Featured Students
Our current featured students are: Beth Vernaleo (Birds), Vicky Hsiao-Wei Tu (Birds) , Genevieve Spanjer Wright (Bats) and Chris Glaze (Birds).
Beth Vernaleo
Beth Vernaleo always had an interest in Biology. As an undergraduate she decided to become a Neuroscience major because she figured it would help her in her quest to become an occupational therapist. That was before she caught the “Research bug”.
After Beth took an upper level Neuroscience research course, she realized she really enjoyed research, her favorite kind involving animal models of learning. It was then she decided to go to graduate school and study birdsong.
Why birdsong?
Birdsong is known to be a useful model for vocal learning and animal communication. It is a field that provides the grounds to answering multiple research questions.
In many songbird species, male birds sing to attract females and to defend their territories from other males. For some of her research projects Beth works with Zebra Finches.
Since becoming a graduate student she has worked at Dr Robert Dooling’s Laboratory of Comparative Psychoacoustics. She started out working on neural mechanisms of song learning and later became interested in perception of song.
Zebra Finch
The zebra finch is an interesting case amongst birdsong species. These birds sing only one song, which they use over and over during their lifetime.
Perception of song
One of the things Beth is looking into is which cues zebra finches pay attention to when they hear song.
Songs consist of brief bursts of sound (syllables) separated by silence (intervals). Each syllable is unique in its acoustic properties. Each syllable and interval in a song has a specific duration that is consistent from song to song. Below is a spectrogram of a motif, which is a specific ordering of syllables that is repeated over and over to form the song. The syllables are labeled as A, B, C, D, E, F, and G.
Research Project
In one set of experiments, Beth is testing whether zebra finches can detect changes made to their song motifs, and what types of changes are most easily detected.
To test this she is modifying two parameters: duration of single intervals and timing of single syllables. More specifically she is increasing the duration of single intervals and reversing single syllables.
By doubling the length of silence between two syllables in a motif, a change is being made to the overall rhythm and timing of song.
In reversing a single syllable, the syllable is flipped in time, and is basically played backwards to the bird. This changes the timing of the syllable’s fine structure, rather than the overall timing of song.
For each target motif, only a single change is made. Some examples of target motifs are below. Changes are marked by asterisks, and can be compared to the original motif.
Experimental setup
Birds are tested using a discrimination experiment.
The testing setup consists of a cage that has two LEDs (red and green) placed on the wall. There is also a speaker overhead through which sounds are played to the bird. The original song motif is played as a repeating background, and the bird must peck the left (red) LED as this motif is being played. At random times, a changed motif is played to the bird instead of the original motif. If the bird detects a change from the original motif, he must peck the right (green) LED. If he does so correctly, he receives a food reward. If he does not detect a change, then he must remain pecking the left LED. By recording which keys the bird pecks during the changed motifs, we can determine which changes the bird can detect, how quickly he can detect these changes, and which changes the bird seems unable to detect.
Preliminary results
So far results suggest that zebra finches are quite good at discriminating syllable reversals, but are poor at discriminating changes to interval durations. Thus, changes to the overall timing and rhythm of song do not seem to be as salient as changes to individual syllables. This may be due in part to how the birds are listening to song. Song contains many different types of information, including frequency and timing information. Perhaps the birds are paying more attention to frequency patterns within the syllables, but less attention to the overall timing of syllables.
Future research
Beth continues to examine the ability of zebra finches to make fine temporal discriminations, using birdsongs in which the spectral content has been replaced by white noise.
Vicky Hsiao-Wei Tu
Have you ever wondered what the birds are “talking” about outside your window? 
Is that flock of birds welcoming you along the hiking trail or does it simply want you to get out of its territory?
Whether animals have “language"is an interesting yet unsolved question. Although human language is fundamentally different from the communication systems of other animals, there are still some similarities, and we can learn more details about one another by studying these parallels.
