I’ve got this great undergraduate student in my lab who is (at least for now) exhibiting all the traits of the ideal student any professor would love to have around: he volunteers ~10+ hours per week here, if no one is around, he finds things to do to teach himself new skills, he reads journal articles on his own accord that are aligned with interests of the lab, AND he’s creative! Lately, he has been trying to tackle a rather substantial challenge I presented to the lab: designing a force plate that can measure forces under granular media.
OK. Here’s the context: I used force plates to measure how hard and in which directions an animal is pushing when it steps on the ground. By combining this with measured movements of the animal that I quantify by analyzing synchronized high-speed video, I can calculate how much power the muscles must produce around each joint in order for the animal to move the way it does. This technology has been commonly used since the mid-1980s or earlier, for locomotion on flat, homogeneously hard surfaces. While we know a lot about how animals move across these types of “lab environment” surfaces, we know far less about movement over natural surfaces that may shift or change squishiness, orientation, etc. with each step.
Granular media, such as sand, therefore, is a particularly interesting material to me, as it actually makes state changes as an animal moves. For example, when sand is sitting undisturbed, it resembles a solid. Yet, when an animal strikes the surface and strokes through it with its foot, the sand actually becomes a fluid for a short while. Any sand that is kicked up during the step is actually acting like a gas! With all this in mind, measuring forces on sand can be a rather challenging conundrum.
In a meeting with my undergraduate student last week, he presented a design to me that involved peppering a surface with a grid of lumps that each could sense fluid movement. Little did he know, what he was showing me was something holding remarkable resemblance to hair cells, sensory receptors that are found in our ears AND in the lateral line system of fishes!
Ever wonder how a school of hundreds of fish manages to… school with such regularity and neat, synchronized prowess? The next time you catch that rainbow trout, sunfish, sailfish, or whatever strikes your fancy, take a close look at the side of their body and you’ll see a series of dashes and dots that run from the rear side of the gill margin all the down to the base of the tail. These little holes mark the external opening to the lateral line, the pressure-sensing secret for fishes.
Within these pores are receptors called neuromasts. Neuromasts look a little like a thimble placed open side down on a table top. Each neuromast is made up of a group of cells called hair cells, named because they (grossly) resemble a Leprocaun troll doll, with different length hair bundles mounted on top.
The entire neuromast is covered in a gelatinous glob, forming a cupula. Deflection of these hair bundles due to changes in fluid flow causes the production of a receptor potential. Deflection direction also stimulates the production of different types and magnitudes of potentials, enabling the fish to determine the direction of the flow. Changes in fluid flow direction or pressure can be due to an underwater obstacle, a neighboring fish, or even a predator, enabling a fish to respond seemingly magically while the approaching object is still far away.
It turns out that my student (I guess not so surprisingly) was not the first person to think of using the neuromast as a biomimetic sensor: a group of scientists at the University of Illinois, Northwestern University, and Institut fur Zoologie have developed nano hair cells they call ALL (for artificial lateral line) that can localize the position of a crayfish placed in a tank. Take a look at their paper, published recently in Bioinspiration and Biomimetics. Call it bias or whatever you want, I am still excited to see where (if anywhere) my student will take this biomimetic idea of his. Let me know if you have any input on how he can potentially use this to invent a new type of force plate technology, and I’ll put you in touch with my student to try to make this a reality!
There are countless stories (legends?) about dedicated faculty who conjure brilliant theories while showering, solve a major molecular roadblock while folding laundry, or dream up innovative interpretations of complex data. Every time I hear a story like this, I have felt… so envious. But then I would brush it off with the thought of, “Well, at least I’m not that big of a geek!”
My research strives to understand how we navigate through our complex universe with seemingly barely a thought. After all, how many of us have stood at the edge of a sidewalk in deep contemplation about how much we must flex our ankle, simultaneously activate our gastrocnemius and tibialis anterior, and then absorb the shock with the thick pad of flesh (and fat) on our heels combined with perfectly choreographed knee flexion, as we step down to the asphalt below? Should we have to go through such deep lines of thought, we would undoubtedly all go the way of the indecisive squirrel that gets 3/4 of the way across the road, just to turn around and throw itself under a tire in an effort to return to safety, rather than continue forth to the opposite side of the road a mere couple leaps away. To return to the point, we regularly move across a myriad of different surface types and barely notice it. To top it off, not only do we do it, there are tons of other organisms out there who do it better than we do. What’s going on?
