This project can be adjusted either for project 1 (internship) or for project 2 (thesis) with an extended version.

How do humans represent action videos? What are the dimensions underlying the behavioral video similarity? For example, humans could judge videos as similar based on the speed of motion, type of motion, or the semantics of the action and scene. To research this topic, we are interested in collecting behavioral data (similarity judgements). Ideally this will be done in an online experiment, to acquire data from a larger number of participants, and a top candidate for the experimental setup is the Odd-One-Out (OOO) similarity.
A direct application of interest for these data is to conduct an analysis similar to Hebart et al., 2020 to uncover the dimensions of the mental representations.

You will be working with Christina Sartzetaki (PhD student) in the lab of Dr. Iris Groen at the IvI (with collaborations at the Psychology department). There will be weekly supervision meetings.

The project including only data collection is more suitable for the shorter project 1 (internship), while the project including both the data collection and the dimensional analysis can be a good fit for the longer project 2 (thesis).

This project might be for you if:
(1) You are interested in learning how to design, conduct, monitor, and collect data from an online behavioral experiment
(2) You are interested in learning how to analyze behavioral data like in Hebart et al., 2020
(3) You are interested in the way humans represent videos

Previous experience with online experiments is beneficial.
Previous experience with Python programming skills is beneficial for the shorter project version (data collection), and essential for the longer project version (data collection + analysis).

Contact c.sartzetaki@uva.nl to apply.

References:
Hebart, Martin N., et al. “Revealing the multidimensional mental representations of natural objects underlying human similarity judgements.” Nature human behaviour 4.11 (2020): 1173-1185.