Birds, especially parrots and songbirds, are commonly used as research models to study vocal communication and compare with human language.
Vicky Hsiao-Wei Tu is particularly interested in the vocalization of the budgerigar (Melopsittacus undulates), a small parrot native to Central Australia. Vocal communication is very important for this species to coordinate social and reproductive behaviors within large flocks. Even more interesting, the acoustic complexity, non-repeating structure, and correlation with intimate social behaviors of warble makes it similar to running human speech. Unfortunately, those same characteristics also make warble extremely difficult to analyze.
Example of a segment of a warble song (spectrogram)
With her experiments Vicky hopes to be able to provide a more complete portrait of budgerigar warble songs in order to further understand the way this species communicates in nature before relating specific properties to those of human nature language.
In the Laboratory of Comparative Psychoacoustics (Dooling lab) , Vicky and her colleagues have already developed a systematic way to efficiently segment long streams of warble into individual elements and then categorize them into eight distinct groups.
Using psychoacoustic methods, Vicky was able to train budgerigars to perform a discrimination task.
Results have shown that budgerigars can discriminate much better between the 8 groups and much worse (if at all) within any one of these groups. This means the birds’ perceptual groups match the groupings created by humans who used purely objective acoustic features of each warble element.
With these eight reasonable and universal categories of warble elements, Vicky has discovered that the relative proportion of different elements in warble of the same bird is distinct, but the distribution is similar across 3 budgerigars.
Vicky and her colleagues continue to provide better understanding of the structure of this complex vocalization in budgerigars by looking at other issues such as whether:
- a bird shows the same proportion of elements on different occasions
- the delivery of elements is ordered
- warble elements occur in isolation as calls (having different functions)
or how birds perceive either warble elements or sequences of warble elements can be pursued in the hope to get more insights of the “syntax” of budgerigar warble when comparing it with human speech.
Read more about Vicky's fascinating research and some more "budgie facts" here.
Genevieve Spanjer Wright (Bats)
The social life of bats
Genevieve Spanjer Wright (Genni) is interested in the social life of bats. In fact, she's so interested in it that she sends them out to dinner together so she can study their behavior. Although that description of her work is a bit of an exaggeration, it's no joke. Genni's work in Cindy Moss' lab focuses on social interaction in bats, animals that are more commonly studied for their ability to echolocate. There's more to bats than echolocation, however. Many species of bats are highly social, roost together in colonies, and use social calls to communicate with one another.
Bats learning from each other
Genni's experiments with bats focus on the role social interaction plays in learning and development. She wants to know if they learn behaviors from each other and, if they do, what role vocal communication may play in the learning process. She is able to study these types of interactions in the lab by using the lab's anechoic flight chamber to compare how quickly untrained bats learn to catch a suspended mealworm when they are paired with a trained partner vs. an untrained partner. She records all of the trials in her experiments with the lab's infrared tracking cameras and records sounds made by the bats with ultrasonic recording equipment. This allows her to study the flight paths of the bats and their social calls to look for evidence of social interaction during the learning process.
The results of Genni's experiments so far look promising. None of the bats paired with other untrained bats have learned to capture the mealworms, but half of the bats paired with a trained partner have learned the task completely, and most of them have had at least partial success.
Learning in the jungle
As a complement to her lab experiments, Genni has traveled to the Smithsonian Tropical Research Institute on Barro Colorado Island in Panama to study social interaction and learning in wild bats. While in Panama, she observes the native fruit eating bats foraging and interacting in the wild, with an eye toward social behavior. In addition, she conducts experiments with wild bats similar to those she performs in the lab. Genni's work in Panama allows her to study broader behaviors that impact the social learning process, such as whether bats choose to follow other bats to food or they prefer to forage alone.
