Wednesday, May 03, 2006

Monday, May 1, 2006 – A practical on electrocomunnication from Dr. Joseph Bastian

Dr. Bastian studies the neurobiology of the weakly electric fish, Apteronotus leptorhynchus. He first gave us an overview of the electrosensory system and then showed us how communication can be mimicked in the lab. Here’s a short summary of the overview:

Many fish are electroreceptive meaning they can sense electric signals in the environment via specialized receptors. Certain species (e.g. sharks) are highly sensitive to electric signals and preferentially use this information to detect prey. (Electrical signals in fish are produced by the muscular potentials and the gill potentials; this is what the fish cue in on.) Even though many fish can use electric signals of other animals, most have a passive electrosensory system. Electric fish have an active electrosensory system in which they produce an electric field around themselves and sense their own signal. There are two types of electric fish, those that produce pulses of electric charge and those that produce it in a continuous wave. In the latter, the electric receptors are tuned to the frequency of that species. This, in effect, gives each sympatric species, its own “channel” to use to communicate.

Electric fish use this specialized sensory system to navigate through the turbid streams they live in and to communicate with each other. In navigation/orientation, when an object is placed near the fish, it disturbs the electric field and an electric “image” is projected onto the skin of the electric fish. This changed is registered by the electroreceptors and results in an increase in action potential frequency that is processed by the brain. When two electric fish are next to each other, they can both sense the electric organ discharge (EOD) of the other. The two overlapping waves produce a beat frequency between the two fish that is dependent on the difference in the two frequencies. These beats can be used as social signals or one fish may try to “jam” the signal of the other fish by producing a signal of the same frequency. Jamming makes fish have a hard time locating objects nearby and most try to avoid it actively. This behavior is called the jamming avoidance response (JAR). The higher frequency signaler will produce an even higher frequency and the lower signaler will produce a lower frequency to increase the difference in EOD frequencies between them. This separates the two signals so that each can be processed with less interference. One study showed an opposite response. When two fish were placed together, instead of increasing the difference in their frequencies, they decreased it to both produce the same frequency. This is called phase locking and may be used so that each fish can still detect predators and prey in the environment. Besides jamming avoidance and phase locking, electric fish can also use rapid changes in frequency and amplitude of the EOD signal to communicate to male or female conspecifics. A. leptorhynchus males produce two such signals called chirps due to the short high frequency EOD burst they make. Type I chirps are much higher in frequency than the male’s normal wave and are thought to be a sexual signal to females. Type II chirps are slightly higher in frequency and are thought to be an agonistic signal to other males. Females also produce type II chirps.

When we went down to the lab, we were able to see (via an oscilloscope) the EOD waveform that an A. leptorhynchus makes. The fish was placed in a small chamber hooked up to two sets of electrodes where we could not only record the EOD the fish was producing, but we could also simulate another fish by sending an electric current of similar frequency through the chamber. When we set the frequency to the same value as the fish in the chamber, we observed the JAR. We also observed several type I and type II chirps. When Dr. Bastian placed a Sternarchorhyncus sp. fish in the chamber, it almost immediately phase locked onto the signal we were sending. Dr. Bastian said he had never seen anything like that before and we were all impressed. Not much is known about this species and so there are many possible answers as to the function of this behavior. This species, along with most of the weakly electric fish, live in a habitat that makes studying the field behavior logistically difficult. However, if you could find a way around those difficulties, it seems like there are a wealth of questions to be answered about the evolution of electrocommunication in this unique group of fish.

Wednesday, April 26, 2006 – Yet more discussion and a letter from Kate Beckingham

So, today we were supposed to settle on an experimental design for our class project. Unfortunately, we spent the entire hour going round and round over the pitfalls of the experimental design for each – actually it was mostly the privacy study. I think it is the more interesting study and as we said on Monday, it is more marketable to a wide audience. Matt thought that there was no way to set up an experimental design that would definitively show whether one of the two fish, the winner or the loser, was controlling the level of privacy. I thought there was and Laura and I kept throwing out ideas on how to improve the experimental design. Matt was not convinced and said that there were too many lab artifacts in the design now to have it mean anything. About 30 minutes after class had ended, I finally defaulted to doing the chemical audience experiment, even though I think it has as many potential flaws as the privacy study. In the end, I think we went as far as we could guessing at what the fish would do in each study. I think we needed to take what we had and do some pilot work to come up with an experimental design that would work.

