Binz et al. propose a general framework for meta-learning and contrast it with built-by-hand Bayesian models. We comment on some architectural assumptions of the approach, its relation to the active inference framework, its potential applicability to living systems in general, and the advantages of the latter in addressing the explanation problem.

Gilead et al. propose an ontology of abstract representations based on folk-psychological conceptions of cognitive architecture. There is, however, no evidence that the experience of cognition reveals the architecture of cognition. Scale-free architectural models propose that cognition has the same computational architecture from sub-cellular to whole-organism scales. This scale-free architecture supports representations with diverse functions and levels of abstraction.

Shifting prey distributions due to global warming are expected to generate dramatic ecosystem-wide changes in trophic structure within Arctic marine ecosystems. Yet a relatively poor understanding of contemporary Arctic food webs makes it difficult to predict the consequences of such changes for Arctic predators. Doing so requires quantitative approaches that can track contemporary changes in predator diets through time, using accurate, well-defined methods. Here we use fatty acids (FA) to quantify differences in consumer diet using permutational multivariate analysis of variance tests that characterize spatial and temporal changes in consumer FA signatures. Specifically we explore differences in Greenland shark (Somniosus microcephalus) FA to differentiate their potential trophic role between Svalbard, Norway and Cumberland Sound, Canada. Greenland shark FA signatures revealed significant inter-annual differences, probably driven by varying seal and Greenland halibut responses to environmental conditions such as the NAO, bottom temperature, and annual sea-ice extent. Uncommon FA were also found to play an important role in driving spatial and temporal differences in Greenland shark FA profiles. Our statistical approach should facilitate quantification of changing consumer diets across a range of marine ecosystems.

Recent research has shown that the verbal information and vicarious learning pathways to fear create long term fear cognitions and can create cognitive biases and avoidance in children. However, it is unlikely that these pathways operate in isolation in the aetiology of childhood fear and the interaction between these pathways is untested. Three preliminary experiments are reported that explore the combined effect of verbal threat information and vicarious learning on self-reported fear beliefs in 7–9-year-old children. Results showed that prior negative information significantly facilitated the effect of negative vicarious learning on children's fear beliefs (Experiment 1); however, there was not a significant combined effect of verbal threat information and vicarious learning when they the information was presented during (Experiment 2) or after (Experiment 3) vicarious learning. These results support the idea that verbal information can affect CS-US associations formed in subsequent vicarious learning events, but contradict the proposal that it can change fear beliefs already acquired through vicarious learning by changing how a person evaluates the vicarious learning episode.