TEDS Data Dictionary

Description

Example images from Perspective missions are shown below. A written description of a mission now follows.

There were 5 Perspective tests ('missions') in the Navigation battery. Each mission had the same rules and followed the same pattern of activity. Each mission comprised a single task. Twins were allowed up to four attempts to succeed in responding correctly in each mission.

Each mission started with a stimulus image depicting a target object within the virtual world. The stimulus was shown only before the first attempt, not before the second, third or fourth attempts, The stimulus was followed by the response in which the twin had to find the target object and click on it, and in which up to four attempts were allowed. While the response depicted the virtual world in the usual way, with a viewpoint at street level, the stimulus depicted the scene in a more unusual way: the viewpoint was from above looking down, and in most missions the colours were distorted giving a more abstract view. In the stimulus image, the target object was highlighted by making it flash (alternating between two monochrome colours). In the response view, the target object had its normal appearance and the twin had to find it by changing the direction of view with mouse movements.

During the stimulus part of the mission, which stayed on screen for a fixed length of time, a twin could not make any input and hence could not change the viewpoint or direction of view or click on any object. During the response part of the mission, as usual, the view could be rotated left or right through 360 degrees (by moving the mouse from side to side) and could be moved up and down (by moving the mouse forwards or backwards), and the mouse could be clicked in an attempt to locate the target. However, the location of the respondent's viewpoint could not be moved (forward or back or from side to side at street level).

Each attempt consisted of a mouse click on a location indicated by a red dot in the centre of the screen. Up to four clicks (attempts) were allowed within the time limit of the mission. If an attempt was correct, such that the red dot was correctly positioned over the image of the target object when clicked, then the mission ended and any remaining attempts were not used. If an attempt was incorrect, then either the next attempt or the next mission would follow.

Each Perspective mission had a time limit of 60 seconds for completion. Within this 60 seconds, there was no time limit on individual attempts in the mission. If the time limit expired before all the attempts had been made, then all remaining attempts were discontinued, the mission would come to an end, and the next mission in the battery would begin. There was no progress bar or timer on screen to help twins keep track of time; however, "hurry up" prompts appeared on screen as the time limit approached.

Feedback was provided after each attempt in order to encourage a twin to keep trying to complete the mission. There were normally only two possible outcomes in each attempt: either it was completed successfully (click on target) or it was failed (click off target). In either case, feedback was provided using text on screen. Success in the attempt, and hence in the mission, was shown by a message such as "Yes!" followed by some contextual elaboration. After a failed attempt, the text was highlighted with an orange background, and included a reminder to try again, e.g. "No, we don't have a match here. Try again!" As mentioned above, an approaching timeout was signalled by "hurry up" messages flashing on the screen; a failure due to timeout was implied by the mission ending without success, and the next mission beginning.

Input in this type of mission consisted simply of mouse movements and mouse clicks. As explained above, mouse movements were used to change the direction of view, and to try to place the red dot over the position of the target object. A mouse click was used to complete each attempt to confirm the position. If a twin did not make any mouse clicks during the 60 second time limit of the mission, then the mission would time out without any meaningful response data being recorded. However, an absence of such input did not necessarily mean that the mission had been left unattended - it was possible for a twin to spend the entire 60 seconds moving the mouse around, looking for the target but not finding it, and not making any clicks. For this reason, there was no "auto-pause" feature in Perspective missions; if the entire mission passed without mouse clicks, then the next mission would simply begin.

In Perspective missions, neither the compass nor the map appeared on screen.

As in all other Navigation missions, twins could pause at any stage by pressing P on the keyboard; this would stop the clock and blank out the image of the streets. A further click on P would then resume the mission.

Images of the 5 Perspective missions, depicting both the initial stimulus image and the response image with the position of the target, are shown in the Table of Test Items below. Example images from missions are shown under the next heading.

Example Images

The images below are screen shots taken from Perspective missions. These images serve as examples of how missions were attempted, with or without success.

The first image, below, is an example of the stimulus shown at the start of each Perspective mission. As in all Perspective missions, the instruction text gives a reminder of the aim: '...flashing object. Find it quickly!'. The image shows a sort of bird's eye perspective of a scene from the virtual world and, as in most missions, the colours have been distorted to give a more abstract view. The target object flashed by alternating between a dark colour and a light colour; in this example, the target is one of the benches, to the right and slightly above the position of the central red dot. This stimulus image, with its text and its flashing target object, remained on screen for a fixed period of time, during which a twin was unable to make input of any kind.

The second image, below, was from the immediate start of the response part of a Perspective mission. After the bird's eye view of the stimulus, the view returned to the usual street-level perspective in the virtual world. The text on screen provided a reminder of the instruction for the response - this text remained on screen throughout the first attempt, but disappeared during subsequent attempts. Note that in this initial view, the target object is not visible - it was necessary to rotate the viewpoint in order to find it. Note also that neither compass nor map were needed in Perspective missions, so they did not appear on screen.

The third image, below, was from the first attempt of a Perspective mission (hence the instruction text is still on screen). This image illustrates the location of the target object (a green bin), with the red dot positioned over it in the middle of the screen ready for a successful click.

