Nonlinear Methods for Psychological and Social Sciences

ABOUT THE WORKSHOP

This workshop provides the theoretical background, as well as hands-on training in R for some of the most popular and broadly-applicable nonlinear analyses methods for the application in psychology and the social sciences. Specifically, we will focus on the following methods: Recurrence Quantification Analyses (RQA), Fractal Analysis (FA), and Convergent Cross-Mapping (CCM).

RQA has a broad applicability, allowing the analysis of quantitative and qualitative changes in time series and sequences, the analysis of coupling properties between two or more systems, as well as the analysis of multidimensional time series. FA can be used to explore fractal scaling relations and long-memory properties of time series. CCM is a method to analyze the strength and direction of influence among two time series allowing the construction of a causal network between a set of variables.

In the psychological and social sciences, these methods are helpful to investigate systems with nonlinear time evolution since they do not make strong assumptions about the data, and allow to capture complex temporal patterns. They have been particularly useful for the analysis of time-series data from naturalistic settings and semi-experimental studies, investigations of response times, as well as the analysis of joint-action data in order to quantify group dynamics such as synchrony and coupling between individuals, as well as leader-follower relationships.

Bring Your Data. While we will prepare lots of exercises and even provide exemplary data sets for the hands-on parts of the workshop, participants are welcome to bring their own data. That way, participants can draft individual analysis pipelines and discuss their data in more detail together with the instructors. Reach out to the instructors throughout the workshop days to get you started, but there will be more time during the last workshop session.

Poster Session. Participants are invited to present a research project or research idea in which nonlinear methods have been, are or could be applied. We encourage you to seize this opportunity to discuss ideas for the application of nonlinear methods with instructors, assistants and fellow participants. Sign up for a poster during the application process; instructions for poster preparations will be sent out after registration.

Social Gatherings. In our opinion, the NLM workshops are not only for learning great new methods, but also for networking. We will therefore organize two social gatherings, where we can get to know each other, and may exchange ideas, opinions or beverage preferences ;-) Participation is voluntary, costs have to be covered by yourself. You can sign up for either or both events during the on-site registration.

REQUIREMENTS

No previous knowledge of the methods discussed in the workshop is required. However, basic knowledge of R is necessary, i.e., specifying working directories, loading and installing packages, loading and saving data, executing lines of code, and performing basic operations in the command line. If you are not familiar with R, we highly recommend working through a tutorial prior to the workshop (e.g., here).

Participants need to bring their own laptops with a current version of R and RStudio being installed in order to follow along during the hands-on sessions.

INSTRUCTORS AND TEACHING ASSISTANTS

Sebastian Wallot

Since 2020, Sebastian is professor for general psychology and research methods at the Leuphana University of Lüneburg, Germany. He did his PhD in experimental psychology at the University of Cincinnati, USA, and worked afterwards as postdoctoral researcher at the Interacting Minds Centre at the Aarhus University, Denmark, and at the Department of Language and Literature at the Max Planck Institute for Empirical Aesthetics in Frankfurt/M., Germany. His research focuses on reading, i.e., on the relation between reading process characteristics and comprehension, joint action, and the development of and application of nonlinear time-series analyses.

Dan Mønster

Originally trained as a physicist, Dan now directs the Cognition and Behavior Lab at Aarhus University, Denmark. There he is associate professor at the Department of Economics and Business Economics where he teaches mathematics, and is part of the Interacting Minds Centre. His research focuses on interpersonal dynamics in group cooperation, using both linear and nonlinear methods to analyze behavioral and physiological data from controlled experiments. Dan has contributed to algorithms for nonlinear analysis and R packages.

Giuseppe Leonardi

During his doctorate in experimental psychology, which he completed in Trieste, Italy, Giuseppe studied applications of Dynamical Systems Theory in psychology and spent two years at the Center for Complex Systems at the Florida Atlantic University, USA. He specializes in dynamical analyses of naturalistic human interactions, with a focus on Recurrence Quantification Analysis. He is now dean of psychology studies at the University of Economics and Human Sciences in Warsaw, Poland.

