Maggot Moves

[Abbott]

[From Bruce Abbott (2016.09.18.1615 EDT]

Alex sent me a link to another of his papers on the movement of fruit-fly maggots; you can find it at https://elifesciences.org/content/4/e06694 This links to a web page displaying the article itself (including graphs) plus a series of videos showing the movements of individual maggots around a gradient. (In most cases the action has been sped up as the maggots move rather slowly.)

Maggots are somewhat worm- or caterpillar-like critters that navigate by raising the head end from time to time, then bending it left and right to sample changes in the sensed perception (e.g., light intensity, odor intensity). The head is then lowered and the maggot crawls forward, realigning its body with the position of the head as it does so. This is similar to ecoli-style movement: presumably if the maggot is controlling for moving up the gradient, a signal of decreasing intensity will soon trigger the equivalent of a “tumble” (changed head orientation) and the maggot heads off in a new direction. If the intensity is increasing, the maggot continues more or less straight ahead. However, a difference from e. coli is that the maggot can sense changes in intensity as it wags its head and thereby choose the new direction based on these intensity changes. E. coli’s changes in direction following a tumble are random.

It occurs to me that the sampling phase takes up time, during which the maggot is stationary. Could this be the origin of the power-law relation between curvature and angular velocity? Because turns tend to occur at such times, the maggot would take more time to execute a turn than to continue moving straight ahead. In other words, its velocity would be lower through the turns than when moving straight.

I haven’t taken the time to read the paper in detail yet, but what I’ve seen of it is quite impressive.

Bruce

[rsmarken]

[From Rick Marken (2016.09.19.1000)]

···

Bruce Abbott (2016.09.18.1615 EDT)–

Â

BA: Alex sent me a link to another of his papers on the movement of fruit-fly maggots; you can find it at https://elifesciences.org/content/4/e06694Â .

RM: This is a really impressive piece of work, Alex. No wonder you think I’m an idiot. This research displays an impressive level of skill in mathematics, biology, neurophysiology and research instrumentation. And it was carried out in a Spanish research lab and written up in impeccable English.

RM: I read it over and I got a vague idea regarding what you found. I think the only thing I can “contribute” from a PCT perspective is just to recommend that you look at the whole process of chemotaxis as control of input rather than control by input. This is a pretty high level suggestion but it might influence the way you go about studying the chemotaxis process.Â

RM: The larval chemotaxis takes place in a loop, of course, so while input affects output (the bending movements) at the same time the output affects input. Assuming the relevant input is, indeed, the chemical gradient through which the larva moves, PCT would see the bending movements as the means of controlling sensed chemical concentration. And if I am reading Figure 1A correctly, it looks like the larvae  move so as to maintain the sensed gradient at some intermediate level; they are not moving directly to the highest concentration (in the center of the Figure). This suggests that, in this situation, at least, the reference for sensed concentration is not to have it constantly increasing but, rather, maintained at some intermediate level. I would imagine that this constancy should be reflected in the neural signal that is presumably proportional to sensed concentration. But then Figure 1C seems to suggest that they head into the gradient and increase sensed concentration quite a bit at first but then bring it down to a relatively constant level. If sensed concentration is, indeed, a controlled variable, it would be interesting to see if the pattern in Figure 1C reflects a change in reference or simply the behavior of the controlled variable in a very sluggish control system.Â

RM: One more thing; I think the chemotaxis process used by the larva is probably similar to that used by E. coli, to the extent that the direction of bend is random. But if I am right about the idea that the larva are controlling sensed concentration relative to a reference value, then this should be taken into account in your efforts to determine the input-output function relating sensed concentration to probability of bending. Presumably the probability of bending should decrease as concentration increases, but only up to a point (the reference point) at which point the probability of bending should start to increase again as concentration further increases.Â

RM: Anyway, it is a lovely piece of research. It would be wonderful if you could apply all these research skills to studying the behavior of these beasts from a control theory perspective -  one that views their behavior as a process of control of input. Then the next paper you write might be titled: “Chemotaxis guides perceptual features extracted dynamically at the sensory periphery”.Â

Best regards

Rick

Â

Â

Maggots are somewhat worm- or caterpillar-like critters that navigate by raising the head end from time to time, then bending it left and right to sample changes in the sensed perception (e.g., light intensity, odor intensity). The head is then lowered  and the maggot crawls forward, realigning its body with the position of the head as it does so. This is similar to ecoli-style movement: presumably if the maggot is controlling for moving up the gradient, a signal of decreasing intensity will soon trigger the equivalent of a “tumble� (changed head orientation) and the maggot heads off in a new direction. If the intensity is increasing, the maggot continues more or less straight ahead. However, a difference from e. coli is that the maggot can sense changes in intensity as it wags its head and thereby choose the new direction based on these intensity changes. E. coli’s changes in direction following a tumble are random.

