Virtual View: research methods

How does one research the influence of landscape and sound on a human? Fortunately a lot of research has gone into finding out how people react to visual landscape stimuli. Most articles I’ve read made use of static pictures, some used video. As pictures can be found in abundance on the web and are easily stored and manipulated I chose static colour pictures as the main visual stimulus.

In most experiments natural landscapes are compared to urban environments with varying amounts of green. Almost always the natural and greener urban scenes have more positive effects on health and affect related variables compared to the urban environments. So it seemed logical not to use the pictures of urban environments. Together with the students I decided on using landscape pictures that were at odds with the most preferred landscape. So that would be: chaotic natural scenes with a restricted view and no deflected vistas or water. As I discussed my experiment setup with Sarah she strongly recommended I’d use a control set of stimuli. That way I could (hopefully) confirm the findings from other experiments and I’d have a contrast set to compare the natural scenes to and hopefully see significant differences between the contrast set and the different landscapes. As the installation will be placed in health care environments I decided to make a set of neutral hospital interiors as a contrast set.

stimuli

The final installation will be an animation so I wanted to use sets of landscapes to mimic a little the animation effect. We decided on sets of 6 images. Then we had to figure out what time the images would need to be shown to have a measurable effect. Not very much could be found in literature about this so the students did some tests, showing the images for different time periods. The result on the heart-rate was very diverse. So I consulted Malcolm and asked him what to make of this. He said the sample was too small to conclude anything. His suggestion was to show people two sets with the images displaying at different lengths and then to ask people what they preferred. He had already pointed out earlier that it does take some time for stimuli to take effect. Unfortunately the students only tested 10 and 25 seconds to compare. From that they concluded that 25 seconds was a bit too long but that people preferred the longer exposure. So we settled for 20 seconds per image. And each set would last two minutes.

Of course a baseline measurement was needed for the heart-rate as well as the self reported data (view below). For the experiment to have any scientific value Malcolm said I needed at least five minutes of baseline measurement. Not to complicate things further Hein advised to not make use of any specific stimulus but just use an empty screen. It would be quite a long time to sit there and do and see nothing, but it would be for the good cause!

As I reported earlier the research on the effects of natural sounds has been a lot more sparse. But as with visual landscapes water was perceived as more pleasant compared to for example mechanical sounds. And aesthetically pleasing an non-threating bird sounds seem to indicate a positive effect on attention restoration and stress reduction. So we used different combinations of water and bird sounds. The hospital interior set was accompanied by sounds from a hospital waiting room.

In this review of health effects of viewing landscapes there’s an extensive list of research and physiological parameters measured. For Virtual View I’m interested in heart-rate and heart-coherence. Further more I would like to know how a certain landscape makes people feel. I want the installation to have a relaxing effect and to positively influence a sense of well-being. For measuring the physiological side I of course use the Heartlive sensor. I measures beats per minute and calculates heart-cohere. The EventIDE software logs the heart data every second and calculates means for every picture.

Not only do I not own a device to measure for example skin conductivity (GSR) I’m also curious about how people feel when watching the sets. So I needed some record of perceived relaxation state and affect. It was not easy to find a (short) questionnaire which measures that. Malcolm pointed me to the Smith Relaxation States Inventory 3 (SRSI3). It is a very interesting and validated inventory but alas consists of 38 items. It doesn’t make sense to ask people 38 questions after two minutes of pictures. The questionnaire may not be modified without consent so I asked Sarah what to do. She suggested to simplify things and just ask people how relaxed they are on a 10 point scale.

She said 10 points are better then five because it is easier to see the middle and it is more fine grained. It gives people the opportunity to pinpoint how they feel. We settled on three questions: I feel at ease, I feel relaxed, I feel joyful and happy. If my installation can make that happen I’m satisfied no matter what the heart-cohere or heart-rate is. All questions are integrated in EventIDE. Carlos, one of the students, made a nice colour feedback on the scale.

The students take notes of remarks the participants make on their experience of the trail. This may also yield interesting results in relation to the experiment data.

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.