www.theseedsofscience.pub/p/the-shape-of-a-thought
2 corrections found
ideas don’t live in individual neurons or even in specific brain regions.
This is too absolute. Human neuroscience has identified highly selective "concept cells" and also specific semantic hubs such as the anterior temporal lobe, so ideas are not only represented in fully distributed, region-unspecific ways.
Full reasoning
The sentence presents a false either/or: that ideas are not represented in individual neurons and not in specific brain regions. The evidence is more mixed.
Single-neuron recordings in humans have repeatedly found "concept cells" in the medial temporal lobe that respond selectively and invariantly to particular people or objects. That is direct evidence that at least some concept-level information can be represented at the level of individual neurons, even if those neurons are embedded in larger distributed codes.
Likewise, semantic-memory research does identify specific brain regions with privileged roles in conceptual representation. Major reviews describe the anterior temporal lobes (ATL) as an amodal or transmodal semantic "hub," and note that damage there disrupts conceptual knowledge across modalities. That directly contradicts the claim that ideas do not live in specific brain regions.
So the more accurate statement would be that concepts are often represented both in distributed population codes and in selective neurons / specialized regions. The article's wording is misleading because it rules out localized representations that the literature explicitly documents.
2 sources
- Concept cells: the building blocks of declarative memory functions
These studies have shown that neurons in the human medial temporal lobe respond in a remarkably selective and abstract manner to particular persons or objects, such as Jennifer Aniston, Luke Skywalker or the Tower of Pisa. These neurons have been named 'Jennifer Aniston neurons' or, more recently, 'concept cells'.
- Where do you know what you know? The representation of semantic knowledge in the human brain
This theory predicts that a lesion of the specific brain region supporting the amodal hub should disrupt all abilities requiring central conceptual knowledge... Patients with semantic dementia... resulting from focal atrophy of the anterior temporal lobes (ATL) bilaterally, show precisely this pattern... therefore, semantic dementia suggests that the ATL supports an amodal hub.
This is how learning works in any neural system.
This universal claim is too strong. Biological learning is not all one prediction-error mechanism: neuroscience also documents core Hebbian, spike-timing-dependent, and homeostatic learning rules that are described separately from explicit expectation-vs-outcome error learning.
Full reasoning
The problem here is the word "any." Neuroscience does not treat all learning in neural systems as a single prediction-error loop in which the system first forms an explicit expectation and then updates from the mismatch.
Two major counterexamples are standard in the literature:
- Hebbian / spike-timing-dependent plasticity (STDP). A major review describes STDP as a Hebbian synaptic learning rule observed across many circuits and species. This rule depends on the timing/correlation of pre- and postsynaptic activity, not on an explicit expectation-versus-outcome comparison.
- Homeostatic plasticity. Another major review emphasizes that homeostatic plasticity is a distinct class of mechanism that stabilizes neural activity and is crucial for development and memory-related processes. It is discussed precisely because Hebbian plasticity alone is insufficient and destabilizing.
Researchers can sometimes reinterpret broad classes of learning in predictive or error-driven terms, but that is a theoretical framing — not an established fact that covers all neural learning. The article's wording turns one framework into a universal claim, which overstates the evidence.
2 sources
- Spike Timing–Dependent Plasticity: A Hebbian Learning Rule
Spike timing–dependent plasticity (STDP) as a Hebbian synaptic learning rule has been demonstrated in various neural circuits over a wide spectrum of species, from insects to humans.
- Homeostatic plasticity in the developing nervous system
However, Hebbian plasticity is probably insufficient to explain activity-dependent development because it tends to destabilize the activity of neural circuits. Here we discuss evidence from a number of systems that homeostatic synaptic plasticity is crucial for processes ranging from memory storage to activity-dependent development.