Oswald Barral defends his PhD thesis on Implicit Interaction with Textual Information using Physiological Signals on March 1st, 2018

M.Sc. Oswald Barral defends his doctoral thesis Implicit Interaction with Textual Information using Physiological Signals on Thursday the 1st of March 2018 at 12 o'clock noon in the University of Helsinki Main Building, Auditorium XIII (Unioninkatu 34, 3rd floor). His opponent is Senior Scientist Anne-Marie Brouwer (Netherlands Organization for Applied Scientific Research (TNO Soesterberg), Netherlands) and custos Professor Giulio Jacucci (University of Helsinki). The defence will be held in English.

Implicit Interaction with Textual Information using Physiological Signals

Implicit interaction refers to human-computer interaction techniques that do not require active engagement from the users. Instead, the user is passively monitored while performing a computer task, and the data gathered is used to infer interactive inputs to the system. For instance, while typing a word in a text processor (explicit interaction), the processor may identify that we have misspelled the word, and automatically correct it (implicit interaction). Therefore, implicit interaction allows additional interactive channels between human and machines, without further burdening the user.

The recording of physiological signals such as users' brain activity or cardiovascular responses is becoming easier as measurement devices become smaller, more accurate, and less intrusive. These signals can then be analyzed to infer users' properties such as their interest or mood in relation to information content. Such properties have great value as implicit interactive measures, as they represent the inner state of the user, and can then be used by systems to better respond to the users' needs.

The main focus of this thesis is to study physiological signals to collect user responses and feedback on digital information in the form of textual content, for implicit interaction. In details, the research investigates: a) physiological signals to indicate how interesting or relevant a piece of text is to the users, b) physiological signals to indicate affective responses (e.g., feeling happy) of the users to textual information, and c) the use of physiological signals for implicit interaction in real time.

The results of the research show that physiological signals are a great source for implicit interaction with textual information, in real time. The work calls for researchers and practitioners to consider the use of physiological signals as implicit inputs for improved information delivery and personalization.

Availability of the dissertation

An electronic version of the doctoral dissertation is available on the e-thesis site of the University of Helsinki at http://urn.fi/URN:ISBN:978-951-51-4071-5.

Printed copies will be available on request from Oswald Barral: oswald.barral@cs.helsinki.fi.


05.02.2018 - 12:05 Pirjo Moen
01.02.2018 - 17:20 Pirjo Moen