Table 2.
Issues and recommendations to improve the replicability of neuroscience, and their recent developments
| Problem | Solution | Developments/examples |
|---|---|---|
| QRPs (e.g., p-hacking, publication bias) | Registered reports improve incentives while preserving career advancement | >300 journals1; PCI-RR; ACD (theme 2) |
| Lack of detail to replicate/generalize studies | Adopt and enforce comprehensive reporting guidelines | ARRIVE; PREPARE; ACD (theme 4) |
| Lack of methods transparency | Share protocols, raw data, and analysis code | OSF; ACD (theme 5) |
| Incorrect analyses (e.g., identifying sex differences) | Improve methodological and statistical training and facilitate collaborations between experimental scientists and statisticians | ACD (theme 1) |
| Lack of incentives to replicate studies, test theories through adversarial collaborations | Incentivize replication studies and adversarial collaborations through funding opportunities | GAC |
| Lack of incentives to engage in open, pro-replication practices | Reward scientists who engage in scientific practices of openness, transparency, and rigor | Hiring, promotion2 |
ACD: Advisory Committee to the Director of NIH Working Group on Enhancing Rigor, Transparency, and Translatability in Animal Research (Gladman, 2021); PCI-RR: Peer Community In-Registered Reports (Eder and Frings, 2021); ARRIVE: Animal Research: Reporting of In Vivo Experiments (Percie du Sert et al., 2020); PREPARE: Planning Research and Experimental Procedures on Animals: Recommendations for Excellence (Smith et al., 2018); OSF: Open Science Framework; GAC: Generative Adversarial Collaborations (2021; Retrieved April 24, 2022, from https://gac.ccneuro.org/); CoS: Center for Open Science.