Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Novice2Expert - a Cognitive Model within a Usability Evaluation Framework

Paper presented at 7. Interdisziplinärer Workshop Kognitive Systeme. Mensch, Teams, Systeme und Automaten 2018, 21.-22. Juni 2018, Braunschweig
 
: Wolf, Karen Insa; Thalappully, Rashik; Wagner, Yves; Wallhoff, Frank; Appell, Jens-E.

:
Volltext urn:nbn:de:0011-n-5073291 (1.6 MByte PDF)
MD5 Fingerprint: 8790fc4f599e903430ef0d6c43f8a64a
Erstellt am: 31.10.2018


2018, 6 S.
Interdisziplinärer Workshop "Kognitive Systeme" <7, 2018, Braunschweig>
Englisch
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IDMT ()

Abstract
The motivation of the usability evaluation framework CogUA (Cognitive Usability Analysis) is to support usability analysis processes of graphical user interfaces (GUI) based on software tools deriving objective and reproducible evaluation parameters in a partly automatic process. In the current prototype status, it consists of five modules: (i) Observer, (ii) Trace Illustrator, (iii) Screen Shot Analyser, (iv) Use Case Detector and (v) Predictor. The framework and its modules are shortly described in the paper. The focus of this paper lies on the fifth module, the Predictor. It is based on a cognitive model to simulate human users while interacting with a GUI. For the investigation, a test application is set up in which a label is presented as a task on the screen. The user, respectively the model, has to find and click the button with the presented label. The derived interaction times based on the model are compared with results of a small user study in order to evaluate the different versions of the cognitive model. Three different scenarios are investigated: (a) the model as a novice, searching each time for the correct button on the screen, (b) the model, which is able to remember the position of buttons of the GUI, and (c) the model, which is able to remember the position of buttons as well as the sequence of the tasks. The interaction time necessary to execute a specific task sequence is reduced from scenario (a) to (b) to (c). The relative reduction of interaction times derived from the user study and predicted by the model is comparable.

: http://publica.fraunhofer.de/dokumente/N-507329.html