On another topic, earlier in the semester we were discussing insect eyes and I noticed in a picture that flies have a bristle between each ommatidium (Bradubury & Verhencamp, figure 9.4). When Kathleen M. Beckingham came later to give a seminar (“Using Genetics to Investigate Gravity Responses in

the Fruit Fly Drosophila melanogaster”), I asked her about whether these bristles were mechanoreceptors. She didn’t know off the top of her head, but she took my email and said she would get back to me. Just Monday, I received her reply. With her permission, here is her email:

Dear Sunny,

in cleaning up my desk recently, I came across the paper with your
question concerning the bristles in the fly eye and whether or not
these are sensory bristles. I sent your query to one of our local
fly eye experts and he tells me the following.

He can't say from actually having looked, but he believes that those
bristles are sensory because he knows that a neuron definitely
differentiates at the base of each of the eye bristle positions and
is present at about 48 hours after pupariation. So unless those
neurons die, which would be unexpected, the eye bristles must be
presumed to be sensory.

Thanks for bringing this up - I've often wondered myself and your
query gave me the energy to investigate.

Kate B.

Kathleen M. Beckingham
Professor of Biochemistry and Cell Biology
Rice University

I have to thank Dr. Beckingham for pursuing this question for me!

Monday April 24, 2006 – More experimental design and self assessment

We decided that the two experiments involving mate choice of females while an audience is present would have too many confounding variables. The biggest problem we faced was that we couldn’t figure out how to assure that males would have an audience of females present, yet not display to them and ignore the females he was supposed to be choosing. So that left us to choose between the chemical audience study and the privacy study, which we will do on Wednesday.

Self Evaluation

How did I do?

Overall, I felt that I had a pretty good grasp of the material and contributed to the class discussions in a way that helped both me and the other students explore new ideas.

How much did I learn?

I think I learned quite a lot. Much of the information we covered I already had some knowledge of. However, this class helped me solidify quite a few things. It was also very refreshing to hear a new perspective on several topics that I had learned before. Other professors I had in the past had very solidified ideas on certain topics and I was expected to accept those ideas as the way it was. Even though most of them were minor things, I liked the evolutionary approach we took in this class and how most topics were treated not as black-and-white, but with shades of grey.

Did the course meet my expectations?

Yes, the course did meet my expectations. I am glad that I took this class. I definitely got out of it what I was looking for and a bit more. I liked the freedom in the class to pursue a topic that just came up. I really feel most of my learning happened at these times.

Did my performance meet my expectations?

I felt my performance at the beginning of the semester was not quite as good as I would have liked it to be. I wasn’t getting as in-depth into the reading as I like to and I know I made a few strange comments (well, stranger than usual). Because I was changing my degree program and applying to another school, I was pre-occupied and time constrained. After spring break, however, things started to become more settled and I had more time to devote to this class. This was nice for me because I have enjoyed the readings as well as the class discussions. I liked feeling prepared and being able to contribute more in class.

What/how much did I contribute?

I think I contributed by asking the questions that came to me from the reading and the lectures. I usually think and therefore ask questions that are a bit off-topic. As I said before, I was happy to be encouraged to do this and I think that this contributed to many class discussions. I also hope that some of my insights as a physiologist contributed to everyone’s learning. As for how much I contributed, I often felt like I was contributing too much. I don’t like awkward silent moments and I usually say what I am thinking. These two factors combined with the fact that other class members often weren’t as vocal, lead to many situations where Matt and I were the only ones talking. I was aware of this and was hoping the others would contribute more as the class went on, but not much changed.

