Periodic Reporting for period 3 - AgeConsolidate (The Missing Link of Episodic Memory Decline in Aging: The Role of Inefficient Systems Consolidation)
Periodo di rendicontazione: 2020-05-01 al 2021-10-31
Episodic memory function declines with age. About half of older people report worries about their own memory function, and objective memory tests confirm that the ability to learn and retrieve new information declines with age, especially after 60 years. Two reasons for this are the ability to encode new information and the ability to retrieve the same information after a certain time both decline with age. However, a third critical process for successful episodic memory – consolidation of the learned material – is almost not studied in the context of aging. The main hypothesis of the current project is that the brain processes of episodic memory consolidation are less effective in older adults, and this can account for a substantial portion of the episodic memory decline in aging.” To test this hypothesis, the projects will address its two objectives: (1) Use recent advances in memory consolidation theory to yield an elaborate model of episodic memory deficits in aging. (2) Use aging as a model to uncover how structural and functional brain changes affect episodic memory consolidation in general. These questions are important for society, given the aging of the population and the increase in cognitive challenges and demand for learning and memory that come with the rapid development and change in for instance information technology. Thus, healthy, happy and productive older adults will depend on well-functioning episodic memory and the current project will help illuminate the conditions and mechanisms for this.
During the first part of the project, we have developed a task for use in fMRI, with the aim of allowing assessment of consolidation processes in the brain. This task is an alternating item-face/item-place association task, and where the participants are learning to associate specific types of stimuli. These classes of stimuli elicit activity in specific known regions of the cerebral cortex. Since the episodic part of these memories thus requires the participant to combine different types of memory together, we can measure how these relevant brain regions communicate before, during and after the learning experience. We will use this information as a window to study the consolidation of memories online and offline. Each participant is scanned several times to allow comparing the same participants across conditions, which yields better experimental control and superior statistical power. For instance, all participants are scanned in the morning and on the evening on one occasion, and in the evening and the following morning on another occasion, allowing us to measure the impact of sleep on memory-consolidation related brain activity. The brain activity during sleep is also measured by use of EEG. The results so far are very promising, and all indications are that our fMRI paradigm works very well with regard to optimal task performance in each age group and that it elicits robust activity in the relevant brain areas.
We have also used efforts to develop, adapt and implement a processing pipeline for multi-modal neuroimaging. This is customized to the project, allows us to run integrated analyses across may different brain imaging techniques that are used for different purposes is the project. This includes task- and resting state fMRI, structural, diffusion tensor imaging (DTI), positron emission tomography (PET) and electrophysiological (EEG) data, which is a critical part of the project. Following recent standards for reproducible neuroimaging, the organization of MRI, PET and EEG-data collected in the project adheres to the Brain Imaging Data Structure specifications (BIDS - https://bids.neuroimaging.io).
Scientifically, the project has so far contributed with several publications in scientific journals, as well as a number of talks and presentations on conferences, meetings, seminars and workshops. Of the results so far, three can be highlighted:
(1) Maintained frontal activity underlies high memory function in aging: In a paper by Vidal-Piñeiro et al (Vidal-Piñeiro D. et al. Maintained frontal activity underlies high memory function over 8 years in aging. Cerebral Cortex, 2019), we directly addressed the question of why some individuals exhibit substantial episodic memory decline with aging while others show preserved function. By combining memory tests, functional fMRI measuring brain activity during encoding of episodic memories, combined with longitudinal measures of brain atrophy, we found that maintenance of brain activity in the frontal cortex during encoding was a primary characteristic of older participants with preservation of high memory function. We believe this activity reflects their intact ability to integrate new and old information in the brain, which likely is a prerequisite for efficient systems consolidation of memories. Since many of the participants were recruited from ongoing research projects, we were able to connect the results from the fMRI experiment to longitudinal data spanning up to eight years back. Thus, we could show that the maintained frontal activity was associated with less brain atrophy and better preservation of scores on neuropsychological tests of episodic memory over several years. In contrast, levels of amyloid-beta protein, associated with Alzheimer’s disease, did not impact the results, suggesting that the observed relationships represents processes of normal aging, not preclinical Alzheimer’s disease.
(2) Development and decline of the hippocampal long-axis during encoding and retrieval of episodic memories: These results were published in a paper by Langnes et al. (Langnes et al. Development and decline of the hippocampal long-axis specialization and differentiation during encoding and retrieval of episodic memories. Cerebral Cortex, 2018). It has been well known since 1957 that hippocampus plays a critical role for episodic memory. Current knowledge is that hippocampus plays a critical role both in encoding, consolidation and retrieval of episodic memories. Thus, understanding the specific role of hippocampus in any memory computation is a critical task in all research on episodic memory. We used task-fMRI to test how anterior and posterior hippocampal activity and connectivity, which is a measure of how different regions in the brain communicates during the execution of a task, contribute to formation and retrieval of accurate episodic memories. We were able to connect the data to an additional sample of children previously examined at the center. This enabled us to test how hippocampal activity contributed to successful episodic memory throughout most of the life. We found evidence for a long-axis encoding-retrieval specialization, where anterior hippocampus was relatively more involved in encoding of memories and posterior hippocampus relatively more involved in retrieval of memories. Interestingly, this functional segregation declined linearly during development and aging, eventually vanishing in the older adults. This was mainly driven by age effects on retrieval-related activity, possibly reflecting less effective consolidation of the memories during the retention period after encoding. We concluded that the hippocampal long-axis differentiation and communication during episodic memory tasks develop rapidly during childhood, are different in older compared with younger adults, and that the age effects are related to task engagement, not the successful retrieval of episodic memories specifically.
