لخّصلي

and zoom in and out. With the baseline 2D system, users can
toggle the display of time labels, pan, and zoom in and out of a
portion of the map.
It is important to note that the purpose of the baseline 2D
system is to provide the space time cube representation with a
reasonable baseline. That is, the space time cube should preferably
beat the baseline in at least some aspect to merit further research
by the information visualization community. The purpose of the
2D baseline system is not to investigate how 2D visualization can
be made more effective. This in itself is an interesting research
question but out of the scope of this paper.
3.1 Space Time Cube System
To perform our investigation, we developed a space time cube
system capable of rendering walking data traces inside a cube (see
Fig. 2). The system has a “measurement” plane that can be moved
up and down along the height axis to make it easier to read when a
particular event occurred. The exact time of the measurement
plane’s current position is displayed to the right of the space time
cube’s display area (not shown in Fig. 2).
The space time cube system is controlled via either the
keyboard or a graphical user interface (GUI). Using the GUI or a
keyboard, the user can rotate, zoom, and move the measurement
plane up or down.
3.2 Baseline 2D System
The baseline 2D system displays walking data traces using
different colors (green, blue, yellow, red; see also Fig. 3). The
colors were the same as in the space time cube system.
The colored line traces indicate different persons, and the labels
indicate the start and end times for a person at a specific point in
the map. Users can toggle the display of labels and zoom in and
out with the keyboard.
4 METHOD
We used a between-subjects experimental design where participants were exposed to one of two conditions: either the space time
cube system or the baseline 2D system.
Often within-subjects experimental design is preferable since

1. variation between conditions is controlled within the participant; and 2) generally, a smaller number of participants are
   required. However, in this experiment, it is plausible that
   participants become increasingly familiar with the material and
   task during the experiment. With a within-subjects design, there is
   a risk that one condition (call it condition A) better aids
   participants in understanding the material and the task than the
   other condition (call it condition B). This asymmetrical skilltransfer effect would in fact penalize the performance of condition
   A when preceded by condition B and unfairly benefit condition B
   when preceded by condition A. To avoid this confound, a betweensubjects design was used, and the number of participants in the
   experiment was increased accordingly ðn ¼ 30Þ.
   4.1 Participants
   Thirty participants, 15 male and 15 female, were recruited from the
   university campus. The participants were screened for color
   blindness. None had any previous experience in using information
   visualization tools. The two groups were gender balanced to the
   extent possible.
   4.2 Apparatus
   The experiment was conducted on two laptops with 1500 screens
   and 32-bit color depth. Although the physical dimensions of the
   laptop screens were identical, the screen resolution varied slightly
   in the vertical dimension. The first laptop had a screen resolution
   of 1,280 1,024 while the second laptop had a screen resolution of
   1,280 800. The laptops were balanced between the conditions