Ralf Cox

Ralf is currently associate professor at the Developmental Psychology group in the Department of Psychology at the University of Groningen, Netherlands. He received his PhD in social sciences at the Radboud University Nijmegen, Netherlands, in 2007. His research addresses the coordinative basis of behavior, cognition and their development. This entails a micro-genetic and complex dynamical systems approach to intra- and interpersonal coordination, as well as the advancement of nonlinear time-series techniques to study this. He has published on motor performance, planning, language development, dyslexia, mother-child, client-therapist and peer interaction, and gesture-speech attunement.

Monika Tschense

During her studies in linguistics, Monika focused on the (neuro-) cognition of language and speech-language pathology. Since November 2018, she is a PhD candidate in psychology at the Max-Planck-Institute for Empirical Aesthetics in Frankfurt/M., Germany, and the Leuphana University of Lüneburg, Germany. Together with Sebastian Wallot, Monika investigates to what extent nonlinear measures are suitable for predicting text comprehension during natural reading based on time series of eye movements, reading times, and electrophysiological measures.

Alon Tomashin

Alon is studying the neurophysiological mechanisms of group interactions in the Social Neuroscience Lab at Bar-Ilan University, advised by Ilanit Gordon. Aiming to predict relationship quality from autonomic nervous system activity, he evaluates interpersonal synchrony and its variation utilizing non-linear methods. Additionally, in collaboration with Sebastian Wallot, he develops extensions to RQA designed to capture the dynamics of social coordination.

SCHEDULE

Monday, July 17th

  8.30   On-Site Registration
  9:00   Welcome!
  9:30   Why bother? Nonlinear methods in psychology and social sciences
11:00   How to deal with complex systems?
12:30   LUNCH BREAK
13:30   Recurrence Quantification Analysis I
15:30   Recurrence Quantification Analysis II
17:00   Research example RQA
19:00   Social Gathering: tba

Tuesday, July 18th

  9:00   Cross-Recurrence Quantification Analysis I
11:00   Cross-Recurrence Quantification Analysis II
12:30   LUNCH BREAK
13:30   Cross-Recurrence Quantification Analysis III
15:30   Multidimensional Recurrence Quantification Analysis
17:00   Research example CRQA or MdRQA (guest talk)

Wednesday, July 19th

  9:00   Parameter estimation
11:00   Parameter exploration
12:30   LUNCH BREAK
13:30   Fractal Analysis I
15:30   Fractal Analysis II
17:00   Research example FA (guest talk)
19:00   Social Gathering: tba

Thursday, July 20th

  9:00   Convergent Cross Mapping I
11:00   Convergent Cross Mapping II
12:30   Lunch Break
13:30   Convergent Cross Mapping III
15:00   Research example CCM (guest talk)
16:00   Poster Session

Friday, July 21st

  9:00   Wrap-up | Q&A
10:00   Data Analysis Session: A: Analyze your own data   |   B: Analyze an example data set
13:00   Good bye!

APPLICATION, REGISTRATION & COSTS

The NLM 2023 is limited to 25 participants. Should there be more applications than slots, participants will be selected based on their motivational statements and CVs.

To apply for the workshop, you have to fill in an application form. You have to enter your personal information, answer some questions about your prior knowledge (meant only for the instructors to get a sense of the group not for participant selection), and upload or enter the following:

  • a short CV (max. 2 pages, PDF format, max. 2 MB);
  • a motivational statement that briefly indicates why you want to participate, and how you intend to use nonlinear methods in your research (max. 1,500 characters);
  • in case you are applying for financial support: a brief statement explaining why you should be considered for funding (max. 1,500 characters).

You can apply from March 1st to April 15th 2023 (23:59 CEST). Please use this application link: www.soscisurvey.de/NLM2023/

Participation in the NLM 2023 is free of charge, but you have to cover costs for travel, accommodation, food, etc. yourself. For a limited number of participants, we can assist with travel and accommodation costs (up to 200€) – in this case, please explain your need for financial assistance in the application form.

WORKSHOP LOCATION

Leuphana Universität Lüneburg
Universitätsalle 1
21335 Lüneburg, Germany

Building: tba | Room: tba

The NLM 2023 will be an on-site workshop, virtual participation is neither intended nor not provided for.
Please check whether you will need a visa for travelling to Germany! Applying for and obtaining the visa is your own responsibility. If you need an invite to get a visa, please indicate so in the application form.