We are encouraged by the many positive commentaries on our target article. In this response, we recapitulate some of the points raised and identify synergies between them. We have arranged our response based on the tension between data and architecture that arises in the meta-learning framework. We additionally provide a short discussion that touches upon connections to foundation models.

Psychologists and neuroscientists extensively rely on computational models for studying and analyzing the human mind. Traditionally, such computational models have been hand-designed by expert researchers. Two prominent examples are cognitive architectures and Bayesian models of cognition. Although the former requires the specification of a fixed set of computational structures and a definition of how these structures interact with each other, the latter necessitates the commitment to a particular prior and a likelihood function that – in combination with Bayes' rule – determine the model's behavior. In recent years, a new framework has established itself as a promising tool for building models of human cognition: the framework of meta-learning. In contrast to the previously mentioned model classes, meta-learned models acquire their inductive biases from experience, that is, by repeatedly interacting with an environment. However, a coherent research program around meta-learned models of cognition is still missing to date. The purpose of this article is to synthesize previous work in this field and establish such a research program. We accomplish this by pointing out that meta-learning can be used to construct Bayes-optimal learning algorithms, allowing us to draw strong connections to the rational analysis of cognition. We then discuss several advantages of the meta-learning framework over traditional methods and reexamine prior work in the context of these new insights.

We introduce the Berlin Numeracy Test, a new psychometrically sound instrument that quickly assesses statistical numeracy and risk literacy. We present 21 studies (n=5336) showing robust psychometric discriminability across 15 countries (e.g., Germany, Pakistan, Japan, USA) and diverse samples (e.g., medical professionals, general populations, Mechanical Turk web panels). Analyses demonstrate desirable patterns of convergent validity (e.g., numeracy, general cognitive abilities), discriminant validity (e.g., personality, motivation), and criterion validity (e.g., numerical and non-numerical questions about risk). The Berlin Numeracy Test was found to be the strongest predictor of comprehension of everyday risks (e.g., evaluating claims about products and treatments; interpreting forecasts), doubling the predictive power of other numeracy instruments and accounting for unique variance beyond other cognitive tests (e.g., cognitive reflection, working memory, intelligence). The Berlin Numeracy Test typically takes about three minutes to complete and is available in multiple languages and formats, including a computer adaptive test that automatically scores and reports data to researchers (http://www.riskliteracy.org). The online forum also provides interactive content for public outreach and education, and offers a recommendation system for test format selection. Discussion centers on construct validity of numeracy for risk literacy, underlying cognitive mechanisms, and applications in adaptive decision support.

Different treatment options for pulmonary hypertension have emerged in recent years, and evidence-based management strategies have improved quality of life and survival in adults. In children with pulmonary vascular disease, therapeutic algorithms are not so clearly defined; this study determined current treatment initiation in children with pulmonary hypertension in participating centres of a registry.

Through the multinational Tracking Outcomes and Practice in Pediatric Pulmonary Hypertension registry, patient demographics, diagnosis, and treatment as judged and executed by the local physician were collected. Inclusion criteria were >3 months and <18 years of age and diagnostic cardiac catheterisation consistent with pulmonary hypertension (mean pulmonary arterial pressure ⩾25 mmHg, pulmonary vascular resistance index ⩾3 Wood units×m2, and mean pulmonary capillary wedge pressure ⩽12 mmHg).

At diagnostic catheterisation, 217/244 patients (88.9%) were treatment naïve for pulmonary hypertension-targeted therapy. Targeted therapy was initiated after catheterisation in 170 (78.3%) treatment-naïve patients. A total of 19 patients received supportive therapy, 28 patients were not started on therapy, and 26 patients (10.7%) were on targeted treatment before catheterisation. Among treatment-naïve subjects, treatment was initiated with one targeted drug (n=112, 51.6%), dual therapy (n=39, 18%) or triple-therapy (n=5, 2.3%), and calcium channel blockers with one targeted medication in one patient (0.5%). Phosphodiesterase inhibitors type 5 were used frequently; some patients with pulmonary hypertension related to lung disease received targeted therapy.

There is a diverse therapeutic approach for children with pulmonary hypertension with a need of better-defined treatment algorithms based on paediatric consensus for different aetiologies including the best possible diagnostic workup.