Mechanisms
Representational Structures · Person Models
Autobiographical Memory
Example
Asked what your twenties were like, what comes back depends on what you're currently trying to be. The same decade yields a story about recklessness or one about persistence. Not because you have forgotten the other events, but because the version being assembled has to fit who you are now.
Autobiographical Memory in Humans
Definition
Memories of one's own past, constructed at retrieval rather than retrieved intact, within a system where current self-goals shape which memories are assembled and how (Conway & Pleydell-Pearce, 2000). The system answers two competing demands at once, an experience-near record of goal activity and a coherent self that persists beyond the present moment, and healthy remembering holds them in balance rather than sacrificing one to the other (Conway et al., 2004).
*Framing note: distinct from Episodic Memory and the rows that decompose an episode's source, time, and context. This row concerns the organisation of one's own past around a self, and the goal-directed control of access to it.*
Neural Substrate
A core network comprising medial and ventrolateral prefrontal cortex, medial and lateral temporal cortex, retrosplenial and posterior cingulate cortex, the temporoparietal junction, and the cerebellum, each activated in at least half of 24 functional imaging studies of autobiographical retrieval (Svoboda et al., 2006). Cerebral activation is predominantly left-lateralised, most strongly in lateral cortex and least in lateral subcortical structures; cerebellar activation runs the other way, predominantly right-sided, which is what crossed cerebello-cortical connectivity would predict. The reach beyond medial temporal regions reflects that autobiographical remembering recruits self-reflection, emotion, imagery, and semantic processing alongside episodic retrieval. Medial prefrontal cortex is reliably engaged and is more active for episodic than for personal semantic autobiographical retrieval, consistent with a role in relating the retrieved event to self-knowledge; the imaging evidence does not establish that the region is necessary for autobiographical memory, and the authors note its activation may partly reflect co-activation with connected medial temporal regions.
AI at a Glance
In humans, autobiographical memory organises one's own past around a self, and the goals that self currently holds govern which parts of that past are reachable at any moment (Conway & Pleydell-Pearce, 2000). On the AI side the closest work is not about a system's own history at all but about fictional-character role play, where visibility-tagged memory layers and persona-scoped prompting are shown to control whether a model answers or refuses questions about facts its character never witnessed (Tang et al., 2026; Wang et al., 2026). The asymmetry runs deeper than performance: two people leaving a conversation each carry their own record of where they were and what they were doing while it happened, and a system has no such record, so whatever past it organises was loaded into it or assembled with a user rather than lived.
Visibility tagging: marking each stored fact with the set of characters permitted to retrieve it, so that facts outside that set cannot be returned however well they match a query (Tang et al., 2026).
Autobiographical Memory in AI
How Far It Has Been Measured
Measurement in AI: Indirectly Measured
AI Memory Performance: Comparison Not Possible
Where It Lives in the System
What Has Been Measured
- Training holds no record of the model's own experiences: ChatGPT (versions 3.5 and 4.0) asked for self-defining memories produced invented, redemptive stories, and readers identified human narratives correctly about 71% of the time and ChatGPT narratives about 60% of the time (Mansfield et al., 2026).
- When GPT-4, Claude Haiku and Claude Opus stood in as simulated participants on the Emotional Recall Task alongside 1,200 people, GPT-4 came closest to human patterns but with less variability; the authors note that LLMs lack autobiographical memory, so only word associations were compared (Carini et al., 2025).
- In saved memory, Generative Agents keep a time-stamped "memory stream" of their own experiences plus reflections and plans, a structure resembling a self organized around goals; in interviews on self-knowledge and memory, the full system was rated more believable (TrueSkill 29.89) than crowdworker-written answers (22.95) or a no-memory version (21.21), though it embellished recalled events (Park et al., 2023).
- That human comparison does not measure autobiographical memory: the 25 crowdworkers watched a replay of the agent's life and inspected its memory store while role-playing it, so they reported someone else's recorded past, and 100 raters scored believability rather than accuracy (Park et al., 2023).
- On LoCoMo, question answering over conversations averaging about 600 turns between agents given personas and timelines of life events, the best model, GPT-4-turbo, scored 51.6 F1 (a 0-100 answer-overlap score) against 18.4 to 42.8 for other models and a reported human score of 87.9; the paper does not report how many people were tested or how, and the task is reading another's conversation rather than remembering one's own life (Maharana et al., 2024).
- Commercial assistants with saved memory recalling a user's past chats (3 to 6 sessions) answered 24.7% to 71.1% of 97 questions correctly (best: ChatGPT with GPT-4o-mini), versus 91.8% when GPT-4o could read the full history in its context window, meaning the text it can see in the current conversation; no human baseline was run (Wu et al., 2025).
