Geography Reference
In-Depth Information
Table 1 (continued)
ΔΕ
L
a
b
R
G
В
00
6
8
10
8
6
94.80
30.00
30.00
201
255
179
85.22
30.00
30.00
174
227
153
73.72
30.00
30.00
143
195
123
61.18
30.00
30.00
110
161
92
52.49
30.00
30.00
87
138
71
46.48
30.00
30.00
72
122
57
4
6
8
10
12
94.80
30.00
30.00
201
255
179
88.32
30.00
30.00
183
236
162
79.31
30.00
30.00
158
211
138
68.49
30.00
30.00
129
180
110
56.71
30.00
30.00
98
149
81
44.69
30.00
30.00
68
118
53
by the test of colour vision with pseudoisochromatic plates. All participants
succeeded in this test.
Respondents were asked to find an area in the map depicted with a dot, match it
with one of the six classes of the legend and mark their answer via questionnaire
which was projected after the experimental stimulus.
Time to answer the question was not limited. The order of stimuli was random-
ized for each single participant in order to avoid the effect of the eye adaptation to
handle such tasks. It was assumed that at the beginning of the experiment partic-
ipants might have more problems to recognize colour differences, while at the end
of the experiment they might show better results (Holmqvist et al.
2011
).
In the last part of the experiment the self-evaluation of users
strategy was
carried out. Volunteers were asked to describe their strategy during solving the
given task and visual searching for the answer.
'
Results
The experiment was analysed in two phases. In the first step only colour schemes
with equal colour distance step were examined. Based on preliminary results of
11 participants, which indicated strong influence of order of classes within the
colour scheme on their correct value identification, three more colour schemes were
added to the experiment and tested. New colour schemes setting were optimized in
the sense of making visual steps between classes unequal. Specification of these
schemes (Table
1b
) was done deterministically. It was believed, that colour
schemes with growing colour distance between classes in the middle (e.g. B, C,
D and E) will cause more efficient solving a spatial task with use of choropleth map.
Results are based on analysing the accuracy of answers and eye-movements
regarding specified areas of interest.
