From sender to receiver: physics and sensory ecology of hearing in insects and vertebrates

All behaviour is guided, or restricted, by the senses. Sense organs have evolved in multiple ways to extract and pre-process information from the external world. However, molecular mechanisms of sense organ specification and their evolutionary origins have remained unclear. We have used closely-related Drosophila species to explore how ears can contribute to evolution - and how evolution, in turn, has shaped ears.

In flies (Diptera), hearing is mediated by Johnston’s Organ (JO) neurons in the second antennal segment (1). In Drosophilids, the spectral tuning of the flies’ antennal ears correlates with the spectral composition of song pulses produced by conspecific males (2). Laser-Doppler vibrometric analysis of sound receiver mechanics and extracellular recordings of compound action potentials from the antennal nerve show that the species-specific auditory tuning is partly the result of variations in the molecular modules for mechanotransduction in JO neurons.

RNA-Seq based transcriptomics of the JOs from six closely-related Drosophila species combined with predictive bioinformatics (i-cisTarget and iRegulon) identified a particular type of transcription factor from the nuclear hormone receptor family as important contributor to inter-specific variation in Drosophilid ears.

Nuclear hormone receptor proteins are also required for normal sex organ development. The investigated mutants showed sexually dimorphic defects in auditory function (both with regard to auditory mechanics and auditory nerve responses). On the sender side of Drosophila acoustic communication, in turn, mutant males displayed severe defects in song production (both in their propensity to produce songs and with regard to song structure). The duality of its contributions presents this nuclear receptor gene as a potential substrate for genetic coupling in the Drosophila acoustic communication system.

Human hearing is enhanced by an active process that amplifies the ear's mechanical inputs several hundredfold, sharpens frequency tuning to allow the discrimination of tones differing in frequency by less than 0.2 %, and compresses six orders of magnitude in the amplitude of sounds into only two orders of magnitude in neural output. In addition, spontaneous otoacoustic emissions emerge from ears in a very quiet environment, an indication that the active process can be so exuberant as to become unstable. Cooperativity between mechanoelectrical-transduction channels confers negative stiffness on the hair bundle, which together with myosin-based adaptation motors elicits a dynamical instability that underlies the active process. Experiments on individual hair bundles indicate that the bundle's operation near this instability, a Hopf bifurcation, accounts for the four characteristics of the active process.

The vestibular type I hair cell and its distinctive calyceal synapse are found only in the inner ears of reptiles, birds and mammals.  Like the cochlea, the type I – calyx synapse may represent adaptations to life on land.  Over the past 20 years, evidence has accrued that these unusual-looking synapses are also functionally remarkable, featuring not just chemical (quantal) transmission of vesicle-bound glutamate from ribbon synapses but also a form of non-quantal transmission that depends on currents through ion channels in dense arrays on presynaptic (hair cell) and postsynaptic (calyceal) membranes.  Quantal and non-quantal transmission filter the transmitted mechanosensory signal in distinct ways and have been recorded both together and separately, suggesting unexpectedly rich possibilities for shaping vestibular inputs to the brain.

Humans excel at selectively listening to a target speaker in background noise such as competing voices. While the encoding of speech in the auditory cortex is modulated by selective attention, it remains debated whether such modulation occurs already in subcortical auditory structures. Investigating the contribution of the human brainstem to attention has, in particular, been hindered by the tiny amplitude of the brainstem response. Its measurement normally requires a large number of repetitions of the same short sound stimuli, which may lead to a loss of attention and to neural adaptation. This talk describes a mathematical method to measure the auditory brainstem response to running speech, an acoustic stimulus that does not repeat and that has a high ecological validity.  This research employs this method to assess the brainstem's activity when a subject listens to one of two competing speakers, and show that the brainstem response is consistently modulated by attention.

How does the brain process acoustic information? Mapping the auditory neural circuits is indispensable to answer this question. The fruit fly is ideally suited for such tasks, with its small brain size and a rich repertoire of genetic tools. Moreover, they use acoustic signals to communicate with each other. Toward comprehensive identification of auditory neural circuits in the fly brain, this study systematically identified the auditory sensory neurons and their downstream neurons. The anatomic and functional analyses revealed frequency segregation at the first layer of the auditory pathway and the convergence of frequency information in the subsequent downstream pathways. Second-order auditory neurons have intensive binaural interactions, raising the possibility that the fly is capable of comparing acoustic signals detected at the left and right ears. Based on analysis, this research established the first comprehensive map of primary and secondary auditory neurons in the fly brain, which are characterized by frequency segregation and convergence, binaural interaction, and multimodal pathways.

Social interactions require continually adjusting behavior in response to sensory feedback. For example, when having a conversation, sensory cues from a partner (e.g., sounds or facial expressions) affect speech patterns in real time. Human speech signals, in turn, are the sensory cues that modify a partner’s actions. What are the underlying computations and neural mechanisms that govern these interactions? To address these questions, lab studies the acoustic communication system of Drosophila. As an advantage, the fly nervous system is relatively simple, with a wealth of genetic tools to interrogate it. Importantly, Drosophila acoustic behaviors are highly quantifiable and robust. During courtship, males produce time-varying songs via wing vibration, while females arbitrate mating decisions. This study discovered that, rather than being a stereotyped fixed action sequence, male song structure and intensity are continually sculpted by interactions with the female, over timescales ranging from tens of milliseconds to minutes – and this research is mapping the underlying circuits and computations. This research has also developed methods to relate song representations in the female brain to changes in her behavior, across multiple timescales. The focus on natural acoustic signals, either as the output of the male nervous system or as the input to the female nervous system, provides a powerful, quantitative handle for studying the basic building blocks of communication.

Dr Christian Sumner, University of Nottingham, UK

Many animals are not thought to be able to combine their vocalizations into structured sequences, as do songbirds, humans and a few other species. Nonetheless, it remains possible that these many animals are able to recognize ordering relationships in sequences generated by ‘artificial grammars’. This talk will explore how understanding the extent of these hidden receptive learning abilities could clarify the neurobiological origins of language. First, an overview behavioral results with structured sequence learning in three primate species: marmosets, macaques and humans. Then a focus on brain imaging results identifying evolutionarily conserved frontal brain regions in macaques and humans involved in predicting events that occur next in a sequence. Finally, results are presented from a new study involving comparative intracranial recordings in humans and monkeys processing the sequences. Overall, the findings indicate that human and non-human primates possess an evolutionarily conserved neural network involved in processing structured auditory input and they provide hints on how the human brain differentiated for language. 

If we are to understand how activity in the brain gives rise to auditory perception and guides behaviour, it is essential to consider the way in which neural processing is shaped both by the sensory and behavioural context in which sounds occur and by lifelong changes in experience that refine or degrade perceptual abilities as a result of learning or hearing loss. This talk will consider the neural circuits and strategies that enable the auditory system to adjust to the statistics of the auditory scene, as well as to longer lasting changes in inputs that result from hearing impairments. In addition to providing insights into the adaptive capabilities of the auditory system, findings indicate that different forms of plasticity may represent therapeutic targets for restoring perceptual abilities following hearing loss.