# Cell death

**URL:** <https://discourse.numenta.org/t/cell-death/5617>\
**Category:** Numenta Theory\
**Created:** [March 13, 2019, 9:05am UTC](https://discourse.numenta.org/t/cell-death/5617 "2019-03-13T09:05:25Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Rodi](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/rodi/32/3720_2.png) [@Rodi](https://discourse.numenta.org/u/Rodi)\
**Post date:** [March 13, 2019, 9:05am UTC](https://discourse.numenta.org/t/cell-death/5617/1 "2019-03-13T09:05:25Z")

</div>

Hi everyone,

Can you help me about cell death?  
I dont understand much about what is cell death in brain, and how spatial pooler can implement a cell death. as i understand, if all connections in receptive field of a column are connected, it means that this column is death.  
please correct me if i am wrong.

Thank you

---

<div class="post-metadata">

**Author:** ![jimmyw](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/jimmyw/32/1485_2.png) [@jimmyw](https://discourse.numenta.org/u/jimmyw)\
**Post date:** [March 13, 2019, 10:57am UTC](https://discourse.numenta.org/t/cell-death/5617/2 "2019-03-13T10:57:13Z")

</div>

Hi Rodi,

I’m not a biologist, but are you talking about the death of neurons due to trauma/Parkinson’s/Alzheimer’s, that sort of thing?

You’re describing some sort of death by overconnection, would you have a link to the biological detail of this?

---

<div class="post-metadata">

**Author:** ![Rodi](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/rodi/32/3720_2.png) [@Rodi](https://discourse.numenta.org/u/Rodi)\
**Post date:** [March 13, 2019, 11:17am UTC](https://discourse.numenta.org/t/cell-death/5617/3 "2019-03-13T11:17:10Z")

</div>

Thank you.  
actually, I read paper about HTM. and I see some simulation about cell death.  
i dont understand this concept, how to implement it in HTM.  
Can you tell me more about it?

---

<div class="post-metadata">

**Author:** ![dmac](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/dmac/32/5842_2.png) [@dmac](https://discourse.numenta.org/u/dmac)\
**Post date:** [March 14, 2019, 5:22pm UTC](https://discourse.numenta.org/t/cell-death/5617/4 "2019-03-14T17:22:40Z")

</div>

Cells die all of the time in The human brain. Some times its random, other times the result of a blow the the head. And people dont even notice because the brain keeps working.

The spatial pooler paper simulates some cells randomly dying to show that the SP keeps working. All good models of the brain should be able to survive a few cells dying, since real brains can too.

I dont know the details of how they modeled cell death, other than: it was random and the cells died after the network was trained.

---

<div class="post-metadata">

**Author:** ![Bitking](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bitking/32/3099_2.png) [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Post date:** [March 14, 2019, 5:28pm UTC](https://discourse.numenta.org/t/cell-death/5617/5 "2019-03-14T17:28:36Z")

</div>

There is some discussion of the general SP operation and the part of the paper regarding failures on this thread:

> [@What does "the receptive field center" mean in SP?](https://discourse.numenta.org/t/what-does-the-receptive-field-center-mean-in-sp/4872):
>
> [QQ%E6%88%AA%E5%9B%BE20181102152138] Sorry for disturbing. I’m reading the paper The HTM Spatial Pooler—A Neocortical Algorithm for Online Sparse Distributed Coding by Cui. While confused that in the part of fault Tolerance, it discuss the “receptive field center” with respectively showing the picture above. What does the “receptive field center” mean?

---

<div class="post-metadata">

**Author:** ![Bitking](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bitking/32/3099_2.png) [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Post date:** [March 14, 2019, 8:20pm UTC](https://discourse.numenta.org/t/cell-death/5617/6 "2019-03-14T20:20:15Z")

</div>

> [@Rodi](#):
>
> if all connections in receptive field of a column are connected, it means that this column is death.

No, I don’t think that this is true. It does mean that the cell will respond to any pattern that turns on the bits in it’s receptive field. This would be a highly unusual learned pattern and the “sparse activation” part of HTM theory should prevent this from happening.

---

<div class="post-metadata">

**Author:** ![rhyolight](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/rhyolight/32/3922_2.png) [@rhyolight](https://discourse.numenta.org/u/rhyolight)\
**Post date:** [March 17, 2019, 12:49am UTC](https://discourse.numenta.org/t/cell-death/5617/7 "2019-03-17T00:49:29Z")

</div>

> [@Rodi](#):
>
> I read paper about HTM. and I see some simulation about cell death.

One of the spatial pooler papers showed tolerance to simulated cell death. I this is what @Rodi is referring to. It is not a part of the Spatial Pooling process, it is just something that SP handles well.