Â

It occurs to me that the sampling phase takes up time, during which the maggot is stationary. Could this be the origin of the power-law relation between curvature and angular velocity? Because turns tend to occur at such times, the maggot would take more time to execute a turn than to continue moving straight ahead. In other words, its velocity would be lower through the turns than when moving straight.

Â

I haven’t taken the time to read the paper in detail yet, but what I’ve seen of it is quite impressive.

Â

Bruce

–
Richard S. MarkenÂ

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

[Alex_Gomez-Marin]

If you want to see two of the first movies I took of Drosophila larvae crawling towards/around an odour source (suspended in one of the circles and out of reach of the animal), so that you can get a feeling for the phenomenon, find them attached.

​
 larva_approach_withoutTracking.mp4

​​
 larva_circle_withTracking.mov

​

···

On Mon, Sep 19, 2016 at 7:00 PM, Richard Marken rsmarken@gmail.com wrote:

[From Rick Marken (2016.09.19.1000)]

Bruce Abbott (2016.09.18.1615 EDT)–

Â

BA: Alex sent me a link to another of his papers on the movement of fruit-fly maggots; you can find it at https://elifesciences.org/content/4/e06694Â .

RM: This is a really impressive piece of work, Alex. No wonder you think I’m an idiot. This research displays an impressive level of skill in mathematics, biology, neurophysiology and research instrumentation. And it was carried out in a Spanish research lab and written up in impeccable English.

RM: I read it over and I got a vague idea regarding what you found. I think the only thing I can “contribute” from a PCT perspective is just to recommend that you look at the whole process of chemotaxis as control of input rather than control by input. This is a pretty high level suggestion but it might influence the way you go about studying the chemotaxis process.Â

RM: The larval chemotaxis takes place in a loop, of course, so while input affects output (the bending movements) at the same time the output affects input. Assuming the relevant input is, indeed, the chemical gradient through which the larva moves, PCT would see the bending movements as the means of controlling sensed chemical concentration. And if I am reading Figure 1A correctly, it looks like the larvae  move so as to maintain the sensed gradient at some intermediate level; they are not moving directly to the highest concentration (in the center of the Figure). This suggests that, in this situation, at least, the reference for sensed concentration is not to have it constantly increasing but, rather, maintained at some intermediate level. I would imagine that this constancy should be reflected in the neural signal that is presumably proportional to sensed concentration. But then Figure 1C seems to suggest that they head into the gradient and increase sensed concentration quite a bit at first but then bring it down to a relatively constant level. If sensed concentration is, indeed, a controlled variable, it would be interesting to see if the pattern in Figure 1C reflects a change in reference or simply the behavior of the controlled variable in a very sluggish control system.Â

RM: One more thing; I think the chemotaxis process used by the larva is probably similar to that used by E. coli, to the extent that the direction of bend is random. But if I am right about the idea that the larva are controlling sensed concentration relative to a reference value, then this should be taken into account in your efforts to determine the input-output function relating sensed concentration to probability of bending. Presumably the probability of bending should decrease as concentration increases, but only up to a point (the reference point) at which point the probability of bending should start to increase again as concentration further increases.Â

RM: Anyway, it is a lovely piece of research. It would be wonderful if you could apply all these research skills to studying the behavior of these beasts from a control theory perspective -  one that views their behavior as a process of control of input. Then the next paper you write might be titled: “Chemotaxis guides perceptual features extracted dynamically at the sensory periphery”.Â

Best regards

Rick

Â

Â

Maggots are somewhat worm- or caterpillar-like critters that navigate by raising the head end from time to time, then bending it left and right to sample changes in the sensed perception (e.g., light intensity, odor intensity). The head is then lowered  and the maggot crawls forward, realigning its body with the position of the head as it does so. This is similar to ecoli-style movement: presumably if the maggot is controlling for moving up the gradient, a signal of decreasing intensity will soon trigger the equivalent of a “tumble� (changed head orientation) and the maggot heads off in a new direction. If the intensity is increasing, the maggot continues more or less straight ahead. However, a difference from e. coli is that the maggot can sense changes in intensity as it wags its head and thereby choose the new direction based on these intensity changes. E. coli’s changes in direction following a tumble are random.

Â

It occurs to me that the sampling phase takes up time, during which the maggot is stationary. Could this be the origin of the power-law relation between curvature and angular velocity? Because turns tend to occur at such times, the maggot would take more time to execute a turn than to continue moving straight ahead. In other words, its velocity would be lower through the turns than when moving straight.

