Information Technology Reference
In-Depth Information
- learning phase 2: the participants then have five minutes to learn the rules of
the earcons by using the sonified menu. This stage is done without the help of the
experimenter;
- test phase: 12 earcons are presented successively; the participants must
associate each earcon with the node to which it corresponds in the hierarchy. Two
new earcons are also presented that correspond to two missing nodes on level 4. The
recall rates are thus calculated.
The results of the experiment show very high recall rates, with an average rate of
81.5%. The strongest rates are obtained for the highest hierarchy levels, and the
three weakest (between 50 and 60%) are obtained for level 4 (the lowest), but no
satisfactory explanation is given. Results also show a very high recognition level for
the two new earcons (83% and 100%).
In a second experiment, Brewster tests the influence of the learning mode on the
recall rate of earcons. Four different types of learning are then tested: individual or
group, with or without sounds in the first and second learning phases. Two test
sessions are put in place: the first just after the learning part and the second a week
later. The result does not show an effect of time between the two sessions and the
“active” learning is the most efficient, such as when subjects use the HMI
themselves.
8.5.4.2. Measure of the number of times keys are pressed
The number of times keys are pressed here corresponds to the number of times
the user presses on a key of the keyboard (e.g. the keyboard of a cell phone) to reach
the item of the menu that he is looking for. This allows us to know whether the user
is carrying out an optimal path between the starting and the arrival points in the
menu.
Leplâtre [LEP 02] proposes a method for evaluating the efficiency of the
sonification of a cell phone menu based on this type of measure. Two participant
groups are compared: Group 1 with sonification and Group 2 without sonification
(i.e. only the key-press beep are present).
The evaluation method is divided into two phases:
- for both groups, the subjects must carry out 56 tasks: actions to carry out in
menus, or elements of the menu to be found. In Group 1 only, the participants carry
out two additional “blind” tests, i.e. the cell phone screen is off and the participants
can only see the keys of the telephone.
Search WWH ::




Custom Search