overlap and differ for different spatial tasks.

 

Speakers:
Ramona Grzeschik
, Bournemouth University, UK: Effects of ageing on landmark recognition 
Michele Burigo, Bielefeld University: The influence of extra-linguistic information on spatial language 
Holger Schultheis, U Bremen: Adjusting our view on perspective taking: scalable representation structures and reference frames 
Christoph Hölscher, Chair of Cognitive Science, ETH Zürich, Switzerland: The components of navigation ability 
Tobias Meilinger, MPI for Biological Cybernetics, Tübingen, Germany: The integration of room views 
Jakub Krukar, University of Münster: Wayfinding through orientation: towards a new paradigm of wayfinding

Symposium S3: Cognition and Manual Action
Dirk Koester and Christian Seegelke, Bielefeld University

The human hand is a marvellous complex and flexible tool and people have always been intrigued by its multifunctional utilization. Many of our daily activities require (intentional) physical interactions with one or both hands and possibly with multiple objects. Consequently, the domain of grasping and object manipulation provides a fascinating field of research for cognitive neuroscience, because both sensorimotor and cognitive functions are involved. In this symposium, we will bring together scientists from different disciplines who will present their latest research from the frontiers of the interaction of sensorimotor and cognitive processes in the context of grasping, using a variety of diverse approaches and methodologies. Specifically, data will be presented on how the central nervous system masters the redundancy in the degrees of freedom for the upper limb by means of kinematic and force synergies during unconstrained grasping. In addition, examination of coordinative patterns between gaze and hand movements using eye tracking provides insights into the role of eye-movements in the planning of sequential actions. Furthermore, the role of working memory and object representations in the planning and control of reach-to-grasp actions will be discussed using behavioural (RT and kinematics), neurophysiological (EEG), and neuroimaging (fMRI) methods. This symposium is an attempt to extend our understanding of grasping and the interactions with other cognitive domains and to contribute to the further development of latest neurocognitive theories of movements and cognition.

Speakers:
Abdeldjallil Naceri, Alessandro Moscatelli, Robert Haschke, Marco Santello, Marc Ernst
Digit position and force synergies during unconstrained grasping 
Anna Belardinelli, Martin Butz: Anticipating object interaction with the eyes and with the hands: perceptual and planning aspects 
Rumeysa Gunduz, Thomas Schack, Dirk KoesterThe neurophysiological interaction between working memory and grasping movements 
Christian Seegelke, Iris Güldenpenning, Julian Dettling, Thomas SchackDistinct effects of visuomotor priming on action preparation and motor programming 
Marc Himmelbach, Mareike GannInteractions of cortical networks for object recognition and object grasping

Symposium S4: Dynamics of Sketching and Sketch Understanding (DySket)
Ahmed M. H. Abdel-Fatah, Haythem O. Ismail, Kai-Uwe Kühnberger

Due to the success of touch interfaces as mainstream tools, cognitively inspired AI research faces the challenge to develop human-computer interfaces that employ the capacity of sketch understanding as a basis for enhanced communication with machines. Sketches outperform languages in more easily drawing on one’s well developed intuitions, especially when spatial relations are of a central concern. They are used in various ways to communicate ideas, to support design processes by externalising ideas, to understand complex relations or processes, and even to support memorisation. However, the recognition or retrieval of sketches by computational tools is generally difficult, and requires long computations or simulation of complex mechanisms, such as spatial reasoning, analogy making, abstraction, learning, etc., which are not as intuitive as the humans’ processing for sketch production or recognition. Moreover, while clearly some pictures are sketches and some are not, it is not equally clear whether some sketches are pictures and some are not. A sequence of gestures, for example, may be accepted as a sketch, while clearly not a picture. And, if a picture is worth a thousand words, does this hold also for sketches?
The DySket symposium aims at more deeply discussing these topics on a scientific, interdisciplinary basis. How do humans conceptualise ideas via sketching? What are the main underlying cognitive mechanisms responsible for recognition? Which parts of a sketch play more significant roles than others? How to build AI models guided by the way humans operate on sketches to perform similar tasks?

