Psychology of Information Processing

Our group works on understanding how humans use their perceptual systems, especially vision, actively during natural extended behavior to guide decisions and actions with their bodies. This leads to the study of how the visual system uses sensory input, forms beliefs about the world by carrying out computations on the basis of its representations, and then carries out actions in goal directed behavior.

Both experimental evidence from the area of active vision as well as theoretical advances in control theory and machine learning show that perception, action, and learning can only be separated under very restricted circumstances demonstrating the importance of the behavioral goal i.e. the current task. Our work therefore emphasizes the necessity to consider perception, cognition, decision making and actions jointly and investigates their interplay.

To make progress in this area we use an integrated approach requiring us to apply different methodological techniques. Our current focus is on:

  • behavioral studies involving measurements of human eye and body movements during complex naturalistic tasks in natural and virtual environments,
  • developing inverse optimal control and inverse optimal decision-making models that allow recovering individual subjects' cognitive, subjective beliefs, intrinsic costs, and internal models,
  • building models of tasks and developing algorithms that learn how to solve these tasks in virtual agents,
  • developing models for the representation and quantification of extended sequential human behavior,
  • simulation of learning algorithms in naturalistic virtual environments,
  • developing learning algorithms with an emphasis on the learning of sensory representations for actions.

Our research group is part of the Centre for Cognitive Science and the Institute of Psychology in the Department of Human Sciences at Technical University of Darmstadt. We are also part of the Hessian Center for AI, hessian.ai, the Ellis Unit Darmstadt, and the Konrad Zuse School of Excellence in Artificial Intelligence, ELIZA.

We are grateful for funding from ERC, DFG, BMBF, and the Excellence Program of the Hessian Ministry of Higher Education, Science, Research and Art.

Members

Publications

Teaching

23 May – we presented a poster on multistep planning of eye movements in visual search at the conference on Curiosity, Creativity and Complexity at the Zuckerman Institute, Columbia University

19 May – Constantin gave a talk in the Symposium on Continuous Psychophysics at the Vision Sciences Annual Meeting in St. Pete Beach, USA

26 April – Congratulations to Susanne Trick, who has been selected as a recipient of the German AI-Newcomer Awards 2023 (#KI-Newcomer*innen) which is awarded every two years by the German Federal Ministry of Education and Research and the German Informatics Society

21 April 2023 – new empirical study out on people learning to adjust their visuomotor interception strategies: Huaiyong Zhao, Dominik Straub, Constantin A. Rothkopf: People learn a two-stage control for faster locomotor interception, Psychological Research

06 April 2023 – two papers have been accepted at the Cognitive Science conference: F. Kadner, H. Willkomm, I. Ibs, C. A. Rothkopf, Finding your way out: planning strategies in human maze-solving behavior and F. Tatai, D. Straub, C. A. Rothkopf, People use Newtonian physics in intuitive sensorimotor decisions under risk.

21 February 2023 – Nils Neupaertl's PhD thesis: Interacting with an uncertain physical world: probabilistic models of human perception and action was selected as best PhD thesis in the department of human sciences in 2022. Congratulations!

21 January 2023 – our abstract Fabian Kessler, Julia Frankenstein, Constantin Rothkopf: 'Optimal control under uncertainty predicts variability in human navigation behavior' has been accepted to COSYNE

05 January 2023 - Florian Kadner and Tobias Thomas are presenting our poster 'Improving saliency models' predictions of the next fixation with humans' intrinsic cost of gaze shifts' at the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

14 December – Constantin gave a talk at the Computational Cognitive Neuroscience seminar at UKE, Hamburg-Eppendorf

01 December – Constantin gave a talk in the SFB 1528 – Cognition of Interaction – Colloquium at the University of Goettingen

03 November – Constantin gave a talk in the Colloquium Cognitive Systems at the University of Ulm

07 October - eLife published a digest about our continuous psychophysics paper entitled: fixing misperceptions.

06 October- Constantin gave a talk at the retreat of the Affective and cognitive mechanisms of specific Internet-use disorders (ACSID) research group

03 October – Constantin gave a colloquium talk at the Bernese talks in Sport Science at the University of Bern, Switzerland

16 September - our abstract: Claire Ott, Inga Ibs, Constantin Rothkopf, Frank Jäkel: 'Leveraging human optimization strategies for automatic explanation generation' has been accepted at the workshop on human behavioral aspects of (X)AI, Birkbeck, University of London

15 September – new paper accepted at the 36th Conference on Neural Information Processing Systems (NeurIPS): Matthias Schultheis, Constantin A Rothkopf, Heinz Koeppl: Reinforcement Learning with Non-Exponential Discounting.

05 September - we are presenting a poster at the biannual conference of the German cognitive science society KogWis in Freiburg: Vojtěch Rada, Julia Frankenstein, David Sedláček, Zdeněk Míkovec, Constantin A. Rothkopf: Towards understanding spatial illusions in architecture – a pilot study exploring factors influencing illusive perspectives.

1 September – Fabian Tatai started as a PhD student- welcome!

15 June - our keynote + tutorial 'Inverse optimal control for continuous psychophysics' has been accepted to the Conference on Computational Cognitive Neuroscience!

24 March – new publication in Nature Machine Intelligence from our collaboration with Kristian Kersting's lab doing psychological experiments with large scale language models: Patrick Schramowski, Cigdem Turan, Nico Andersen, Constantin A. Rothkopf & Kristian Kersting: Large pre-trained language models contain human-like biases of what is right and wrong to do

17 March – we have been awarded an ERC Consolidator grant by the European Research Council for our proposal ''ACTOR – Towards a computational account of natural sequential behavior“! Here the university's press release.