Â

I haven’t taken the time to read the paper in detail yet, but what I’ve seen of it is quite impressive.

Â

Bruce

Richard S. MarkenÂ

“The childhood of the human race is far from over. We
have a long way to go before most people will understand that what they do for
others is just as important to their well-being as what they do for
themselves.” – William T. Powers

–

[_Boris_Hartman]

I’m sorry Alex I’ll answer blindly on Ricks’ post  as I couldn’t open your post. So I don’t know whether you answered Bruce or Rick.  I don’t know why my Outlook is blocking your message.

···

From: Richard Marken [mailto:rsmarken@gmail.com]
Sent: Monday, September 19, 2016 7:00 PM
To: csgnet@lists.illinois.edu
Subject: Re: Maggot Moves

[From Rick Marken (2016.09.19.1000)]

Bruce Abbott (2016.09.18.1615 EDT)–

BA: Alex sent me a link to another of his papers on the movement of fruit-fly maggots; you can find it at https://elifesciences.org/content/4/e06694 .

RM: This is a really impressive piece of work, Alex. No wonder you think I’m an idiot. This research displays an impressive level of skill in mathematics, biology, neurophysiology and research instrumentation. And it was carried out in a Spanish research lab and written up in impeccable English.

RM: I read it over and I got a vague idea regarding what you found. I think the only thing I can “contribute” from a PCT perspective is just to recommend that you look at the whole process of chemotaxis as control of input rather than control by input. This is a pretty high level suggestion but it might influence the way you go about studying the chemotaxis process.

RM: The larval chemotaxis takes place in a loop, of course, so while input affects output (the bending movements) at the same time the output affects input.

HB : It seems Rick that you can’t be stopped. I hope that Alex will not notice your RCT mistakes or at least he’ll not consider them. I’ve got really good impresiion of him.

All together your contribution is mostly a little better this time, but stlil contain your basic RCT arguments.Â

I don’t understand how you always slip at the same aspects of PCT which are clear. I really don’t beleive that nobody noticed that your life is happening »at the same time«. Conttol events never happen at the same time…

Bill P (LCS III) : When the states of the variables are expressed as small whole numbers it would seem to work in imagination perfectly and instantly. If the disturbances produces 5 units of effect on the controlled variable, and the controller produces (with a properly time-delay) -5 units of effect it follows that a net effect….etc…

HB : Bill literally wrote »properly time-delay«. So you are definitelly not talking about PCT but RCT. How can you control anything at the same time ? I think you could stop writing nonsense. Barb do something. I just hope that members on CSGnet don’t beleive that they can see that somebody is shooting at them and they can remove at the same time from the bullet trajectory. And I hope they don’t beleive to Rick’s fary tale that they can »protected from disturbances« (bombs, bullets, knives, etc) what is probably equally to Rick’s beleive that evrything in the loop happen at the same time.

RM : Assuming the relevant input is, indeed, the chemical gradient through which the larva moves, PCT would see the bending movements as the means of controlling sensed chemical concentration. And if I am reading Figure 1A correctly, it looks like the larvae move so as to maintain the sensed gradient at some intermediate level; they are not moving directly to the highest concentration (in the center of the Figure). This suggests that, in this situation, at least, the reference for sensed concentration is not to have it constantly increasing but, rather, maintained at some intermediate level. I would imagine that this constancy should be reflected in the neural signal that is presumably proportional to sensed concentration. But then Figure 1C seems to suggest that they head into the gradient and increase sensed concentration quite a bit at first but then bring it down to a relatively constant level. If sensed concentration is, indeed, a controlled variable, it would be interesting to see if the pattern in Figure 1C reflects a change in reference or simply the behavior of the controlled variable in a very sluggish control system.

HB : What by your oppinion could be »controled variable« here but »sensed concentration« ? Is there any other possibility ?

RM: One more thing; I think the chemotaxis process used by the larva is probably similar to that used by E. coli, to the extent that the direction of bend is random.

HB : Why didn’t you write that direction of band is »controlled« if by your oppinion »behavior is process of control« in general sense ? So behavior can be also random, it’s not always controlled in RCT ? So generally speaking »behavior is not control« ?

RM:Â But if I am right about the idea that the larva are controlling sensed concentration relative to a reference value, then this should be taken into account in your efforts to determine the input-output function relating sensed concentration to probability of bending. Presumably the probability of bending should decrease as concentration increases, but only up to a point (the reference point) at which point the probability of bending should start to increase again as concentration further increases.

HB : Presumably I would think that probability of bending depends on difference between sensed concentration (perceptual siignal) and reference value.

RM: Anyway, it is a lovely piece of research. It would be wonderful if you could apply all these research skills to studying the behavior of these beasts from a control theory perspective - one that views their behavior as a process of control of input.

HB : Here we goa again. Behavior does not control input. Behavior is no kind of control proccess not in larva, neither in E.Colli and not in you.Â

But Bill suggested also other possible explanation of how chemotaxis work in E.Colli.

RM : Then the next paper you write might be titled: “Chemotaxis guides perceptual features extracted dynamically at the sensory periphery”.

Best,

Boris

Best regards

Rick

Maggots are somewhat worm- or caterpillar-like critters that navigate by raising the head end from time to time, then bending it left and right to sample changes in the sensed perception (e.g., light intensity, odor intensity). The head is then lowered and the maggot crawls forward, realigning its body with the position of the head as it does so. This is similar to ecoli-style movement: presumably if the maggot is controlling for moving up the gradient, a signal of decreasing intensity will soon trigger the equivalent of a “tumble� (changed head orientation) and the maggot heads off in a new direction. If the intensity is increasing, the maggot continues more or less straight ahead. However, a difference from e. coli is that the maggot can sense changes in intensity as it wags its head and thereby choose the new direction based on these intensity changes. E. coli’s changes in direction following a tumble are random.

It occurs to me that the sampling phase takes up time, during which the maggot is stationary. Could this be the origin of the power-law relation between curvature and angular velocity? Because turns tend to occur at such times, the maggot would take more time to execute a turn than to continue moving straight ahead. In other words, its velocity would be lower through the turns than when moving straight.

I haven’t taken the time to read the paper in detail yet, but what I’ve seen of it is quite impressive.

Bruce

–

Richard S. Marken

“The childhood of the human race is far from over. We have a long way to go before most people will understand that what they do for others is just as important to their well-being as what they do for themselves.” – William T. Powers

[chadtgreen]

[Chad Green (2016.09.20.1151 EST)]

CG: I received Alex’s message with the movies embedded in the e-mail. Here are the contents of his e-mail:

AG: If you want to see two of the first movies I took of Drosophila larvae crawling towards/around an odour source (suspended in one of the circles and out of
reach of the animal), so that you can get a feeling for the phenomenon, find them attached.

Larva approach without tracking

https://drive.google.com/file/d/0B6VC9UCXd0kyYUkyUmtlUmZvY1U/view

Larva circle with tracking

https://drive.google.com/file/d/0B6VC9UCXd0kyV2ZReHI1YXpqMXc/view

···

Chad T. Green, PMP

Research Office

Loudoun County Public Schools

21000 Education Court

Ashburn, VA 20148

Voice: 571-252-1486

Fax: 571-252-1575

“We are not what we know but what we are willing to learn.â€? - Mary Catherine Bateson

From: Boris Hartman [mailto:boris.hartman@masicom.net]
Sent: Tuesday, September 20, 2016 11:43 AM
To: csgnet@lists.illinois.edu
Subject: RE: Maggot Moves

I’m sorry Alex I’ll answer blindly on Ricks’ post as I couldn’t open your post. So I don’t know whether you answered Bruce or Rick. I don’t know why
my Outlook is blocking your message.

From: Richard Marken [mailto:rsmarken@gmail.com
]
Sent: Monday, September 19, 2016 7:00 PM
To: csgnet@lists.illinois.edu
Subject: Re: Maggot Moves

[From Rick Marken (2016.09.19.1000)]

Bruce Abbott (2016.09.18.1615 EDT)–

BA: Alex sent me a link to another of his papers on the movement of fruit-fly maggots; you can find it at
https://elifesciences.org/content/4/e06694 .

RM: This is a really impressive piece of work, Alex. No wonder you think I’m an idiot. This research displays an impressive level of skill in mathematics, biology, neurophysiology and research instrumentation. And it was
carried out in a Spanish research lab and written up in impeccable English.

RM: I read it over and I got a vague idea regarding what you found. I think the only thing I can “contribute” from a PCT perspective is just to recommend that you look at the whole process of chemotaxis as control
of input rather than control by input. This is a pretty high level suggestion but it might influence the way you go about studying the chemotaxis process.

RM: The larval chemotaxis takes place in a loop, of course, so while input affects output (the bending movements) at the same time the output affects input.

HB : It seems Rick that you can’t be stopped. I hope that Alex will not notice your RCT mistakes or at least he’ll not consider them. I’ve got really
good impresiion of him.

All together your contribution is mostly a little better this time, but stlil contain your basic RCT arguments.

