9 ms·
> Ants use the rate at which they meet and smell other ants, or the chemicals deposited by other ants, to decide what to do next. A neuron uses the rate at whic
by anonytrary 7y ago
> Ants use the rate at which they meet and smell other ants, or the chemicals deposited by other ants, to decide what to do next. A neuron uses the rate at which it is stimulated by other neurons to decide whether to fire. In both cases, memory arises from changes in how ants or neurons connect and stimulate each other.
Main takeaway, very interesting analogy. Ant colonies are great examples of complex systems with emergent large-scale behavior. Indeed the same could be said about networks of neurons. Interesting to think of an ant colony as the sum of oscillations of signals.
> Every morning, the shape of the colony’s foraging area changes, like an amoeba that expands and contracts.
Sounds like an emergent macroscopic "heartbeat" of the colony.
> In an older, larger colony, each ant has more ants to meet than in a younger, smaller one, and the outcome is a more stable dynamic.
It makes sense that small perturbations would temporarily morph the heartbeat, but would probably snap back into the default oscillation pretty quickly. It would be interesting to see if a small colony is equally resilient to small perturbations as a large colony is to large perturbations, keeping some adjusted ratio of the perturbationSize/colonySize constant.
> individual ants live at most a year.
This comes as a surprise to me.
- awb 7y agoIs it very different from a large scale construction project? The group knows how to build the structure even if no one individual can explain every detail. And to an observer they'd see humans arrive each morning, spread out to accomplish tasks and then return to their cars.
- ogig 7y agoIn large scale human projects someone, or at least a small group of leaders, will know almost every detail and delegate. Ant colonies have no clear leadership, just a pheromone voting and decision system.
- grawprog 7y agoMy last job was probably more ant colony like than most. We had jobs, someone quoted them, someone measured them, someone put them in the schedule, several of us did the jobs, the finished jobs were delivered and installed. All without anyone really delegating. We just followed the schedule and the paperwork. Even the boss didn't always know the details of the schedule or what we did. It all flowed pretty smoothly most of the time. We all just kind of knew what to do. It kind of broke down a bit when things went wrong though, it was was hard to figure out sometimes where and when problems happened, and in that case someone usually made an authoritative decision, but it wasn't that often.
- aphextim 7y agoThat works from what I've observed at my company for jobs on a timetable < 6 months, or a shutdown/repair industrial type job. Once you start getting into 5 phase, 2.5 year long projects it helps having a dedicated PM overseeing every detail and coordinating between the engineer/owners/superintendents. Heaven forbid you get into a government contract where now you have 10x the paperwork/submittals and RFI's compared to a private job, which it really does help to have one person pretty much memorize the spec book and know where to find everything when needed.
- jakear 7y agoEven more surprising for me is that the queen can live 30+ years.
- iRobbery 7y agoto my surprise i still watch this youtube channel after seeing a video by accident https://www.youtube.com/user/AntsCanada https://www.youtube.com/user/AntsCanada
- chrisco255 7y agoI can't believe how popular that channel is. 3.2 million subscribers...for ant videos!
- bayesian_horse 7y agoI got over his way of story telling eventually. ;-) But yes, he does great work...
- cttet 7y agoYeah, me too. I followed him for almost a year but in the end I felt like he tried to make video too frequently and the information concentration is too low.
- sitkack 7y agoI wish I could watch this in my sleep.
- james_s_tayler 7y agoRIP Fire Nation.
- thelookingglass 7y agoShe's giving Elizabeth a run for her money
- saurik 7y ago> This comes as a surprise to me. Were you expecting them to live longer or shorter? (I would likely have guessed six months.)
- rland 7y ago>Ant colonies are great examples of complex systems with emergent large-scale behavior. Indeed the same could be said about networks of neurons. The same can be said of human organizations, too. Organizations have distinct behaviors ("company culture" comes to mind) which can be totally out of the control of individuals, if the org is big/complex enough. God I love ants so much. Such a philosophically interesting creature.
- bayesian_horse 7y agoI think many worker ants live two years or more, a queen can live 15-30 years, depending on species and luck. The difference in the contact between ants and between neurons is that it's not the same ants that contact each other. A larger pool of workers also means the colony is less likely to suffer catastrophic setbacks. For example when they send a good portion of the pool to a promising food source and then those get washed away, fall prey or whatever.