- Comparison is not possible: a 2026 preprint built an explicitly Conway-style store of life periods, general events and specific events for AI personas but judged role-play realism rather than memory, and no study located gives AI a human autobiographical paradigm (Autobiographical Interview, cue words, Autobiographical Memory Test) about its own past with documented human results on a common scale (Tang & Guo, 2026).
Where a Mechanism Can Live
04 Product
The whole configuration, and the only thing a benchmark observes directly.
01
Weights
Across deployments
02
Context Window
Within a session
03
Memory Layer
Across sessions
Which of the three a result belongs to is inferred from it, not seen in it.
Read more
Cerebellum
Cerebellum
The densely folded structure behind the brainstem holding most of the brain's neurons in a uniform, repeating circuit.
The cerebellum sits under the occipital lobes behind the pons and medulla, attached by three pairs of peduncles. It is about a tenth of brain volume yet contains roughly four-fifths of all neurons, largely because of its granule cells. Its cortex uses the same circuit everywhere, which makes it unusually tractable: input arrives as mossy and climbing fibers, output leaves through Purkinje cells to the deep nuclei. Classically a structure of motor coordination and timing, it is now widely implicated in cognition and language on imaging and lesion evidence, though what it contributes beyond motor control is actively argued.
Cerebral cortex
Division in the cerebral hemisphere
The folded sheet of grey matter covering each hemisphere, two to four millimetres thick and layered rather than uniform.
The cortex is a continuous sheet of neurons and local circuitry spread over the hemisphere's surface, thrown into gyri and sulci so that roughly two-thirds of it lies buried in the folds. Most of it (the neocortex) has six layers distinguished by cell type and by which connections arrive and leave; older allocortex has three or four. Areas differ in layer thickness, cell packing and connectivity, which is what cytoarchitectonic maps such as Brodmann's record. Where any given area's boundary actually falls, and how closely those maps correspond to functional divisions, is still argued over; modern multimodal parcellations disagree with Brodmann in places.
Cerebrum
Division
The paired hemispheres sitting above and around the brainstem, making up roughly four-fifths of the human brain by mass.
The cerebrum is the largest part of the brain: two hemispheres of folded cortex wrapped around deep grey masses and the white matter that connects them. It develops from the telencephalon, the forwardmost part of the embryonic forebrain, and in humans accounts for most of the brain's volume. The corpus callosum joins the hemispheres beneath the longitudinal fissure that separates them. "Cerebrum" is a gross anatomical container rather than a functional unit, so claims are rarely made about it as a whole; in practice it names an address, and the working divisions are the lobes, areas and nuclei inside it.
Medial prefrontal cortex
Cortex in the prefrontal cortex
The inner wall of the prefrontal cortex, facing the midline between the two hemispheres and wrapping around the corpus callosum's front end.
The medial prefrontal cortex is the flat inner face of the frontal lobe, running from near the frontal pole back to the genu of the corpus callosum, where it meets the cingulate. It is usually divided into dorsal and ventral halves with different connections: the ventral part to amygdala, hypothalamus and brainstem, the dorsal part to lateral frontal and temporal cortex. It is a hub of the rodent memory-consolidation literature and of the human self-reference and default-mode literatures. The term is used loosely and often includes tissue another author would call cingulate.
Medial temporal lobe
Cortex in the temporal lobe
The inner fold of the temporal lobe (hippocampus, the cortices wrapped around it, and the amygdala), removed together in the classic amnesia cases.
A regional grouping rather than a single structure: the hippocampal formation, the entorhinal, perirhinal and parahippocampal cortices that feed it, and, on some definitions, the amygdala. It became the canonical "memory region" because the surgeries and infarcts that produce dense anterograde amnesia (H.M. most famously) take out most of it at once. Whether the parts constitute one memory system or several with different jobs is an open argument, and lesions in patients often cross the boundaries that would settle it.
Posterior cingulate cortex
Cortex in the cingulate cortex
The rear section of the cingulate, above the splenium of the corpus callosum and above the retrosplenial cortex in the callosal sulcus.
The posterior cingulate cortex sits above and behind the splenium of the corpus callosum, bordering the precuneus above and the retrosplenial cortex below; its dorsal and ventral halves have different connections and different response profiles. It has one of the highest resting metabolic rates in the brain and is among the first regions to show reduced glucose metabolism in Alzheimer's disease. It is a core node of the default mode network and is engaged during autobiographical memory retrieval and imagining the future. Whether its activity reflects internally directed thought, the breadth of attention, or a more general scene-construction role is not settled.