I don’t understand how you always slip at the same aspects of PCT which are clear. I really don’t beleive that nobody noticed that your life is happening
»at the same time«. Conttol events never happen at the same time

Bill P (LCS III) : When the states of the variables are expressed as small whole numbers it would seem to work in imagination perfectly and instantly. If the disturbances
produces 5 units of effect on the controlled variable, and the controller produces (with a properly time-delay) -5 units of effect it follows that a net effect….etc

HB : Bill literally wrote »properly time-delay«. So you are definitelly not talking about PCT but RCT. How can you control anything at the same time
? I think you could stop writing nonsense. Barb do something. I just hope that members on CSGnet don’t beleive that they can see that somebody is shooting at them and they can remove at the same time from the bullet trajectory. And I hope they don’t beleive
to Rick’s fary tale that they can »protected from disturbances« (bombs, bullets, knives, etc) what is probably equally to Rick’s beleive that evrything in the loop happen at the same time.

RM :

Assuming the relevant input is, indeed, the chemical gradient through which the larva moves, PCT would see the bending movements as the means of controlling sensed chemical concentration. And if I am reading Figure 1A correctly, it looks
like the larvae move so as to maintain the sensed gradient at some intermediate level; they are not moving directly to the highest concentration (in the center of the Figure). This suggests that, in this situation, at least, the reference for sensed concentration
is not to have it constantly increasing but, rather, maintained at some intermediate level. I would imagine that this constancy should be reflected in the neural signal that is presumably proportional to sensed concentration. But then Figure 1C seems to suggest
that they head into the gradient and increase sensed concentration quite a bit at first but then bring it down to a relatively constant level. If sensed concentration is, indeed, a controlled variable, it would be interesting to see if the pattern in Figure
1C reflects a change in reference or simply the behavior of the controlled variable in a very sluggish control system.

HB : What by your oppinion could be »controled variable« here but »sensed concentration« ? Is there any other possibility ?

RM: One more thing; I think the chemotaxis process used by the larva is probably similar to that used by E. coli, to the extent that the direction of bend is random.

HB : Why didn’t you write that direction of band is »controlled« if by your oppinion »behavior is process of control« in general sense ? So behavior
can be also random, it’s not always controlled in RCT ? So generally speaking »behavior is not control« ?

RM: But if I am right about the idea that the larva are controlling sensed concentration relative to a reference value, then this should be taken into account in your efforts to determine the input-output function relating
sensed concentration to probability of bending. Presumably the probability of bending should decrease as concentration increases, but only up to a point (the reference point) at which point the probability of bending should start to increase again as concentration
further increases.

HB : Presumably I would think that probability of bending depends on difference between sensed concentration (perceptual siignal) and reference value.

RM: Anyway, it is a lovely piece of research. It would be wonderful if you could apply all these research skills to studying the behavior of these beasts from a control theory perspective - one that views their behavior
as a process of control of input.

HB : Here we goa again. Behavior does not control input. Behavior is no kind of control proccess not in larva, neither in E.Colli and not in you.

But Bill suggested also other possible explanation of how chemotaxis work in E.Colli.

RM : Then the next paper you write might be titled: “Chemotaxis guides perceptual features extracted dynamically at the sensory periphery”.

Best,

Boris

Best regards

Rick

Maggots are somewhat worm- or caterpillar-like critters that navigate by raising the head end from time to time, then bending it left and right to sample changes in the sensed perception
(e.g., light intensity, odor intensity). The head is then lowered and the maggot crawls forward, realigning its body with the position of the head as it does so. This is similar to ecoli-style movement: presumably if the maggot is controlling for moving
up the gradient, a signal of decreasing intensity will soon trigger the equivalent of a “tumbleâ€? (changed head orientation) and the maggot heads off in a new direction. If the intensity is increasing, the maggot continues more or less straight ahead. However,
a difference from e. coli is that the maggot can sense changes in intensity as it wags its head and thereby choose the new direction based on these intensity changes. E. coli’s changes in direction following a tumble are random.

It occurs to me that the sampling phase takes up time, during which the maggot is stationary. Could this be the origin of the power-law relation between curvature and angular velocity?
Because turns tend to occur at such times, the maggot would take more time to execute a turn than to continue moving straight ahead. In other words, its velocity would be lower through the turns than when moving straight.

I haven’t taken the time to read the paper in detail yet, but what I’ve seen of it is quite impressive.

Bruce

–

Richard S. Marken

“The childhood of the human race is far from over. We have a long way to go before most people will understand that what they do for others is just as important to their well-being as what they do for themselves.” – William T. Powers