A small, computationally flexible network produces the phenotypic diversity of song recognition in crickets

How neural networks evolved to generate the diversity of species-specific communication signals is unknown. For receivers of the signals one hypothesis is that novel recognition phenotypes arise from parameter variation in computationally flexible feature detection networks. We test this hypothesis in crickets, where males generate and females recognize the mating songs with a species-specific pulse pattern, by investigating whether the song recognition network in the cricket brain has the computational flexibility to recognize different temporal features. Using electrophysiological recordings from the network that recognizes crucial properties of the pulse pattern on the short timescale in the cricket Gryllus bimaculatus, we built a computational model that reproduces the neuronal and behavioral tuning of that species. An analysis of the model's parameter space reveals that the network can provide all recognition phenotypes for pulse duration and pause known in crickets and even other insects. Phenotypic diversity in the model is consistent with known preference types in crickets and other insects, and arise from computations that likely evolved to increase energy efficiency and robustness of pattern recognition. The model's parameter to phenotype mapping is degenerate-different network parameters can create similar changes in the phenotype-which likely supports evolutionary plasticity. Our study suggests that computationally flexible networks underlie the diverse pattern recognition phenotypes and we reveal network properties that constrain and support behavioral diversity.

Vocal learning in songbirds: the role of syllable order in song recognition

Songbird vocal learning has interesting behavioural and neural parallels with speech acquisition in human infants. Zebra finch males sing one unique song that they imitate from conspecific males, and both sexes learn to recognize their father's song. Although males copy the stereotyped syllable sequence of their father's song, the role of sequential information in recognition remains unclear. Here, we investigated father's song recognition after changing the serial order of syllables (switching the middle syllables, first and last syllables, or playing all syllables in inverse order). Behavioural approach and call responses of adult male and female zebra finches to their father's versus unfamiliar songs in playback tests demonstrated significant recognition of father's song with all syllable-order manipulations. We then measured behavioural responses to normal versus inversed-order father's song. In line with our first results, the subjects did not differentiate between the two. Interestingly, when males' strength of song learning was taken into account, we found a significant correlation between song imitation scores and the approach responses to the father's song. These findings suggest that syllable sequence is not essential for recognition of father's song in zebra finches, but that it does affect responsiveness of males in proportion to the strength of vocal learning. This article is part of the theme issue ‘Vocal learning in animals and humans’.

EEG-based decoding and recognition of imagined music

ABSTRACTWhen listening to music, the brain generates a neural response that follows the amplitude envelope of the musical sound. Previous studies have shown that it is possible to decode this envelope-following response from electroencephalography (EEG) data during music perception. However, a successful decoding and recognition of imagined music, without the physical presentation of a music stimulus, has not been established to date. During music imagination, the human brain internally replays a musical sound, which naturally leads to the hypothesis that a similar envelope-following response might be generated. In this study, we demonstrate that this response is indeed present during music imagination and that it can be decoded from EEG data. Furthermore, we show that the decoded envelope allows for classification of imagined music in a song recognition task, containing tracks with lyrics as well as purely instrumental tasks. A two-song classifier achieves a median accuracy of 95%, while a 12-song classifier achieves a median accuracy of 66.7%. The results of this study demonstrate the feasibility of decoding imagined music, thereby setting the stage for new neuroscientific experiments in this area as well as for new types of brain-computer interfaces based on music imagination.

A small, computationally flexible network produces the phenotypic diversity of song recognition in crickets

How neural networks evolve to recognize species-specific communication signals is unknown. One hypothesis is that novel recognition phenotypes are produced by parameter variation in a computationally flexible “mother network”. We test this hypothesis in crickets, where males produce and females recognize mating songs with a species-specific pulse pattern. Whether the song recognition network in crickets is computationally flexible to recognize the diversity of pulse patterns and what network properties support and constrain this flexibility is unknown. Using electrophysiological recordings from the cricket Gryllus bimaculatus, we built a model of the song recognition network that reproduces the network dynamics as well as the neuronal and behavioral tuning for that species. An analysis of the model’s parameter space reveals that the network can produce all recognition phenotypes known in crickets and even other insects. Biases in phenotypic diversity produced by the model are consistent with the existing behavioral diversity in crickets, and arise from computations that likely evolved to increase energy efficiency and robustness of song recognition. The model’s parameter to phenotype mapping is degenerate – different network parameters can create similar changes in the phenotype – which is thought to support evolutionary plasticity. Our study suggest that a computationally flexible mother network could underlie the diversity of song recognition phenotypes in crickets and we reveal network properties that constrain and support behavioral diversity.

Vocal recognition suggests premating isolation between lineages of a lekking hummingbird

Species with genetically differentiated allopatric populations commonly differ in phenotypic traits due to drift and/or selection, which can be important drivers of reproductive isolation. Wedge-tailed sabrewing (Campylopterus curvipennis) is a species complex composed of three genetically and acoustically differentiated allopatric lineages that correspond to currently recognized subspecies in Mexico: C. c. curvipennis (Sierra Madre Oriental), C. c. pampa (Yucatán Peninsula), and C. c. excellens (Los Tuxtlas). Although excellens is taxonomically recognized as a distinct species, there is genetic evidence that lineages excellens and curvipennis have diverged from each other later than pampa. In this study, we experimentally tested C. c. curvipennis song recognition as a major factor in premating reproductive isolation for lineage recognition. To this end, we conducted a song playback experiment to test whether territorial males of one C. c. curvipennis lek discriminate among potential competitors based on male songs from the three lineages. Males of curvipennis responded more aggressively to songs of their own lineage and excellens, than to songs of the most divergent lineage pampa, as evidenced by significant differences in a variety of intensity and latency response variables. This indicate that the pampa male song does not represent a competitive threat as curvipennis and excellens songs, in which divergence and song recognition represent premating reproductive isolation between these isolated lineages. However, the acoustic limits between curvipennis and excellens might be attenuated by gene flow in case of secondary contact between them, despite the strong and relatively rapid divergence of their sexually selected song traits.