Nature Human Behaviour最新文献

We study the relationship between the Fridays for Future climate protest movement in Germany and citizen political behaviour. In 2019, crowds of young protesters, mostly under voting age, demanded immediate climate action. Exploiting cell-phone-based mobility data and hand-collected information on nearly 4,000 climate protests, we created a highly disaggregated measure of protest participation. Using this measure, we show that Green Party vote shares increased more in counties with higher protest participation (n = 960). To address the possibility of non-random protest participation, we used various empirical strategies. When we examined mechanisms, we found evidence for three relevant factors: reverse intergenerational transmission of pro-environmental attitudes from children to parents (n = 76,563), stronger climate-related social media presence by Green Party politicians (n = 197,830) and increased local media coverage of environmental issues (n = 47,060). Our findings suggest that youth protests may initiate the societal change needed to overcome the climate crisis.

Variation in the efficiency of extracting calorie-rich and nutrient-dense resources directly impacts energy expenditure and potentially has important repercussions for cultural transmission where social learning strategies are used. Assessing variation in efficiency is key to understanding the evolution of complex behavioural traits in primates. Here we examine evidence for individual-level differences beyond age- and sex-class in non-human primate extractive foraging efficiency. We used 25 years (1992–2017) of video of 21 chimpanzees aged ≥6 years in Bossou, Guinea, to longitudinally investigate individual-level differences in stone tool use efficiency. Data from 3,882 oil-palm nut-cracking bouts from >800 h of observation were collected. We found reliability in relative efficiency across four measures of nut-cracking efficiency, as well as a significant effect of age. Our findings highlight the importance of longitudinal data from long-term field sites when investigating underlying cognitive and behavioural diversity across individual lifespans and between populations.

Decisions form a central bottleneck to most tasks, one that people often experience as costly. Previous work proposes mitigating those costs by lowering one’s threshold for deciding. Here we test an alternative solution, one that targets the basis of most choice costs: the idea that choosing one option sacrifices others (mutual exclusivity). Across 6 studies (N = 565), we test whether this tension can be relieved by framing choices as inclusive (allowing selection of more than 1 option, as in buffets). We find that inclusivity makes choices more efficient by selectively reducing competition between potential responses as participants accumulate information for each of their options. Inclusivity also made participants feel less conflicted, especially when they could not decide which good option to keep or which bad option to get rid of. These inclusivity benefits were also distinguishable from the effects of manipulating decision threshold (increased urgency), which improved choices but not experiences thereof.

Replications are important for assessing the reliability of published findings. However, they are costly, and it is infeasible to replicate everything. Accurate, fast, lower-cost alternatives such as eliciting predictions could accelerate assessment for rapid policy implementation in a crisis and help guide a more efficient allocation of scarce replication resources. We elicited judgements from participants on 100 claims from preprints about an emerging area of research (COVID-19 pandemic) using an interactive structured elicitation protocol, and we conducted 29 new high-powered replications. After interacting with their peers, participant groups with lower task expertise (‘beginners’) updated their estimates and confidence in their judgements significantly more than groups with greater task expertise (‘experienced’). For experienced individuals, the average accuracy was 0.57 (95% CI: [0.53, 0.61]) after interaction, and they correctly classified 61% of claims; beginners’ average accuracy was 0.58 (95% CI: [0.54, 0.62]), correctly classifying 69% of claims. The difference in accuracy between groups was not statistically significant and their judgements on the full set of claims were correlated (r(98) = 0.48, P < 0.001). These results suggest that both beginners and more-experienced participants using a structured process have some ability to make better-than-chance predictions about the reliability of ‘fast science’ under conditions of high uncertainty. However, given the importance of such assessments for making evidence-based critical decisions in a crisis, more research is required to understand who the right experts in forecasting replicability are and how their judgements ought to be elicited.

Artificial intelligence (AI) technologies are rapidly advancing, enhancing human capabilities across various fields spanning from finance to medicine. Despite their numerous advantages, AI systems can exhibit biased judgements in domains ranging from perception to emotion. Here, in a series of experiments (n = 1,401 participants), we reveal a feedback loop where human–AI interactions alter processes underlying human perceptual, emotional and social judgements, subsequently amplifying biases in humans. This amplification is significantly greater than that observed in interactions between humans, due to both the tendency of AI systems to amplify biases and the way humans perceive AI systems. Participants are often unaware of the extent of the AI’s influence, rendering them more susceptible to it. These findings uncover a mechanism wherein AI systems amplify biases, which are further internalized by humans, triggering a snowball effect where small errors in judgement escalate into much larger ones.

Although the use of short self-report measures is common practice in biobank initiatives, such a phenotyping strategy is inherently prone to reporting errors. To explore challenges related to self-report errors, we first derived a reporting error score in the UK Biobank (UKBB; n = 73,127), capturing inconsistent self-reporting in time-invariant phenotypes across multiple measurement occasions. We then performed genome-wide scans on the reporting error score, applied downstream analyses (linkage disequilibrium score regression and Mendelian randomization) and compared its properties to the UKBB participation propensity. Finally, we improved phenotype resolution for 24 measures and inspected the changes in genomic findings. We found that reporting error was present across all 33 assessed self-report measures, with repeatability levels as low as 47% (childhood body size). Reporting error was not independent from UKBB participation, evidenced by the negative genetic correlation between the two outcomes (r_g = −0.77), their shared causes (for example, education) and the loss in self-report accuracy following participation bias correction. Across all analyses, the impact of reporting error ranged from reduced power (for example, for gene discovery) to biased estimates (for example, if present in the exposure variable) and attenuation of genome-wide quantities (for example, 21% relative attenuation in SNP heritability for childhood height). Our findings highlight that both self-report accuracy and selective participation are competing biases and sources of poor reproducibility for biobank-scale research.