- hansbo 7y ago> Main takeaway, very interesting analogy. Ant colonies are great examples of complex systems with emergent large-scale behavior. Indeed the same could be said about networks of neurons. Interesting to think of an ant colony as the sum of oscillations of signals. Indeed. I've long thought that an ant colony should be seen as a single individual, rather than a group. One part which can procreate, like the reproductive system. Another which can fight off invaders, like white blood cells, or perhaps muscles. The anthill, in turn, is like a body; constructed by the cells and neurons, and protecting the system as a whole.
- pure-awesome 7y agoThis point is made in the book "The Soul of the White Ant" by Eugène Marais, and is perhaps the primary thesis of the book. It is a fascinating book, and I can recommend reading it, though having been published in 1925, it is possible some of the information is out-of-date. https://www.amazon.com/Soul-White-Ant-Complete-Unabridged/dp/1508988714 https://www.amazon.com/Soul-White-Ant-Complete-Unabridged/dp... "White Ant" actually refers to termites, though the same principles apply. The book was originally published in Afrikaans. According to Wikipedia: > [The book] was plagiarised by Nobel laureate Maurice Maeterlinck, who published La Vie des Termites (translated into English as The Life of Termites or The Life of White Ants), an entomological book,[3] in what has been called "a classic example of academic plagiarism" by University of London's professor of biology, David Bignell.[4] https://en.wikipedia.org/wiki/Eug%C3%A8ne_Marais#Theft_of_his_work https://en.wikipedia.org/wiki/Eug%C3%A8ne_Marais#Theft_of_hi... Following reference [4] led to this page which I unfortunately don't have the time to read in its entirety right now, but from a skim seems to have some interesting further information on termites: https://web.archive.org/web/20070915005006/http://www.biology.qmul.ac.uk/research/staff/bignell/Inaugural.htm https://web.archive.org/web/20070915005006/http://www.biolog...
- wh1t3n01s3 7y agoThe hardcover version is only 814.57$
- 7y ago
- bobosha 7y ago>Main takeaway, very interesting analogy. Ant colonies are great examples of complex systems with emergent large-scale behavior. Indeed the same could be said about networks of neurons. Interesting to think of an ant colony as the sum of oscillations of signals. Or perhaps like a Fourier decomposition of a complex waveform. Whereby each ant essentially becomes a constituent "wave" in a complex signal.
- K0SM0S 7y agoThat's my intuition for the brain, as well. Something to do with composition and relative independence, from which arise incredible complexity, "multi-dimensional" processing — it's a big, big graph. You might have to consider several such "waves" in ant colonies — maybe one electrically defined for a certain type of information; another chemically defined for another domain; etc.
- anonytrary 7y agohttps://en.wikipedia.org/wiki/Kuramoto_model https://en.wikipedia.org/wiki/Kuramoto_model
- K0SM0S 7y agoImmensely interesting food for thought. Thanks!
- 6510 7y agoWe don't really know how old workers get. 1-3 years is the estimate for red wood ants. For Pharaoh Ant workers it's 70 days.
- Psyladine 7y ago>Indeed the same could be said about networks of neurons. The idea of memory as persistent echoes of neuron firings in spacing & intensity is fascinating.
- loceng 7y agoThis reminded me of a theory I've been developing, I don't know if others have had it or have it, however an emergent behaviour that I've had some inklings may be true for humans relating to how old the parents are when having children. If having children at an older age (whether older age of male, female, or both) it could mean that resources and society are more stable - and perhaps another factor that there's more time to spend with, raise the children - giving them more guidance and having more nuanced knowledge to pass down, and so evolution may have selected for successful births at an older age toward creating children with different characteristics, behaviour, abilities.
- blotter_paper 7y agoI don't have a source, but I recall seeing a pop-sci news story claiming that men who have children at an older age have sons with higher than average life expectancy. IIRC they controlled for lifespan of the father, but I don't remember any word about controlling for fertility/sperm-count of the father in later years.
- scarface74 7y agoI would think having kids younger, as long as the parents are not resource constrained - stable financially - would be more advantageous. The parents would be around and healthy to help their kids emotionally and/or financially through their young adult years and the kids would be less likely to have to take care of their parents while trying to get their own life off the ground.