Search

Animals can perform complex and purposeful behaviors by executing simpler movements in flexible sequences. It is particularly challenging to analyze behavior sequences when they are highly variable, as is the case in language production, certain types of birdsong and, as in our experiments, flies grooming. High sequence variability necessitates rigorous quantification of large amounts of data to identify organizational principles and temporal structure of such behavior. To cope with large amounts of data, and minimize human effort and subjective bias, researchers often use automatic behavior recognition software. Our standard grooming assay involves coating flies in dust and videotaping them as they groom to remove it. The flies move freely and so perform the same movements in various orientations. As the dust is removed, their appearance changes. These conditions make it difficult to rely on precise body alignment and anatomical landmarks such as eyes or legs and thus present challenges to existing behavior classification software. Human observers use speed, location, and shape of the movements as the diagnostic features of particular grooming actions. We applied this intuition to design a new automatic behavior recognition system (ABRS) based on spatiotemporal features in the video data, heavily weighted for temporal dynamics and invariant to the animal's position and orientation in the scene. We use these spatiotemporal features in two steps of supervised classification that reflect two time-scales at which the behavior is structured. As a proof of principle, we show results from quantification and analysis of a large data set of stimulus-induced fly grooming behaviors that would have been difficult to assess in a smaller dataset of human-annotated ethograms. While we developed and validated this approach to analyze fly grooming behavior, we propose that the strategy of combining alignment-invariant features and multi-timescale analysis may be generally useful for movement-based classification of behavior from video data.

Human language, as well as birdsong, relies on the ability to arrange vocal elements in new sequences. However, little is known about the ontogenetic origin of this capacity. Here we track the development of vocal combinatorial capacity in three species of vocal learners, combining an experimental approach in zebra finches (Taeniopygia guttata) with an analysis of natural development of vocal transitions in Bengalese finches (Lonchura striata domestica) and pre-lingual human infants. We find a common, stepwise pattern of acquiring vocal transitions across species. In our first study, juvenile zebra finches were trained to perform one song and then the training target was altered, prompting the birds to swap syllable order, or insert a new syllable into a string. All birds solved these permutation tasks in a series of steps, gradually approximating the target sequence by acquiring new pairwise syllable transitions, sometimes too slowly to accomplish the task fully. Similarly, in the more complex songs of Bengalese finches, branching points and bidirectional transitions in song syntax were acquired in a stepwise fashion, starting from a more restrictive set of vocal transitions. The babbling of pre-lingual human infants showed a similar pattern: instead of a single developmental shift from reduplicated to variegated babbling (that is, from repetitive to diverse sequences), we observed multiple shifts, where each new syllable type slowly acquired a diversity of pairwise transitions, asynchronously over development. Collectively, these results point to a common generative process that is conserved across species, suggesting that the long-noted gap between perceptual versus motor combinatorial capabilities in human infants may arise partly from the challenges in constructing new pairwise vocal transitions.

Exploratory variability is essential for sensory-motor learning, but it is not known how and at what time scales it is regulated. We manipulated song learning in zebra finches to experimentally control the requirements for vocal exploration in different parts of their song. We first trained birds to perform a one-syllable song, and once they mastered it we added a new syllable to the song model. Remarkably, when practicing the modified song, birds rapidly alternated between high and low acoustic variability to confine vocal exploration to the newly added syllable. Further, even within syllables, acoustic variability changed independently across song elements that were only milliseconds apart. Analysis of the entire vocal output during learning revealed that the variability of each song element decreased as it approached the target, correlating with momentary local distance from the target and less so with the overall distance. We conclude that vocal error is computed locally in sub-syllabic time scales and that song elements can be learned and crystalized independently. Songbirds have dedicated brain circuitry for vocal babbling in the anterior forebrain pathway (AFP), which generates exploratory song patterns that drive premotor neurons at the song nucleus RA (robust nucleus of the arcopallium). We hypothesize that either AFP adjusts the gain of vocal exploration in fine time scales, or that the sensitivity of RA premotor neurons to AFP/HVC inputs varies across song elements.

Motor sequences are formed through the serial execution of different movements, but how nervous systems implement this process remains largely unknown. We determined the organizational principles governing how dirty fruit flies groom their bodies with sequential movements. Using genetically targeted activation of neural subsets, we drove distinct motor programs that clean individual body parts. This enabled competition experiments revealing that the motor programs are organized into a suppression hierarchy; motor programs that occur first suppress those that occur later. Cleaning one body part reduces the sensory drive to its motor program, which relieves suppression of the next movement, allowing the grooming sequence to progress down the hierarchy. A model featuring independently evoked cleaning movements activated in parallel, but selected serially through hierarchical suppression, was successful in reproducing the grooming sequence. This provides the first example of an innate motor sequence implemented by the prevailing model for generating human action sequences. DOI: http://dx.doi.org/10.7554/eLife.02951.001, eLife digest Anyone who has ever lived with a cat is familiar with its grooming behavior. This innate behavior follows a particular sequence as the cat methodically cleans its body parts one-by-one. Many animals also have grooming habits, even insects such as fruit flies. The fact that grooming sequences are seen across such different species suggests that this behavior is important for survival. Nevertheless, how the brain organizes grooming sequences, or other behaviors that involve a sequence of tasks, is not well understood. Fruit flies make a good model for studying grooming behavior for a couple of reasons. First, they are fastidious cleaners. When coated with dust they will faithfully carry out a series of cleaning tasks to clean each body part. Second, there are many genetic tools and techniques that researchers can use to manipulate the fruit flies' behaviors. One technique allows specific brain cells to be targeted and activated to trigger particular behaviors. Seeds et al. used these sophisticated techniques, computer modeling, and behavioral observations to uncover how the brains of fruit flies orchestrate a grooming sequence. Dust-covered flies follow a predictable sequence of cleaning tasks: beginning by using their front legs to clean their eyes, they then clean their antennae and head. This likely helps to protect their sensory organs. Next, they move on to the abdomen, possibly to ensure that dust doesn't interfere with their ability to breathe. Wings and thorax follow last. Periodically, the flies stop to rub their legs together to remove any accumulated dust before resuming the cleaning sequence. Seeds et al. activated different sets of brain cells one-by-one to see if they could trigger a particular grooming task and found that individual cleaning tasks could be triggered, in the absence of dust, by stimulating a specific group of brain cells. This suggests each cleaning task is a discrete behavior controlled by a subset of cells. Then Seeds et al. tried to stimulate more than one cleaning behavior at a time; they discovered that wing-cleaning suppressed thorax-cleaning, abdomen-cleaning suppressed both of these, and head-cleaning suppressed all the others. This suggests that a ‘hierarchy’ exists in the brain that exactly matches the sequence that flies normally follow as they clean their body parts. By learning more about how the brain coordinates grooming sequences, the findings of Seeds et al. may also provide insights into other behaviors that involve a sequence of tasks, such as nest building in animals or typing in humans. Following on from this work, one of the next challenges will be to see if such behaviors also use a ‘suppression hierarchy’ to ensure that individual tasks are carried out in the right order. DOI: http://dx.doi.org/10.7554/eLife.02951.002