Temporal Signal Processing Mechanism In The Auditory Cortex

Time-varying temporal features are fundamental characteristics of natural sounds such as human speech and animal vocalizations. Neural representation of temporal features begins at the auditory periphery where auditory-nerve fibers faithfully represent fine details of complex sounds in their temporal discharge patterns ISmith 1980, Palmer 1982, Joris 1992, Wang 19931. At subsequent nuclei along the ascending auditory pathway, precision of this temporal representation degrades gradually, due to biophysical properties of neurons along the ascending pathway and temporal integration of converging inputs from one station to the nextl Frisina 1990, Langner 1987, Creutzfeldt 1980]. It has been shown that neurons in the auditory cortex can only be synchronized to temporal variations of sequential stimuli at a rate far less than 100 Hz [Scbreiner 1988, Eggermont 1994, Gaese 1995, Bieser 1996]. The functional implication of this limited temporal response rate is largely unknown. By using temporal variations introduced by a variety of means, we found that neurons in the auditory cortex of alert monkeys discharge maximally when successive stimulus events arrive at a particular rate, corresponding to a temporal modulation frequency. For individual neurons, this preferred temporal modulation frequency remains largely constant regardless of how temporal variations are introduced, in amplitude or in frequency domain. These observations suggest an intrinsic temporal integration mechanism for auditory cortical neurons that is applied to all of its time-varying inputs. Such a mechanism can serve as the base for segmenting streams of sounds in auditory scene analysis. We further showed that for populations of neurons in the awake auditory cortex, the distribution of preferred temporal modulation frequencies is centered at 25-30 Hz. Interestingly, two major types of species- specific vocalizations produced by the primate used in this study (callithrix jacchus jacch us) oscillate near this cortically preferred temporal modulation frequency, suggesting a possible link between temporal characteristics of communication sounds and their neural encoding mechanisms in the auditory cortex.