Retrosplenial cortex
Cortex in the cingulate cortex
A small strip of cortex tucked behind and below the splenium of the corpus callosum, Brodmann areas 29 and 30.
The retrosplenial cortex lies immediately behind the splenium (the thickened rear end of the corpus callosum), buried in the callosal sulcus, and is substantially smaller in humans than its prominence in rodent work suggests. It connects the hippocampal formation and anterior thalamus with parietal and visual cortex, making it a waypoint between memory and spatial systems. Damage produces topographical disorientation in which landmarks are recognized but cannot be used for heading. Rodent and human literatures use the term for somewhat different tissue, since the human region is displaced by the expansion of surrounding cortex.
Temporal lobe
Lobe in the cerebral hemisphere
The lobe below the lateral fissure, running forward to the temporal pole and back into parietal and occipital cortex without a sharp boundary.
The temporal lobe sits beneath the lateral fissure, folded into superior, middle and inferior temporal gyri laterally, with the fusiform and parahippocampal gyri on its ventral surface and the hippocampal formation curled inside its medial wall. Heschl's gyrus, buried in the fissure's floor, carries primary auditory cortex. Its posterior boundary against parietal and occipital cortex is conventional rather than anatomical. The lobe is associated with auditory processing, object and face recognition along its ventral stream, semantic knowledge at its pole, and, through the medial structures, declarative memory.
Temporoparietal junction
Cortex
The territory where the posterior temporal lobe meets the inferior parietal lobule, named by convention rather than by any boundary.
The temporoparietal junction is the cortex around the meeting of the superior temporal sulcus, the angular gyrus and the supramarginal gyrus; it is a region of convenience with no agreed cytoarchitectonic borders, and coordinates published under the name do not converge on a single site. On the right it is central to the reorienting of attention and is damaged in most cases of hemispatial neglect; on both sides, and especially a more posterior part of it, it is among the most reliable activations in theory-of-mind tasks. Whether attention and mentalizing engage the same tissue or adjacent subregions is an active dispute.
Ventrolateral prefrontal cortex
Cortex in the prefrontal cortex
The lower lateral prefrontal cortex, overlapping the inferior frontal gyrus and Brodmann areas 44, 45, and 47.
The ventrolateral prefrontal cortex runs along the inferior part of the lateral frontal surface, below the inferior frontal sulcus, and overlaps the inferior frontal gyrus almost entirely, so in the left hemisphere it shares tissue with what is called Broca's area. It connects strongly with temporal cortex, giving it a different input profile from its dorsolateral neighbor. It is engaged during retrieval of specific memories from semantic stores, response inhibition, and selection among competing alternatives. The right-hemisphere inhibition account and the left-hemisphere selection account developed in separate literatures, and whether one control operation covers both is not settled.
Carini et al. (2025)
Exploring the Link Between the Emotional Recall Task and Mental Health in Humans and LLMs
In
Information, 16(12), 1057
Abstract
The ability of large language models to recall human emotions provides a novel opportunity to investigate links among memory, affect, and mental health. This study explores whether the Emotional Recall Task (ERT), a free word-association paradigm, can reveal cognitive markers of distress in both humans and large language models (LLMs). Using spreading activation simulations grounded in cognitive network science, we examined how the recall of emotional concepts (e.g., stress, anxiety, and depression) relates to psychometric measures of well-being and personality. In Study 1, correlations were tested between activation dynamics and clinical scales (DASS-21, PANAS, and Life Satisfaction) in...
Find it
Conway & Pleydell-Pearce (2000)
The construction of autobiographical memories in the self-memory system.
In
Psychological Review, 107(2), 261-288
Abstract
The authors describe a model of autobiographical memory in which memories are transitory mental constructions within a self-memory system (SMS). The SMS contains an autobiographical knowledge base and current goals of the working self. Within the SMS, control processes modulate access to the knowledge base by successively shaping cues used to activate autobiographical memory knowledge structures and, in this way, form specific memories. The relation of the knowledge base to active goals is reciprocal, and the knowledge base "grounds" the goals of the working self. It is shown how this model can be used to draw together a wide range of diverse data from cognitive, social, developmental,...
Find it
Conway et al. (2004)
The Self and Autobiographical Memory: Correspondence and Coherence
In
Social Cognition, 22(5), 491-529
Abstract
A modified version of Conway and Pleydell-Pearce's Self Memory System (SMS) account of autobiographical memory and the self is introduced. Modifications include discussion of a fundamental tension between adaptive correspondence (experience-near sensory-perceptual records of goal activity) and self-coherence (a more abstracted and conceptually-rich long-term store of conceptual and remembered knowledge). This tension is examined in relation to each SMS component—the episodic memory system, long-term self, and the working self. The long-term self, a new aspect of the model, consists of the interaction of the autobiographical knowledge base and the conceptual self. The working self,...
