Georg F. Striedter - US grants

The general objective of the proposed research is to understand how auditory information can be used to modify vocal output and thereby learn new vocalizations. Vocal learning has evolved independently in humans, songbirds and parrots, but has been studied extensively only in songbirds, where auditory responses can be recorded throughout most of the vocal motor system. In parrots, anatomical studies indicate that the vocal motor system receives inputs from auditory structures, but physiological recordings have thus far revealed no auditory responses within the vocal motor system. This discrepancy between the anatomical and physiological data suggests that in parrots auditory information may be gated out of the vocal motor system whenever vocal learning is not appropriate. If this hypothesis is correct, then manipulations of the putative gating pathway should facilitate the recording of auditory responses within the vocal motor system. Three specific experiments are proposed: 1. In order to confirm that auditory responses are indeed absent from the vocal motor system under normal conditions, complex auditory stimuli will be played to urethane anesthetized budgerigars, which are small parrots, while recording extracellular responses of neurons in both vocal motor and auditory structures. This experiment will provide baseline data for the second experiment. 2 In order to test the gating hypothesis directly, the putative gating pathway will be either inactivated or stimulated, depending on whether the gate is inhibitory or facilitatory in function. Inactivation will be accomplished by iontophoretic infusion of muscimol or lidocaine. Stimulation will be mediated by electrical pulses that are temporally correlated with the presentation of complex auditory stimuli. If the gating hypothesis Is correct, then one of these manipulations should facilitate the recording of auditory responses in the vocal motor system. 3 Bilateral excitotoxic lesions of the putative gating pathway will be used to assess its behavioral functions. According to the gating hypothesis, this manipulation should lead either to increased vocal plasticity or to a gradual deterioration of all learned vocalizations, depending on whether the gate is inhibitory or faciliatory, respectively.

9604299 Striedter The ability to learn complex vocalizations has evolved only in humans, songbirds, and parrots. This project investigates how the neural circuits controlling vocalization have changed during the evolution of vocal learning. Particular emphasis is placed on an area of the midbrain, as the function of this region is hypothesized to have changed significantly as the capacity for vocal behavior moves from being genetically coded to being dependent on learning. Neuroanatomical analysis of the inputs and outputs of this area, observations of the behavioral effects of damage to this area, and electrophysiological examination of the activity of midbrain neurons are all used to test comparative analysis is then used to reconstruct how neural circuitry and behavioral function of the vocal control system change during evolution. In doing so, this is one of the first studies to examine how historical constraints can influence current brain function. The results of this study enhance our understanding of the brain's systems for controlling vocal behavior, including human speech, as well as improve our understanding of how nervous systems in general evolve to accomodate complex new behavioral abilities within the constraints of preexisting circuits with preexisting functions. Such research is important in advancing knowledge about how natural and artificial learning systems acquire increasingly complex abilities as a function of experience.

Behavioral Functions of Vocal Imitation in Parrots

PI: Georg F. Striedter
co-PI: Nancy T. Burley

Why do some animals imitate complex sounds? This question has most frequently been addressed in songbirds, which learn their songs primarily to defend a territory and/or attract a mate. These explanations may not hold for the parrots, however, because parrots have evolved their remarkable imitative abilities independently of the songbirds. So, why do parrots imitate sounds? Recent data from the Striedter laboratory suggest that parrot vocal learning plays a role in pairbond formation. Specifically, when male and female budgerigars are placed in pairs, the males consistently imitate the contact calls of the females with whom they are paired, while the females retain their original calls. This sexual asymmetry in imitative behavior suggests that males imitate females in order to influence female mate choice. If this hypothesis is correct, then three predictions should hold: 1) Sexual selection should have led to an asymmetry in the vocal learning abilities of male and female budgerigars. This hypothesis will be tested by comparing how quickly all-male and all-female groups of budgerigars develop shared contact calls. If males are "better" at vocal learning, then vocal convergence should occur more quickly among males than among females. 2) Males should preferentially imitate females whom they are courting. This hypothesis will be tested by determining whether the time it takes for a male to learn a female's call is inversely correlated with his interest in this female, as measured by the frequency of other, well-known courtship behaviors. The female's attractiveness will be manipulated by painting her cere, which is brown in mature females but pale blue in immature females. 3) Female budgerigars should associate preferentially with males that have learned to imitate them. This hypothesis will be tested by determining whether females give more courtship displays towards unfamiliar males that already know their call (because they were tutored by another female that shares the test female's call) than towards males that do not yet know the test female's call. In a complementary experiment, some male budgerigars will be rendered imitation-impaired by selectively severing the connection between their auditory and vocal motor systems. The proposed experiments could significantly impact the field of "sexual selection" and "female choice" because most prior studies focused on female choice for overt physical traits - not learning ability. The proposed experiments also pave the way for a more mechanistic analysis of why male budgerigars imitate female calls.