Speakers:
Malumbo Chipofya
, U Münster: Sketchmapia - A framework for recognition, interpretation and visualization of sketch maps, and integration of sketch maps and metric maps 
Stefan Schneider, University of Osnabrück (UOS), Germany: Mental object manipulation to generate sketches 
Oliver Kutz, Free University of Bozen-Bolzano, Italy: Image schemas, concept invention, and generalization 
Kirsten Bergmann, University of Bielefeld, Germany: Social sketching – Depicting gestures in multimodal communication
Zoe Falomir Llansola, University of Bremen, Germany: Image understanding using sketching and qualitative descriptors
Kai-Uwe Kühnberger, University of Osnabrück (UOS), Germany: The role of concepts in sketch understanding

Symposium S5: Social Perception
Tobias Schlicht, Ruhr-Universität Bochum

In the debate about social understanding, the possibility of directly perceiving the mental states of others has been proposed as an alternative to cognitively more sophisticated strategies as formulated by theory-theory and simulation-theory. What is at stake in this debate is whether the psychological processes underlying social understanding have to be conceived of as perceptual or cognitive or inferential. In this symposium, various questions concerning the possibility of direct social perception of states like emotions and intentions will be addressed. Shaun Gallagher (University of Memphis), one of the chief proponents of social perception will clarify how perception and other minds should be conceived for social perception to be possible and defend this proposal in the light of recent objections. J. Suilin Lavelle (University of Edinburgh) investigates whether proponents of a theory-based approach are committed to the claim that mental states are unobservable. She defends a theory-based approach by exploring what kinds of psychological state can be considered observable if one is to take this approach, focusing particularly on whether non-folk psychological states can be perceived. Finally, Tobias Schlicht (Ruhr-Universität Bochum) then discusses social perception against the background of the debate between an enactive perspective on perception and cognition and a more traditional representationalist perspective on these capacities. Based on a notion of perception in terms of predictive coding, the focus is on the structure and content of the mental representations underlying social perception and possible functions for social perception.

Speakers:
Shaun Gallagher
, Memphis: Perceiving the embodied mind
J. Suilin Lavelle, Edinburgh: Which psychological states can we see?
Tobias Schlicht, Bochum: On the nature and function(s) of social perception

Symposium S6: Mental Files in Cognitive Science: Core cognition, Concepts and Mindreading
Albert Newen, Ruhr-Universität Bochum

Although the idea of mental files as a tool in mental representations of the human mind is not a new one in Cognitive Science, there are remarkable fruitful new usages of mental files to explain several phenomena which were waiting quite long for an adequate treatment. The symposium aims to present some of the key phenomena and argue that the framework of mental files could help to make progress in understanding them. The areas of application to prove the fruitfulness of the perspective of using mental files include: 1. the development of concepts in ontogeny, 2. modes of presentation and co-reference, 3. the development of the ability to pass the explicit false belief task and 4. the development of the core agency system (according to Susan Carey and E. Spelke).

Speakers:
Albert Newen
, Ruhr-Universität Bochum: Mental files and concepts
Francois Recanati, Institut Jean Nicod, Paris: Mental files in a dynamic perspective
Josef Perner, University of Salzburg: Mental files theory of mind
Joulia Smortchkova, Ruhr-Universität Bochum: Core agency cognition: from object-files to agent-files

Symposium S7: PROSOCRATES: Problem Solving, Creativity and Spatial Reasoning in Cognitive Systems
Ana-Maria Olteteanu and Zoe Falomir, University of Bremen

Problem-solving, computational creativity, human creative cognition and spatial cognition are topics often treated separately, despite their major potential for synergies. Problem-solving has been approached in different ways by artificial intelligence and the study of human cognition. For example, humans face ill-structured problem-solving very often, however processes to tackle such problems, like the use of re-representation, are rarely implemented in cognitive agents.
Computational creativity focuses on building creative artificial systems capable of creative feats similar to those achieved by humans and modes of evaluation that can be used to assess such systems. However, the processes and representations in the field are rarely compared to those used by humans.
Human creative cognition investigates the way humans solve a multiplicity of creative tasks, from the simple (coming up with an alternative use for an object) to the complex (solving insight problems), asking questions about process. However, no tools and frameworks exist for implementing computational approaches to test hypotheses in a unified manner.
Finally, spatial cognition is known to contribute to the development of abstract thought, and have a role in insightful problem solving. However, the role of spatial cognition is rarely studied in conjunction with creativity, and with a cognitive systems perspective, aimed at implementing working models.
The topical focus of this symposium is to bring these disciplines together, by bringing in dialogue specialists from each of the fields, as to produce new theoretical tools, approaches and methodologies for the study of problem-solving, creativity and spatial reasoning in cognitive systems.

Speakers:
Bipin Indurkhya, Jagiellonian University, Krakow, Poland: Reflections on using a creativity-stimulating technique for conducting creativity workshops
Kai-Uwe Kühnberger, Institute of Cognitive Science, Universität Osnabrück: Challenges and directions for making cognitive systems creative
Christian Freksa, BSCC, Universität Bremen: The special role of real space for cognition revisited
Anna Fedor, Hungarian Academy of Sciences: Insight and evolution
Ana-Maria Olteteanu, BSCC, Universität Bremen: Re-representation in cognitive systems
Zoe Falomir, BSCC, Universität Bremen: Qualitative reasoning models to help solving spatial ability tests