Find it
Maharana et al. (2024)
Evaluating Very Long-Term Conversational Memory of LLM Agents
In
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 13851-13870
Abstract
Paraphrase, not the publisher's wording. Prior work on long-term open-domain dialogue evaluates model responses over no more than five chat sessions, leaving the effectiveness of long-context models and retrieval augmentation in genuinely long-term conversation untested. The authors introduce a machine-human pipeline that generates high-quality very long-term dialogues by grounding LLM agent conversations in personas and temporal event graphs, with agents able to share and react to images. Human annotators verify and edit the output for long-range consistency and grounding. The resulting dataset, LoCoMo, supports an evaluation benchmark covering question answering, event summarisation,...
Find it
Mansfield et al. (2026)
What might we learn about autobiographical narrative processing from Artificial Intelligence?
In
Humanities and Social Sciences Communications, 13(1)
Find it
Park et al. (2023)
Generative Agents: Interactive Simulacra of Human Behavior
In
Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology, 1-22
Abstract
Believable proxies of human behavior can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication to prototyping tools. In this paper, we introduce generative agents: computational software agents that simulate believable human behavior. Generative agents wake up, cook breakfast, and head to work; artists paint, while authors write; they form opinions, notice each other, and initiate conversations; they remember and reflect on days past as they plan the next day. To enable generative agents, we describe an architecture that extends a large language model to store a complete record of the agent’s experiences using natural...
Find it
Svoboda et al. (2006)
The functional neuroanatomy of autobiographical memory: A meta-analysis
In
Neuropsychologia, 44(12), 2189-2208
Abstract
Autobiographical memory (AM) entails a complex set of operations, including episodic memory, self-reflection, emotion, visual imagery, attention, executive functions, and semantic processes. The heterogeneous nature of AM poses significant challenges in capturing its behavioral and neuroanatomical correlates. Investigators have recently turned their attention to the functional neuroanatomy of AM. We used the effect-location method of meta-analysis to analyze data from 24 functional imaging studies of AM. The results indicated a core neural network of left-lateralized regions, including the medial and ventrolateral prefrontal, medial and lateral temporal and retrosplenial/posterior...
Find it
Tang et al. (2026)
Staying In Character: Perspective-Bounded Memory For Book-Based Role-Playing Agents
In
arXiv
Abstract
Recent LLM role-playing systems build character agents from novels by extracting characters, scenes, and relations. Yet long-narrative role-playing suffers from two failures: Factual Overreach, where shared retrieval or parametric memory lets a character use facts outside its perspective, and Stylistic Monotony, where profile descriptions flatten a character into a fixed voice. To address these failures, we propose REVERIEMEM, a three-layer memory architecture for book-based character agents. The episodic layer stores first-person scene memories; the semantic layer stores visibility-tagged facts; and the personality layer stores situation-dependent speech and behaviour patterns. For...
Find it
Wang et al. (2026)
Memory-Driven Role-Playing: Evaluation and Enhancement of Persona Knowledge Utilization in LLMs
In
Findings of the Association for Computational Linguistics: ACL 2026, 23475-23510
Abstract
A core challenge for faithful LLM role-playing is sustaining consistent characterization throughout long, open-ended dialogues, as models frequently fail to recall and accurately apply their designated persona knowledge without explicit cues. To tackle this, we propose the Memory-Driven Role-Playing paradigm. Inspired by Stanislavski's "emotional memory" acting theory, this paradigm frames persona knowledge as the LLM's internal memory store, requiring retrieval and application based solely on dialogue context, thereby providing a rigorous test of depth and autonomous use of knowledge. Centered on this paradigm, we contribute: (1) MREval, a fine-grained evaluation framework assessing four...
Find it
Wu et al. (2024)
LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
In
arXiv
Abstract
Recent large language model (LLM)-driven chat assistant systems have integrated memory components to track user-assistant chat histories, enabling more accurate and personalized responses. However, their long-term memory capabilities in sustained interactions remain underexplored. We introduce LongMemEval, a comprehensive benchmark designed to evaluate five core long-term memory abilities of chat assistants: information extraction, multi-session reasoning, temporal reasoning, knowledge updates, and abstention. With 500 meticulously curated questions embedded within freely scalable user-assistant chat histories, LongMemEval presents a significant challenge to existing long-term memory...
Find it