Programme
Invited Speakers
Prof. Steven Sloman – Brown University (USA)
Prof. Guillaume Deffuant – Irstea (Fr)
Dr. Josè Ramasco – University of the Balearic Island (ES)
Steering Commettee
Prof. Roger Whitaker – University of Cardiff (Uk)
Dr Rosapia Lauro-Grotto - University of Florence (IT)
Dr Andrea Passarella - Institute of Informatics and Telematics (IIT) - CNR, Pisa (IT)
Jennifer Willies – Project Manager, Awareness coordination action - Napier University (Uk)
Organizing Commettee
Dr Franco Bagnoli - Department of Physics, University of Florence (IT)
Dr Andrea Guazzini – Department of Education and Psychology, University of Florence (IT)
Dr Giovanna Pacini - Department of Physics, University of Florence (IT)
Dr Giorgio Gronchi - Department of Health Sciences, University of Florence (IT)
Dr Emanuele Massaro - Department of Information Engineering, University of Florence (IT)
Dr Elisa Guidi - Department of Information Engineering, University of Florence (IT)
Dr Cristina Cecchini - Department of Information Engineering, University of Florence (IT)
June 18th
RECOGNITION Dissemination Workshop
From Cognitive Activity to Artificial Self Awareness
Florence, 18th June, 2013.
Agenda
Venue: Palazzo Nonfinito, Via del Proconsolo 12, Florence.
(June 18th 2013)
9.00 - Welcome and Salutation
Dr. Franco Bagnoli - Centre for the Study of Complex Dynamics (CSDC), University of Florence, (IT)
9.15 - Title: An overview of the RECOGNITION project
Speaker: Prof. Roger Whitaker - Cardiff University, (UK)
9.40 - Morning Session
Title: A Tri-Partite model of Cognition: Decision Making and Problem Solving processing models
Chair: Dr. Rosapia Lauro-Grotto - University of Florence & CSDC, (IT)
9.45 - Title: The Right Way to Think About Heuristics in Cognition
Abstract: I will present the dual systems perspective on human thought and describe the place of heuristics in that view. I will then describe a promising normative framework for causal reasoning, Causal Bayesian Networks. I show a couple of limitations of the framework for modeling human judgment. First, people tend to neglect alternative causes when reasoning predictively but not diagnostically. Second, people typically know less about causal system than they think they do. I report studies showing this is true in politics and that shattering the illusion leads to more moderate attitudes and can reduce donations
to political advocacy groups. The illusion is not present in those who score well on the Cognitive Reflection Test.
Speaker: Prof. Steven Sloman - Brown University, (USA)
10.40 - Title: Computational models in psychology: A tri-partite model of cognition for dual process theory
Abstract: Computational models represent a natural bridge between cognitive psychology and ICT. Different computational approaches have been applied to model reasoning and decision making. However, these approaches show several limitations in describing dual process theories of cognition. We describe a tri-partite model of the cognitive systems that can address part of these limitations. This model can be thought as a general framework based on the idea that the cognitive system relies on predictions based on memories that are continuously generated either based on the information gathered from the senses or from the knowledge that characterizes the system. Future applications to both psychology and ICT are discussed.
Speaker: Dr Giorgio Gronchi - University of Florence & CSDC, (IT)
11.15 - Coffe Break
11.30 - Title: Influential Neighbours Selection for Information Diffusion
Abstract:The problem of maximizing information diffusion through a network is a topic of considerable recent interest. A conventional problem is to select a set of any arbitrary k nodes as the initial influenced nodes, so that they can effectively disseminate the information to the rest of the network. However, this model is usually unrealistic in online social networks since we cannot typically choose arbitrary nodes in the network as the initial influenced nodes. We introduce a new problem called Influential Neighbour Selection problem to select a node’s neighbours to efficiently disseminate its information. In the initial study in above, constant probability value is used for the decision making whether the information would get propagated at each node in the network. We have extended our study by exploring decision making behaviour using an experimental study on the photo rating together with eye tracking.
Speaker: Dr Eiko Yoneki - University of Cambridge, (UK)
11.55 - Title: Cognitive Heuristics for Data and Knowledge Dissemination in Opportunistic Networks
Abstract: In the convergence of the Cyber–Physical World, user devices will act as proxies of humans in the cyber world. Such devices will be required to act in a vast information landscape, asserting the relevance of data spread in the cyber world, in order to let their human users become aware of the content they really need. This is a remarkably similar situation to what the human brain has to do all the time, when deciding what information coming from the surrounding environment is interesting and what can simply be ignored. The brain performs this task using the so-called cognitive heuristics, i.e. simple, rapid, yet very effective schemes. In this presentation, we show a series of a new approaches that exploit the cognitive heuristics framework for developing self-adaptive systems that deals with effective data and knowledge dissemination in an opportunistic network scenario.
Speaker: Dr Matteo Mordacchini - Institute of Informatics and Telematics - CNR, (IT)
12.20 - Title: The Human Layer in Decision-Making in Networked Environments
Abstract: The emergence of intelligent sensing and communication technologies fosters the generation and dissemination of huge amounts of information that collectively enriches people's awareness about their environment and its resources. With this information at hand, users then decide how to access these resources to best serve their interests. However, situations repeatedly emerge where the users' welfare is better satisfied by the same finite set of resources and the uncoordinated access to them gives rise to tragedy of commons effects and serious congestion problems. In this context, the high-level question we address is how efficiently the competition about the resources is resolved under different assumptions about the way the users make their decisions. The users are first viewed as strategic perfectly informed software agents that make fully rational decisions attempting to minimize the cost of accessing the acquired resource. We then exploit insights from experimental economics and cognitive psychology to model agents of bounded rationality who either do not possess perfect information or cannot exploit all the available information due to time restrictions and computational limitations.
