Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Cognitive computational models of information sampling under threat
Information sampling is a requirement in many threat-related scenarios. This project deals with optimal information sampling in an ethological computer game, and approximative shortcuts that biological agents may use.
Keywords: Computational models, cognitive models, general psychology, computational neuroscience
**Background**
Biological agents are often thought to use specific cognitive strategies when under acute threat. Our lab's interest is in a computational understanding of such strategies. Information sampling is a requirement in many threat-related scenarios. While humans often sample information in a statistically optimal fashion, they may have to resort to crude shortcuts when under threat to survival. To emulate this situation, we have developed a simple computer game that allows measuring people's information sampling behaviour. A normative model describes how people should behave in this game.
**Your project**
You will analyse an existing data and compare it to the model's predictions. You will then implement and test a range of model variants that account for simplifications and model-free heuristics. Depending on the results, you will revise the initial model and modify the computer game to best capture cognitive strategies that people use. The project can be adapted to a 12 month UZH or 6 month ETH MSc thesis.
**Your profile**
You have a background in computer science, physics, engineering, neuroinformatics, or related fields, and a strong interest in statistical modelling and cognitive systems. Mathematical formalism has a meaning for you. You are experienced with standard software for data analysis (e.g. in Python, R, or MATLAB).
**Background** Biological agents are often thought to use specific cognitive strategies when under acute threat. Our lab's interest is in a computational understanding of such strategies. Information sampling is a requirement in many threat-related scenarios. While humans often sample information in a statistically optimal fashion, they may have to resort to crude shortcuts when under threat to survival. To emulate this situation, we have developed a simple computer game that allows measuring people's information sampling behaviour. A normative model describes how people should behave in this game.
**Your project** You will analyse an existing data and compare it to the model's predictions. You will then implement and test a range of model variants that account for simplifications and model-free heuristics. Depending on the results, you will revise the initial model and modify the computer game to best capture cognitive strategies that people use. The project can be adapted to a 12 month UZH or 6 month ETH MSc thesis.
**Your profile** You have a background in computer science, physics, engineering, neuroinformatics, or related fields, and a strong interest in statistical modelling and cognitive systems. Mathematical formalism has a meaning for you. You are experienced with standard software for data analysis (e.g. in Python, R, or MATLAB).
Analyse existing data with computational cognitive models, develop more sophisticated models, possibly implement new experiments.
Analyse existing data with computational cognitive models, develop more sophisticated models, possibly implement new experiments.
Supervision:
Prof. Dr. Dominik R. Bach
dominik.bach@uzh.ch
www.bachlab.org
Supervision: Prof. Dr. Dominik R. Bach dominik.bach@uzh.ch www.bachlab.org