NEWS: Preliminary evaluations, of my PhD research, have shown that the joint use of cognitive and emotional processing factors have a significant impact on computer-mediated information spaces and can really improve humans’ comprehension capabilities and performance, mediate their cognitive overload, and create strong correlations amongst these dimensions.
VISION: To develop a dynamic comprehensive personalization filter that will be based on the automatic extraction of humans’ most intrinsic values, namely, User Perceptual Preference Characteristics, and serve as the prime element of any entity (Web-based system, advanced training system, mobile system, virtual environment, spacecraft, aircraft, robot, etc.), liable to accept any type of semantic content, alter and return it, adjusted to the unique characteristics and preferences of the user. User Perceptual Preference Characteristics (UPPC) are all “the critical factors that influence the visual, mental and emotional processes liable of manipulating the newly information received and building upon prior knowledge, that is different for each user or user group. These characteristics determine the visual attention, cognitive and emotional processing taking place throughout the whole process of accepting an object of perception (stimulus) until the comprehensive response to it”.
STEPS:
- An extensive research of the Cognitive Psychology domains related to visual, cognitive and emotional processing.
- The development of a three-dimensional model composed of the following optimized parameters: (a) Visual attention, cognitive processing (speed of processing, control of processing and working memory span), (b) cognitive styles, and (c) emotion control (core anxiety, application specific anxiety, current anxiety, and emotion regulation). All these factors have a specific impact into the computer-mediated information space (and more specifically on text, images, time availability, information quantity, links – learner control, navigation support, additional navigational support, aesthetics etc.) and have been measured during my PhD research, with appropriate psychometric tools / tests running under a common platform.
- The creation of a comprehensive user profile, that is composed of the “traditional” user characteristics (i.e. knowledge, goals, background, experience, preferences, activities, etc.), device / channels characteristics (i.e. device screen size, bandwidth, input method etc.), and the UPPC (cognitive and emotional parameters).
- The development of an open and interoperable Web-based architecture, namely AdaptiveWeb (http://www3.cs.ucy.ac.cy/AdaptiveWeb, positive tested in a controlled simulation environment with more than 200 threads), that is based on the abovementioned characteristics (profile) and is composed of 5 interrelated components: The User Profiling Construction component; the Semantic Web Editor (design phase); the Adaptation and Personalization component that implements the “mapping rules” process for content alteration; and the AdaptiveWeb User Interface (AdaptiveInteliWeb).
- Preliminary evaluation of the system and model’s impact into the computer-mediated information space with a sample of 232 users (in a controlled eLearning environment). Outcome: (a) High joint positive impact of these factors in the computer-mediated information space, since there is a significant increase at subjects' performance, and (b) high correlation amongst the dimensions of the various factors of the model.
- Further evaluation conducted extending the scope the system / model on more generic Web structures; eServices/eCommerce experiments carried out with a total sample of 144 users. Outcome: (a) users in the personalized condition were far more accurate in providing the correct answer for each task mediating their cognitive overload, (b) users in the personalized condition were significantly faster at task completion, and (c) users’ satisfaction has been impressively increased while navigating to the personalized environment as opposed to the conventional one.
FUTURE RESEARCH OBJECTIVES:
- Measuring users’ emotional state with the automatic extraction of body signals (blood pressure, galvanic skin etc.) in relation to users’ performance. Subsequent research objectives: (a) Identify the effect of the rules formulated by the combination of data extracted from sensors and the psychological state of a user, onto the actual content (i.e. marketing, eHealth, etc.) that could be delivered to him/her; (b) Define the Intelligent User Interface and rules that will alter the content to be delivered based on the signals taken and the individual differences (as well as current emotional state) of the users. This will signify the importance of adopting the incoming content in such a way that will be most suitable for the users; (c) Further identify rules with regards to the quantity or the form data need to be altered and delivered, so to be more efficient and effective when conveyed to each user in terms of comprehension and information assimilation; (d) Combine users’ anxiety levels, taken with the use of biometrical sensors metrics (blood pressure, galvanic skin etc.), with various cognitive typologies and the potential mapping rules of these values to any content, for delivering a personalized and adapted results to their current emotional states.
- The construction of a Semantic Web editor for automated development of personalized content.
- Further evaluate the impact of cognitive aspects into computer-mediated environments with the use of an Eye-tracker device.
- To further investigate the constraints and challenges arise from the implementation of such issues and implement the abovementioned model into more advanced computer-mediated entities in order to further evaluate the validity of their impact.
CONTRIBUTION:
The innovation of this research is that for the first time, to our knowledge, the introduction of the comprehensive user profile concept extends the notion of the “traditional” one, in such a way that incorporates the User Perceptual Preference Characteristics, which serve as the primal personalization filtering element of any type of content. This approach emphasizes on critical factors that influence the visual, mental and emotional processes, under a common model, that mediate or manipulate new information that is received and built upon prior knowledge, respectively different for each user or user group. The UPPC model is ‘open’, expandable, can be implemented in any platform, and / or computer-mediated entity.
|