23 February - Dominik Straub received one of two John I. Yellott Travel Awards for Vision Science, which are given annually to a graduate student or postdoctoral researcher attending the Vision Sciences Annual Meeting to present research 'that provides new quantitative insights into the human visual system' for the contribution: Dominik Straub, Constantin Rothkopf. An analysis method for continuous psychophysics based on Bayesian inverse optimal control. Congratulations!

08 February 2022 – we are participating in the Vision Sciences Society annual meeting with three talks and one poster: Florian Kadner, Tabea A Wilke, Thi DK Vo, David Hoppe, Constantin A Rothkopf: Trade-off between uncertainty reduction and reward collection reveals intrinsic cost of gaze switches; Dominik Straub, Constantin A. Rothkopf: An analysis method for continuous psychophysics based on Bayesian inverse optimal control; Tobias Thomas, David Hoppe, Constantin A. Rothkopf: Measuring the cost function of saccadic decisions reveals stable individual gaze preferences

28 January 2022 – we are participating in COSYNE: Dominik Straub, Matthias Schultheis, Constantin Rothkopf: Inferring implicit sensorimotor costs by inverse optimal control with signal dependent noise

21 January 2022 – new paper accepted at the 25th International Conference on Artificial Intelligence and Statistics (AISTATS): Susanne Trick, Constantin A Rothkopf: Bayesian Classifier Fusion with an Explicit Model of Correlation

12 January 2022 – Constantin gave a talk at the invited symposium on 'active learning' at the Budapest CEU Conference on Cognitive Development (BCCCD)

08 December 2021 – new paper out: N. Neupärtl, F. Tatai, C. A. Rothkopf: Naturalistic embodied interactions elicit intuitive physical behaviour in accordance with Newtonian physics, Cognitive Neuropsychology

30 November 2021 – Constantin gave a talk in the nip-colloquium in Tübingen: From ideal observers to bounded actors through inverse optimal control

01 November 2021- Philipp Hummel joined the lab as PhD student- welcome!

28 September 2021 – new paper accepted at the 35th Conference on Neural Information Processing Systems (NeurIPS): Matthias Schultheis, Dominik Straub, Constantin A Rothkopf: Inverse Optimal Control Adapted to the Noise Characteristics of the Human Sensorimotor System

24 August 2021 – Constantin gave a talk 'Open problems in modeling naturalistic sequential behavior' at the Special Symposium in honor of Michael F. Land at the European Conference on Visual Perception, ECVP

July – August 2021 – Inga participated in the Oxford Machine Learning Summer School (OxML 2021)

05 – 23 July 2021 – Fabian participated in NeuroMatch Academy: Computational Neuroscience

12 May 2021 – we are presenting AdaptiFont, our human-in-the-loop font that adapts its shape to the way readers interact with it at CHI, the Conference on Human Factors in Computing Systems. Watch the talk on youtube.

01 February 2021 - we have been successful in two cluster grant applications. The collaborative cluster ”The adaptive Mind“ with the universities of Giessen and Marburg investigates the principles and computational basis for the balance between stability and change in human adaptability.
The second cluster, ”The third wave of AI", in collaboration with AI center hessian.ai, the Goethe University, Frankfurt and the University Mainz, focuses on the development of a new generation of AI systems that are more human-like.

04 December 2020 – Our collaborative LOEWE Research Cluster WhiteBox has been approved for funding by the Hessian State Ministry of Higher Education, Research and the Arts. WhiteBox is joining the twin disciplines AI and Cognitive Science. Starting 2021, an interdisciplinary team will develop whitebox methods to foster human understanding of natural and artificial intelligence and its results.
29 September 2020 – Constantin gave a talk 'Reverse engineering behavioral costs and benefits with inverse reinforcement learning' in the Workshop 'Inferring and testing optimality in perception and neurons' at the Bernstein Conference
15 September 2020 – Constantin is part of the ELLIS (European Laboratory for Learning and Intelligent Systems) faculty of the newly launched ELLIS Unit at TU Darmstadt
31 August 2020- we are part of the founding members of the Hessian Center for AI
06 July 2020 - new article in the popular science magazine 'Spektrum der Wissenschaft' together with Albert Newen, Nele Russwinkel and Martin V. Butz
24 June 2020 – Constantin gave a talk as part of smartstart's 'midsummerbrains- computational neuroscience from my point of view'
15 June 2020 – Niteesh Midlagajni joined the lab as PhD student- welcome!
02 June 2020- Inga Ibs joined the lab as PhD student- welcome!
15 May 2020 – invited keynote at the virtual Symposium on Explainable Machine Learning for Biomedical Data Analysis and Care Robotics, Universität Oldenburg, Germany
01 April 2020 – the German Federal Ministry of Education and Research (BMBF) has approved our collaborative grant PlexPlain with Kristian Kersting, Frank Jaekel, and Florian Steinke
16 March 2020 - the lab is closed due to the Covid-19 pandemic
15 February 2020 – Fabian Kessler has joined the lab as PhD student- welcome!
10-14 February 2020 – Constantin is teaching at the Minds and Machines Workshop in Cognitive Science, Artificial Intelligence and Education at the University of Luxembourg together with Pedro Cardoso-Leite, Paul Schrater, and Christophe Schommer

Older News