Flexible perception: functional plasticity mechanisms in the human brain

Breaking the camouflage of a nearby object allows a casual rambler to detect a snake in the grass, and a more experienced hiker to determine whether it is a venomous European Adder or harmless grass snake. The visual processes of (1) detecting and segmenting targets from cluttered backgrounds, and (2) discriminating whether similar features belong to the same or different objects are central to many everyday visual activities. Practice in these tasks is known to make us better: training and experience improve our core skills in visual recognition. Yet the functional brain architecture that supports these processes, and their plasticity in normal and abnormal function, is poorly understood.
Here we exploit recent technological advances in magnetic resonance imaging to trace how the brain changes with learning at much finer resolution than previously possible. We focus on the visual cortex that is known to process information registered by the eyes to extract meaningful patterns. In particular, imaging using 7 Tesla magnets allows us to separate middle from deep layers of the visual cortex. Middle layers are known to receive inputs from the eyes and process simple object properties (e.g. position, orientation). In contrast, deep layers are thought to receive signals from higher brain centres that assign meaning to sensory information and allow us to interpret complex scenes and make decisions. The key challenge is to understand whether learning changes our judgements by modifying the processing of simple visual features as they enter the brain from the eyes, or whether it acts at later stages that interpret ambiguous sensory patterns and assign objects into meaningful categories. Changes in brain patterns deep in the cortex would suggest that learning boosts our perceptual skills by altering the way sensory information is interpreted at later-rather than earlier-stages of processing.
We then test whether these changes in the brain's responses produced by training relate to changes in the brain's neurochemistry. We will exploit advances in MR imaging of metabolites to measure GABA, the primary neurotransmitter that the brain uses for suppressing rather than exciting its neurons. GABA has been shown to play a critical role in human brain development. Yet, its role in boosting perception through training and facilitating judgements remains largely unexplored. Here, we will trace how GABA changes during training and whether these changes relate to our ability to improve perceptual skills through training. We will also trace whether these changes in the brain's neurochemistry link with changes in brain function across areas in the visual brain that are involved in processing and interpreting sensory information. These studies will allow us to understand how learning alters the balance in the brain's chemical signals (excitation vs. inhibition) to boost the brain's flexibility and capacity to perform in everyday perceptual tasks.
Taken together, exploiting technological advances that allow us to trace how the human brain changes through learning and experience in a non-invasive manner is key for understanding the brain's capacity for plasticity and flexible behaviour. Understanding the brain's basic mechanisms for learning is critical for developing lifelong learning programmes that target speciliased circuits to boost performance through training in early years education but also across the lifespan.

The human's brain capacity for sensory plasticity has been studied mainly in the context of neurodevelopment (i.e. critical periods) and pathology (e.g. amblyopia) with interventional approaches (e.g. sensory deprivation) that result in drastic brain re-organisation. Yet, understanding the brain plasticity mechanisms that mediate subtler changes in perceptual judgments through shorter-term experience and training remains a challenge.
Here we focus on two perceptual skills at the core of visual recognition: the ability to detect the features of an object from cluttered backgrounds and discriminate whether they belong to the same or different objects. Learning and experience have been suggested to facilitate this ability to translate complex patterns of visual information into perceptual decisions. We will exploit methodological advances in high-field (7T) brain imaging to investigate the functional and neurochemical brain plasticity mechanisms that boost our ability to make perceptual judgments. We will test the hypothesis that perceptual learning is implemented by feedback and inhibitory mechanisms that re-weight sensory information across stages of processing (from early to higher visual cortex). In particular, the high resolution of 7T imaging allows us to measure functional signals in different cortical layers. We will test whether learning alters fMRI activation patters in deep-rather than middle-layers in the visual cortex, consistent with feedback processing. Further, advances in MR Spectroscopy enable us to test the role of GABA-the primary inhibitory neurotransmitter for brain plasticity-in perceptual learning. We will test whether learning-dependent changes in GABA relate to changes in functional brain activity and improved behavioural performance in perceptual tasks. Investigating these core mechanisms of brain plasticity will advance our understanding of how the brain optimises its capacity to support adaptive behaviour through learning and experience.

Who might benefit?
The main non-academic beneficiaries who may benefit from the research are educators, professionals from health and social services and policy makers. We will exploit opportunities for knowledge transfer to local authorities (e.g. schools, healthcare services) and companies that develop i) MR technology for applications in healthcare, ii) software for advanced analysis of high-field imaging data, iii) software for cognitive assessment and training tools.

How might they benefit?
Our findings will advance our understanding of the link between learning and human brain plasticity. Our work emphasises the potential for neural flexibility and high cognitive performance in adulthood, highlighting the importance of lifelong learning. Promoting lifelong learning is important for training and re-training later in life and cognitive remediation in healthy ageing or disease (e.g. neurodevelopmental, and neurodegenerative disorders). Thus, there are potential implications for diagnosis and training in education and healthcare as well as wider societal benefits through impact on public policy.

While the proposal is squarely aimed at basic research, there is potential for knowledge transfer in the following domains:
1. Design of screening and training tools of learning ability across the life span in health and disease. Understanding the fundamental brain mechanisms that mediate learning ability is critical for developing: i) diagnostic tools that are more sensitive than standard scales, ii) training programmes that target specific neural functions and have potential for remediation. In this context and with seed funding from the Impact Acceleration Account, we are pursuing a partnership with companies that develop user-friendly screening and intervention tools (i.e. Cambridge Cognition, Peak).
2. Implementation of MR technology to healthcare applications focusing on neuroimaging. Protocols for high-field neuroimaging of cognitive function and brain plasticity have the potential to serve as the basis for the development of advanced diagnostic MRI tools in healthcare (e.g. neurosurgery, brain repair). Our current developments on MR Spectroscopy have informed diagnostic imaging tools in disease (e.g. Motor Neuron Disease) and large-scale clinical trials on stroke rehabilitation. Our collaboration with Cambridge's NIHR Biomedical Research Centre and Siemens (MR Technology) will facilitate further the clinical translation of these technological developments.
3. Development of advanced computational tools for the analysis of high-field neuroimaging data. The collaboration with Brain Innovation provides an avenue for healthcare applications. In particular, advancing user-friendly analysis tools for neuroimaging will facilitate their application to clinical practice. This has the potential to result in more detailed and precise diagnosis based on high-resolution data that capture not only brain structure but also functional capacity.
4. Insights gained from the proposed research would be expected to lead to optimisation of learning interventions across the lifespan. In particular, UK Government policy is firmly focussed on improving lifelong educational programmes, as UK populations are growing older. This potential for translation of our basic science findings into wider societal benefits has been recognised through our previous work that has contributed to shaping UK government policy (Foresight Project: Future of Skills and Lifelong Learning).