 |
|
Janssen, J.H. and Westerink, J.H.D.M. and van den Broek, E.L.
(2009)
An affective music player: Methods and models for physiological computing.
Report Philips Research TN-2009-00180
Koninklijke Philips Electronics N.V., Eindhoven, The Netherlands.
Full text not available from this repository. AbstractAffective computing is embraced by many to create more intelligent systems and smart environments. In this thesis, a specific affective application is envisioned: an affective
physiological music player (APMP), which should be able to direct its user's mood.
In a first study, the relationship between music, mood, and physiology was assessed, using both experimenter-selected and personal music for mood inductions. Hypotheses were based on the mood behavior model (MBM), which predicts mood effects on autonomic activity during task related demand appraisals (informational) and mood regulation (directive). Our results showed mood-congruent autonomic activity for skin temperature and electrodermal activity. We support the MBMs informational mood impact but argue for a negative-mood-only directive mood impact. Taking this into account, the MBM remains useful for predicting mood congruent autonomic activity.
Second, mood inference from physiology was integrated in an affective loop. However, a meta-analysis of the literature showed that it is problematic to infer affect from physiology in real world applications. Therefore, a framework was developed with directives on increasing the validity of physiological research and on using physiological signals in applications.
Third, the APMP is developed, using personalized probabilistic models of physiological effects of music, incorporating the principle of initial values. This comprised three steps: (1) data gathering of participants during real-world music listening, (2) personalization
of the APMP, and (3) real-world validation of the APMP. The successful validation shows that the APMP concept is open for practical application.
The success of the APMP indicates the rapid evolvement of affective computing that will soon come to fruition, augmenting lifestyle and technology. This provides a promising future for artificial and ambient intelligence, and marks a new era in consumer product development. | Item Type: | External Report |
|---|
| Research Group: | EWI-HMI: Human Media Interaction |
|---|
| Research Program: | CTIT-NICE: Natural Interaction in Computer-mediated Environments |
|---|
| Uncontrolled Keywords: | emotion, mood, music, methods, model, affect |
|---|
| ID Code: | 18384 |
|---|
| Deposited On: | 28 July 2011 |
|---|
| International: | Yes |
|---|
| More Information: | statistics |
|---|
Export this item as: To correct this item please ask your editor Repository Staff Only: edit this item
|
|
|
Nederlands
Contact
Sitemap
Search
Organisation