"Nosognosia: unrealistic optimism and the denial of illness"
Professor Peter Brugger, University Hospital Zurich, CHF 140'925.- (3 years)

Aim of the doctoral thesis by Corinne Tamagni was to investigate how the human brain builds a positively biased picture of one's own health state. The first part of the section "own contributions" is dedicated to the study of positive illusions in the process of self-evaluation. The second part introduces two experiments on lateralized emotional processing, with a special focus on the influence of emotion on spatial attention.

Anosognosia designates the inability to recognize one's own illness. Representing one of the most striking and enigmatic forms of agnosia, no model is currently able to account fully and comprehensively for the phenomenon in its scope and complexity. Unlike previous investigations focusing on dysfunction, the present thesis emphasizes the study of the normal function impaired in anosognosia, a function reasonably labeled nosognosia. Nosognosia is the function that continuously monitors our own well-being, against the background of interoceptive and external changes. It comprises not only the detection of signs of illness, but also the ability to realistically estimate one's future health state. Many studies showed that the latter is generally biased towards optimism and that healthy people display a strong tendency to view themselves as less likely than others to turn ill. This positive illusion is called unrealistic optimism.

It seems that there is a qualitative similarity between clinical confabulatory behavior and unrealistic optimism. Conceptualizing the optimistic attitude inherent to nosognosia as a form of "prospective anosognosia", study 1 supports this view. In the same way it abolishes anosognosic denial, left-ear cold water irrigation reduced unrealistic optimism for one's own future health state in healthy subjects. This finding indicates the right parieto-insular cortex – activated by this procedure - as a key structure for the appraisal of one's own vulnerability to illness.

Study 2 and 3 investigate brain lateralization of emotions, exploring the interaction between emotional processing and spatial attention. We found a modulation of spatial attention by emotional states and traits with a strong association between negative affect and leftward hemispatial attention. The results validate the predominant role of the right hemisphere in negative emotions and indicate the importance of the parietal lobes in the processing of non-spatial contents.

Together, the findings of the experiments with healthy volunteers are consistent with clinical observations from patients with unilateral brain lesions as well as with previous findings collected in healthy populations. They corroborate the existence of functional hemispheric asymmetries in both self-awareness and emotional processing and support the view of a right hemisphere specialization for both the ability to recognize one's own illness and to process negative emotions.

" Interactively controlled language as a bridge between law and artificial intelligence "
Dr. Stefan Höfler, University Zurich, CHF 93'718.-

The field of Artificial Intelligence and Law investigates how methods of artificial intelligence can be employed in the legal domain. One main obstacle to progress in the field is the fact that legal texts are written in natural language, while AI methods are based on formal logic. A complete and precise automatic translation of one form of representation into the other is as yet not feasible due to natural language ambiguity.

The aim of the project was to investigate whether the syntax and semantics of legislative language (specifically, Swiss legislative language) can be controlled so that it can be automatically translated into formal logical representations and still remains natural and expressive enough for practical use in legislative drafting.

In a first step, existing legislative texts (and drafts of such texts provided by the editorial board of the Swiss Federal Chancellery) were analysed with regard to potential ambiguities contained therein. Two types of ambiguities were distinguished: ambiguities that pose a problem to automatic processing but are usually not noticed by human readers, and ambiguities that lead to misinterpretations even in humans. The methods of controlled natural language were then applied to both types of ambiguities: the use of some linguistic constructions that can lead to ambiguity were either prohibited (construction rules), or the constructions were assigned default interpretations that determine the sense in which they are to be used (interpretation rules). This method was used to develop a linguistic standard consisting of a number of well-defined conventions that reduce ambiguity in legislative language and thus facilitate its automatic processing.

In order to ensure that the developed standard would not deviate from existing legislative language too much, construction and interpretation rules were, wherever possible, designed to reflect (a) existing domain-specific conventions, (b) the style guidelines issued by various cantonal and federal authorities, (c) frequency distributions occurring in existing legislative language. Thus, additional control mechanisms were established for a range of phenomena that are frequently sources of ambiguities; these mechanisms are explained in more detail in the publications that resulted from the project.

The project showed that, while the chosen approach cannot solve all problems that the automatic semantic processing of legislative language faces, it can get some major obstacles out of the way. A whole range of lexical, syntactic and semantic ambiguities can be prevented by employing controlled language. However, the fact that research in the area of deontic logic has as yet not been able to represent the content of legislative texts in a complete and adequate way will continue to pose a problem. The acceptability of a standard such as the one developed in the project will furthermore depend on how closely the defined rules manage to reflect actual language usage. With regard to this issue, in-depth corpus-based studies in Swiss legal linguistics would be desirable.

The project has yielded two technical reports, four conference contributions, one journal article and two invited talks. As suggested by the cogito foundation council, a follow-up proposal was submitted to the Swiss National Foundation and has been accepted to grant the funds for another three years.

"The potential of ancient sacred natural sites for a novel approach to nature conservation"
Dr. Claudia Rutte, University of Berne Fr. 67'200.-

The conservation of biodiversity has been mainly based on protected areas. However, the exclusion of local people is believed to be one of the reasons why protected areas are often ineffective. More recently, conservation initiatives involve local communities in the management of protected areas. The implementation of participatory forms of governance also shows shortcomings and in many cases paper parks have been created without much participation of the communities. An alternative conservation strategy to such top-down approaches could build on supporting existing community-based conservation areas, such as sacred natural sites. Sacred landscapes exist in many countries and are especially prevalent in biodiversity rich regions.