Monday April 17, 2006 – More brainstorming and a lesson on side bias

So we further discussed the potential experimental designs. I think that the chemical communication experiment would be the easiest logistically, but with the large potential of producing negative results or relatively uninteresting results. On the other hand, I think the other experiments might have more interesting outcomes, but are harder logistically. For the two where we would use a video playback of females as the audience, there is concern that the male would spend all of his time courting the playback female instead of the stimulus female. The privacy study also has its challenges in that it is a study where you have two animals that can control the level of privacy/publicity they have while fighting. This confounds the study in that there is no subject animal and no stimulus animal – the movements of each are dependent on the other and determining which is contributing what to the interaction could be difficult (especially with Betta splendens as the model).

Along with discussion of the experiments, we got an impromptu lesson in how to design an experiment to reduce the effect of side bias. Ingo told us of his preferred approach: have a priori rules that identify exact what constitutes a side bias, then switch the stimuli for every individual. If the animal spends >80% of the time on one side, it has a side bias and can be thrown out of the analysis. However, the rules for determining whether an animal has a side bias must be set before the trials begin.

Wednesday, April 12, 2006 – Brainstorming for class project

We began throwing out ideas for our class experiment on audience effects. I think we were all a bit unsure as to what was possible. We wanted an experimental design that would be fairly (and not too costly) to set up and run. Yet, in the hopes that we might follow in the footsteps of the Losey et al. (1986) graduate class, we wanted something that might break new ground. After wrestling with it for a while, since asexual Amazon mollies copy the mate choice of Mexican mollies, I asked whether we could fool Amazon mollies into making the “wrong” choice by letting her eavesdrop onto a Mexican molly choosing that male. We came up with several experimental designs, but in the end, we had not made much progress in deciding on one. So for Monday, we are to come up with the relevant background for each potential experiment, the putative set-up and potential criticisms that we might receive from reviewers. Here is the link to the forum:

Monday, April 10, 2006 – More on Eavesdropping

We discussed the information content of signals that have many components. How do you know which component carries the information and which component(s) are secondary? We discussed how this is true of many chemical signals some components of which are “sticky” and keep the scent there for a certain time and others which are volatile to increase the area of broadcast for the signal. I wonder if there is a way that animals can “hide” components of a signal that might point to a metabolic disorder or if, due to the way chemical signals are produced, there is only a certain degree to which one could “fake” an attractive chemical signal.

We also discussed whether eavesdropping could be costly to the eavesdropper if they were caught observing an interacting pair. In most human societies, eavesdropping is considered rude or bad behavior and often people (especially children) are admonished for listening in on a conversation of which they are not a part. Is this truly a human phenomenon or are there other animal societies like ours where there is a cost to eavesdropping?

Monday, April 3, 2006 – Animal Communication Networks, Matos and Schlupp

We started reading and discussing audience effects in class. An audience is any third party that can receive a signal that was actually intended for another individual. This third party can be visible (an audience) or hidden (an eavesdropper) to the individuals communicating. It can be a same sex or opposite sex conspecific or a heterospecific (predator, parasite or competitor for resources). The third party can also be actually present, likely present or just have been present over evolutionary history to effect the communicating pair’s interaction. This is a rather new way to conduct research in the field of animal communication because historically, experiments in this field have been done on two individuals, the signaler and the intended receiver. However, how often do animals in the wild interact in these dyadic pairs? In most animals that I can think of, it seems unlikely that many signals would be just between two individuals, especially in a species that is colonial or social. I think that research in this area may enlighten and possibly overturn much of our previous knowledge about how animals communicate.

Wednesday, April 5, 2006 – More on Audience Effects

Today we discussed how the audience member can use the information gained from watching two individuals interact.

1. They can determine their ability to win an interaction with either party and then decide whether to stay in the area or leave.

2. Observing the interaction can physiologically prime the audience member for a similar interaction (e.g., mating, fighting).

3. The audience member can determine information about the social structure (i.e., who grooms who in a primate group).

Coming from a different perspective of behavioral endocrinology, I think there are many studies that relate to the physiological priming component, but are not necessarily framed from that evolutionary context of eavesdropping/audience effects. One of the studies I had mentioned in class was that female goats mount each other (probably for an evolutionary audience, although this was not studied), which results in increased sexual behavior from males that are watching. Although there may be other dominance or social signaling occurring between the females, this interaction acts as a priming signal for sexual behavior in males. Since some in the class were reticent to believe my retelling of this study, here is the reference and selected parts of the abstract.

