The Termites and the Pendulum

 

The Termites and the Pendulum 1 

It is well known that termites display adaptive social behavior under varying environmental conditions. Different types of individuals (i.e., castes) show distinctive phenotypes specialized in specific colony tasks. The whole colony is coordinated by interactions among colony members.2

 New unpublished work suggests greater complexity in the termite caste system, greater communications abilities, as well as greater termite intelligence than previously assumed. This new work is summarized below.

 Worldwide there are about 3000 species of termites including about 360 species living in Australia. The new subterranean species, not yet named, was found living on a great wood--mud ball suspended on a 100 foot termite built vine anchored to the top of a cave in New South Wales. When initially discovered the ball was swinging along an orbit with a periodic length of about 7 feet.

 Building upon the work of bacteriologists3, the investigators developed a cyberMite, a mobile robot shaped as a typical termite of this new species but equipped with multiple sensors and bathed in “friendly” pheromones. Given that termites are nearly blind, the cyberMite was immediately accepted and several were inserted into the colony at various locations, including the nursery and the Queens chamber.

 The cyberMites transmitted all termite generated sounds, electrical signals, digitized pheromone samples, and video of associated behavioral actions. The collected data was analyzed using natural language machine learning techniques4(NLML).

Below are key findings, organized by time of discovery.

Week 1:

The NLML system identified over 2000 “words” of the termite’s language where individual words were either pure sounds, electrical signals, pheromones, or combinations of these elements.

The system also mapped the morphology of the language. The language displayed all the features of a combinatorial communication system where two or more signals are combined to produce an effect that neither signal alone can produce. Combinatorial communications systems are associated with higher species, but this may represent another possible exception5

 

Week 2:

The cyberMites located in the nursery were the first to successfully mimic the termite language6. Soon, all cyberMites were able to communicate with other colony members.


 

Week 3:

By analyzing conversations (i.e., the exchange of messages between termites) the investigators were able to identify several new castes and caste interactions:


(a)  Fact-caste: Termites in this caste talked exclusively about facts and the history of events related to the colony. For example, they knew the temperature of the nest currently as well as any past day of the year. They would reply to questions from any cyberMite but only if the question called for a factual answer.


(b)  Analyst-caste: Those in this caste responded only to questions about the future or about the probability of occurrence of some event. For example, if a member of the fact-caste gave an analyst-termite the current population of the colony, the current birth rate (about 30,000/day) and the current death rate, that analyst would give the estimated population of the colony for any given future day. For the most part estimates from different analyst-termites were in fairly close agreement.


(c)  Policy-caste: Those in this class only responded to questions about what future action should be taken when they were given a consensus prediction from the analyst-caste. For example, if the consensus prediction supported a high likelihood of a food shortage, a member of the policy caste would offer a set of actions aimed at increasing the future food supply. Policy answers from different analyst-termites were often in close agreement, but sometimes they were far apart. Continuing the above example, a second policy analyst might recommend reducing the birth rate rather than increasing the food supply. Only consensus positions were brought to the Queen, who would then instruct the soldier-caste to carry out the policy.

 

It was observed that neither peer pressure, nor other castes, nor the Queen would withhold (or grant) resources or services to any policy-caste member based upon the policies offered by that member. Members of the policy-caste would change their positions but only when presented with sufficient facts and logic.

 

Week 8:

Members of the fact-caste were routinely asked about the state of the nest. Recent answers suggested the overall health of the nest was declining. Temperature, birth rates, and food supplies were on a slow downward trend. Several analyst-caste members asked for more details before making policy recommendations. One previously unreported detail emerged. In the past 30 days, the period length of the nest orbit had decreased in length by about 0.2%. After further inquiries to the fact-caste it was discovered the decrease in period length could be traced back over a long duration of at least three Queen generations (about 70 years).

 

Weeks 9 to 12:

Many analysts estimated the nest would, within the near future, cease to swing. There was a growing consensus over the assumption that the decreasing swing length was the primary cause

of the overall decline in the health of the nest. The analysts were not in agreement about how to fix the decreasing swing length. The fact-termites showed there was a strong correlation between the increasing mass of the nest and the decrease in the swing period. Some analysts suggested the period could be corrected by reducing the mass of the nest (the “massites”).

Other analysts suggested altering the length of the vine that secured the nest to the cave ceiling would correct the period. These analysts were called “lengthites”.

 

Week 13:

A third group among the policy-termites emerged. They did not deny the facts, but they did not agree with the assumptions that (a) the primary cause of the decline in nest health was due to decreasing period length and (b) changing either the period length or the nest mass would return the nest to health. They argued that their opponents were basing their policies on correlations, not causes and moreover there was insufficient evidence to warrant causal conclusions. The policy advocated by this group was to abandon the nest and establish a new colony elsewhere. This third group was called “migrators”.

 

Week 20:

After intense and widespread conversations no consensus between the three groups had emerged. Without a consensus nothing could be presented to the Queen and no corrective actions would be taken. The Queen was growing impatient. The policy-caste decided that each group should look deeper into the implications of their recommended policies and report their findings to all members of the policy community.


