Information Technology Reference
In-Depth Information
- two independent variables (categorized):
- the sonification condition: three conditions (C1 to C3, specified previously),
- the experimental task: six tasks (T1 to T6, specified previously);
- three dependent variables (measured):
- the duration of the realization of tasks,
- the perceived driving comfort,
- the perceived frequency of the gaze diversion.
Variance analyses (95% confidence interval) on independent variables allowed
us to define which have a significant influence on data gathered. These analyses
correspond to an experimental plan with repeated measures, of which the inter-
subject variables identified are the gender and age class. Moreover, Fisher tests
enabled us to compare different variable modalities (e.g. hierarchical library versus
vocal library versus
minimal library for the variable sonification condition).
Furthermore, it is worth noting that the analysis of data concerns the latest of the
three conditions tested by participants (the order of conditions being evenly
distributed to all subjects) in order to overcome a presentation order effect and
concentrate on the situation where the subject is familiarized with the system as well
as is possible.
The observations thus made for the three types of measures, enable us to
respectively come to the following conclusions:
- An almost significant effect of sonification on the time to complete tasks is
observed [p=0.047]. The Fisher test shows that the only difference is in the vocal
library (LIB3), which has a longer execution time than the minimal library (LIBm)
[p=0.02]. However, the hierarchical library (LIB1/LIB2) has similar times to vocal
and minimal libraries [p=0.9 and p=0.08, respectively] (see Figure 8.12, for the
graphic representation of data).
- A significant influence of the sonification condition on the perceived level of
comfort [p<0.0001]. Participants judge their driving to be more comfortable with the
vocal library (LIB3) than when auditory guiding is deactivated (LIBm) [p<0.0001].
