Insect behaviour

[John]

Two recent aspects of insect behaviour struck me as intiguing.
I was sitting on a bench a few feet from a stone wall behind me, wearing a canvas vest with light brown leather trimming. I was surprised to see two bees land on the trimming and try and do things to it with their mouth-parts.
Then I looked at the wall. It was covered with small patches of brown lichen, about the same size and colour as my leather trimmings, and dozens of bees were collecting nectar or pollen or something from these patches. So the two bees had ignored negative sense data from taste and smell, and just relied on visual memory, and perhaps distance-from-hive. I understand that bees collect from different plants at different times of the year, and that bee-keepers ship hives a long distance to pollinate crops that the bees would not otherwise see. So trial and error based on taste must be how they identify plants, but my experience suggests that they then learn visual cues which under some circumstances seem to dominate.

In the second case I was watching a dragon-fly patrol up and down a pond. I could see it dodging left-and-right, and up-and-down to pick off flies along its path. Then I saw a fly drop away to the right so that the dragon-fly missed it. It then followed a path which was a mirror-image of a question-mark, returning to catch the fly which was now on its left-hand side. Given that the predator and prey probably depart from exact symmetry in eye-sight, reflexive behaviour and the effect of wind, that seemed a good strategy to improve the chances of catching the fly on the return path. I assume this behaviour is built into the neuronal circuit from birth.

[David]

Glad to see more comments here. I've been looking for a place to discuss this stuff.

I discount the idea of behavior through genetics because no amount of genetic info would be enough to cover all scenarios an organism will go through. A great amount of additional genetic info would exist too, covering possibilities that might happen, just in case. The brain is little bitty the way it is. Where would it be stored? How would genetics be represented in neural pathways? I think this is a nebulous myth yet to be completely thrown out. Brains do by nature what we consider complex math. We need to explore analog computers in contrast to digital computers to better understand brain functionality, which is what the brain is.

[John]

I've got a third example, concerning a colony of leaf-cutter ants in the London Natual History Museum.
They set up a long branch of a tree, between the nest and a collection of leaves on twigs. The branch supported two streams of ants going out, and coming back, each carrying a section of leaf that they had cut.
When an ant got back to the nest, it stood outside whilst ants from the nest appeared to examine the leaf with their antennae, often rejecting 3 or 4 until they found one that fitted their apparent needs. The leaf was taken off the other ant, who then went out again along the branch.
One ant dropped its leaf during return to the nest, but apparently the sequence it was going through was not sensitive to such an occurrence. At the nest it waited whilst ants from the nest approached and rejected it after the antennae usage. I watched for 15 minutes before getting bored. Presumably when the lights go down all ants return to the nest and reset their sequencers for the next day.
From an evolutionary perspective, I suppose the extra neurons necessary to check the leaf has not been dropped, are statistically better used for other purposes.

I heard a talk a few years ago from a researcher reporting that a different species of ant could either walk on the forest floor, or along branches. When returning from a search for food along branches, it then waggled its antennae at another ant for a time interval proportional to the path length. If it had used the forest floor, there was no such proportionality, presumably since it just gave a bearing relative to the sun, like bees do.

My earlier bee example interested me because of the "funny writing" texts that people are circulating.
e.g "Tihs wtiring cna be raed enev tghouh it lkoos fnnuy."
Reading this seems to require that we suspend operation of some components of the reading mechanism. Just like the two bees were suspending taste and smell in identifying the lichen.
The similarity seems to talk to an underlying similarity in brain organisation, which is surprising.

[John]

Wikipedia has an article
"Bee learning and communication" which demonstrates quite complex operant conditioning in bees. I suppose that if a bee keeps visiting patche of lichen of similar size and colour all day, where it gets pollen, it can become "overtrained" and ignore contrary stimuli. Like when we absent-mindedly drive our route to work on a week-end. So maybe this is just a similarity in short-term operant conditioning that we are seeing, rather than a neuronal similarity.

[David]

I like behaviorism. Some say it isn't valid anymore and reductionism explains it away. But all are abstract ideas, including what reductionism provides. Behaviorism is no different. Behaviorism also has years of empirical research behind it. That's what legitimizes it - not whether it's reductive, but whether empirical research backs up ideas. Concerning neurons, pathways, and hormons, we need abstract ideas because that's what level we work on. Behaviorism is still a very workable model.

The similarity of behavior in ants easily leads one to think genes do that in the brain. But genes do provide antennae and other body parts that fit a role or lifestyle. From that it's very easy for them to fall into the behavior.

[John]

I think it was Chomsky who gave behaviourism a bad name, by demonstrating the complex recursive nature of language. People had previously talked about speech-stimulus and speech-response which from our viewpoint now seems obviously over-simplistic. Clearly what we hear is compared with our internal ontologies that describe our experience of things in the world around us. (Functional-MRI, and priming studies in cognitive psychology show this). There can then be a lot of complex processing before a response is given.
The problem with early behaviourism was that it did not recognise this complexity, probably because of the simplistic nature of machines of the time, which (subconsciously) formed the basis for (largely unconscious) models of the brain.
Nowadays we have digital computers with extensive databases, and neural nets, machine learning, rule and logic-based software, and Bayesian statistics. We do not expect a brain's reaction to be simple, and expect it also to be partially knowledge-driven.
I cannot explain how a piece of behaviour can be inherited, but it clearly can. Otherwise a new-born would be unable to regulate its oxygen levels or its blood-pressure. Operation of the eye requires us to coordinate 5 muscles in each eye to achieve binocular fusion, and to use the convergence angle to set focusing in the ciliary muscles. And all this goes on in the face of high-speed saccades and microsaccades that apply the very narrow high-resolution part of the retina to different parts of the image in turn. A new-born takes a while to perfect this, but it is difficult to imagine that there is not already a basic control system present at birth.
We all use dozens of well-trained response sequences every day, some of which we had to learn with difficulty under conscious supervision.

[David]

I hear what you're saying. I just heard "Mind in Life" with Evan Thompson (BSP 89) today, and I think he helps my case to some degree. I see the body as an environment for the mind and structuring the development of it. Genetics are obviously involved for the way the brain gets constructed, but as to individual functioning of neurons toward specific activity, I don't see how genetics apply much.

I saw a nature show once, challenging the idea beavers build dams by instinct. It was asserted they cognitively place each twig, and it would not be possible for their little brains to hold enough info to build dams. It has to do with a priori reasoning and determinism on our part, it seems. People seem unable to accept dynamics unfold.