(1)  The massites reported their calculations showed that even with up to a 50% reduction or 50% increase in nest mass, changes in period length, if any, would be temporary. Eventually, the nest would cease to swing.


(2)  The lengthites reported their calculations showed similar results. Neither up to a 50% reduction nor 50% increase in vine length would result in permanent period length changes. Eventually, the nest would cease to swing.

(3)  The migrators reported there were no suitable locations within a reachable distance from the current nest. Any migration of the nest would have low probability of success. In addition, only the Queen and a few soldiers would be able to attempt the risky journey. Once the Queen left the nest there would be no new children. The entire remaining colony would perish in a matter of weeks.

 

Week 22:

A small number of policy-termites suggested a way to restore the orbit and presumably restore the nest’s health, was to apply an outside force to the nest. They felt safe in ignoring possible orbital variations such as any eccentricity of the nest's orbit or any precession or tilt of its rotational axis. They proposed attaching a second vine to the nest. The new vine would loop over a horizontal outcrop at a point many feet away from the current anchor point. At the far end of the new vine they would build an enclosure to hold the weight of many rocks. Once the added rocks moved the nest to greater than a 60° angle, the new vine would be detached from the nest. There was debate over the angle to be achieved before detaching the new vine. If too small, it might have little effect. If too great, it might jolt the nest and cause irreparable damage, or worse, it might sever the original supporting vine. After further analysis it was agreed that the angle should be 57.4°.


This “outside force policy” gained rapid acceptance. It was presented to and approved by the Queen.

 

Week 30:

Work on the new vine was completed. Rocks were added to the enclosure until the desired release angle was achieved. The new vine was detached. The period length was 52 feet, the same period length that was recorded 10 Queen-lifetimes ago.  

 

Weeks 40-60:

Temperatures are improving and the health of the nest is on an upward trend. So far, the outside force policy had resulted in a successful "great reset". It remains to be seen if this trend will continue.

=================================

QuestionsBefore answering you may wish to refresh your knowledge about Earth's energy budget here.

(1)  Talking into account all orbital variations (i.e., the eccentricity of Earth's orbit around the sun, the precession, and tilt of its rotational axis), does there exist one or more orbital variations that would change Earth's Climate trend to one more favorable than our current trend. If so, in what feasible ways could any of these more favorable variations be achieved?

(2)  Freeman Dyson  Suggested that an advanced extra-terrestrial civilization might obtain energy by enclosing a star inside a sphere.  A less ambitious effort for us would be to launch “umbrellas” that block the sun’s radiation from reaching parts of the earth. These could be put into geostationary orbit over the North and South poles.  Is this a feasible and economical way to impact climate change?  Similar, space based ideas have been proposed, see https://ieeexplore.ieee.org/document/1512527   

                                                             Endnotes                                           

1 This article is entirely true, except for the parts that might be fictional. The reader is responsible for determining to which category any sentence belongs. Moreover, any similarities to the Sociology of Human Climate change or to thinking about Climate change issues is almost accidental. The author is not responsible for inferences or faulty assumptions about climate change made either by the author or the reader of this article. This article was partially inspired by “High-Fidelity Record of Earth’s Climate History Puts Current Changes in Context.” https://phys.org/news/2020-09-high-fidelity-earth-climate-history-current.html


2 Watanabe et al., “Social Interactions Affecting Caste Development through Physiological Actions in Termites” https://www.frontiersin.org/articles/10.3389/fphys.2014.00127/full


    3 (a) Kylilis et al., “Tools for Engineering Coordinated System Behaviour in Synthetic Microbial Consortia.” https://www.nature.com/articles/s41467-018-05046-2 ,

    (b)  “They remember: Communities of microbes found to have working memory” https://phys.org/news/2020-04- microbes-memory.html,

    (c)  Bagnères and Hanus, “Communication and Social Regulation in Termites.” http://link.springer.com/10.1007/978-3-319-17599-7_11  

    (d)  Farsad, Guo, and Eckford, “Tabletop Molecular Communication.” journals.plos.org/plosone/article?id=10.1371/journal.pone.0082935

    4 (a) Freitag and Roy, “Unsupervised Natural Language Generation with Denoising Autoencoders.” http://arxiv.org/abs/1804.07899    (b) Collobert and Weston, “A Unified Architecture for Natural Language Processing: Deep Neural Networks with Multitask Learning.” https://thetalkingmachines.com/sites/default/files/2018- 12/unified_nlp.pdf

    5 Scott-Phillips et al., “Combinatorial Communication in Bacteria.” https://journals.plos.org/plosone/article?id=10.1371/journal.pone.009592

    6 Zaadnoordijk, Besold, and Cusack, “The Next Big Thing(s) in Supervised Machine Learning.” http://arxiv.org/abs/2009.08497

Comments

Popular posts from this blog

The Age After Wealth

Stigmergy