Female-female mounting among goats stimulates sexual performance in males.

Shearer MK, Katz LS. Horm Behav. 2006 Feb 24

The hypothesis that female-female mounting is proceptivity in goats, in that male goats are aroused by the visual cues of this mounting behavior, was tested. Once a week, male goats were randomly selected and placed in a test pen in which they were allowed to observe one of six selected social or sexual stimulus conditions. Viewing mounting behavior, whether male-female or female-female, increased the total number of sexual behaviors displayed, increased ejaculation frequency, and decreased latency to first mount and ejaculation, post-ejaculatory interval, and the interval between ejaculations. We conclude that male goats are aroused by the visual cues of mounting behavior, and that female-female mounting is proceptivity in goats.

Monday, March 23, 2003 – Special Guest, John Endler

Wow. What an incredible visit from one of the great scientists in animal communication. Dr. Schlupp had said a short conversation with Endler was like reading a textbook and it was. Endler was a wealth of knowledge and easy to talk to as well. He had a unique perspective and would bring out points in a topic that no one else had. For example, I asked him about my research. I had always thought that the signaling that happened between Gulf pipefish was largely visual. Since Endler is so knowledgeable about visual communication systems, I was geared up to talk to him about the sexually dimorphic banding, the temporary barring that females display during competitions and the possibility of UV signaling in the species. However, I mentioned that they make random, intermittent small clicks and Endler got really excited about the possible role of auditory communication in this fish that relies heavily on crypsis. I also mentioned that males and females do a display that looks like a shudder – males during sexual interactions and females during both sexual and competitive interactions. Since the movement is largely in the trunk area where the barring is on females, I thought the shudders drew visual attention to that area. However, Endler thought there was likely a large component of lateral line signaling going on in the shudder. This makes sense because the animals are displaying laterally in close proximity, within a couple of centimeters, during this interaction. I now have several new lines of research for the future.

Another topic Endler talked about that I found very interesting was the use of carotenoids in color signaling. He said that many color signals in animals are likely due to deposition of waste products. He said that deposition of these waste products in the tissues is more efficient than to remove them via the kidneys. He told us that the hue in guppies is a product of metabolic function; the carotenoids are deposited in the tissues to balance the production of pterapines. This, then, is a direct cue to the animal’s metabolic function. I like this explanation better than I do the hypothesis that the coloration is a signal of immune function. Previous studies on this latter hypothesis have ended in inconclusive results. This could be due to the fact that immune function is correlated with (but not directly resultant from) metabolic function. I am curious to see how further research on this topic turns out.

Tuesday, May 02, 2006

Wednesday, March 22, 2006 - Dr. Martin Plath’s presentation on non-visual communication in a cave dwelling fish

Martin presented some of his research on the evolution of mate choice when species move from a surface-dwelling habitat to a cave-dwelling habitat. He conducted shoaling and mate choice tests with a surface species and a cave species in the light, in the dark, and with a solid, clear barrier or a mesh barrier between fish. He showed that the surface species does show shoaling behavior in the dark. The cave form only shoals when a wire mesh is the barrier. He showed that surface females can detect the presence of males in the dark and spend more time near males than the empty chamber. However, only the cave fish show a preference for larger males in the dark. He also did a rearing/common garden experiment that showed the sensory shift to non-visual communication in cave fish was due to phylogeny, not ontogeny.

Martin only distinguished between visual and non-visual communication in his experiments. Although he demonstrated that both surface and cave fish do use non-visual communication, he did not make any suggestions as to which sensory systems might be primarily responsible for the non-visual communication. He mentioned in the beginning that most fish sense non-visual cues through the ears, nose, taste buds, lateral line, and electrosensory receptors. From the time course in Martin’s mesh-barrier trials, it is likely that the cave fish uses the lateral line to communicate in the dark. It is easy to see how the fish could judge the size of a potential mate or rival in this fashion. Martin and I were talking later about chemicals that could temporarily disable the lateral line. I started considering it for my studies as well. It would also be interesting to know if these or other fish rely on electrosensory receptors in the absence of visual cues. Although rather off topic, I thought one of the most interesting things Martin said was that fish have taste buds along the side of their body. Although I didn’t find much on them, here are some references:

