Periodic Reporting for period 3 - EVENTS (MAKING SENSE OF THE WORLD: COGNITIVE AND NEURAL PROCESSES UNDERPINNING HOW WE COMPREHEND, PREDICT AND REMEMBER EVENTS)
Reporting period: 2022-10-01 to 2024-03-31
During our waking lives we are continuously exposed to a vast amount of information about the world around us. Yet somehow we make sense of this information and we consciously experience a coherent and ordered world, where life proceeds in a sequence of events with recognisable beginnings and ends. How the human mind manages to re-process continuous experience into these event-units is remarkably poorly understood. To date, the field has been held back by the significant methodological challenges to studying complex mental processes operating in naturalistic situations. The EVENTS project seeks to address these challenges in an ambitious and interdisciplinary programme of research, involving behavioural studies (including immersive virtual reality), cutting-edge functional MRI and neuropsychology in specialised populations. Across a series of studies, EVENTS aims to establish how information processed in independent neural modules is combined within a mental “event model”, which is an overarching representation of the important features of any given situation. The project is discovering how event models are updated and how they are instantiated in the brain. EVENTS is also defining how event models shape our perception and memory of everyday situations and how they interact with stored knowledge. Finally, we will integrate these novel findings with previous disparate lines of evidence into a neurocognitive model of event processing. The knowledge generated by EVENTS will have far-reaching impact across the social, cognitive and neuro- sciences, shedding light on long-standing debates about how we internally represent the external world, how beliefs about the state of the world interact with how we perceive and remember events, and on how we perceive the passage of time. Moreover, the development of a detailed cognitive and neural model of event processing will represent a vital step towards a mechanistic account of conscious experience.
This mid-term report overs the first 30 months of a 5 year programme of work which aims to address all of the goals of the EVENTS project. The main methods of the project involve cognitive testing (in healthy adults and adults with brain disease), functional MRI scanning and computational modelling. This mixed methods approach capitalizes on the strength of each, with the aim that conclusions made on the basis of one can be tested on another. In this way, we will provide a comprehensive and generalizable account of the psychological and neural processes that underpin event cognition. In line with the planned evolution of the work over the course of the project, we have already conducted extensive cognitive testing in healthy participants and several fMRI studies. One study involving adults with brain disease is well underway and another has ethical approval pending. Computational modelling, inspired by the new behavioral findings, will be undertaken in the next phase of the project.
The PI published a review of event memory which summarized the state of the art and provided a theoretical framework for understanding event processing. They also organised and presented at a symposium at an international conference on memory about how knowledge affects how we perceive and remember events.
The first major aim of EVENTS was to define what the key dimensions of events are and to establish how they are represented in the brain. In an fMRI study we demonstrated that familiar and expected elements of an event dominate its representation in the brain. This reveals a fundamental property of how brains tend to use predictable information to “scaffold” our neural representation of an event and thereby enable the retrieval of more idiosyncratic elements. In another study we showed that events can be demarcated by changes in the goals of people present. A third study investigated the role of surprise in determining how well events are remembered and whether observing a surprising action triggers the neural markers commonly seen at event boundaries.
The second aim of EVENTS was to characterize how knowledge influences how events are perceived as they unfold. In a study that involved a large-scale online behavioral study and an fMRI study, we showed that knowledge constrains how people activate relevant conceptual knowledge in order to interpret a narrative story-line appropriately. We used Google’s Universal Sentence Encoder to calculate text-based measures of similarity in people’s recalled memories as well as cutting-edge fMRI analyses to both show that individuals who share the same prior knowledge of an event activate conceptual knowledge – as evidenced both neurally and behaviorally – in a more similar way to those who do not. In a separate line of inquiry, we are conducting two neuropsychological studies to establish how individuals with memory problems are able to comprehend and immediately recall short narrative events.