How do humans and animals imitate sounds? This question is central to the broader question of how humans learn to speak a language, but it is difficult to answer in humans because their nervous systems are relatively impervious to experimental study. Because of this difficulty, Dr. Striedter and his collaborators have used budgerigars, which are small parrots, to test a specific model of how vocal imitation works. The model specifies that a previously identified high-level auditory region in the brain of budgerigars sends a reinforcement (or reward) signal to a high-level vocal control center whenever the bird produced a sound that resembles the desired (i.e. target) vocalization. By trial and error, and selective reinforcement, the system should gradually learn to produce the target sound. To test this hypothesis, two kinds of experiments are proposed. First, the high-level auditory region is lesioned (i.e. chemically damaged) during the time that a bird is "trying" to learn a remembered sound. According to the hypothesis, such a lesion should prevent the learning. Second, the auditory region will be stimulated electrically every time the bird produced a vocalization that matches some acoustic criterion set by the experimenter. According to the hypothesis, this should amount to experimenter-driven selective reinforcement and, over time, cause the bird to produce vocalizations that closely match the experimenter's expectation. Collectively, these experiments will lead to a detailed mechanistic understanding of how parrots "parrot". Combined with other prior work in the Striedter laboratory on the behavioral significance of vocal imitation, this work will lead to a comprehensive account of how and why birds imitate sounds. This work fulfills the public's desire to understand how and why birds sing, and furthers the neuroscientists' more specific quest of understanding how complex brains learn complex things.

The principal aim of the proposed research is to determine what evolutionary changes in developmental mechanisms have caused parrots and songbirds to evolve disproportionately large forebrains. This work is important because disproportionately large forebrains evolved several times independently (e.g., parrots, songbirds, primates, dolphins and whales) and seem to make their bearer's "smart" in a variety of ways. Yet almost nothing is known about how evolution creates enlarged forebrains. Developmental biologists can make genetically modified animals that have unusually large brains because they vary in some aspect of brain development, such as having reduced levels of neuronal cell death. However, these mutant animals tend to die as embryos and never become "smarter". The approach in the proposed work is to inquire of Nature how it manipulated brain development to create "smarter" animals with enlarged forebrains. Specifically, brain development will be compared in a parakeet, a zebra finch and a quail, asking which cellular or molecular mechanisms differ between these species. If Nature used the same mechanism to produce enlarged forebrains in parrots and songbirds, then this suggests the existence of a constraint that must be satisfied in order to obtain well-functioning enlarged forebrains. Knowledge of this constraint would, in turn, inform future attempts to mimic some aspects of brain evolution in laboratory experiments. It would also improve understanding of the mechanisms that underlie the evolution of increased forebrain size and increased intelligence within the human lineage.

The proposed work will support the training of two PhD students and at least four undergraduate students who do independent research. The PI is committed to broadening participation of students at all levels and is involved in community outreach programs and monthly series for K-12 students.

The forebrain is proportionately larger in humans than in other mammals. Similarly, the forebrain is proportionately larger in parrots and songbirds than in other birds. These species differences in adult brain proportions have been well described and are thought to account for species differences in behavioral complexity and intelligence. Almost completely unknown, however, are the developmental mechanisms that generate such species differences. Previous work from the Striedter laboratory has shown that forebrain enlargement in parrots and songbirds occurs because the forebrain's precursor cells in these species proliferate for a longer period of time, thereby generating a larger forebrain precursor pool. Although this is a powerful mechansim for enlarging a brain region, other species may enlarge a brain region by other mechanisms, such as changing the spatial patterns of gene expression in young embryos or changing the rates at which precursor cells divide. The proposed research explores these alternative mechanisms by comparing brain region sizes, patterns of gene expression, and rates of cell division across young embryos of different bird species, including parakeets, quail, chickens, and ducks. If one or more of these parameters differs between the examined species, then evolution is free to vary brain proportions through several different developmental mechanisms, rather than constrained to utilize just one. More generally, the findings will clarify some of the rules that govern brain evolution. An important long-term goal is to manipulate brain development in ways that follow these rules and, thus, mimic the natural evolutionary changes. Such experiments are exciting because they will allow for the testing of evolutionary hypotheses. Overall, the proposed work will motivate and train at least one graduate student and several undergraduates performing independent research. It will also excite and educate the general public, who will be exposed to it through public lectures and outreach to student groups.

Understanding how brains function to specify behavior, thoughts and memories of an organism is a major challenge of the twenty first century. Brain initiatives in the US and other countries tend to focus on the human brain, but understanding how neural circuitry and activity control behavior in non-human species may also yield important insights into brain function. Of course, exploring the neural circuitry, activity patterns responsible for behavior in many different animal species requires large amounts of work. Therefore, careful planning for these efforts is required to maximize their impact on the field. To facilitate this planning process, this workshop will convene a group of 40 experts in comparative neurobiology, brain research and allied disciplines to debate critical questions, such as whether one should functionally map entire brains or subsystems, whether to focus on closely related animal species or distant relatives, how to integrate developmental with adult data, how to identify equivalent brain components in seemingly very different brains across species, and how to integrate information about brain structure with data on brain function and behavior. Workshop participants will discuss these questions in small groups and then report their answers to the entire group, which will synthesize those answers into a coherent research agenda with short- and long-term goals. An important outcome of this workshop will be a written report or published journal article that will offer guidance to both researchers and funding agencies. To maximize the workshop's impact, video recordings of key sessions will be disseminated through a publicly accessible web site, which will also host the report and some related documents. One quarter of all workshop participants will be postdoctoral researchers. They are expected to make important contributions and, simultaneously, gain valuable experience. A strong effort will be made to include women and members of underrepresented groups to broaden participation in this important area of neuroscience.