Study type: retrospective database analysis
N: 5732 EMDR VR home-use sessions, with settings kept constant throughout the session

Background and aim
Within EMDR, working-memory load is considered a core mechanism through which emotionality of trauma-related memories can decrease. VR makes it possible to parameterize stimulus characteristics such as speed and trajectory of a moving visual stimulus. In clinical practice, this raises the question which settings are, on average, associated with larger reductions in Subjective Units of Distress, and to what extent settings can be used for personalization. The aim of this analysis was to use a large routine dataset to examine how different speeds and trajectories relate to changes in SUD within sessions.

Methods and design
A retrospective analysis was conducted on 5732 EMDR VR sessions from a subset of the Psylaris database. Only home-use sessions were included so that the chosen stimulus settings remained constant throughout the session. SUD was measured at the start and at the end of each session on a 0 to 10 scale. The analysis compared mean SUD reduction between speed settings 0.75, 1, 1.25, and 1.5 and between trajectory types linear, bouncing, and elliptical.

Outcome measures
The primary outcome measure was the mean SUD reduction per session, defined as SUD at start minus SUD at end. In addition, category counts were considered in interpretation because some settings occur less often and therefore yield less stable estimates.

Results
The default speed of 1 showed the largest mean SUD reduction. Speeds of 1.25 and 1.5 showed very similar effects, but on average slightly less than speed 1. A lower speed of 0.75 was associated with a smaller reduction in SUD. The number of sessions with speeds below 1 was relatively limited, so this finding should be interpreted cautiously. For trajectory, the elliptical trajectory was on average associated with the largest SUD reduction, followed by linear and bouncing trajectories. Here as well, the number of sessions with an elliptical trajectory was lower, and follow-up research is needed to confirm these differences statistically robustly.

Interpretation and clinical relevance
The findings support that stimulus settings within EMDR VR are not neutral but are related to changes in SUD. This is clinically relevant because clinicians can start with an evidence-informed default setting that performs favorably on average while remaining attentive to individual response. In EMDR, the goal is not maximal load but optimal load. Both overload and underload can make a session less effective. The data therefore support a responsive approach in which SUD course and client feedback guide adjustments. The indication of a potentially favorable elliptical trajectory is mechanistically interesting because trajectory complexity may steer attention and influence working-memory load.

Limitations
This is an observational analysis without randomization. Selection biases may occur, for example if certain settings are chosen more often for specific types of clients or memories. The analysis uses SUD as an outcome but does not link to long-term symptom improvement. In addition, some settings are underrepresented, which limits precision.

Conclusion
Based on thousands of sessions, speed 1 appears to be a robust candidate as a default setting for SUD reduction within EMDR VR, with indications that an elliptical trajectory may offer additional potential. At the same time, the analysis underscores the importance of personalization: optimal settings may differ per client. The findings provide concrete empirical support for the parameterizable character of Psylaris EMDR modules and form a basis for controlled follow-up studies.

Figure 1. Working-memory load during EMDR VR

Figure 2. SUD reduction by speed

Table 1. SUD statistics by trajectory