There have been some recent studies by Jaeggi et al showing "growth mindset", "effort" etc etc are indeed important factors. But these are minor in effect size at best, based on my interpretation of their graphs. For a universally highly effective approach like 2x or 3x rate of normal progress with equal training dose, there has to be an incorporation of strong dopaminergic stimulus through truly remarkable game design(which arguably is much, much more difficult compared to programming the actual task itself), along with a combination of additional interventions which are shown to improve neuroplasticity/learning rate to some degree(HIIT, HIFT, NSDR, tDCS, rTMS, tPBM, balance training based vestibular-neuroplasticity activation, juggling based PFC-premotor circuit activation, closed eye mini-rest periods for better encoding, sensory gamma stimulation, nootropics, etc etc).