• 17:30 - 18:30
• Bristol-Myers Squibb Lecture Theatre, Yusuf Hamied Department of Chemistry

Viruses generally get a very bad press. Deservedly so. Viruses cause serious epidemics and pandemics in humans and threaten food security by causing crop and livestock diseases. However, of the more than 10³⁰ virus particles that exist at any instant, few endanger us, plants, or animals. Most infect bacteria and archaea and without these viruses Earth’s biosphere would collapse.

Viruses are the largest group of emerging plant pathogens. But paradoxically, many confer benefits on plants that may increase survival and reproductive success. These apparently mutualistic effects may be conditional, for example, by aiding host survival under conditions of drought or cold or, more surprisingly, by helping plants attract pollinators.

Certain viruses, notably ‘partitiviruses’ and ‘endornaviruses’ may have travelled so far on an evolutionary path as to become obligate plant symbionts. These viruses are transmitted exclusively via seed and pollen and are not infectious in a conventional sense (i.e., not transmitted by invertebrates, wounding etc.), and induce no obvious disease. It has been suggested that these ‘persistent’ viruses are mutualists and might be useful in crop breeding. Recent experiments exploring these ideas will be discussed.

• 17:30 - 18:30
• Bristol-Myers Squibb Lecture Theatre, Yusuf Hamied Department of Chemistry

For decades, psychologists have characterised young children as little scientists, a comparison made famous by the 1999 book The Scientist in the Crib. A core claim of this perspective is that children (and other animals such as chimpanzees, dogs, and birds) use abstract theories to navigate the world, much like researchers discovering and applying general laws.

This idea is striking when contrasted with contemporary philosophy of science. Within this discipline, there is a consensus that scientists rarely understand the world through universal laws; instead, they operate in a piecemeal, pluralistic way. The world is tangled and dappled. To successfully predict and explain it, scientists often adopt a ‘case-worker’ approach: sifting through a collection of models to determine which one applies here and now. If scientific success relies on this case-worker strategy rather than through the application of general laws, then we should reassess how children and animals learn to navigate the causal world.

In this talk, I develop a case-worker account of how we navigate our everyday physical environments. I draw on embodied and predictive processing accounts of cognition and action to do so. Under this approach, generative models affect how an agent probes the world through actions such as touch or moving their eyes. These models are, in turn, continuously selected and adjusted based on contextual sensory information. Ultimately, this approach shows that successful causal understanding need not appeal to general laws. We do not need to reject the idea of the child as a scientist, but we do need to update our notion of what this means.