Motivation

Most stud­ies of pres­ence in vir­tu­al real­ity have taken place in aca­dem­ic labor­at­or­ies sim­il­ar to the ones we have here at UCL. The advant­age of this is that par­ti­cipants in the study get almost exactly the same exper­i­ence and thus the exper­i­ment­al con­di­tions are well con­trolled.

Now that con­sumer vir­tu­al real­ity is reach­ing many more users, in the Pres­ence Exper­i­ment for VRJam we are ask­ing wheth­er we can run an exper­i­ment with people using their own equip­ment. We lose a lot of con­trol over how the per­son exper­i­ences the sim­u­la­tion, but we poten­tially get access to many more par­ti­cipants and a broad­er range of par­ti­cipants.

We were also motiv­ated to devel­op the app and these pages to intro­duce new vir­tu­al real­ity users to some of the con­cepts that might help us under­stand and improve vir­tu­al real­ity.

Experiment Script

Please do not read any fur­ther if you intend to try the applic­a­tion.

If you haven’t been able to try the Pres­ence Exper­i­ment applic­a­tion, then what the per­son exper­i­ences is the fol­low­ing:

  • The par­ti­cipant first enters an envir­on­ment that presents a short ques­tion­naire. This ques­tion­naire cap­tures some basic par­ti­cipant inform­a­tion, includ­ing gender, games play­ing exper­i­ence and vir­tu­al real­ity exper­i­ence.
  • The par­ti­cipant enters a bar scene. They are sat in a chair against a table.
  • After few seconds, there is a per­form­ance by a sing­er.
  • Dur­ing the per­form­ance anoth­er audi­ence mem­ber knocks a table caus­ing an object to fall towards the user.
  • There is brief applause as the sing­er fin­ishes.
  • The par­ti­cipant is then trans­por­ted to anoth­er scene to answer a series of ques­tions about the exper­i­ence.
  • The exper­i­ence fin­ishes with a brief sum­mary of the exper­i­ment for the user.

Conditions

There are eight dif­fer­ent ver­sions of the exper­i­ence. Each par­ti­cipant is ran­domly assigned to one of these ver­sions. There are three vari­ables.

Vari­able 1: The user has a self avatar or not.

Screen Shot 2015-05-12 at 17.22.54

Vari­able 2: The sing­er looks at the user or not:

Screen Shot 2015-05-12 at 17.23.08

Vari­able 3: We attempt to induce an illu­sion of body own­er­ship by hav­ing the sing­er ask the user the tap along to the beat and (if they have an avatar) anim­at­ing the avatar tap­ping along to the beat, or there is no induc­tion.

Screen Shot 2015-05-12 at 17.23.27

See also our Mak­ing Of Video on this page.

Discussion

The exper­i­ment will test four hypo­theses:

  1. That the user hav­ing a body increases pres­ence
  2. That the sing­er facing the user will increase pres­ence
  3. That the illu­sion of tap­ping along to the song will increase pres­ence
  4. That the tap­ping induc­tion will increase a body own­er­ship illu­sion

The first two are well estab­lished in the aca­dem­ic lit­er­at­ure on pres­ence through pre­vi­ous exper­i­ments. We aim to show that such res­ults can be rep­lic­ated in this type of exper­i­ment­al set up.

The third and four are more explor­at­ory. They are based on the fam­ous rub­ber hand illu­sion first demon­strated by Botvinik and Cohen.  In this illu­sion, a par­ti­cipant has the illu­sion that a rub­ber hand is part of their own body (see this instruct­ive video). In the ori­gin­al paper, the illu­sion is induced by syn­chron­ous tap­ping on the rub­ber hand and real hand. Slater et al. were the first to show that a sim­il­ar illu­sion could be induced in vir­tu­al real­ity, again by syn­chron­ous tap­ping on the vir­tu­al hand and real hand. Yuan and Steed then demon­strated that the tap­ping induc­tion was unne­ces­sary, and that a vir­tu­al real­ity with motion tracked hands could induce a sim­il­ar illu­sion: the explan­a­tion being that the user sees the vir­tu­al hand move as their real hand does.

How­ever the Gear­VR doesn’t have a hand track­er. So for this envir­on­ment we are see­ing if the user can self-induce the illu­sion by tap­ping their hand on their own leg.