Final Report Summary - CONSTRUCTIVEMEM (Emergence and decline of constructive memory – Life-span changes in a common brain network for imagination and episodic memory)
In this project, we have examined how episodic memory develops through childhood and declines with higher age. Children, adults and older adults have undergone extensive memory and neuropsychological testing as well as magnetic resonance imaging (MRI) of the head. A major task in the project has been to better understand the brain changes that are responsible for these changes in memory function. Among the most important conclusions are that imagination and recall abilities are related in development, and supported by development of communication of the so-called default mode network of the brain. This network is a so-called task-negative network, and characterized by increased activity when the participant is not engaged in specific external tasks. Further, we found that development of episodic memory through childhood can be predicted from the rate of brain maturation in the same participants, especially hippocampal subfields, which mature at different rates. Similarly, rate of volume or thickness reductions in older adults are related to reductions of episodic memory function, even in healthy adults with very low risk of Alzheimer’s disease. Also, we have studied the effect of communication between major brain networks, and the effect variations in this communication has on episodic memory function over time. We found that differences in communication between the well-known memory structure hippocampus and the rest of the cortex, as well as between and within large-scale cortical networks, have consequences for memory function. This moves beyond focusing on critical brain structures into looking and specific overarching patterns of communication within the cerebral cortex. Thus, both functional and structural aspects of brain change seem to be critical for optimal episodic memory function through life. Finally, it is well known that hippocampus needs to be active for successful encoding of a memory. In this project, we were able to show that although high hippocampal encoding activity is necessary for later recall of a memory, it is not sufficient for securing long-term storage of a memory. Rather, successful memory encoding were also characterized by increased hippocampal-neocortical connectivity, meaning that the communication between hippocampus and posterior cortical, i.e. visual regions, seems critical for our ability to encode a memory with sufficient details for later retrieval. Importantly, this was only seen for memories that lasted for at least 6 weeks, not for memories that were tested after 1.5 hours. We also showed that a key to good long-term memory is efficient decoupling of two large-scale brain networks, i.e. the default mode network and so-called task-positive networks. Thus, by systematically testing memory over more extended time intervals than those usually employed in the literature (weeks vs. hours), we were able to detect differences in how the brain encode and retrieve episodic memories.