Genni hopes that her work will lead to a better understanding of how bats learn and how flexible their behaviors are. She would also like to provide the scientific community with insights into social learning in a broader context, which could allow for a better understanding of the advantages and evolutionary pressures that lead to social grouping in animals.On the summer of 2007 she had a chance to do so with the aid of C-CEBH when she attended a mammology conference and presented a poster on her work.
Keeping bats out of harm's way
In addition to her work on social learning, Genni has conducted a set of experiments to test a bat-deterrent device. The device broadcasts broadband noise, including high-frequency sound in the same range as bat's echolocation calls. The device is intended for use in deterring bats from areas where wind turbines, a potentially fatal hazard for bats, are in use. Initial results with the deterrent look promising. Similar devices may eventually be used to prevent many bat deaths. If Genni ever manages to decipher the social calls of bats, perhaps someone should remind them to say thank you.
For more information on Genni's research and a list of her publications, see visit the Batlab's website.
NEW! Check out this article on our own Genni Spanjer Wright.
Chris Glaze (Birds)
A new tool for neuroscience
One could say that Chris Glaze's work requires an astounding attention to detail. Chris works in Professor Todd Troyer's lab, where they study song learning in the zebra finch. In the course of their research, the lab has recorded thousands of zebra finch songs from finches in all stages of development, from those just learning to sing to fully developed adults. That extensive pool of data allows Chris to study birdsong using an interesting new tool – time.
Chris's research focuses on the incredibly small variations found in different repetitions of a zebra finch's song. Although a zebra finch ostensibly always sings the same song, the vast amount of data at Chris' disposal allows him to examine minute variations in the time structure of the song. The reasoning behind this approach is simple, but subtle. The idea is that variations in song timing are produced by fluctuations in the operation of neural circuits responsible for producing song. Thus, by studying the variations in song timing, Chris should be able to make inferences about the neural circuits responsible for song production. In addition, any accurate model of song production should reproduce the patterns seen in the behavior.
One of the reasons Chris' approach to studying birdsong is so new is because such an analysis would not have been possible a decade ago. To perform a detailed study of such small timing variations requires the storage and analysis of large quantities of data, a process not available until recent years. In Dr. Troyer's lab, significant effort has gone into creating tools for recording, storing, and analyzing bird song. Much of Chris' data analysis is done automatically by computer. Chris screens the results manually to be sure the data was processed correctly, and provides the human brain necessary to interpret the results.
Birds are not music boxes
One of the results of Chris' research has been the discovery that when a song is longer or shorter than average, the individual components that make up the song all tend to be correspondingly longer or shorter as well. This indicates that there is likely some sort of global timing mechanism which affects all aspects of song production. Such a global timing mechanism would be in line with a recently proposed “music box” model of song production, in which the sounds that make up a song (and the spaces between them) are generated using a single “clock” for timing. However, global tempo changes do not tell the whole story about song production.
Having determined that sounds within the song (called syllables) and the gaps between them tend to stretch and compress along with the song, Chris went on to ask whether or not syllables and gaps stretched equally as the overall song length changed. What he found was that as a song's length changed, gaps and syllables both changed in length with the song, but the length of the gaps changed more than the length of the syllables. This difference in elasticity between syllables and gaps suggests that they have separate representations within the brain of the zebra finch, a feature not accounted for in the “music box” model of song production.
From measurements to models
Currently, Chris is working on a model for song production which can account for the patterns he has seen within his data thus far. He is interested in more than just song production, however. Chris also wants to understand how zebra finches learn to sing their songs. All of the data for his experiments so far come from mature birds with fully developed songs. Eventually, Chris plans to perform a similar analysis of song production in juvenile birds who are in the late stages of learning their songs. He hopes that by examining the variability of juvenile songs as they “solidify” he will be able to create a model for juvenile song production and use it to hypothesize mechanisms of song development. With luck, his research will teach us more about sensorimotor learning in general and perhaps even provide insight into speech development in humans, a process which in some respects parallels birdsong learning.
For more information on Chris and his research, see his web page here.
