direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Page Content




We are proud to announce that our BMBF (Federal Ministry of Education and Research) funded project AVALANCHE (Adaptive, Virtual Assistant for AvaLAnche WarNings for CHaracter PropErties) will start spring 2018.

The project AVALANCHE is part of the Software Campus program, a leadership development program for outstanding PhD students of computer science.

In this context, Sabine Prezenski will lead the project AVALANCHE, supported by our project partner Scheer Holding.

The goal is to investigate the dynamic decision-making processes of freeriders in field studies and in VR simulation in order to avoid dangerous decisions by applying persuasive strategies.

We are looking for interested students from the fields of computer science, psychology and human factors.



The aim of the project is to develop and test an interaction concept for wearable companion systems for the usage in high-risk situations.

The intention of the project is to support users in their subjective decision-making process by issuing warning and pointing out alternative actions. The companion system should consider both the situational influences on and the personality traits of its users; these should be prepared in a way so that potential dangerous decisions are stopped and behavioral changes induced. Techniques of persuasive communication should be applied to accomplish this.

Backcountry skiing and snowboarding is the use case for the interaction concept. Thus, warnings for users entering high-risk avalanche terrain need to be developed. These warnings are adapted to the personality profile of the users. The interaction concept is usable in different devices such as smartphones and smartwatches, as well as in augmented reality devices, e.g. smart ski goggles.

Technically speaking, avalanches warning apps are highly functional already and they help backcountry skiers assess the avalanche risk adequately. However, most avalanche accidents occur despite the correct assessment of avalanche danger, because wrong decisions are made. Both well-studied decision-making mistakes (overconfidence bias, e.g. nothing will happen to me, anchor heuristics e.g., last time nothing happened to me), as well as personality factors (e.g. level of risk taking) are important. To a large extent, avalanche accidents occur due to errors in the decision-making process of backcountry skiers.



Decisions in real environments are influenced by the expertise, personality and state of the decision-maker and the complexity of the situation; they are also often made under time pressure. Heuristics, biases and distortions lead to shortened information processing and thus the risk of making a potentially dangerous decision.

This project will examine whether personalized, persuasive warnings can positively influence the subjective dynamic decision-making process and the decision-making strategy in naturalistic environments.

Measurement of personality traits will be achieved by using data mining from e.g. social network data. Based on these characteristics, convincing communication will be designed.



The project will run until 31st of July 2019. The project is led by Sabine Prezenski and is supported by Kenneth zur Kammer, Martin Krabbe and Meret Vollmann.



This Project is supported by the Software Campus program, a leadership development program for outstanding PhD students of computer science. It is funded by the BMBF (01IS17052)


Martin Krabbe and Kenneth zur Kammer gave a workshop on our Project at the Human Factors Summer School in Berlin on October 12th.

Sabine Prezenski gave a presentation at ISSW in Innsbruck on October 9th. 


First Place of the Research to Marketchallenge. We won the first place of the Research to Marketchallenge in the categorie digital with the project Avalanche!


Prezenski, S.,Russwinkel, N. and Win, K.T. (Oktober 2018), in Proceedings of the International Snow Science Workshop, Innsbruck, Austria. (pp. 835-837).



Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions