VR Flight Experience Reduces Fear of Heights by Action-Based Prediction of Safety

A research group led by FUJINO Misako and HARUNO Masahiko at the National Institute of Information and Communications Technology (NICT, President: TOKUDA Hideyuki Ph.D.), has demonstrated that experiencing active flight in VR allows individuals to predict that they can transition to a safe state even if they fall from a height, thereby reducing fear responses. This discovery challenges the traditional understanding that fear extinction necessarily requires repeated exposure to fear-inducing stimuli.

In this study, the researchers compared a group of participants who actively experienced low-altitude VR flight (Flight Group) with a control group who passively watched recordings of the flight experience. The Flight Group showed significantly greater reductions in both physiological (skin conductance response, SCR) and subjective (self-reported fear score, SFS) fear responses when walking on a virtual plank at high altitude compared to the Control Group. Furthermore, among the Flight Group participants, those who more strongly felt "I can fly, so falling is not dangerous" exhibited a greater reduction in fear responses.

These results suggest that "action-based prediction" can reduce fear responses without relying on repeated exposure, potentially offering a new approach to fear extinction.

The findings were published online on May 13, 2025, in the Proceedings of the National Academy of Sciences (PNAS).

Traditionally, fear extinction has been understood to occur through repeated exposure to fear-inducing situations, allowing individuals to gradually learn that the situation is safe. However, predicting “I can transition to a safe state through self-initiated actions” may also serve as an effective mechanism for reducing fear. In this study, the researchers focused on whether experiencing active flight in VR which creates an "action-based prediction" that "even if I fall, I can fly," can reduce physiological and subjective fear responses or not.

Participants with a tendency toward fear of heights first completed a virtual body ownership task to become familiar with their VR avatars, then performed a high-altitude plank-walking task (First Plank Task) in VR at approximately 300 meters above ground level (see Figure 1).

Physiological fear responses were measured via skin conductance responses (SCR) using electrodes attached to the participants' fingers, and subjective fear responses were assessed using a self-reported fear score (SFS) on an 11-point scale.

Participants were then randomly assigned to one of two groups:

Afterwards, both groups completed the Second Plank Task, and SCR and SFS were measured again.

Analysis of SCR data showed that while both groups exhibited reduced fear responses in the Second Plank Task compared to the First, the Flight Group showed a significantly greater decrease in SCR than the Control Group (see Figure 2).

Similarly, analysis of the SFS data revealed that both groups showed lower subjective fear in the Second Plank Task, but the Flight Group exhibited a significantly greater reduction than the Control Group (see Figure 3).

These findings were confirmed across two independent experiments (Experiment 1 and Experiment 2).

Overall, the results indicate that actively flying at low altitudes in VR contributed to a greater reduction in fear responses to heights, possibly because participants predicted that they could transition to safety even if they fell.

To further test this hypothesis, the researchers conducted a multivariate regression analysis using questionnaire scores collected after the VR tasks. They found that the degree to which participants felt "falling would not be dangerous because they could fly" (Safety Prediction Score) significantly contributed to the reduction in SCR among Flight Group participants (see Table 1).

This study suggests that the prediction of transitioning to safety through self-action could serve as a new mechanism for fear extinction, distinct from traditional exposure-based methods. Going forward, researchers plan to investigate whether this VR flight experience can provide long-term reductions in fear responses in real-world settings for individuals with acrophobia and whether it can be applied as an effective treatment strategy.

This study was supported in part by the following projects: JST CREST Project "Elucidation and application of multi-world predictive coding in cyber society" (Principal Investigator: Masahiko Haruno, JPMJCR22P4), JST Moonshot R&D Program "Realization of an Avatar-Symbiotic Society Where Everyone Can Perform Activities Freely Without Constraints of Body, Brain, Space, and Time by 2050" (JPMJMS2011) and JSPS Grant-in-Aid for Transformative Research Areas (A) "Multidimensional Biology: Decoding and Manipulating Brain Dynamics for Behavioral Change" (22H05155)
 

All experiments were approved by the NICT Ethics Committee, and all participants provided informed consent after being briefed on the experimental procedures.

Fear extinction refers to the process by which fear responses formed through past experiences gradually weaken through repeated safe exposure to fear-inducing contexts. In the case of fear of heights, repeatedly experiencing standing safely at high altitudes leads to the formation of new learning that "high places are not dangerous," counteracting the original memory association of "height = danger" and thereby suppressing fear responses. Exposure therapy, a treatment for phobias and PTSD, is based on this mechanism.

When humans experience fear or stress, the sympathetic nervous system becomes activated, causing a small amount of sweat to be secreted from areas like the palms. The skin conductance response refers to changes in the skin’s electrical conductance associated with this sweating. SCR is widely used in research on fear conditioning and fear extinction as a physiological indicator of fear responses in humans.