Eye and Pen allows to replay the data previously recorded like if it was done with a tape recorder. So, one may have the “movie” played forward or backward. But it is also possible to apply filters to go “trough” the data based on criteria, for example to jump from pause to pause…
The video below shows the simple play (forward) of a simultaneous record of pen and eye mouvements. Location of the eye’s “line of sight” is marked with a cross shaped icon, wheras pen location is symbolized with… a pencil icon. When the pen left the tablet (“up” pause), a strait line joins this point to the place where the pen will be in touch with the tablet again. As an help, the final product is drawn in blue as a watermark. Then we can see what will be written.
In the second half of the video, you’ll a parallel processing sequence: the subjects continues to write while his eyes jump back to the prime text.
In a first step, one may display the final product (text):
Eye and Pen can display a “static”preview of a protocol, i.e. to display a representation of all the events of a certain type on the production. In the following example, events writing pauses are represented.
A writing pause duration can be displayed as a circle whose diameter width is in relationship with the pause duration: the longer the pause is, the larger the diameter will be. The circle is centered at the very location where the pen stopped or leaved the paper.
The next example is taken from the same experiment. Only pauses with a duration longer than 50 milliseconds are considered.
Eye movements display
What you’ll see here is a static representation of eye mouvements recorded while writing with an eyetracker. The repesentation show below is a temperature map where the red areas are locations where the subject’s eyes stayed the most.
A first hands-on
A guided tour of Eye and pen in video – 12 min.
Using a background picture and keyboard to ease analysis in Eye and Pen. [ 3 min. ]
What are sequences, what are they good for, how does it work ? [ 7 min. ]
Using templates for sequences descriptions. [ 4 min. ]
Words separation is a convenient semi-automated tool to locate words boundaries (based on pauses, time and distances). From there, you can generate sequences, Areas of Interest and data coding. [ 8 min. ]
Simple case words separation and later data extraction can be performed with keyboard shortcut, paving the way for automation with EPKeys. [ 4 min. ]
Eye and Pen works with a single protocol (file) at a time. To save time, one shoud industrialize actions as much as possible, i.e. try to find which actions can be automatically repeated on a bunch of files.
Let’s take a simple example. 2 groups of 25 participants wrote two texts each. Then, we have got 100 protocols (files) to process.
If we wish to export the pauses data of theses productions (for analyse into a statistical analyses software or a spreadsheet), this means repeating a 100 times the same operation “open a file”, select menu Extractions, sub-menu Pauses, then give a file name, click on Save.
This is time and energy that can be better used, for other tasks for example.
What if someone else can do it for you ?
This is the point where EPKeys comes in.
EPKeys is a kind of robot that “blindly” repeats an action. You tell it what to do (what keyboard key presses it has to produce) and on which folder’s content.
EPKeys can repeat almost any keypress you can do in Eye and Pen. Thus, you may use it to repeat actions other than data extractions, such as filtering, fixations building, words separation, etc.
To industrialize data extractions, depending on what will be done next, there is two ways: either extract the data of each protocol into a separate file, or extract all data into a single file (see Extractions/Preferences/General to allow adding into an existing file).
If willing to directly build a single data matrix, then the better way is to extract all into a single file, and that’s it.
But, on the other hand, if you need to retrieve specific data from results, it may be more handy to create a result file for each protocol, then retrieve these specific data with MakeGroup, to produce a single matrix. This may avoid to have to write macro in a spreadsheet or so.
Let’s take an example.
A task was to write two words, amongst which, only the second is of interest. Specifically, we’re intrested in the trade-off between tracing time and pausing time within this word. Words have been written into “slots”, thus are physically well separated.
The most time saving method is then to use EPKEYS to automatically “separate words”, next extract “tracing dynamics” data, one result file per protocol.
Finally, using MakeGroupe, row and columns of interest can be automatically picked-up in each file to create new single result file containing a ready-to-be-analysed matrix.