The next image (below) shows the same view immediately after success in the attempt - the feedback message ("Great!") indicates that the mission has been completed.

The next image, below, shows what happens after an unsuccessful attempt. A mouse-click has been made somewhere other than over the target object. The feedback message, highlighted with an orange background, tells the twin that the attempt was unsuccessful and encourages the twin to try again with another attempt.

The next image (below) illustrates the "hurry up" prompt that flashed repeatedly on the screen as the 60-second mission time limit approached. If a mission timed out before it could be completed successfully, then the next mission would start without further feedback. Note that the instruction text no longer appears at the bottom of the screen, as this image was taken from the 2nd or 3rd attempt in the mission.

The final image (below) illustrates the "last chance" prompt that flashed repeatedly on the screen during the 4th and final attempt of a mission. This image was taken just after a failed third attempt, hence the feedback text at the bottom of the screen - this would disappear a few seconds later.

Web Test Rules

All five Perspective missions followed these rules:

• Tasks
  Each mission comprised a single compulsory task.
• Attempts
  Up to 4 attempts were allowed to complete each mission (and hence each task). Each mission ended either on successful completion, or after 4 failed attempts, or after a mission timeout.
• Timeout rule
  A time limit of 60 seconds applied to each mission. As the time limit approached, text warnings would appear on screen ('hurry up') but there was no other indication of the passing of time. When the mission timed out, it would simply end and the next mission would start.
• Widgets
  Neither the compass nor the map appeared on screen for any tasks.
• Pausing
  The mission could be paused at any time by pressing P on the keyboard. The image of the virtual world was then replaced on screen by text explaining that the mission was paused, and that pressing P again would resume. Each mission could be paused indefinitely in this way, as often as a twin wished.
• No auto-pausing
  In the event of a mission passing without any input mouse-clicks, either because it had been left unattended or because the twin had not made an attempt, then it would time out. However, there was no auto-pause in this mission type, so the following mission would then automatically start.
• Attempt outcomes
  The possible outcomes for each attempt of a mission were as follows:
  • Success: clicked on the target object
  • Failure: clicked elsewhere
  • Attempt 2/3/4 skipped after success in an earlier attempt
  • Attempt discontinued after mission timeout (could occur in during attempt 1 with no input clicks or during attempt 2/3/4 after one or more input clicks)
  These outcomes, along with times, were recorded in the raw data and could be converted into task and mission scores.

Dataset Item Variables

The raw item variables relating to each Perspective mission, as listed in the table below, were generated automatically by programs on the web server. For the sake of completeness, the table also lists derived variables relating directly to a Perspective mission or its component attempts. Note that the dataset only includes data for completed missions. If a twin started a mission but left it unfinished, then the data for that mission were deleted from the dataset.

Note that the raw outcome variables for each attempt (rcnpsXaZcm1/2, rcnpsXaZer1/2) have been dropped from the dataset in favour of a single more informative attempt outcome variable (rcnpsXaZr1/2). Note also that reaction times (rcnpsXaZrt1/2) have been dropped from the dataset because they are identical to the respective completion times (rcnpsXaZct1/2) for Perspective attempts. The start and end dates/times have not been retained in the dataset.

In all variable names, X denotes the mission number (1 to 5) and Z denotes the attempt number (1 to 4).

Table of Test Items

Each Perspective mission is illustrated below using two images followed by the feedback text that was given on successful completion.

For each mission, the first image illustrates the initial stimulus, with a viewpoint from above ground level, the instruction text, and in some cases modified image colours (monochrome or negative). In real time during the game, the target object flashed by alternating between dark and light colours; this flashing cannot be depicted in these static images, so an arrow has been added to indicate the position of the target object.

The second image for each mission illustrates the response, with a viewpoint at ground level and normal image colours; the second image also shows the correct position of the target object, located by the red dot in the middle.

Perspective mission 1

Stimulus image (with arrow added to show target object):

Response image:

Feedback text on success: "Great! It contained a dangerous canister. A recovery squad will take care of it."

Perspective mission 2

Stimulus image (with arrow added to show target object):

Response image:

Feedback text on success: "Right! This light bulb was being used by a rival agency to communicate through Morse code."

Perspective mission 3

Stimulus image (with arrow added to show target object):

Response image. (Note this is from the second attempt, hence the instruction is not shown. The instruction for the first attempt was "Look around and zoom on the object that was flashing.")

Feedback text on success: "Yes! A rival agent has put a bug under that bench. Our recovery squad will get rid of it."

Perspective mission 4

Stimulus image (with arrow added to show target object):

Response image:

Feedback text on success: "Well done! The flag contained a coded message. We have decrypted it and got the instructions for your next mission."

Perspective mission 5

Stimulus image (with arrow added to show target object):

Response image. (Note this is from the second attempt, hence the instruction is not shown. The instruction for the first attempt was "Look around and zoom on the object that was flashing.")

Feedback text on success: "Yes! Ok, we have located the object, but found nothing dangerous. Let's move on!"