idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
Can virtual agents strengthen the trust of people with a migration background in the police? A research team from the University of Würzburg has investigated this. The results surprised even those responsible.
Intelligent virtual agents can help to strengthen the trust of people with a migration background in institutions such as the police. This is the key finding of a new study by scientists at Julius-Maximilians-Universität Würzburg (JMU). Computer science professor Birgit Lugrin was responsible for this study. She holds the Chair of Computer Science V at JMU; so-called Socially Interactive Agents are one of her main areas of research.
The background to this study is the fact that around 24 million people with a migration background currently live in Germany. It's good if they also trust the local legal and government system. Positive contact with the authorities – and therefore also with police officers – can play an important role in this. If a person with Turkish roots, for example, also meets a police officer of the same origin, this could play a particularly important role in strengthening trust in the authorities, according to the Würzburg research team.
Virtual agents are a cost-effective reinforcement
However, the proportion of employees with a migration background in the German police force is comparatively low. At the same time, the police often lack sufficient staff to carry out more public relations work in addition to their actual duties. The use of “intelligent virtual agents” could solve this problem. For those who can't imagine what a virtual agent is: These basically resemble characters in a computer game that move through virtual worlds and are behaving in a human-like manner.
“Intelligent virtual agents that are designed to interact with people in a natural and intuitive way are relatively inexpensive to develop and can potentially be used on a large scale,” says Lugrin, describing the advantages of these virtual helpers. They could therefore be a welcome addition to the police's public relations work. Another advantage is that the agents can also be designed to look and speak very similarly to people with a migration background.
An interactive exchange in the digital world
Whether virtual agents with a migration background are actually able to strengthen the trust of real people with the same background in the police: Lugrin and her team investigated this as part of an experimental study. They recently presented the results at the International Conference on Intelligent Virtual Agents of the Association for Computing Machinery.
“We developed an interactive scenario in which our participants interacted with a police officer on the computer. In one case, the virtual agents were recognizable as typically German; in the other, their appearance and language distinguished them as people with a mixed cultural background,” says Lugrin. The test subjects themselves were all people living in Germany with Turkish roots.
How empathetic is the digital counterpart?
Before and after the online meetings, the researchers used questionnaires to record important parameters of their participants' attitudes towards the police. These included aspects such as trust in the police in general and the effectiveness of their work. Their assessment of the extent to which police officers act fairly and equally when dealing with different population groups was also recorded.
With regard to the virtual agent, the team determined how the study participants perceived him or her. This involved aspects such as similarity, warmth, competence and empathy – and, of course, trust in the digital counterpart.
Trust is growing – in every case
The analysis of all of this data shows that trust in the police could indeed be increased by interacting with a virtual police officer. However, the team was unable to confirm one of its main hypotheses: “We had expected that group similarity, i.e. an identical cultural background, would have a stronger positive effect on trust,” explains Birgit Lugrin. In the experiments, however, trust grew regardless of whether the study participants encountered a “typical German” agent or one with a migrant background.
However, according to Lugrin, this does not invalidate this thesis. “In psychology, numerous research studies have shown that people perceive third parties more positively if they resemble them,” explains the scientist. Many factors can influence what constitutes similarity, such as religion, political views or cultural background.
Accordingly, Lugrin is convinced that further studies are needed to investigate this aspect in more detail. Overall, however, the key finding for her remains: “Virtual agents are an effective tool for positive personal interaction with authorities.”
Prof. Dr. Birgit Lugrin, Chair of Computer Science V (Socially Interactive Agents), +49 931 31-84602, birgit.lugrin@uni-wuerzburg.de
Enhancing Trust towards the Police through Interaction with Virtual Agents - Investigating the Ingroup Effect with Mixed-Cultural Individuals. Birgit Lugrin, Elisabeth Granal, Maximilian Baumann, Anastasia Fiolka, and Tobias Haase. ACM ISBN 979-8-4007-0625-7/24/09, https://doi.org/10.1145/3652988.3673944
"Can I help you?” In the study, participants were asked to interact with virtual police officers
Informatik V
University of Wuerzburg
Merkmale dieser Pressemitteilung:
Journalisten, Wissenschaftler
Gesellschaft, Informationstechnik, Medien- und Kommunikationswissenschaften
überregional
Forschungsergebnisse
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
