Week 3

As you might have noticed by now, my week starts on Wednesday: the day I meet with Karsten to discuss my progress. Last week’s meeting was again very fruitful: Karsten introduced me to Zotero, a powerful reference manager. I downloaded Zotero yesterday, including the Chrome plug-in, and it seems like a wonderful tool indeed! I immediately started to import all the previously downloaded papers and, although it took me the entire afternoon, I am convinced that it will be well worth the effort in the long run.

Actually, my anti-productivity streak already began last weekend, when I was staying at my girlfriend’s house. While I was searching for new papers about physiological sensor technologies on Saturday morning, the house’s internet connectivity caved. Thanks Belgacom, for the thesis-free weekend… (My girlfriend’s parents got to exchange their router for a new one on Monday, so all is well now.)

That is not all folks… To continue this ‘dog ate my homework’ story: I noticed my Nexus 5 was acting strange lately, so I decided to return it to the store, since it is still under warranty now. I also argued that I will be relying more on my Android phone as the thesis progresses, so returning the phone in the beginning of the semester seemed like the best way to go. Fortunately, I was able to use an Android phone from the HCI department (i.e. Samsung S II) on which I could test my small accelerometer app.

To conclude this blog post, I will actually say something about the thesis content for a change. The most important thing I learned up until now is the general understanding in the academic world that the automatic detection of emotions requires as many input signals as possible. After thoroughly reading up on the physiology of emotions, I managed to extract a compact list of physiological hotspots (and corresponding measurement techniques/devices) that could serve as an indicator for certain types of emotions:

  • Heart: electrocardiography, photoplethysmography, …
  • Blood: pulse oximeter, blood pressure sensor, …
  • Respiration: respiratory transducer, temperature transducer, …
  • Body temperature and sweat: thermometer, electro-dermal activity sensor, …
  • Muscle tension: electromyography, accelerometer, …
  • Pupillary response: eye tracking, …
  • Brain activity: functional magnetic resonance imaging, electroencephalography, hemoencephalography, …

The challenge now is to review all available sensors that are fit for the job of automatic (preferably mobile) emotion detection: a very interesting but huge task indeed. I already found and skimmed some papers regarding EEG, ECG, EDA, PPG and blood pressure sensors, but there is still a lot of research to be done.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s