'Logic as a tool'
An interview with Nina Gierasimczuk
Nina Gierasimczuk is a Post-Doc researcher at the Institute for Language, Logic and Computation (ILLC). She obtained her MA in Philosophy at the Warsaw University, and a PhD in 2010 at the ILLC, under supervision of Johan van Benthem and Dick de Jongh. Her current VENI research focuses on formal learning theory, dynamic epistemic logic, and belief revision. She is also actively involved in research into the role of logic and logical modeling in cognitive science.
‘I began my higher education studies in Warsaw; I enrolled in an interdisciplinary MA program in humanities. The scheme gave me the chance to choose all sorts of University courses, creating my own individual curriculum. I focused mostly on the interdisciplinary agenda of logic. My first choice was philosophy, especially epistemology and analytical philosophy. I complemented the studies with psychology, mathematics, and computer science. It might seem surprising, but I also studied history of art.’
‘I moved to the Netherlands in 2006. I initially applied for a PhD position at the ILLC, but funny enough I lost the competition to my partner. I was then anyway called on the phone by a member of the hiring committee and advised to come to the Netherlands; my profile turned out to be interesting enough. I took it at face value, came to Amsterdam, and after a year of non-funded research I was given a two-year contract. In the final year I got a job in the Talenten Kracht project and obtained additional funding for teaching at Amsterdam University College. All the pieces came together in the end of 2010, when I finished and defended my thesis. Immediately after that, in 2011, I moved to Groningen, where I was a offered a post-doc position at the Artificial Intelligence Department.’
‘I find the ILLC environment extremely stimulating and motivating. Maybe that’s because we have so many strong individuals of similar academic and scientific profile. So many logicians–smallest differences in their frameworks become important. Everyone’s fighting to establish their identity and their independent research agenda. As a result environment is very dynamic.’
Learning patterns in groups
‘In 2013, I got a VENI grant for my project ‘Learning from each other. Formal analysis of multi-agent learning’. The aim is to see how being in a group affects learning. Learning strategies studied in computational linguistics, but also in formal epistemology are by and large oriented towards single agent learning. In psychology, there is much research showing that the presence of peers can enhance learning in certain situations. We also know of cases when being in a group impairs learning. An interesting question is what kind of roles subjects assume in group-learning situations and why. How individual and group knowledge changes in social networks is a very current problem. For example, the most popular model of group learning, classroom situation, is very problematic in this respect. Take for instance a case of pluralistic ignorance: social situation in which students after a lecture are asked whether they understood everything and stay silent even though there is still quite some things they could ask. I want to analyse these multi-agent effects in formal learning theory, extending results from my PhD towards the convergence of group knowledge.’
‘The first step is to overview similar phenomena that have been already studied, for example in psychology, epistemology, philosophical logic, and sociology. Then make some logical and computational sense of it, find common denominators, specify game-theoretical, linguistic, and knowledge related aspects. Finally, I hope to link to the existent frameworks of learning and see where the agenda should be pushed forward.’
‘I have been involved in adding logical tasks to ‘Rekentuin.nl’ (Math Garden), which is an online educational training system for children, used massively by kids in the Netherlands. Children learn mathematical operations, like addition and multiplication, in online games. My role was to develop and help implementing a logical reasoning game. The game we chose is a simplified version of Mastermind. It is often said that logical thinking is a matter of pure talent rather than learning, but I do not believe that. Working with the Rekentuin team, especially with the psychologists Han van de Maas and Maartje Raaijmakers helped me keeping in touch with psychological, empirical reality. Here one can see learning in action, learning patterns, but also error patterns. In this way, we can discover whether there is logic to learning, in the way people acquire new skills and knowledge. And this seems to be the case: one of the results of the project shows clear-cut error patterns. This suggests that there are certain stages of trained abilities, certain common thresholds that children have to overcome in order to get to the ideal performance. So this sheds some light not only on the logical abilities of children, but also on the learning of them.’
‘Before I started university, one of the options was to apply to fine arts academy. Now I still do make art, mostly painting, drawing, and some design. Sometimes those things come quite close to my academic life. I designed quite some covers for scientific books, Johan van Benthem’s books, my colleagues PhD theses. And whenever there is some sort of quasi-artistic task to be done in close academic proximity, I take care of that, and find it quite rewarding.’
‘I do not really divide my time into free time and work. Almost everything I do scientifically I like doing, so the transition is fluent. Art is no exception here; I am becoming increasingly interested in cognitive science of art. Art and history of art has a lot to say about perception, in particular about visual perception. How do we reason with visual stimuli? What makes us like certain types of pictures and dislike others? Finally, more towards social epistemology, what makes certain types of art popular? I am currently reaching out to Rietveld Academy trying to find out how to match my logical/computational agenda with the study of art and visual perception, but this is really something that I am just beginning to understand and explore.’
‘I have spent a big part of my life as an expat, and I do not have a big problem with that, it feels quite natural for me. Amsterdam is a great place to live in and it’s quite close to Poland, in many respects. Being an expat academic couple is another issue on its own. It is often a long and tiresome fight to be stationed in one place. And since now it is getting harder and harder to get a permanent job anywhere in Academia, this is a real issue. I observed quite a lot of prejudice towards couples in the academic world. Sometimes it can be especially harmful towards women, because of the negative stereotypes in male dominated communities. Criticism based on stereotypes, though often expressed unintentionally, can resonate for a very long time. They can also make one underachieve in their work. My personal strategy, not always successful, is to distance myself from those thoughts, and focus on work. I also try to fight explicit and point out implicit bias whenever I see it.’
The potential of logic
‘It is a popular slogan that logic is a study of human reasoning. We should be aware that this idea of logic being just a description of how people reason is naïve, and it leads to all sorts of modelling problems. For example, if you assume that people are just logical machines, they should in principle know all logical consequences of everything they believe. This is unrealistic; a combination with certain types of processing restrictions could help here. Such are the computational ones. And here is where logic has a new role to play, as it is used extensively in studying the notions of computability and complexity.’
‘In my opinion, mathematical logic has a great potential as a tool for cognitive science. In my joint work with psychologists logic turned out to be essential in the pre-modelling phase, when experiments where designed, and then again later when results needed to be interpreted. Moreover, logic has languages that can fittingly describe all kinds of abstract structures. As a result, it could be one of the tools for investigating how human mind work on a qualitative level. For example, modal logics are used in computer science to verify processes, algorithms; to test whether they are well designed and will be successful. I think that this way of thinking may well be applied in cognitive science.’