Mobile interaction describes the interaction of human users with mobile devices outside of stationary settings like a desktop workstation. Since the introduction of modern smartphones a decade ago, the interaction with mobile devices has changed: users more and more communicate while being on the go. Hence, revisiting the fundamentals of technology-supported communication becomes necessary. When interacting with a device in the field, errors increasingly happen, since the user has to monitor their contextual surroundings, which binds mental resources. This thesis explores the interplay between the user’s mental state (which results from limited mental resources), the contexts in the field, and mobile interaction. A particular focus is put on the occurrence of errors, which are examined on three levels: knowledge-based mistakes, rule-based mistakes, and skill-based slips [Rea90]. These errors can disrupt mobile communication. To explore the interplay as mentioned above, and the occurrence of errors, the prevalent method of this thesis are field studies. Since existing frameworks do not suffice to support mobile field studies as required by this thesis, the Open Source CoConUT field study framework is developed. Overall, in this thesis, two field studies and two laboratory studies are conducted. The laboratory studies address aspects that cannot be tested in the field. Knowledge-based mistakes during mobile communication are assessed in a laboratory study since an extensive qualitative analysis is required. The occurrence of rule-based mistakes under stress is also tested in the laboratory since the probands had to remain seated. Finally, two consecutive field studies investigate the interplay of context, internal state, and interaction in commute-like situations. A second study places its focus on skill-based slips. Findings reveal that at present users do not stop for typing on their smartphones any more during walking outdoors, despite physical activity and increased stress leading to a higher typing error slips. Stress also has a negative influence on the occurrence of rule-based errors. Additionally, erroneous mental models can lead to disrupted mobile communication due to incomplete or false knowledge. Overall, the CoConUT framework proves to be reliable to support mobile field studies, and an overview of the gathered sensor data is given. Consumer wearables like chest belts are considered to be a robust and affordable solution for measuring the user’s arousal as an indicator of their internal state. To conclude, the thesis provides essential contributions to understand mobile interaction in the field after the widespread adoption of mobile devices and puts a particular focus on the occurrence of different types of errors.

Top