Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
Regardless of country or discipline, publications are an expectation—if not a requirement—to obtain a PhD. In this issue, PhD students, academics and external stakeholders describe how this imposes detrimental consequences but also provides benefits for individuals and the scientific community, including clear calls for future improvements. The discussion is amplified with more contributions on our community forum.
Publications are commonly used to evaluate PhD students’ aptitude and have the appeal of a single, ‘objective’ measure. A collection of World Views in this issue, however, suggests that this creates only an illusion of true meritocracy. Not only assessments but PhD training per se require substantive improvements to benefit science and scientists.
Classic avoidance learning leads to a dilemma: if an animal always avoids a cue that lead to a negative outcome, it will never learn anything new about the cue and outcome. A new study suggests that a protected childhood period helps resolve that dilemma: children actually prefer to explore aversive cues but only do so if a parent is present.
Many theories have been put forward to explain how different sound systems evolved. Whether differences in vocal tract shape play a role has so far remained unclear. Dediu et al. document subtle differences among four broad ethnolinguistic groups. Using computer simulations, they demonstrate how differences can be amplified over time, leading to diverse vowel systems.
Wahl et al. present palaeoenvironmental, epigraphic and archaeological evidence that suggests that the Maya engaged in tactics akin to total warfare earlier and more frequently than previously thought.
Why are people so often overconfident? Schwardmann and van der Weele show that people self-deceive into higher confidence if they have the opportunity to persuade others for profit and that higher confidence aides persuasion.
Would you rather lose your job to a robot or a human? Granulo et al. show that people’s preference for humans to take on the jobs of humans reverses when they consider their own jobs: when it comes to themselves, humans prefer being replaced by robots.
Young children switched to a preference for an aversive conditioned stimulus if acquisition occurred in the presence of their parent. Results suggest that early learning systems are constructed to permit modification by parental presence.
Lee et al. show people's biases in social perception can be explained merely by the structure of their social networks, without assuming biased cognition. Social perception biases can be explained by homophily of personal networks and minority-group size.
Chenoweth and Belgioioso describe the momentum of protest movements as the product of the number of participants (mass) and concentration of events in time (velocity). Higher momentum is associated with a higher probability of irregular leader exit.
Lieder et al. leverage artificial intelligence to redesign our to-do lists into games that make us more productive. Four experiments suggest that their approach can help people make better decisions, overcome procrastination and prioritize better.
Does anatomy affect cross-linguistic differences? Using computer models of the vocal tract, Dediu et al modelled how vowels are learned and transmitted across generations. Simulations show how variations in the hard palate contribute to phonetic diversity.
Anxiety is characterized by altered responses under uncertain conditions. Aylward et al. show that these altered responses are due to anxious individuals updating their behaviour in response to negative outcomes more quickly than non-anxious individuals.