jump to navigation

The other psychophysiological signals April 28, 2010

Posted by mikkosalminen in Blogroll.
Tags: , , ,
add a comment

Besides the already covered EEG we are also planning to record some other psychophysiological signals when studying digital reputation in different experimental settings. We plan to use at least electrodermal activation (EDA), facial electromyography (EMG), and electrocardiography (ECG).  The dimensional theory of emotion places emotions in a two dimensional space defined by valence (unpleasant Рpleasant) and arousal (calm Рexcited). Valence can be effectively studied with the facial EMG. The cheek muscle and the periocular muscle are usually activated during positive emotions, like during smiling. The frowning muscle is, of course, related to negative emotions. For the measurement of arousal, EDA, in practice sweating of the hands, is a useful index. The interpretation of these signals is somewhat straightforward, at least when compared to heart rate, or ECG more generally. The problem is that the activity of the heart is regulated by many processes; the ECG has been used to study stress, arousal, valence, attention, and orientation reflex, for example. So it is good to have other signals to accompany in the interpretation.

There are various other signals that could be also recorded, like breathing rate, movement of the subject by accelerometers or tracking the gaze or changes in the pupils. However, there are limits to how much wires we can attach to the subjects. On the other hand, having more than just one or two signals collected is not advisable either. The interpretation is almost always easier when there are more psychophysiological signals to look at. Since we are applying the psychophysiological method to a new area, the study of digital reputation, we cannot know beforehand which signals would be the most useful. After the first experiments we have at least some guidelines for selecting the best signals for further studies of digital reputation.

EEG in the study of reputation? March 18, 2010

Posted by mikkosalminen in Blogroll.
Tags: , ,
1 comment so far

One part of our Digital Reputation project is to conduct laboratory experiments using the psychophysiological methods. This entry focuses on one of the physiological signals: the electroencephalography, EEG, and on how we are going to use it in the study of digital reputation.

EEG is the oldest of the brain research methods, the more modern being MEG, PET, and fMRI, for example. The strength of EEG is its applicability to more ecologically valid settings. Mobile, light-weight, recording systems allow data collection in real life situations. In addition, EEG has good temporal resolution; brain electrical activity can be studied with one millisecond accuracy. One weakness is lower spatial resolution when compared to some other methods. Activity of sub-cortical brain structures is best studied with other methods than EEG. Like other psychophysiological methods, also EEG is sensitive to noise, e.g. by movement (body movement and eye blinks).

In practice EEG is recorded with an electrode cap and also some conductive gel is needed to establish connection between the scalp and the electrodes. The system records voltage changes between each recording electrode and a reference electrode. The number of electrodes usually varies between one and 256, a general rule of thumb is that the more electrodes you have the better the spatial resolution.

There are two main lines of research for quantifying EEG; either studying event related potentials (ERP) or studying power changes in different frequency bands. The power changes can also be event-related (event-related desynchronization, ERD; and event related synchronization, ERS) or on level of longer-term tonic activity. In frequency analyses the focus is often on changes of power on delta (0-4 Hz), theta (5-7 Hz), alpha (8-12 Hz), beta (12-20 or 30 Hz), and gamma (36-44 Hz) bands. Various studies have found different EEG oscillatory responses to different emotional, cognitive and motor processes. Our interest focuses on those responses that tell something about emotions and motivational processes; we aim to study which psychophysiologically observed emotional responses are related to good and bad reputation.

For example, we will study the frontal alpha asymmetry of the EEG when viewing material related to good and bad reputation companies. Frontal alpha asymmetry is probably the most studied EEG measure of emotion and motivation. Studies that have examined relationships between different individual difference measures and resting EEG activity have shown that asymmetrical activation of the anterior cortical regions seems to influence emotional responding. Relatively higher left frontal activity, either as a state or a trait, indicates a propensity to approach a stimulus, whereas relatively higher right frontal activity indicates a propensity to withdraw from a stimulus. The underlying assumption in frontal asymmetry studies is that activity in the alpha range (8-13 Hz) is inversely related to underlying cortical processing; it is typical that alpha power decreases when the underlying cortical systems engage in active processing. It must be kept in mind, that these frontal asymmetries are not measures of positive or negative affects per se, but they tap a broader motivational tendency towards approach-related or withdrawal-related behaviors and emotions. Our working hypothesis is that good reputation is related to approach tendency.