**5. Conclusions**

This work analysed the current state-of-the-art in implicit measure-based emotion recognition elicited by HMDs, and gave a perspective using a systematic and aggregated analysis that can guide future research. After two decades of little research analysing emotions using HMDs in combination with implicit measures, mostly undertaken through the physiological arousal responses of the ANS, in recent years, an inflexion point has been reached. The number of papers published is increasing exponentially, and more emotions are being analysed, including valence-related states, more complex biomedical signal processing procedures are increasingly being performed, including EEG analyses and other behavioural measures, and machine-learning algorithms are being newly applied to develop automatic emotion recognition systems. The results sugges<sup>t</sup> that VR might revolutionise emotion elicitation methods in laboratory environments in the next decade, and impact on affective computing research, transversely in many areas, opening new opportunities for the scientific community. However, more research is needed to increase the understanding of emotion dynamics in immersive VR and, in particular, its validity in performing direct comparisons between simulated and real environments.

**Author Contributions:** Conceptualisation, J.M.-M.; methodology, J.M.-M.; formal analysis, J.M.-M.; investigation, J.M.-M.; writing—original draft preparation, J.M.-M.; writing—review and editing, J.M.-M., C.L., J.G. and M.A.; visualisation, J.M.-M.; supervision, C.L., J.G. and M.A.; project administration, J.M.-M.; funding acquisition, J.G. and M.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by European Commission, gran<sup>t</sup> number H2020-825585 HELIOS. **Conflicts of Interest:** The authors declare no conflict of interest.
