Effectiveness of Guidance Cues for Visual Targets in Virtual Environments

How can we help users navigate virtual reality without breaking their sense of immersion? This mixed-method study evaluates passive (rear-view mirror) and active (audio and haptic) cues in a VR scavenger hunt, finding that haptics are the most effective, followed by audio, while rear-view mirrors prove unsuitable for such environments and tasks.

mirrorcondition

Rear-view mirror condition

Method

Task Design

Five key steps including:
task, spatial environment, 3D modeling, equipment & interactions, and conditions.

Study Design

Four key steps including:
participant recruitment, setting, study introduction, practice session & task, survey.

task design

  • Task

    The task was an engaging scavenger hunt in a museum, where participants were asked to find seven specific paintings among the many on display.

  • spatial environment

    Google map, AutoCad, Rhinoceros 3D

    The virtual environment recreated a section of the Louvre Museum's Denon wing, featuring rooms 700–702 and 710–712 to ensure a balanced layout. Approximately 30% of the paintings were removed to maintain participant focus.

  • 3D modeling

    Rhinoceros 3D

    Rooms were modeled in Rhinoceros 3D, highlighting key architectural elements while omitting intricate details to avoid distracting participants.

  • Equipment, Interactions

    Unity

    The model was imported into Unity, where seven target paintings were placed across rooms. Participants used a Meta Quest Pro with a counter showing collected paintings and a task list to aid memory.

  • Conditions

    Unity

    Four sensory cue conditions were tested: 1)Control Condition: 90° front field of view, no cues, 2)Mirror Condition: Rear-view mirror extending vision behind participants, 3)Audio Condition: Beep sound within a 5-meter radius of targets, 4)Haptic Condition: Controller vibration within a 5-meter radius of targets. Each participant experienced only one condition, maintaining task consistency across groups.

study design

  • Participants

    Qualtrics

    The study included 71 university participants (30 female, 41 male), aged 18-35, screened for eligibility based on vision and motion sensitivity. Participants varied in gaming experience, with only two having prior, minimal familiarity with the Louvre.

  • Setting

    Experiments were conducted in a distraction-free lab space, where participants navigated the virtual environment seated, using controllers for movement and interaction.

  • Target Introduction, Practice Session, Task

    Unity

    Participants were first shown a printed list of seven target paintings, along with a brief introduction to each one, including their names and stories. This gave them time to memorize the paintings and provided context to make recall easier, as the focus was not on testing memory or cognition. In the VR environment, they could also access the list at any time in their left hand for reference. Next, they completed a three-minute practice session in a simple VR setting, where they learned the controllers and task requirements, seeing only the target paintings. Finally, they began the main task, which was recorded in a timed 20-minute session.

  • Survey

    After completing the task, participants filled out a survey on task enjoyment, immersion, cue effectiveness, and willingness to try similar tasks, followed by the NASA-TLX, Presence Questionnaire, and Simulator Sickness Questionnaire.

Quantitative Findings and Analysis

Key evaluation metrics included (1) frequency of task list checks, (2) total task completion time, and (3) responses from three survey questionnaires. The findings revealed that guidance cues influenced participants' attention orientation, with some unexpected results. As expected, active cues (audio and haptic) reduced the frequency of task list checks. However, the rear-view mirror unexpectedly led to more frequent task list checks and longer completion times, possibly due to its placement or the added cognitive load from additional input, which may have increased the overall workload.

quanttables_image

Qualitative Findings and Analysis

Survey responses were grouped into four categories—effective, challenging, engaging, and time awareness—and distilled into two key themes: effectiveness and engagement.
Effectiveness: Participants in the audio and haptic conditions consistently praised the cues, noting they helped prevent missed paintings. In contrast, about half of those in the mirror condition rarely used the mirror, with only 6% relying on it frequently. Many participants in the control and mirror conditions described the task as challenging, while those in the audio and haptic conditions found it more manageable.
Engagement: Guidance cues boosted engagement across all groups. Enjoyment was highest in the audio (100%) and haptic (86%) conditions, where cues helped participants stay focused. Time perception varied, with most participants in the audio (76%) and control (70%) conditions experiencing a "loss of time," reflecting deep engagement in the task.

qualfigure_image

lessons learned

Among all the things I learned during this project, one of the most transformative was honing my time management skills. I balanced a wide range of tasks—from 3D modeling to game interaction development, literature review, and pilot tests—while continuously adapting my workflow to meet shifting priorities.

Along the way, I gained valuable insights into working with human subjects, particularly in the intricate world of VR, where precision, attention to detail, and clear eligibility criteria are essential to ensure safety and reliable outcome. Perhaps the most challenging and rewarding lesson was cultivating patience and consistency; recruiting and testing 71 participants, with each session taking about an hour, was demanding. Remaining as attentive and consistent with each participant, from the first to the last, required endurance, as the study spanned two months of daily sessions. This journey not only strengthened my skills but also deepened my passion for innovative, human-centered design and research.

Go to Portfolio Two