The EVENTS project has so far resulted in 5 research papers, one review paper and 10 presentations at national or international scientific meetings. Publications arising from the first 30 months of the project have already been cited over 30 times. While each publication is important, it is the collective body of evidence from bespoke paradigms that will really enable the EVENTS project to have a major impact on our understanding of event cognition. Outreach events have included presenting our research to MP’s at the UK Houses of Parliament, as well as public events at the University of Sussex. In the next half of the project we will build on the progress we have made so far, continuing to focus on the two aims above but also addressing our third and final aim, to develop a comprehensive neurocognitive model of event cognition.
The PI published a review of event memory which summarized the state of the art and provided a theoretical framework for understanding event processing. They also organised and presented at a symposium at an international conference on memory about how knowledge affects how we perceive and remember events.
The first major aim of EVENTS was to define what the key dimensions of events are and to establish how they are represented in the brain. In an fMRI study we demonstrated that familiar and expected elements of an event dominate its representation in the brain. This reveals a fundamental property of how brains tend to use predictable information to “scaffold” our neural representation of an event and thereby enable the retrieval of more idiosyncratic elements. In another study we showed that events can be demarcated by changes in the goals of people present. A third study investigated the role of surprise in determining how well events are remembered and whether observing a surprising action triggers the neural markers commonly seen at event boundaries.
The second aim of EVENTS was to characterize how knowledge influences how events are perceived as they unfold. In a study that involved a large-scale online behavioral study and an fMRI study, we showed that knowledge constrains how people activate relevant conceptual knowledge in order to interpret a narrative story-line appropriately. We used Google’s Universal Sentence Encoder to calculate text-based measures of similarity in people’s recalled memories as well as cutting-edge fMRI analyses to both show that individuals who share the same prior knowledge of an event activate conceptual knowledge – as evidenced both neurally and behaviorally – in a more similar way to those who do not. In a separate line of inquiry, we are conducting two neuropsychological studies to establish how individuals with memory problems are able to comprehend and immediately recall short narrative events.
The EVENTS project has so far resulted in 5 research papers, one review paper and 10 presentations at national or international scientific meetings. Publications arising from the first 30 months of the project have already been cited over 30 times. While each publication is important, it is the collective body of evidence from bespoke paradigms that will really enable the EVENTS project to have a major impact on our understanding of event cognition. Outreach events have included presenting our research to MP’s at the UK Houses of Parliament, as well as public events at the University of Sussex. In the next half of the project we will build on the progress we have made so far, continuing to focus on the two aims above but also addressing our third and final aim, to develop a comprehensive neurocognitive model of event cognition.
The EVENTS project has already made important progress in establishing how everyday experiences are understood at the time and remembered later. Publications arising from the first 30 months of the project have already been cited over 30 times. While each publication is important, it is the collective body of evidence from bespoke paradigms that will really enable the EVENTS project to have a major impact on our understanding of event cognition.
The COVID-19 pandemic forced us to rapidly develop novel methods for online testing of memory for events. This has presented two major challenges: first, how to present complex stimuli such as movie clips online, and second, how to record responses from participants, such as locations on a screen, frames of a movie, or detailed descriptions of events. We have successfully tackled these challenges and our laboratory website now hosts examples of studies that have used all of these methods. We have collected data from around 2000 participants. We have already shared our expertise in online testing using complex bespoke tasks with research collaborators. Upon completion of our studies, we will make all code used to carry out online testing freely available.
The COVID-19 pandemic forced us to rapidly develop novel methods for online testing of memory for events. This has presented two major challenges: first, how to present complex stimuli such as movie clips online, and second, how to record responses from participants, such as locations on a screen, frames of a movie, or detailed descriptions of events. We have successfully tackled these challenges and our laboratory website now hosts examples of studies that have used all of these methods. We have collected data from around 2000 participants. We have already shared our expertise in online testing using complex bespoke tasks with research collaborators. Upon completion of our studies, we will make all code used to carry out online testing freely available.