Density and distribution of external taste buds in cyprinids

Andreas Gomahr1, Margit Palzenberger1 and Kurt Kotrschal1

Issue: Volume 33, Numbers 1-2

Date: January 1992

Pages: 125 - 134

Taste bud types in fishes II. Scanning electron microscopical investigations on Xiphophorus helleri Heckel (Poeciliidae, Cyprinodontiformes, Teleostei)

Klaus Reutter1 Contact Information, Winrich Breipohl3 Contact Informationand Gerhard J. Bijvank5

Issue: Volume 153, Number 2

Date: November 1974

Pages: 151 - 165

Cutaneous taste buds in cod

R. Harvey1* and R. S. Batty1

Journal of Fish Biology
Volume 53(1) Page 138 - July 1998

Monday, March 20, 2006 – Bradubury & Verhencamp, Chapter 17

The topic of chapter 17 is the costs and contstraints on signal evolution. In class we discussed necessary and incidental costs, which helped clarify these topics for me. Necessary signal costs are the essential costs for producing the signals and the structures used to produce the signals. Incidental costs are those that are incurred (by the receiver) when there is disagreement, miscommunication or manipulation between the individuals communicating. The receiver could suffer costs from lost energy, time or information.

We then discussed the constraints on the type of signals that can be produced and received. Physiological constraints within the animal and physical limits in the environment limit signals and sensory systems. For example, an animal’s eyes are either specialized to have high resolution or they are good at motion detection, but they can’t be good at both. We briefly discussed the eyes of spiders. Spiders have several pairs of eyes that are specialized for different functions. Some see in low light, some see in daylight and others are specialized for motion detection. I began wondering about the evo/devo of these eyes. Was there a gene duplication (or several in succession) that resulted in a duplicate set of eyes, the redundancy of which could then be acted upon by natural selection? How did these multiple eye sets evolve and why don’t more animals have them? If you look at the phylogeny of spiders (or arachnids), what is the pattern of eye specialization? Is it a linear progression of specialization or are there many innovations across the branches? I found a good site about spider eyes: and although it doesn’t talk about phylogeny, it gives good details about the types of eye specializations for the environment and behavior of the species. I will leave the link for you to read if you like, but I have to share this part:

Search-light eyes
Net -casting spiders have eight eyes, but in one genus, Deinopis, two of the rear eyes (PME) are enormously enlarged. Their great, curved lenses face forward like twin search-lights, giving the spiders a rather menacing appearance (the 'ogre-faced spiders').

The two biggest eyes are specialised for providing outstanding low-light night vision. They have enormous, compound lenses that give a wide field of view and gather available light very efficiently. The lenses have an F number of 0.58 which means they can concentrate available light more efficiently than a cat (F 0.9) or an owl (F 1.1). Each night a large area of light sensitive membrane is manufactured within these eyes (and rapidly destroyed again at dawn).

Can you believe that last line! What an incredible specialization!

Wednesday, March 8, 2006 – Bradubury & Verhencamp, Chapter 16

Today we discussed chapter 16 in class and we went over some of the issues that had come up in the reading. We talked about signal evolution and how most signals likely evolved in response to natural selection pressures. As long as those traits have variability, there is room for that signal to evolve in a different context.

We discussed how pre-adaptations in receivers have to arise in a different context, likely a natural selection context. Since this pre-adaptation could have evolved in a common ancestor, using phylogenetic analyses is a powerful way to support a pre-existing female preference. This reminds me of the study done by Cisnerso, Bass and colleagues where the peak sensitivity of the auditory neurons changed seasonally in female plain fin midshipman fish. During the breeding season when estrogen levels are high, females’ peak sensitivity is tuned to the frequency range of the hums of preferred males. During the non-breeding season when adults move into deep water and estrogen levels are low, females’ peak sensitivity is tuned to a much lower frequency, likely that of their main predators. This relatively new study highlights the fact that sensory systems can have plasticity. Do receivers in most species go through this kind of seasonal plasticity, and if so, what does that do to our current understanding of sensory exploitation?

