intracellular responses to auditory stimuli revealed that, in a particular population of cells of the ventral nucleus of the lateral lemniscus (VNLL) of rats, fast inhibition occurred before the first action potential. population of auditory nerve fibers. This was because the broadband onset noise, also termed spectral splatter, was suppressed by the fast onset inhibition. This mechanism has the potential to greatly improve the clarity of the representation of the harmonic content of certain kinds of natural sounds. Introduction The auditory brainstem receives input from the auditory nerve, and provides projections mainly to the auditory thalamus, which in turn Prox1 projects to the cortex. However, the auditory brainstem is usually not simply a relay. Nuclei of the brainstem are involved in processing behaviorally important sound cues [1]. Information in the auditory nerve is usually partly carried by the relative timing of action potentials, and these sub-millisecond cues are most accurately decoded early in the sensory pathway [2, 3]. The auditory brainstem predominantly consists of circuits of neurons that have low membrane time constants, capable of decoding the temporal information that is usually intrinsic to sound stimuli [4]. A characteristic 42971-09-5 IC50 of some neurons within the auditory brainstem is usually that of onset inhibition [5]. This is usually a brief hyperpolarization that precedes the first action potential and has been described in the inferior colliculus [6, 7, 5, 8], the ventral nucleus of the lateral lemniscus [5], and the cochlear nucleus [9]. observations of this effect have previously led to a number of conjectures: Fast inhibition in T-stellate cells of rats provided by D-stellate cells, was shown to postpone spikes that were coincident with the inhibition [9, 10]. This may mean that the first spikes in a population of neurons become more temporally aligned, assisting in the lateral integration of the information carried by these spikes. Onset inhibition may act as an event reference, increasing the information content of first spike latency [5, 11]. The rebound from inhibition occurs with a particular delay, leading to a spike only if the excitatory input coincides with the timing of this rebound. If the excitatory input does not coincide with this delay then the spike is usually not produced. This process would create a sensitivity to first spike latency, a property that may represent some important features of sound, such as intensity. Onset inhibition may create direction sensitivity for frequency sweeps [12, 2]. This hypothesis also suggests that rebound from the inhibition makes a neuron sensitive to first-spike delay. Combining this with lateral synaptic connections may create frequency sweep direction selectivity. Onset inhibition may form a component of a mechanism that is usually sensitive to the duration of brief sounds [13]. The onset inhibition prevents the neuron from firing for some short period of time at the beginning of a sound. This means that only sounds of a certain duration would create activity in that particular neuron. In this investigation, we hypothesize that onset 42971-09-5 IC50 inhibition assists in the suppression of broadband spectral splatter. This spectral platter, which occurs at the beginning of any sound with a sharp onset, contains very little information about the harmonic content of the sounds. The harmonic component of the sound stimulus could become more prominent if this element of the stimulus is usually suppressed. This suppression process may be particularly effective when the sound is usually made up of a stream of very brief harmonic components, such as those that occur in speech. Most of the conjectures above are not mutually exclusive and most are specific to particular regions of the auditory brainstem. In particular, the idea that the inhibition suppresses the first spike does not necessarily contradict the possibility that the inhibition could act instead to delay the first spikes in other circumstances [5] or in a individual neural circuit. In this investigation, our hypothesis was tested by developing a model of a cellular microcircuit in the ventral nucleus of the lateral lemniscus (VNLL). It has been previously postulated [5] that the inhibition in this region is usually provided by inhibitory interneurons that are driven by octopus cells of the posterior ventral cochlear nucleus, which are known to project to the VNLL [14]. The hypothesis was addressed by obtaining and analyzing experimental intracellular data 42971-09-5 IC50 from the VNLL of rats. A computational model of the VNLL circuit was established and speech-like sounds were used as the stimuli. It was possible to create a population of model VNLL cells and observe their collective response to the stimuli. By using a model, it was possible to manipulate the delays present in the circuit and more clearly demonstrate the conversation between excitation.