Multimodal imaging investigation of the brain regions supporting different stages of human memory

Lead Research Organisation: University of Cambridge
Department Name: Psychology

Abstract

We generally take it for granted that we can remember people or objects that we have encountered before, but it is thought that this everyday function of retrieving information from our store of memories may involve a number of separate stages. Which brain areas might be involved in these different stages is a question that has not yet received a satisfactory answer, largely because of limitations in the techniques that have traditionally been used. Recent advances allow us to combine information from different techniques (specifically, functional magnetic resonance imaging or fMRI, and magnetoencephalography or MEG) to identify more precisely where in the brain, and when, different cognitive processes take place. It is proposed to use these techniques in experiments that involve healthy volunteers to determine the networks of brain areas that work together to support the retrieval of information from memory. By observing patterns of brain activity during the different stages of memory retrieval (e.g., processing occurring prior to an attempt to search memory, or the checking of information that has been retrieved from memory), we can improve our understanding of which aspects of memory function are impaired, and which may be spared, in patients with brain dysfunction.

Technical Summary

The primary aim of the proposed research is to identify the brain networks that are engaged during the different cognitive processing stages of retrieving information from long-term memory. It is proposed that the operating dynamics of these networks can be established over a millisecond timescale using a multimodal imaging approach that combines functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) with recent advances in analysis of effective connectivity. This proposal will address a key issue in cognitive neuroscience that has remained unresolved because previous studies have limitations that make it difficult to determine which region of the brain is important for which stage of memory retrieval. Combining the multimodal neuroimaging techniques of fMRI and MEG will provide an unprecedented spatial and temporal characterisation of the brain regions such as prefrontal cortex that are involved in different stages of retrieval, including the preparatory processing occurring prior to a retrieval attempt and the post-retrieval monitoring and evaluation of recovered information. Further analyses examining the effective connectivity between brain regions will reveal the functional networks, both within prefrontal cortex and including other areas like the medial temporal lobe and parietal cortex, that enable successful operation of these processes.

Publications

10 25 50

publication icon
Brandt VC (2014) Did I turn off the gas? Reality monitoring of everyday actions. in Cognitive, affective & behavioral neuroscience

publication icon
Buda M (2011) A specific brain structural basis for individual differences in reality monitoring. in The Journal of neuroscience : the official journal of the Society for Neuroscience

publication icon
Dobbins IG (2012) Use of explicit memory cues following parietal lobe lesions. in Neuropsychologia

publication icon
Fornito A (2012) Competitive and cooperative dynamics of large-scale brain functional networks supporting recollection. in Proceedings of the National Academy of Sciences of the United States of America

 
Description The project sought to characterise the strategic processes involved in different stages of long-term memory retrieval by combining the spatial resolution of functional magnetic resonance imaging (fMRI) with the temporal resolution of electroencephalography and magnetoencephalography (EEG/MEG). In the first year of the project, a post-doctoral research associate was recruited (Dr Zara Bergström) and, in collaboration with Dr Richard Henson at the nearby MRC Cognition and Brain Sciences Unit, Dr Bergström and I developed methods to address the considerable technical challenges of integrating the complementary strengths of fMRI and source-localised EEG/MEG.



The multimodal imaging approach taken involved collecting fMRI and EEG/MEG data from the same participants on two separate occasions while they undertook a recollection task that involved retrieving different types of source information. We then used the advanced EEG/MEG distributed source localisation techniques we developed to look for convergence between independent fMRI and EEG/MEG localisation. Notable consistency was observed between the frontal and parietal brain regions identified with each imaging modality, providing novel evidence about how these regions interact over time to support source recollection. These results were replicated and extended in a subsequent experiment which found that activity in some of the identified regions is dependent on the type or the number of memory features that have to be evaluated. These findings have been written up as part of two published journal articles so far, including a substantial paper that describes in detail the methodological approach developed during the course of the grant.



A second strand of research involved applying the statistical method of spatial independent component analysis (ICA) to the neuroimaging data to characterise interactions between brain regions and identify the operating dynamics of large-scale functional brain networks that support recollection. The findings demonstrated that increased cooperation between two frontal-parietal systems, the default mode network (DMN) and the external attention system (EAS), is associated with more rapid recollection. These cooperative interactions were shown to be facilitated by a dynamic reconfiguration of components of the DMN to enhance integration with prefrontal and parietal regions. These results shed new light on the contribution of these well-established large-scale brain networks to memory, and were published in a high-impact journal. Similar ICA methods applied to our EEG data to examine the separable influence of automatic and controlled processes on recollection have also recently been published.



In addition, we capitalized on the MRI data collected during the project to investigate the structural brain basis of individual differences observed in performance on our recollection tasks. We identified that variability in a prefrontal cortex structural landmark, the paracingulate sulcus (PCS), might be important. The PCS is one of the last structural folds to develop before birth and, for this reason, varies greatly in size between individuals in the general population. We found that healthy, apparently cognitively intact, adults whose MRI scans indicated bilateral absence of the PCS were significantly less accurate on our recollection tasks than people with a prominent PCS on at least one side of the brain. These findings generated considerable media interest when they were published.
Exploitation Route Improving our understanding of which aspects of memory function are impaired, and which may be spared, in patients with brain dysfunction could inform the development of rehabilitation strategies that are targeted at the cognitive abilities that remain functional in these patients, thus contributing to improving their quality of life. The findings from this project have clinical implications for patients with memory difficulties, such as those with amnesia, in whom it has traditionally been difficult to isolate which of the many processes that support memory are affected. Because technical limitations have meant that acquiring multimodal neuroimaging evidence has not been possible before, the research findings can inform efforts to distinguish both where in the brain and when in time specific memory impairments have their impact. Other ways in which the research can be exploited are to provide fresh spatiotemporal constraints on the development of cognitive and neural models of human memory, incorporating novel insights into how different stages of memory retrieval are instantiated in the brain.



More generally, the project has helped contribute to the development of sophisticated multimodal imaging analysis methods which will be capitalized on by neuroimaging researchers in other domains.
Sectors Healthcare

URL http://www.memlab.psychol.cam.ac.uk/
 
Description Brain memory finding may help schizophrenia research 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Reuters news article on our research

no actual impacts realised to date
Year(s) Of Engagement Activity 2011
 
Description Keeping track of reality 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach local
Primary Audience Media (as a channel to the public)
Results and Impact University press release describing new study of the brain explaining why some of us are better than others at remembering what really happened.

no actual impacts realised to date
Year(s) Of Engagement Activity 2011
 
Description Memory and the brain 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Public lecture to accompany 19th Experimental Psychology Society Prize

no actual impacts realised to date
Year(s) Of Engagement Activity 2012
 
Description Memory errors are all in the groove 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Article about our research in the Guardian

no actual impacts realised to date
Year(s) Of Engagement Activity 2011
URL http://www.theguardian.com/science/neurophilosophy/2011/oct/07/1
 
Description Why you shouldn't blame your colleagues for stealing your ideas 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Daily Mail news article on our research

no actual impacts realised to date
Year(s) Of Engagement Activity 2011