Reviews & Analysis

A new algorithmic tool developed by Rotaru and colleagues can more accurately predict crime events in US cities. Predictive crime modelling can produce powerful statistical tools, but there are important considerations for researchers to take into account to avoid their findings being misused and doing more harm than good.

Proposals to fight online misinformation range from gently encouraging users to consider the accuracy of information (‘nudges’) to bans and removing content. Using modelling techniques, we find that these interventions are unlikely to be effective in isolation, but that a combined approach can achieve a significant reduction in the spread of misinformation.

A randomized controlled trial of approximately 4,500 households in Botswana during the COVID-19 pandemic was conducted to investigate the effectiveness of using low-tech learning interventions during school closures. A simple combination of phone tutoring and SMS messages substantially improved learning in primary school children in a cost-effective manner.

Human neonates discriminate vowel sounds played forward, as in normal speech, from their waveform reversal after five hours of exposure on the first day of their life. The neural dynamics supporting this rapid perceptual learning indicate a primitive brain mechanism similar to the language-processing network of adults.

Recent studies have suggested that attention switches rhythmically, with up to eight cycles per second. An article by Brookshire calls these results into question, showing that widely used methods in the field do not separate rhythmic from arrhythmic structure.

We developed a new approach that uses high-frequency mobile phone data to measure internal displacement after violent events. We used this approach to study the impact of violence in Afghanistan, highlighting how patterns of internal displacement depend on the nature of the violence experienced.

An enduring puzzle in evolution is the maintenance of costly traits. Šaffa et al.¹ examine phylogenetic evidence for the origins of genital mutilation/cutting (GM/C) in human societies, and find that these practices probably emerged multiple times during the past 5,000–7,000 years, and that female GM/C arose only after male GM/C was present in a society.

This study tested the hypothesis that negativity bias — giving disproportionately more attention and decision weight to negative than to positive stimuli — is associated with right-wing political ideology. Across five distinct studies and multiple measures of ideology, the results provide no consistent evidence that people with right-wing ideology have a stronger negativity bias.

Past research has put forward competing hypotheses about the determinants of the evolvement of romantic love, including it being a consequence of economic development or the result of transmission of culture. A new large-scale empirical study by Baumard et al.¹ puts these different hypotheses to the test.

The probability of the emergence of SARS-CoV-2 vaccine-resistant variants depends on the number of daily infections permitted by society, and the rate and penetrance of vaccination. Rapidly vaccinating all eligible people while maintaining strict physical distancing measures can prevent the evolution of vaccine resistance.

Developing theories by designing experiments that are aimed at falsifying them is a core endeavour in empirical sciences. By analysing 365 articles dedicated to the study of consciousness, Yaron et al.’s study¹ shows that there is almost no dialogue between the four main theories of this elusive phenomenon and gives us an interactive database with which to probe the literature.

Despite substantial attempts to end the global trade in elephant ivory, illegal transnational networks continue to operate. A new study by Wasser and colleagues¹ uses genetic matches between related elephants to reveal the scale, interconnectedness and audacity of traffickers in illegal ivory.

To make sense of what we hear, the brain must integrate information over time. How is this temporal integration orchestrated in human auditory cortex? A new study by Norman-Haignere and colleagues¹ introduces a promising method to estimate neural integration windows and demonstrates that category-specific versus more generic computations operate on distinct timescales.

It is seemingly in the interest of high-income countries to prioritize vaccinating their own population against COVID-19, despite it being immoral. However, mathematical modelling by Ye et al.¹ shows that this approach offers only limited, short-term benefits, whereas equitable vaccine distribution would substantially curb the emergence and spread of new variants.

Low-carbon innovations in technology and behaviour are increasingly prevalent, but they are not always equitable. This Review examines how such innovations can introduce and perpetrate inequalities, and discusses ways to ensure that a low-carbon future is both sustainable and equitable.

Similarities and differences between deep learning models and primate vision have been the focus of recent research. Audition is comparatively less-studied. A new report describes the emergence of human-like auditory perception in a deep neural network, and suggests a promising way to relate perceptual behaviour to specific aspects of the environment.

To interact safely with our environment, we must be able to judge our confidence in what we perceive. But what cues do we use to compute perceptual confidence? Geurts et al.¹ decode brain activity and show that perceptual confidence is based on the distribution of sensory uncertainty, combining uncertainty driven by the input and the visual system.

Syndemic theory considers how social inequalities drive disease interaction. A new study uses a mixed-methods approach to examine how stress interacts with multiple diseases to affect quality of life in Soweto, South Africa.

Scientific progress depends on researchers updating their beliefs when new evidence arises. McDiarmid and colleagues show that psychologists adjust their beliefs after seeing new results from a replication project. While updating is less than a Bayesian model would justify, it is not undermined by personal investment.

The coming years are likely to see slowing economic growth, which has significant consequences for developed democracies. This Perspective by Burgess et al. considers the implications of slowed growth and proposes a guided civic revival approach to addressing challenges.