(3) The brain signature of elaboration during encoding of episodic memories: In this paper (Amlien et al., Elaboration Benefits Source Memory Encoding Through Centrality Change. Scientific Reports, 2019), we addressed the question of how the brain changes it activity patterns during encoding episodic information using elaborative strategies. We believe that how information is encoded into the long-term store is critical for how it is consolidated, and thereby for the accurate retention of the information after longer time intervals. A classic theory in psychology holds that information that is deeply processes, which means that it receives a larger amount of cognitive resources during encoding, will be better integrated and consolidated in our long-term store of existing memories. We do not know, however, how processing depth affects communication patterns within the network of interconnected brain regions working together to ensure successful encoding of the episodic information. In this work we used graph theory, which is a mathematical framework used to model pairwise relations between objects. Based on the brain activity, we calculated so-called centrality indices, which represent the brain regions' relative importance in the memory network. We found that centrality changes in specific cortical regions, that is right middle frontal gyrus, right inferior parietal lobule and left superior frontal gyrus, were positively related to semantic elaboration during encoding. In these same regions, centrality during successful episodic memory encoding was related to performance on the episodic memory task, which was a further indication that these centrality changes reflect processes that support memory encoding through deep elaborative processing. We performed similar analyses for congruent information, which is information that fits into existing knowledge structures. Along with processing depths, this is considered a major factor in consolidation-enhancing encoding. We found no relationship between centrality changes and congruity. Thus, we concluded that while elaboration and congruity have similar beneficial effects on consolidation of episodic memory content, the cortical signatures of these processes are probably different.
We have also used efforts to develop, adapt and implement a processing pipeline for multi-modal neuroimaging. This is customized to the project, allows us to run integrated analyses across may different brain imaging techniques that are used for different purposes is the project. This includes task- and resting state fMRI, structural, diffusion tensor imaging (DTI), positron emission tomography (PET) and electrophysiological (EEG) data, which is a critical part of the project. Following recent standards for reproducible neuroimaging, the organization of MRI, PET and EEG-data collected in the project adheres to the Brain Imaging Data Structure specifications (BIDS - https://bids.neuroimaging.io).
Scientifically, the project has so far contributed with several publications in scientific journals, as well as a number of talks and presentations on conferences, meetings, seminars and workshops. Of the results so far, three can be highlighted:
(1) Maintained frontal activity underlies high memory function in aging: In a paper by Vidal-Piñeiro et al (Vidal-Piñeiro D. et al. Maintained frontal activity underlies high memory function over 8 years in aging. Cerebral Cortex, 2019), we directly addressed the question of why some individuals exhibit substantial episodic memory decline with aging while others show preserved function. By combining memory tests, functional fMRI measuring brain activity during encoding of episodic memories, combined with longitudinal measures of brain atrophy, we found that maintenance of brain activity in the frontal cortex during encoding was a primary characteristic of older participants with preservation of high memory function. We believe this activity reflects their intact ability to integrate new and old information in the brain, which likely is a prerequisite for efficient systems consolidation of memories. Since many of the participants were recruited from ongoing research projects, we were able to connect the results from the fMRI experiment to longitudinal data spanning up to eight years back. Thus, we could show that the maintained frontal activity was associated with less brain atrophy and better preservation of scores on neuropsychological tests of episodic memory over several years. In contrast, levels of amyloid-beta protein, associated with Alzheimer’s disease, did not impact the results, suggesting that the observed relationships represents processes of normal aging, not preclinical Alzheimer’s disease.
(2) Development and decline of the hippocampal long-axis during encoding and retrieval of episodic memories: These results were published in a paper by Langnes et al. (Langnes et al. Development and decline of the hippocampal long-axis specialization and differentiation during encoding and retrieval of episodic memories. Cerebral Cortex, 2018). It has been well known since 1957 that hippocampus plays a critical role for episodic memory. Current knowledge is that hippocampus plays a critical role both in encoding, consolidation and retrieval of episodic memories. Thus, understanding the specific role of hippocampus in any memory computation is a critical task in all research on episodic memory. We used task-fMRI to test how anterior and posterior hippocampal activity and connectivity, which is a measure of how different regions in the brain communicates during the execution of a task, contribute to formation and retrieval of accurate episodic memories. We were able to connect the data to an additional sample of children previously examined at the center. This enabled us to test how hippocampal activity contributed to successful episodic memory throughout most of the life. We found evidence for a long-axis encoding-retrieval specialization, where anterior hippocampus was relatively more involved in encoding of memories and posterior hippocampus relatively more involved in retrieval of memories. Interestingly, this functional segregation declined linearly during development and aging, eventually vanishing in the older adults. This was mainly driven by age effects on retrieval-related activity, possibly reflecting less effective consolidation of the memories during the retention period after encoding. We concluded that the hippocampal long-axis differentiation and communication during episodic memory tasks develop rapidly during childhood, are different in older compared with younger adults, and that the age effects are related to task engagement, not the successful retrieval of episodic memories specifically.