Speaker: Dr Evangelia Kokolaki - National & Kapodistrian University of Athens, (GR)
12.55 - Social Lunch
14.00 - Title: An overview of the AWARENESS coordination action
Speaker: Jennifer Willies - Edinburgh Napier University, (UK)
14.25 - Afternoon Session
Title: Sociophysics of the human virtual dynamics: Theoretical approach and experimental results
Chair: Dr. Andrea Passarella - Institute of Informatics and Telematics - CNR, (IT)
14.30 - Title: The Leviathan model: Absolute dominance, generalised distrust and other patterns emerging from combining vanity with opinion propagation
Abstract: We propose an opinion dynamics model that combines processes of vanity and opinion propagation. The interactions take place between randomly chosen pairs. During an interaction, the agents propagate their opinions about themselves and about other people they know. Moreover, each individual is subject to vanity: if her interlocutor seems to value her highly, then she increases her opinion about this interlocutor. On the contrary she tends to decrease her opinion about those who seem to undervalue her. The combination of these dynamics with the hypothesis that the opinion propagation is more efficient when coming from highly valued individuals, leads to different patterns when varying the parameters. In one of the patterns, absolute dominance of one agent alternates with a state of generalised distrust, where all agents have a very low opinion of all the others (including themselves). We provide some explanations of the mechanisms behind these emergent behaviors and finally propose a discussion about their interest.
Speaker: Prof. Guillaume Deffuant - IRSTEA - Paris, (FR)
15.25 - Title: A microscopic model for social influence in spatially extended electoral processes
Abstract: Influence among individuals is at the core of emergent social phenomena such as the diffusion of innovations, social learning, the dissemination of ideas, beliefs or behaviors. Different approaches have been proposed to integrate inter-agent influence in social models. However, a satisfactory confront between empirical data and model predictions has not been reached yet. Here we advance in this direction by introducing spatial and population diversity into an opinion dynamics model. The model is organized with a metapopulation structure in which individuals' mobility provides a proxy for social context and peer imitation accounts for social influence. Our model reproduces several features observed in vote-shares of the US presidential elections at different geographical levels. It recovers, for example, a Gaussian-like distribution of the party vote-shares with a changing average but a constant dispersion, and spatial correlations falling logarithmically with distance. This approach offers a method to contrast microscopic mechanisms with macroscopic society-wide information.
Speaker: Dr. Jose Javier Ramasco - University of the Balearic Islands. (ES)
16.05 - Coffe Break
16.20 - Title: Human Virtual Interaction: The Small Group Dynamics
Abstract: The aim of this work is to present an effective experimental framework for the study of small groups dynamics. We will present a series of experiment investigating the interactions and the relationship network occurring in a virtual small group dynamics. Through three different experimental conditions (i.e. social constraints) with an increasing degree of social complexity, we have explored the communication dynamics characterizing such a system, and we have assessed the cognitive strategies used by the subjects to represent the social environment. Our findings show a complex interdependence between the perception and the mental representation (i.e. affinity) of the others, and the communication dynamics occuring in a virtual small group interaction. The results follow the main assumptions of the Lewin’s Field Theory, showing a complex interaction between opinion, personality and environment, due to the effect of the contextual dynamics (i.e. experimental condition) both, on the local dynamics (i.e. members), as weel as on the global dynamics (i.e. group dynamics).
Speaker: Dr. Alessandro Cini - University of Florence & CSDC (IT)
16.45 - Title: Revealing the structure of complex networks by cognitive inspired information processing algorithm
Abstract: The problem of community detection is relevant in many scientific disciplines, from social science to statistical physics. A large variety of computational heuristics, some with a long history, have been proposed for the identification of communities or, alternatively, of good graph partitions. We have focused on a psychology and social network-inspired approach which may be useful for further strengthening the link between social network studies and mathematics of community detection. The main issue is that in most cases it is very difficult to have a clear definition of community structure because of people tend to rely in different communities and in different sub-communities at the same time. Our local approach is based on an information diffusion process considering node as agents capable of take decision. We show how by adopting a dynamical perspective towards community detection it is possible to detect both, the right community in which nodes belong to, as well as the right global network structure from the local point of view. After reviewing UNIFI's publications in this field I will show the application of the method to a dynamic social interaction "game" as a joint publication with the CNR.
Speaker: Dr. Emanuele Massaro - University of Florence & CSDC (IT)
17.10 - Title: The personality of places
Abstract: The use of the Five Factor Model to classify human personality is well established in the field of Psychology. In this talk, we explore how participatory ICT systems could be used to build an analogous model to represent the personality of individual venues or neighbourhoods. Taking inspiration from the concept of our "sense of place", we approach the problem using cues from individuals' behaviour, their social relationships and properties of the local environment. The resulting "collective spatial awareness" can benefit in the effective processing, acquisition and exchange of spatially relevant information for decentralised and autonomous applications.
Speaker: Dr. Stuart M. Allen - Cardiff University, (UK)
17.35 - Round Table
Title: From Cognitive Activity to Artificial Self Awareness: open challenges and new insights
Moderator: Dr. Franco Bagnoli
Panelists: S. Sloman, G. Deffuant, R. Whitaker, J. Willies, R. Lauro-Grotto, A. Passarella, M. Ӧnen
19.00 - End of the workshop