Pixel1.gif (51 bytes)
Pixel1.gif (51 bytes)
Pixel1.gif (51 bytes) Main Page Pixel1.gif (51 bytes)
About DSP Laboratory
Pixel.gif (52 bytes)
Contact Us
Pixel.gif (52 bytes)
Go to FIU's Homepage


 Pixel1.gif (51 bytes)


Curve.gif (104 bytes) Pixel1.gif (51 bytes)

Comparative Analysis of Noninvasively Monitored Biosignals for Affective Assessment of a Computer User

Pixel1.gif (51 bytes)

"Comparative Analysis of Noninvasively Monitored Biosignals for Affective Assessment of a Computer User", (2009)
Gao Y., Barreto A. and Adjouadi M.

ABSTRACT: Future human-computer interactions could be enhanced by enabling the computer to detect the userís emotional states (e.g. stress), and adjust its interaction behavior appropriately. We are attempting this affective recognition task through the processing of a variety of biosignals measured noninvasively from the computer user, such as the Pupil Diameter (PD), the Galvanic Skin Response (GSR), the Blood Volume Pulse (BVP), etc. In this study, we compare the innovative processing of PD signals for affective sensing using an Adaptive Interference Canceller (AIC), with the H-infinity time-varying (HITV) adaptive algorithm to the processing of signals traditionally used for this purpose (GSR and BVP). The comparison is made in the framework of an experiment that elicits mild levels of mental stress in the experimental subjects during controlled segments by inclusion of incongruent Stroop word presentations. The GSR signal is monitored by means of surface skin electrodes attached to two fingertips and the BVP is measured through an infrared reflectance photoplethysmograph placed in a third fingertip of the left hand of the subject. Both these analog signals are digitized simultaneously with the PD signal which is obtained directly as a digital sequence from an eye gaze tracking apparatus. In total, 10 feature signals were extracted from the monitored and processed biosignals. The discriminating power of these 10 features is compared through their Receiver Operating Characteristic curves.