To quantify movement across surfaces, my workhorse is something called kinematic analysis. This basically boils down to marking up an animal with a bunch of points (traditionally drawn on with correction fluid and a fine-tip Sharpie marker), filming with a high speed video, and then going through the record frame-by-frame clicking on each individual point across sometimes as many as 600 frames with 10-15 points per frame. Since our motion analysis is done in 3D, this means we then continue on to repeat this on a second camera (and in some instances, a third). Needless to say, this is a very time-consuming way to collect data, even when it is slightly automated with custom-written tracking software (e.g., see Ty Hedrick’s DLTdataviewer MATLAB program).
We recently purchased a super-duper auto-tracking high speed system to film our animal of choice: lizards. The system consists of six cameras that we mount in a ring, which automatically track reflective markers as the lizard runs through a calibrated filming volume. The amazing part of this system is that it literally can track up to 400+ points so accurately and quickly that we have all our tracked data the second the animal is done running through the field of view. The drawback is that we require markers covered in reflective 3M tape, which gets expensive fast, at $0.40 per mostly non-reusable marker. We are currently marking our animals up with 35-40 points, so burn through points at a rather high rate.
At least one reason for the high cost of these points is that they are literally individually hand-wrapped. Yet, for us “non-professionals” in the way of sticker-wrapping beads, this can be a risky endeavor as the point is useless if it does not effectively reflect light. Additionally, wrapping beads can be extremely time intensive and mentally dull for anyone — including one of our undergraduate helpers — to do. As a result, I have been in the midst of mental conniptions as of late, trying to figure out an inexpensive, easy way to build markers… all the while watching the dollars fall away as the lizards soak in their water bowls after a hard day’s exercise, destroying the markers so that they are unusable in future studies.
I woke up early this morning — too early by anyone’s standards at 4.30AM — and my brain started whirring. Perhaps it was due to the various discussions we have had in this course, or possibly it was due to one of our recent chapters from Janine Benyus’s book on Biomimetics, but I started seeing bubbles in my half-dream, half-awake state, and gradually began to realize (to my growing horror) that I had become a bona fide geek. These “bubbles” were actually phospholipid balls that were oriented with hydrophobic tails facing inwards and hydrophillic heads facing outwards. It was as if I had returned to Intro Bio my freshman year, only this time, it had really gotten in my head.
As I was lying there confused, disturbed, and exhausted but still determined to stay asleep enough to continue this thought process while awake enough to remember it, I realized that if there were some way to attach reflective glass beads to the hydrophobic heads, I could drop this solution into water to form balls that I could use as quick and easy, highly-reflective markers. Of course, I would also need to find a way to make them solidify so they maintain their shape in air… but that’s just my cynical, awake inner voice talking.
I started catching creepy crawlies before I could walk — or at least that’s what my parents insist I did. When I was two, I caught a spider in my preschool sandbox by cupping it with a baby food jar. I remember that spider just barely fit into the mouth of the jar because it was HUGE; or at least I certainly thought it was. With this in mind, it’s probably not hard to believe that there are few things that make me squirm… except daddy long legs.
To be totally clear, contrary to popular notion, daddy long legs are not spiders. They arise from an entirely different order and lack many of the segmentation and life history patterns observed in spiders.
Daddy long legs are characterized by having eight, long, spindly legs surrounding a bulbous body (although there are some short-legged forms). Their legs can sometimes be so thin that their much heavier body seems to float through the air as they gallump about their daily chores. When they have eyes, they only have one pair, which are placed sideways rather than facing forward. These eyes do not functionally project a usable image; so they have modified their second set of legs to effectively act as antennae to help them tap their way around as they either quietly ambush or actively chase down their prey. Unlike most other arachnids, which live off a liquid diet, daddy long legs consume their prey in chunks.
All these characteristics make daddy long legs rather sinister; yet, whenever my hair stood on end as a daddy long leg made an appearance in the room, I found myself inexplicably drawn to the same question over and over and over again: How do these animals support such a heavy body on such skinny legs?
I started looking in to this question several years ago and was stunned by what I found. Despite having such skinny legs, Schultz (2000) showed that the muscles are both numerous and complex. While the bulk of the muscle lies close to the hard, carapacial body, long tendons extend all the way to the tippy tips of the legs (often 50+ segments away!!), enabling fine, prehensile motion. This prehensile motion is used to help them climb thin structures such as grass blades, enabling them to wrap their leg completely around a single blade! More detailed studies by Guffey and colleagues (2000) on the microscopic morphology of these leg tips showed that there indeed is only a single tendon that extends to the toe tip, enabling prehensile motion. I couldn’t help but wonder how such a complex, prehensile motion across so many segments could be possible by means of a single tendon, and how this type of design could be applied industrially for highly mobile, exploratory devices… Thoughts? Does anyone know if something like this already exists in true mechanical models?