Monday, March 6, 2006 – Bradubury and Verhencamp, Chapter 16

The topic of chapter 16 is signal evolution. One of the areas I find most intriguing is that of sensory exploitation where signalers use signals that exploit the pre-existing biases in the receiver’s sensory system. According to the theory, for example in mate choice, females have a preferred optima for a signal (for reasons other than mate choice); males that produce signals closest to this optima will attract the most mates, even if females would not have otherwise chosen that male. But what drives the pre-exisiting sensory bias in the receivers? I guess in many cases, some force (or several) of natural selection drives the bias. Can natural selection explain all biases or are there other mechanisms that could also be driving the evolution? If female frogs preferred a lower call than males could produce, then males evolved the ability to produce calls that were at the females’ optima, would the evolution of that trait stop there? Is this where most sexually selected traits are in species that we observe today or are there other mechanisms that continue to drive the females’ optima still lower? The chase-away sexual selection hypothesis states that females evolve a more extreme preference to avoid exploitation, and then males evolve more exploitative signals. In this scenario, it is sexual conflict over acquiring the best mate that drives females’ optima to a more extreme point. However, this hypothesis is not very well supported and still has many critics.

I also wondered how many signals are just “advertisement” for other less conspicuous signals. I think we usually assume that the most conspicuous signals are those with the most information content, or those that are more likely to be honest signals. However, many signals may have evolved simply to gain the attention of other individuals. These attention-drawing signals may be a means to direct the receiver to other signals that do have more information content.

A note on the chronology of posts

I made this blog a couple of weeks ago and then promptly started having problems with it. So, I went back to my computer log. I think I have found the problem and I will now transfer my other entries onto the blog. However, they are going to be a bit out of order, so I have titled them with the date I originally wrote them.

Wednesday, April 19, 2006

Wednesday, April 19, 2006 - The Viewer program and its use in animal behavior trials

After an overview and a demonstration of the Viewer program, I just wanted to put down a few ideas on how the playback/Viewer system combo can be used in animal behavior trials. For those that need an explanation, the Viewer system is a computer program that helps you acquire, record and analyze video of animal behavior trials. The program allows you to specify the experimental parameters such as start/stop times, length of trial, specific areas where data will be scored/recorded, etc. The best part of the program is that it is able to use background contrast technology to actually track the animal. From that, it can give you the total time the animal spent in an area, the velocity at which the animal moves and even recognize some basic behaviors. The program can be set up in advance and all of the data can be gathered in the absence of the experimenter - a huge time saving device!

The video playback technique can be used in combination with the Viewer program to provide a standardized stimulus to the subject, then record the behavior in response to the animal(s) in the video. A video recording is made of a stimulus animal - a potential mate, competitor, audience, predator, parasite, etc., then played back on a video screen that is adjacent to the trial arena. This gives the experimenter the ability to provide a consistent pre-determined stimulus to the subject animals across trials - something that can not be done when using live animals as the stimulus. As long as the animals are responding in the same fashion as they naturally would, this type of stimulus reduces the variability in the study and gives the researcher more power to detect differences in the behavior in response to the stimulus.

Although I am sure this system could be used for a variety of species and set-ups, it seems especially conducive to trials that could be performed in a tank or an arena. So, it is ideal for studies of insects, fish, small to medium sized reptiles, small mammals and possibly birds that can be kept in relatively small enclosures. I guess depending on your technology, you could also apply the Viewer program to certain invertebrates and even bacteria! The time scale over which you record data can also be extended to far greater lengths or times of the day than are logistically possible for most human recorders (i.e. a continuous 48 hour period or the middle of the night). This latter function is limited only by the amount of hard drive space you have. This could shed light on certain animals that are sedentary or have limited periods of activity that are unpredictable. For example, you might want to study behavior of snakes over several days. Snakes often spend most of their time curled up in one spot, but you may be able to determine relevant behavior patterns by collecting data over a longer period of time. Studies such as this are usually undesirable if a human has to be the one to do all of the recording and are thus not conducted very often. However, using the Viewer program, you may be able to elucidate behaviors that are important ecologically, physiologically or evolutionarily.

Here is a link to the Viewer software website if you want to check it out further: