What Is Different About the Brain in Highly Superior Autobiographical Memory?
- 演化之聲
- 1 day ago
- 6 min read
In 2006, a woman originally referred to by the pseudonym AJ, later identified as Jill Price, became the first thoroughly documented modern case of hyperthymesia. She could accurately recall an extraordinary number of dates from her life, describe what she had experienced on those days, and connect them with relevant public events. This phenomenon is now formally known as highly superior autobiographical memory (HSAM). The exceptional ability of individuals with HSAM is largely confined to memories of their own lives: they do not necessarily perform unusually well on standard memory tests, nor do they rely on deliberately trained mnemonic techniques. The central purpose of studying HSAM is therefore to uncover how the brain enables personal experiences to remain rapidly and richly accessible many years after they occurred.
Autobiographical memory is the long-term memory of one's own life experiences and related self-relevant information. It is supported by the coordinated activity of multiple brain regions, among which the hippocampus and the medial prefrontal cortex have attracted the greatest attention. Major functions of the hippocampus include the formation and retrieval of episodic memories (Note 1), memory consolidation, and pattern completion (Note 2). It can therefore contribute to the reconstruction of past event scenes by recombining information about people, places, time, and sensory details into an experience that can be consciously recollected. The medial prefrontal cortex contributes to the control, selection, integration, and contextual evaluation of information during memory retrieval. It is also involved in the formation and use of schemas (Note 3), as well as in coordination within the hippocampal-neocortical system. These functions allow it to integrate long-term personal experiences and use cues to locate relevant events from the past.
In one study, individuals with HSAM and normal memory controls were asked to recollect older or more recent personal experiences, such as the first time they went to a concert or the last time they took a train. The older events recalled by participants had occurred, on average, approximately 19 years before the study, whereas the more recent events had occurred about two years earlier. Participants pressed a button as soon as they had identified the requested memory, allowing the researchers to distinguish the access phase, in which a memory is located, from the subsequent phase of mentally reliving the event. Their brain activity was monitored using functional magnetic resonance imaging (fMRI). Initial analyses showed that individuals with HSAM accessed their memories more quickly and later reported more details. During the initial access phase, they also showed greater activity than normal memory controls in the left hippocampus and left medial prefrontal cortex. These findings suggest that their autobiographical memory circuitry can use a cue to locate the target memory more efficiently.
A subsequent study reanalysed the same neuroimaging dataset using multivoxel pattern analysis (MVPA) (Note 4). The researchers examined whether the patterns of activity formed by multiple voxels within the ventromedial prefrontal cortex differed when participants recollected events from different periods of their lives. In normal memory controls, activity patterns in the ventromedial prefrontal cortex did not clearly distinguish remote memories from more recent ones. In individuals with HSAM, however, the two categories remained clearly discriminable. This finding indicates that, even many years later, the ventromedial prefrontal cortex in HSAM can represent remote and recent autobiographical experiences through distinct activity patterns, reducing the likelihood that earlier life events become blurred together with later memories.
Differences associated with HSAM are also present when individuals are not deliberately trying to recall the past. In a 2024 study, resting-state functional magnetic resonance imaging (rs-fMRI) was used to compare 12 individuals with HSAM and 29 normal memory controls while they rested quietly. The analysis measured the degree to which spontaneous fluctuations in blood oxygenation signals were synchronized across different brain regions. Individuals with HSAM showed stronger functional connectivity between the medial prefrontal cortex and the visual cortex and anterior cingulate cortex; between the posterior cingulate cortex and the medial prefrontal cortex, inferior parietal lobule, and parahippocampal cortex; and between the medial temporal lobe and the posterior cingulate cortex and parahippocampal cortex. The researchers then calculated the total strength of each region's connections with the rest of the brain. In HSAM, the medial prefrontal cortex, anterior cingulate cortex, and retrosplenial cortex showed higher values. Thus, even in the absence of active autobiographical retrieval, brain regions involved in self-related experience, contextual reconstruction, and personal memory retrieval maintain stronger synchronized activity and greater whole-brain connectivity in HSAM.
Current research also suggests that HSAM may involve a different form of memory filtering. Some functional connectivity findings indicate weaker connections between the medial temporal lobe and regions of the salience network, including the anterior insula and anterior cingulate cortex, in individuals with HSAM. The salience network helps the brain evaluate which information is important and deserves priority in processing. If this filtering system operates differently in HSAM, seemingly unimportant everyday fragments that would normally fade from memory may instead be retained for much longer. At present, however, this interpretation remains an inference drawn from neuroimaging connectivity results; direct experimental evidence sufficient to establish a causal relationship is still lacking.
Structural magnetic resonance imaging studies, on the other hand, have not yet identified a stable anatomical feature that can reliably distinguish HSAM. For example, hippocampal volume has not consistently been found to be abnormal. Functional differences appear more repeatedly than structural ones. Nor does HSAM mean that memory is entirely error-free: like individuals with typical memory, people with HSAM can be influenced by misleading information and form false memories.
For now, these neuroimaging findings provide important clues to the mechanisms underlying HSAM, but they do not yet establish its complete cause. Future studies combining longitudinal follow-up, measurements of memory formation and rates of forgetting, and more direct functional tests may clarify why some people's personal experiences remain so richly detailed and accessible many years later, while the memories of most people gradually recede into indistinct fragments of the past.
Note 1: Episodic memory refers to long-term memory for personally experienced events, typically including information about time, place, people, and context. Its formation and retrieval are highly dependent on the hippocampus.
Note 2: Pattern completion is the brain's ability to reconstruct a complete memory or body of information from partial cues. This mechanism supports the rapid recognition and recollection of past experiences.
Note 3: A schema is a cognitive structure or knowledge framework that the brain uses to organize and interpret experience. Derived through abstraction from past experiences, it includes expectations, rules, and associations relevant to particular situations, helping individuals interpret new information, fill in missing details, and make classifications and inferences efficiently.
Note 4: Multivoxel pattern analysis is an analytical method for fMRI data that examines the spatial pattern of activity across multiple voxels simultaneously. A voxel is the smallest unit in a three-dimensional image, comparable to a pixel in a two-dimensional image. By using machine-learning or statistical classification approaches to detect distributed neural representations, MVPA can reveal fine-grained information that traditional univariate analyses may fail to identify.
Author: Shui-Ye You
References:
Orwig W et al. (2024). Cortical hubs of highly superior autobiographical memory. Cortex.
Santangelo V et al. (2020). Enhanced cortical specialization to distinguish older and newer memories in highly superior autobiographical memory. Cortex.
Talbot J et al. (2025). Highly Superior Autobiographical Memory (HSAM): A Systematic Review. Neuropsychology Review.
Comments