Brain pathways for visually-guided defence behaviours

Much of our daily lives, and those of other animals, involves reacting instinctively to events. For example if we see a snake while we are walking in the forest we either freeze or escape. In this project we aim to understand the brain pathways that allow the selection and execution of these instinctive behaviours in animals. Because these pathways emerge early in development, and are shared by humans and other animals, distinguishing the roles of older ("primitive") pathways can be of broad relevance for understanding brain function.

The instinctive reaction to freeze or to escape in the presence of a potential threat is called a defence behaviour. These defence behaviours are not panic. For example, analyses of evacuations during fires show that people do not run randomly but return the way they came. Similarly, animals running away from a potential threat do not run randomly, but straight to a place of refuge if it is available. We have recently discovered that rodents, like humans, use vision to guide their selection of defence behaviour. We found that simply manipulating the visual stimulus that is presented to animals guided them to freeze or escape.

In this project we will use this simple visual manipulation to understand how signals from the eye are used to guide defence behaviours in rodents. We will test the hypothesis that both threat discrimination and defence behaviour are supported by a small area of the mid-brain called the superior colliculus. We will establish how outputs of the eye reach this brain area, how those signals are analysed, and how they are transformed into behavioural outputs. We will test the hypothesis that this brain area works with other navigational guidance systems in the brain to guide behaviour and enable escape to a safe place of refuge. The final outcome will shed light on how the brain links vision to action, in a system likely common to all mammals.

The project is important because evolutionary older pathways are likely to have preserved function in humans, especially for rapid, instinctive behaviours, and may be particularly important in visual functions following cerebral stroke or neonatal brain injury. As well as providing fundamental knowledge about how brains support normal behaviour, the proposed study may therefore shed light on why some visually-guided behaviours are possible in people lacking conscious vision - a phenomenon often known as blindsight. It may also guide the development of interfaces or autonomous devices that need to interact with biological organisms. Finally, it may help the design of the built environment to be informed by an understanding of how we interact with the world in the presence of potential threats.

The aim of this project is to discover how the brain uses vision to guide defence behaviours, a vital set of behaviours that are shared across many species. Defence behaviours are thought to depend on circuits through the mid-brain superior colliculus (SC), but how these circuits discriminate potential threats and organise subsequent behaviour is not known.

We have recently discovered that rodents choose to freeze or escape depending on the nature of a threatening visual stimulus: a small disc moving overhead ('sweep') causes mice to freeze; an expanding disc ('loom') causes mice to rapidly escape to refuge. Our discovery suggests that subcortical circuits are capable of organising a simple sensorimotor transformation - converting vision (sweep vs. loom) into different defence behaviours (freeze vs. escape).

Our specific hypotheses are:
1. Visually-guided escape and freeze can occur independently of cortical vision.
2. Sweep and loom visual stimuli are differentially encoded by sensory pathways in the SC.
3. The SC interacts with navigational guidance systems to direct escape in the direction of a refuge.
4. Neurons in SC link the visual stimuli to both escape and freezing.

We will use behavioural measurements, causal interventions, and measurements of neural activity at key points in the circuit that links visual sensory signals to defence behaviour. The major outcomes are as follows. First, if defence behaviours do not require visual cortical pathways, then this specifies a subcortical circuit. Second, if circuits in SC link different visual stimuli to different behaviours, then this defines a sensorimotor transformation for a behaviour that is conserved across many species. Third, if finding refuge requires guidance systems in the brain, then this implicates the SC in cognitive as well as reflex behaviours. These findings will therefore take us a significant step forward towards understanding how ancient visual pathways guide behaviour.

Academic impact:
Our approach brings together sensory, behavioural and navigational disciplines that have previously had little connection, and therefore offers to be of substantial interest to a large proportion of the neuroscience community. We expect that it is therefore likely to be the subject of invited talks at leading international institutions, for the researcher as well as the primary and co-investigators. To ensure that the proposed work has the widest possible academic impact, we have requested funds to present it at the leading international conferences. In addition, at the end of Year 2 we plan to organise a symposium that brings together research in related fields of conscious and unconscious visual processing, technology, urban design and navigation research. This symposium would aim to disseminate work related to instinctive behaviour to an audience that could find wider applications of the behavioural paradigms.

Societal impact:
Understanding how instinctive behaviours are guided is of clear potential importance to the design of transport, care homes and other public spaces. Because these behaviours recruit ancient subsystems of the mammalian brain, the proposed research offers potential insight into how the brain's alert systems are used to guide behaviour, including whether real threats drive escape or freeze, in humans as well as animals. Academic studies of escape dynamics are already used to inform the design of public spaces, but there has been little assimilation with contemporary neuroscience. To encourage a multi-disciplinary approach with real potential impact on urban design, our proposal includes pathways to collaboration and debate with urban environmental designers and relevant regulatory bodies. In addition, the behavioural experiments are straightforward in design and analysis, and they therefore offer an experimental entry point for undergraduate students. Indeed, a co-author of the paper reporting our initial discovery was a third-year undergraduate student in the psychology programme. By helping us engage with future students in a similar way, this proposal would facilitate understanding of animal experiments among our large undergraduate programme.

Economic and technological impact:
The proposed research offers a clear avenue to economic and technological impact. Recent advances in autonomous systems, including self-driving cars, bring digital and biological organisms into contact during potentially dangerous events. Knowledge of how instinctive escape or freeze behaviours are implemented in biological systems, and what information is used to inform their execution, may be incorporated into digital organisms to avoid escalation of danger to biological organisms, and potentially to reduce that danger. Pathways to developing commercial applications of this knowledge require interest from potential industrial partners. To do this we will start by working with a partner company, whose aim is to provide practical insight to organisations that wish to understand how people interact with the built environment, to translate our discoveries to human behaviours and explore their potential commercial implications.