The Effect of Virtual Environments on Stress Levels

Abstract

I forbindelse med dette projekt, har man fundet ud af at adskillige mennesker lider af, eller oplever symptomer på stress. Derfor er målet for dette projekt at finde en måde hvorpå stress kan reduceres. Gennem research blev forskellige måder at måle stress undersøgt og overvejet, samt hvilke metoder der allerede bliver brugt i forhold til at reducere stress. Man fandt ud af at natur miljøer ofte bliver brugt som en måde at fremme afslapning, samt nedsætte stress. Ud fra dette blev den følgende problemformulering sat op: “Hvilke elementer, i et virtuelt miljø, kan bruges til at fremme afslapning?” Tre eksperimenter blev udført hvori to forskellige virtuelle miljøer (et simpelt rum og en skov scene) blev testet imod hinanden med en implementeret Montreal Imaging Stress Test(MIST). I det første eksperiment blev der ikke fundet nogen signifikant forskel mellem det simple rum og skovscenen. Nulhypotesen; “En skovscene er ikke bedre til at nedsætte stress i deltagere når de udfører en mental stress opgave.” Kunne derfor ikke modbevises. Men fordi der er en forskel mellem de forskellige målinger under forsøget i forhold udgangspunkterne og de forskellige miljøer, kan det konkluderes at MIST testen havde success med at stresse deltagerne. Da undersøgelser viste at natur burde have en effekt i forhold til at reducere stress, blev det besluttet at forbedre de virtuelle miljøer og gentage eksperimentet. I andet eksperiment blev der ikke fundet nogen signifikant forskel mellem de virtuelle miljøer. Derfor kunne nulhypotesen (som var den samme som blev undersøgt i første eksperiment) heller ikke modbevises for dette eksperiment. I det tredje eksperiment blev der implementeret lyd for at skabe en bedre fordybelse i de skabte miljøer. Nulhypotesen kunne ikke modbevises med de udregnede p-værdier. Dog viste et mindre antal af variabler en signifikant forskel mellem udgangspunkterne for målingerne og målingerne for miljøerne, sammenlignet med de forrige eksperimenter. Dette kunne være et tegn på at lyd har en effekt i forhold til at nedsætte stress. Ud fra disse eksperimenter, hvoraf nul-hypoteserne ikke kunne modbevises, kan det ikke være sikkert at natur har en effekt i forhold til at nedsætte stress. Dog kan det godt være påvirket af indlevelsen i miljøerne. Ydermere er der indikationer der peger på at lyd godt kunne være effektivt i forhold til at reducere stress. Derfor skal yderligere undersøgelser på dette emne opmuntres.

In this project, it was found that several people suffer from or experience symptomsof severe stress. Therefore, the aim was to find a way to reduce stress. Through the background research, different ways of measuring stress werelooked into, as well as what methods are already used in relation to reducingstress. It was found that nature environments are commonly used as a way toinduce relaxation, while also reducing stress. From this, the following problemstatement was made: "What elements in a virtual environment are responsible for inducing relaxation?" Three experiments were conducted, in which two different virtual environments (a simple room and a forest scene) were tested against each other withan implemented Montreal Imaging Stress Task. For the first experiment, no significant difference were found between the simple room and the forest scene. The null-hypothesis; "A forest scene is not better at decreasing stress in participants when performing a mental stress task." could therefore not be refuted. As there was a difference between the baselines and the environments, it could however be found that the MIST did in fact stress people. As the background research showed that nature should have an effect in relation to reducing stress, it was therefore decided to improve upon the environments and repeat the experiment. For the second experiment, no significant difference between the virtual en-vironments were found. Therefore, the null-hypothesis (the same as for the firstexperiment) could not be refuted during this experiment either. For the third and last experiment, audio was implemented in order to create a better immersion within the scenes. The null-hypothesis could not be refuted with the calculated p-values. However, less variables showed a significant difference between the baselines and the environments, compared to the previous experiments. This might indicate that audio could have an effect in relation to reducing stress. From these three experiments, in which the null-hypotheses could not be refuted, it is not certain that nature has an effect in relation to reducing stress. However, this might have to do with the immersion of the virtual environments. Additionally, there might be indications towards audio being effective in reducing stress, and therefore further research on this topic are to be encouraged.

A master thesis from Aalborg University

The Effect of Virtual Environments on Stress Levels