耳鼻咽喉科: オープンアクセス

抽象的な

Neural Mechanisms of Fine-Grained Temporal Processing in Audition

Masashi Sakai*

Our auditory percepts do not necessarily correspond to an immediately present acoustic event but, rather, is the outcome of processing incoming signals over a period of time. For example, when acoustic pulses are periodically delivered at >20-40 ms intervals, individual signals are clearly heard as discrete events, whereas at ≤ 20-40 ms intervals, the same signals are perceptually merged together. Psychophysicists have adopted the concept of a “temporal grain” defined by a 20-40 ms time frame to explain the above phenomenon: when successive signals fall into different temporal grains, each signal is perceptually “resolved” as a series of discrete events. However, when the signals fall within the same temporal grain, they are perceptually integrated into a single continuous event. Such temporal grain is lost after bilateral ablation of the primary auditory cortex (AI). Neurophysiology studies on humans and animals support the view that this corresponds to the cut off interval (~30 ms) for AI neurons to generate discharges with time-locking to individual signals (i.e., stimulus-locking response); at shorter intervals, the neurons only generate a single discharge cluster at the onset of the signal train which is often followed by suppression. Such temporal behavior was captured well by our neurocomputational model [1] which incorporates temporal interplay among (1) AMPA-receptor-mediated EPSP, (2) GABAA-receptor-mediated IPSP, (3) NMDA-receptor-mediated EPSP, (4) GABAB-receptor-mediated IPSP in the AI neuron along with (5) short-term plasticity of thalamocortical synaptic connections. Ramifications from these findings are discussed in relation to language impairment.