With the advent of free search came the sale of personal shopping information and the rise of mining search choices for psychological insights into the individual shopper. With the advent of virtual worlds will come an even deeper probing of the mind of the individual user. This is why I've asked questions and suggested researchers in VR (not the Metaverse Mavens: those are software hucksters looking to ride a wave to make a buck) spend time studying the literature on cueing (eg, onset and offset cues). It was obvious when doing the research for HumanML that behavioral modification would be an outcome of applying the techniques of situation semantics. That is a basic tenet of GUI design from day one: interact and evolve or devolve.
which is weird science for predictive analysis but not outside the realm of possibility given suitable knowledge of applying cueing
which suggests to me that the theories of neuroscience could be tested and applied.
The rule of the road on the web is if it can be done it will be done and someone will finance it, and then sell the results.