Contribution of local- and circuit-based mechanisms to adaptive coding in the auditory midbrain

The ability to code accurately the prevailing sensory information is critical to an animal's survival, and brain mechanisms have evolved that enable this information to be extracted from the environment in an optimal manner. As part of this ability, many sensory neurons appear able to adjust their neural tuning in 'real-time' to take account of the unfolding sensory environment. This so-called 'adaptive coding' often results in improved accuracy with which the ambient sensory environment is coded. How this is achieved in the brain remains to be determined, both in terms of cellular mechanisms and in terms of brain circuits that contribute to adaptive coding in any one brain centre. Our experiments will investigate three specific aspects of adaptive coding. First, they will determine mechanisms within neurons themselves that contribute to adaptive coding for the unfolding sound intensities present in a sound. Second, they will assess the contribution of cortical feedback on adaptive coding in lower brain centres. Although sensory information is thought to 'flow' from the peripheral hearing sensors in the ears towards higher brain centres such as the cortex, a substantial number of cortical feedback circuits are also known to exist. The function of these feddback circuits remains unclear, but preliminary evidence from my laboratory suggests that one role is to modify adaptive coding in the auditory midbrain, increasing the time taken by neurons to optimise their sensitivity to sound intensity when the prevailing sound enviroment suddenly changes. Finally, experiments will also examine how specific adaptive coding is for the stimulus feature currently being heard, assessing whether or not adapting a neuron's response to one stimulus parameter (e.g. sound intensity) also means that the responses are adapted to sound-source location, for example.

This proposal will investigate brain mechanisms in the auditory pathway that contribute to the ability of sensory neurons to adjust their sensitivity to take account of the prevailing sound environment. Experiments will combine extracellular and patch-clamp recordings in vivo with an information-theoretical approach to assess the contribution of ascending and descending (cortical) pathways in the central auditory pathway to adaptive coding, particularly for sound intensity, but also for sound-source location. Experiments will also examine the stimulus-specificity of adaptive coding, assessing whether or not the internal adapted state, which is thought to be contribute to enhanced coding of the prevailing stimulus environment, transfers between different stimulus features. Two specific hypotheses will be tested. First, we will assess both local (cellular) and circuit (cortical feedback) contributions to adaptive coding for sound intensity using a combination of in vivo patch-clamp recordings, extracellular recordings and cortical cooling. Second, we will test the hypothesis that adaptive coding is stimulus non-specific, and can 'transfer' between stimulus features. Notwithstanding the presumed importance of adaptive coding in processing sensory information, no study to date has addressed the question of whether the history of sensory stimulation or the history of neural response is the critical factor in adaptive coding. This argument is often considered moot; the brain's 'access' to any stimulus, including the values held by various parameters of that stimulus, can be assessed only through the responses evoked. In this sense, response history is stimulus history. Nevertheless, the notion that neurons adapt their response to take account of stimulus history ignores the possibility that an adapted state to which a neuron is subject can 'transfer' to another stimulus feature, rendering adaptation more an internally-defined, rather than stimulus-defined, state.