To facilitate interdisciplinary research and exchange among researchers interested in understanding and strengthening the link between nature and culture, an online database on sacred natural sites was developed. SANASI is accessible at www.sanasi.org and contains information on different aspects of sacred natural sites around the earth linked to geography and ecology, belief system, management, current threats, and economic values. Data for SANASI are sourced from publications in scientific journals covering different disciplines within life sciences and social sciences, from reports of national or international organisations, from books, and from the internet. The SANASI advisory board is currently working on a data policy to also include information shared by local community members (based on a Free, Prior and Informed Consent).

In May 2011, data on more than 200 sites in 25 countries have been included in SANASI. It is a long-term initiative and collaborations with the IUCN Specialist Group on Cultural and Spiritual Values of Protected Areas (CSVPA), the Indigenous and Community Conserved Areas (ICCA) registry, and Alliance of Religions and Conservation (ARC) are planned.

Conference "Social decision making: bridging economics and biology"
Prof. Laurent Lehmann, Université Neuchâtel, Fr. 18'000.-

Topic of the conference
There is considerable conceptual overlap between the fields of evolutionary biology and economics. Both economics and evolutionary biology are based on similar game-theoretic foundations, making predictions on individual's decisions based upon economic analyses that account for the costs and benefits of particular actions, which depend on the actions taken by others. It is only by regarding behavior as an outcome of both an evolutionary and a lifetime dynamical process, which results in the formation of an individual's preferences who has limited memory, limited cognitive ability, and use social information, that we will gain a better understanding of why individuals take the decision they take. Evolutionary biologist and economist have thus many things to learn from each other, in particular in the domain of social decisions making (decisions on behaviours regarding other individuals), which was the main underlying theme of our conference.

The conference aimed to improve understanding of why individuals take the decisions they do and stimulate dialogue between economists and evolutionary biologists. Four sessions paired leading economists and evolutionary biologists to discuss specific topic so as to explore the common interests between the fields. Professor Alan Grafen gave the first talk of the conference, reviewing key publications by economists linking natural selection and utility maximization. The next invited talks, by Professor Peyton Young and Professor David Stephens, both explored social decision making and social learning, and contrasted the two subjects from an economic, and a biological perspective. In the third session, Colin Camerer discussed the advances in neuroeconomics and in particular that models with limited levels of reasoning (k-level of reasoning) predict successfully human behavior in experimental games, while Arnon Lotem presented experimental results from sparrows illustrating the flexibility of individual learning. In the last session, Professor Paul Seabright, highlighted recent experiments, which bridged the overlap between biology and economics in understanding costly signalling, and using human smiles as a model example. In the closing talk of the conference, Professor John McNamara highlighted that allowing for genetic variation in a population can significantly alter the predictions of standard evolutionary models, and highlighted areas where economists may be able to assist.

Goal of the conference
The major aim of the conference was to create an environment that would be conducive to informal dialogue, mutual learning and the development of new collaborations between evolutionary biologists and economists on the topic of decision-making in social interactions (e.g., any game theoretic or social evolution situation). Therefore world-class experts from both disciplines described their work while avoiding discipline-specific jargon in various sessions (e.g., bounded rationality, learning dynamics, experimental biology and economics), each of which combined an evolutionary biologist and an economist doing related work so as to promote the bridging between approaches. Attendees were a mixture of evolutionary biologists and economists, with an emphasis on attracting younger researchers, at the start of their careers. The goal was also to offer the opportunity to build long-lasting collaborations in the field of decision-making in social interactions.

Most important results
The conference helped stimulate the important dialogue between different disciplines. 67 participants from the USA, UK, Australia, South Africa, Russia, Austria, Netherlands, Belgium, Germany, France, Italy, Hungary, Sweden, Finland, Spain and Switzerland attended. In addition to the talks by seven, very eminent, invited speakers, there were twenty-six 20-minute contributed talks and 10 posters. The oral presentations and poster session reflected the mixture of the attendees, and the conference also served as a platform to disseminate the latest work in social decision making, and stimulate discussion between the disciplines.

It was generally agreed in discussions, both after the talks, and at other points during the conference, that there was a lot of common ground between evolutionary biology and economics, and that there is much to be gained from working together and discussing mutual ideas. Economists who attended the conference were surprised about the extent to which maths was used in biology and the scale and scope of experiments biologists conducted. Several economists, who conduct experimental work, said they would like to learn more about experimental design and data collection techniques used by biologists. They noted that experimental work in their field is in its infancy and biologists have an expertise they can benefit from. Most economists expressed a wish to learn more about biology. Both experimental and theoretical biologists highlighted that a major future challenge for biology is to understand the individual-level diversity of the expression of behaviours, specifically the evolution of flexible learning rules. The consensus was that economics, which is concerned more with proximate, real-time explanations of behaviour than biologists (who tend to focus on ultimate, adaptive explanations) would have a lot to offer. All participants left the conference stimulated with new ideas, and there is no doubt that interesting and topical collaborations will result from the interactions.