(3) The brain signature of elaboration during encoding of episodic memories: In this paper (Amlien et al., Elaboration Benefits Source Memory Encoding Through Centrality Change. Scientific Reports, 2019), we addressed the question of how the brain changes it activity patterns during encoding episodic information using elaborative strategies. We believe that how information is encoded into the long-term store is critical for how it is consolidated, and thereby for the accurate retention of the information after longer time intervals. A classic theory in psychology holds that information that is deeply processes, which means that it receives a larger amount of cognitive resources during encoding, will be better integrated and consolidated in our long-term store of existing memories. We do not know, however, how processing depth affects communication patterns within the network of interconnected brain regions working together to ensure successful encoding of the episodic information. In this work we used graph theory, which is a mathematical framework used to model pairwise relations between objects. Based on the brain activity, we calculated so-called centrality indices, which represent the brain regions' relative importance in the memory network. We found that centrality changes in specific cortical regions, that is right middle frontal gyrus, right inferior parietal lobule and left superior frontal gyrus, were positively related to semantic elaboration during encoding. In these same regions, centrality during successful episodic memory encoding was related to performance on the episodic memory task, which was a further indication that these centrality changes reflect processes that support memory encoding through deep elaborative processing. We performed similar analyses for congruent information, which is information that fits into existing knowledge structures. Along with processing depths, this is considered a major factor in consolidation-enhancing encoding. We found no relationship between centrality changes and congruity. Thus, we concluded that while elaboration and congruity have similar beneficial effects on consolidation of episodic memory content, the cortical signatures of these processes are probably different.
Main progress beyond state of the art is represented by (1) methodological advances related to the fMRI paradigm developed to assess memory consolidation-related processes, as well as the adaptation and implementation of the brain imaging processing pipeline, as well as (2) the scientific advancements described in the section on project achievements above.
Further, the expected results until the end of the project will be centered round central themes of the project, the most immediate expected results will include:
- We will gain more insights into potential processes and mechanisms of the initiation of consolidation at the systems level, by systematically comparing stimulus offset vs. stimulus onset activity, thereby trying to disentangle encoding-specific activity from the initial phases of consolidation, and map whether such activity differs between younger and older adults.
- We will measure connectivity within and across well-established functional networks, as well as from target regions of interest such as the hippocampus and its subfields, and compare connectivity between pre- vs. post-encoding rest periods, both at the levels of regions, networks as well as network configurations. By comparing activity just before vs. after an encoding task, we will capture evidence of the initial consolidation processes immediately following a sequence of encoding items. The strength of this approach is that we follow the continuous and spontaneous brain activity during rest, which is the brain state where consolidation likely is effective.
- Successful retrieval of episodic memories are associated with patterns of cortical activity that partially overlap with the activity patterns seen during initial encoding of the same information, which is referred to as “cortical reinstatement”. Since systems consolidation is assumed to work through repeated strengthening of associations between distributed cortical representations, we expect spontaneous activity patterns occurring during rest from time to time will overlap with the cortical reinstatement effect.
We will systematically test effects of age on all brain imaging and behavior parameters in the project, and expect these to be able to account for some of the observed decline of episodic memory capabilities in aging.
Further, the expected results until the end of the project will be centered round central themes of the project, the most immediate expected results will include:
- We will gain more insights into potential processes and mechanisms of the initiation of consolidation at the systems level, by systematically comparing stimulus offset vs. stimulus onset activity, thereby trying to disentangle encoding-specific activity from the initial phases of consolidation, and map whether such activity differs between younger and older adults.
- We will measure connectivity within and across well-established functional networks, as well as from target regions of interest such as the hippocampus and its subfields, and compare connectivity between pre- vs. post-encoding rest periods, both at the levels of regions, networks as well as network configurations. By comparing activity just before vs. after an encoding task, we will capture evidence of the initial consolidation processes immediately following a sequence of encoding items. The strength of this approach is that we follow the continuous and spontaneous brain activity during rest, which is the brain state where consolidation likely is effective.
- Successful retrieval of episodic memories are associated with patterns of cortical activity that partially overlap with the activity patterns seen during initial encoding of the same information, which is referred to as “cortical reinstatement”. Since systems consolidation is assumed to work through repeated strengthening of associations between distributed cortical representations, we expect spontaneous activity patterns occurring during rest from time to time will overlap with the cortical reinstatement effect.
We will systematically test effects of age on all brain imaging and behavior parameters in the project, and expect these to be able to account for some of the observed decline of episodic memory capabilities in aging.