Abstract
The literature discusses causes of low reproducibility of scientific publications. Our article adds another main cause-uncritical adherence to accepted research procedures. This is evident in: (1) anachronistically requiring researchers to base themselves on theoretical background even if the studies cited were not tested for reproducibility; (2) conducting studies suffering from a novelty effect bias; (3) forcing researchers who use data mining methods and field-based theory, with no preliminary theoretical rationale, to present a theoretical background that allegedly guided their work-as a precondition for publication of their findings. It is possible to increase research validity in relation to the above problems by the following means: (1) Conducting a longitudinal study on the same participants and only on them; (2) Trying to shorten the time period between laboratory experiments and those on humans, based on cost-benefit considerations, anchored in ethical norms; (3) Reporting the theoretical background in a causal modular format; (4) Giving incentives to those who meet the above criteria while moderating the pressure for fast output.