This is an interesting study, and it raises some fascinating questions when viewed through the lenses of Edelman and Halliday. Here are some key takeaways:
-
Memory Encoding vs. Recall (Edelman’s Perspective)
Edelman’s Theory of Neuronal Group Selection (TNGS) emphasises how neural circuits develop through selectional processes. If infants' hippocampi are already encoding memories, then neuronal groups are forming stable patterns. However, if recall is the problem, it suggests that later developmental changes alter the ‘search criteria’ for accessing these early patterns. Essentially, the neural maps that develop later in life may not align with the ones used during infancy. -
Memory and Meaning-Making (Halliday’s Perspective)
From an SFL perspective, early memories are likely encoded with meaning potential that does not fully align with later linguistic and conceptual structures. A newborn’s semiotic system is far more embodied and tied to immediate sensory-motor experiences, whereas an adult’s semiotic system is far more abstract and mediated by language. This mismatch could explain why memories remain inaccessible—they are ‘stored’ in a system that lacks the later-developed organisational frameworks for retrieval. -
Individuation and Memory Access
The individuation framework might also apply here. Infant memories are part of the child’s early individuation process—they are experiences before the system has been fully shaped by the broader social semiotic system. Later, as individuation unfolds and meaning potentials shift, early memories may no longer be ‘translatable’ into the adult’s current meaning system. They still exist as neuronal traces but are effectively ‘outside’ the accessible meaning potential. -
Potential-Instance Collapse in Memory Retrieval
If memory is a semiotic phenomenon (as in Halliday) and also a material one (as in Edelman), then memory retrieval could be seen as an instance of potential being actualised—but only if the system’s meaning potential allows it. Infant memories might be like quantum superpositions: encoded but unobservable because the ‘measurement system’ (i.e., the adult brain’s retrieval strategies) is incompatible with them.
Overall, this study reinforces the idea that memories are not simply ‘stored’ like files in a database but are dynamic, developing alongside cognitive and semiotic systems. Early experiences shape later meaning potentials, even if they cannot be recalled as discrete events.
Yet, there are a few issues in the way the researchers frame their interpretation, particularly in how they conceptualise memory, recall, and the role of developmental changes. Here are the main problems:
-
Over-reliance on a Storage-Retrieval Model
The researchers frame memory as something that is "encoded" and later "retrieved," as if it were a static object placed in storage. This ignores the dynamic, reconstructive nature of memory. Edelman’s Theory of Neuronal Group Selection (TNGS) suggests that memory is not a simple retrieval of stored information but a reactivation of selected neuronal patterns in a changing system. Memories are actively reconstructed each time they are recalled, meaning that if early memories cannot be accessed, it’s not just a ‘search term’ mismatch—it’s a fundamental shift in the way memory itself is structured over time. -
Assumption of ‘Intact but Inaccessible’ Memories
The idea that early memories are still "there" but merely "unreachable" is an oversimplification. This assumes that memories are stored in a fixed state and that the only issue is a failure in retrieval. However, developmental changes in the brain—especially the radical reorganisation of neural networks—suggest that early memories may not just be hidden but may have been restructured or even overwritten by later experiences. If meaning potential changes over time, the instances of early memory may no longer exist in a form that can be instantiated. -
Neglect of Meaning-Making in Memory Formation
From an SFL perspective, the researchers do not fully account for the fact that memory is not just about storing sensory impressions but about making meaning from experience. If infant memory is structured differently from adult memory due to differences in semiotic systems, then early experiences may not be encoded in a way that aligns with later meaning-making processes. Their focus on "hippocampal activity" as proof of memory formation ignores how these memories might have functioned semiotically in infancy versus adulthood. -
Uncritical Use of Rat Studies as a Parallel
The study cites optogenetic experiments in rats as evidence that infant memories may persist into adulthood. However, this assumes that rat and human memory systems function in analogous ways, which is not necessarily the case. Rats lack the same semiotic development as humans, so their early memories might be more continuous across their lifespan. The transformation of human memory over time is tightly linked to language and socialisation, which fundamentally change the meaning potential available for recall.
Alternative Framing
A better way to interpret the results would be:
- Infant memory formation is real, but memory itself is dynamic and shaped by developmental changes in both the brain’s neural architecture (material order) and its semiotic system (semiotic order).
- The inability to recall early memories is not simply a retrieval failure but a reflection of the way meaning-making systems evolve over time.
- Memory is best understood not as an isolated object in storage but as a set of meaning potentials that shift with individuation and experience.
Their study is valuable in demonstrating that early memory encoding happens, but their interpretation of why those memories become inaccessible is too mechanistic and needs a more developmental, semiotic, and reconstructive perspective.
