![](bg8.png)
generic tasks groups, which can be further assessed
based on the research question in scope. The appli-
cability for the driver take-over task has been exem-
plary demonstrated by having subdivided the generic
task groups further. The presented method can be tai-
lored or extended to meet the specific requirements of
the respective research questions, thus making it suit-
able for a technologically independent usage in vari-
ous test setups.
6.2 Discussion
When following the presented methodology, re-
searchers should consciously and attentively docu-
ment instructions about operations provided to the test
persons to avoid unintended, implicit side effects on
the study itself. This is especially relevant for test
environments making use of abstracted vehicles, like
driving simulators of various types, since their tech-
nical implementation goes by with an explicit or im-
plicit tailoring of stimulus material provided to the
test person.
To maintain comparability regarding the between-
subjects factors within the own study as well as render
the study comparable with others, researchers should
minimize the potential for subjective interpretation of
their given instructions by the test persons. Therefore,
a structural analysis of the task under examination is
required, which can be performed by following the
framework presented in this paper.
In general, to achieve comparability within the
own study, the test execution should be formalized by
using a reproducible way of providing the instructions
about operations to each test person. In order to en-
able comparability with other studies, as well as gen-
erate transparency over potential unintended side ef-
fects, it is recommended to devote special attention to
the detailed documentation of the instructions about
operations.
In addition to that, the validity and robustness of
the obtained results can be increased by formalizing
the interaction with the test persons, e.g. by utilizing
written instructions and reduce communication to an
absolute necessity. The difficulty with this is to find
a balance between the formalization of the commu-
nication without making the test persons feeling un-
comfortable or treated impolitely, which can have an
impact on the results of the examination, respectively.
It is recommended to investigate if a gas pedal in-
teraction causing the ego vehicle to accelerate could
be a valid way to solve the driver take-over task in
the respective scenario. Providing access to this ad-
ditional stimulus material can lead to a better under-
standing of driver take-over maneuvers in different
scenarios. Furthermore, inconvenience with the au-
tomation system, i.e. by subjectively driving too slow,
can be qualitatively measured and assessed as a bene-
ficial side effect.
6.3 Outlook
In a subsequent analysis, the test method and -design
of studies investigating the driver take-over task has to
be examined further to gain a holistic understanding
of the research results obtained in this field. There-
fore, the introduced methodology should be imple-
mented in existing and upcoming studies to verify its
suitability. Aspects worthwhile to investigate could
be meta-goals and objectives the test persons shall
achieve as well as the correlation between typical
research questions and technical implementations of
test setups the respective studies are conducted in.
Additionally, the applicability of the technologically
independent approach presented could be verified by
conducting a study based on the methodologies intro-
duced in this work.
Furthermore, deviating performances of test per-
sons within the presented task groups could be
mapped to established models of human error func-
tions. Understanding the origin of the individual be-
havior of each test person contributes to the success-
ful design of future human-machine interfaces for the
driver take-over task.
REFERENCES
Banks, V. A., Eriksson, A., O’Donoghue, J., and Stanton,
N. A. (2018). Is partially automated driving a bad
idea? Observations from an on-road study. Applied
Ergonomics, 68:138–145.
Bubb, H., Bengler, K., Gr
¨
unen, R. E., and Vollrath, M.
(2015). Automobilergonomie. Springer Fachmedien
Wiesbaden, Wiesbaden.
Bundesanstalt f
¨
ur Straßenwesen (2012). Einflussfaktoren
auf das Fahrverhalten und das Unfallrisiko junger
Fahrerinnen und Fahrer. Berichte der Bundesanstalt
f
¨
ur Straßenwesen, Heft M 229(Mensch und Sicher-
heit).
Capallera, M., de Salis, E., Meteier, Q., Angelini, L., Car-
rino, S., Khaled, O. A., and Mugellini, E. (2019). Sec-
ondary task and situation awareness, a mobile applica-
tion for conditionally automated vehicles. In Proceed-
ings of the 11th International Conference on Auto-
motive User Interfaces and Interactive Vehicular Ap-
plications: Adjunct Proceedings, pages 86–92, New
York, NY, USA. ACM.
Diaper, D. and Stanton, N. (2003). The Handbook of Task
Analysis for Human-Computer Interaction. Lawrence
Erlbaum Associates, Mahwah, New Jersey, 1 edition.
Influences of Instructions about Operations on the Evaluation of the Driver Take-Over Task in Cockpits of Highly-Automated Vehicles
79