To find out more about the podcast go to The Conversation Weekly.
Below is a short summary and detailed review of this podcast written by FutureFactual:
Bees Do Math: Why Ecologically Irrelevant Experiments in Animal Cognition Matter
Bees can be trained to discriminate numbers such as odd versus even, revealing that animal cognition can be explored beyond immediate ecological tasks. In this Conversation Weekly interview, Scarlett Howard explains why ecologically irrelevant experiments matter for understanding intelligence, learning, and evolution, and how these experiments can also inform technology and conservation. She describes how bees are trained to land on reward platforms, how rewards and mild aversion are used, and how researchers control for non-numerical cues to test true arithmetic ability. The discussion also touches on colleagues' reactions, funding realities, and the broader value of basic science for science and society, not just direct human benefits.
Introduction: a scientist pushing the boundaries of research
Scarlett Howard, a Monash University lecturer and bee cognition researcher, discusses a blended program of ecologically relevant work and what she calls ecologically irrelevant experiments. Her plea is that studying numerosity and problem solving in bees can illuminate general principles of intelligence, learning, and evolution, as well as offer inspiration for bio-inspired technology. The conversation situates her bee research within a broader scientific curiosity about what animals can learn when the ecological demands of foraging and pollination are not the sole focus.
"they were relevant to other areas of science and also in our understanding of the evolution of intelligence" - Dr Scarlett Howard
What makes an experiment ecologically irrelevant and why it matters
Howard defines ecologically irrelevant experiments as those designed and interpreted without clear links to the life history or ecology of the species under study. Even when a bee is free to fly and visit a lab setup, a task that asks it to categorize odd versus even numbers is not obviously connected to its natural foraging decisions. Yet she argues that such studies can yield generalizable insights about cognition, learning strategies, and how brains solve problems, which can drive advances in neuromorphic computing and our understanding of the evolution of intelligence. The boundary between relevant and irrelevant is fluid; a finding may later reveal ecological significance under environmental change or in combination with other tasks.
"we do it powered by a sandwich, and bees do it powered by a drop of nectar" - Dr Scarlett Howard
Bees as a model and what it takes to teach them math
The researcher describes how honey bee colonies are used, with hives placed in fields or reserves and sugar feeders that recruit bees back to the lab. Stimuli such as printed cards showing numbers or shapes are presented on a circular screen; the bees are trained to land on the correct platform to receive sugar water, while wrong choices are followed by a bitter taste. After initial learning, researchers test what the bees have learned by removing rewards to ensure that their choices reflect genuine numerical discrimination rather than simple reward seeking. Variation among individual bees matters: some learn quickly, others slowly or not at all, which can be advantageous for a colony facing changing environments.
"each bee learned differently, and there was no singular aha moment" - Dr Scarlett Howard
Individual variation and the lack of a single aha moment
Howard details how her team tracks individual bees, marks and follows them, and randomizes stimuli to prevent cueing. She notes that the absence of a population-wide aha moment challenges the idea that cognition should be understood as a uniform capability. The findings highlight adaptive learning strategies, where some bees excel in certain tasks while others respond differently, providing flexibility for the hive in the face of environmental variability.
"There is a lot of value to these seemingly ecologically relevant experiments and, I think they're quite important to science overall" - Dr Scarlett Howard
Broader implications, funding, and how to talk about this research
Howard argues that even ecologically irrelevant cognition studies have relevance for science and society, including potential benefits for technology and our understanding of evolution. She emphasizes that not every project needs to be immediately tied to a human problem, and that such research can be a testbed for how cognition works under different conditions, such as temperature changes or habitat loss. The discussion also touches on funding pressures, the human tendency to demand direct practical outcomes, and the idea that basic science can yield long-term, unforeseen benefits.
"Just because something might not have relevance to ecology, it doesn't mean there's no relevance to science or society as a whole" - Dr Scarlett Howard