Cognitive Control: Bridging individual and collective cognition to drive preference and choice in invasive ants

The ultimate aim of this project is to gain fundamental insights into collective cognition and apply them to the emerging global challenge of invasive animal control. Invasive ants are ecologically devastating, economically damaging, and almost impossible to control. Ants are protected physically within their nest and behaviourally by social immunity. However, their cognitive abilities are almost universally ignored, and offer novel angles of attack. Applying behavioural economic and psychological concepts, I will open the new field of Cognitive Control of invasive animals.

We will start by gaining fundamental insights into individual Argentine ant (Linepithema humile) learning and decision-making. Then, individual choice will be steered using behavioural economic and cognitive interventions. Psychological effects, such as conditioned taste aversion, which may cripple current alien species management, will be tested and overcome. Finally, we will use neuroactives (e.g. caffeine) to improve learning and manipulate preference.

We will then I will take the individual-level manipulations on to the colony level, in order to gain deep insights into collective cognition. Experimental manipulation of collective choice using cognition-based methods are very limited. The few available studies all focus on species which use linear recruitment, where signal effectiveness is proportional to signal quantity. By contrast, all major invasive ant species use non-linear mass recruitment. Non-linear systems are expected to behave much more like individuals than linear systems when challenged with cognitive manipulations, but this prediction has never been tested. We will provide the first empirical tests of such predictions: Using both cognitive and chemical manipulations, we will attempt to steer preference and increase consumption of target food sources by whole insect colonies.

The PhD student will carry out learning assays on individual ants. Then, they will develop a high-throughput system for testing collective decision-making on multiple ant colonies simultaneously. Using image and video analysis, they will develop tools to automatically collect data from these parallel experiments. Alternatively, in collaboration with colleagues, we use advanced machine vision techniques to trace the network structure of ant colonies, and attempt to influence it.