Feature-dependent sensitive periods in the development of complex sound representation

Abstract

Simple tonal stimuli can shape spectral tuning of cortical neurons during an early epoch of brain development. The effects of complex sound experience on cortical development remain to be determined. We exposed rat pups to a frequency-modulated (FM) sweep in different time windows during early development, and examined the effects of such sensory experience on sound representations in the primary auditory cortex (AI). We found that early exposure to a FM sound resulted in altered characteristic frequency representations and broadened spectral tuning in AI neurons, whereas later exposure to the same sound only led to greater selectivity for the sweep rate and direction of the experienced FM sound. These results indicate that cortical representations of different acoustic features are shaped by complex sounds in a series of distinct sensitive periods.

Figure 1.

Exposure to frequency-modulated sounds in the third and fourth time windows alters sweep direction selectivity. A, B, Raster plots of responses to FM sounds recorded from high frequency selective neurons in naive and FM-exposed animals, respectively. The red horizontal bars denote the 30 ms response windows. Note the selective response to downward-modulated sounds in B. C, Representative cortical maps of the χ²-based sweep direction selectivity. Direction selectivity index was calculated from responses to upward and downward sweeps at a rate of 90 octaves/s. Experimental animals were exposed to downward sweeps in four different time windows. Animals that had heard the downward sweeps in window 4 showed a negative shift of the selectivity index, indicating that the neurons were more selective for downward sweeps than those of the other groups.

Figure 2.

Exposure to frequency-modulated sounds alters sweep direction and rate selectivity. A, Sweep direction selectivity index (SDSI) at different sweep rates. Exposure to 80 octaves/s downward FM sounds (marked by the dashed line) in W3 and W4 resulted in downward shifts of SDSI near the exposed sweep rate. B, χ²-based sweep direction selectivity. C, Histogram of best FM rates that activate the strongest responses in cortical neurons. More neurons in W3 and W4 animals became tuned to −90 octaves/s, which is near the exposure FM rate. D, The ratio of neuronal responses (number of spikes in a 30 ms window) to downward versus upward FM sweeps. The ratio is greater for W3 and W4 animals than for the other groups. *p < 0.05.

Figure 3.

Exposure to downward FM sounds in the first time window alters the cortical CF map. A, Representative cortical CF maps of naive, window 1, window 2, window 3, window 4, and adult animals. B, CF distribution along the tonotopic axis. Recordings from four animals were included for each group. C, Percentage primary auditory cortical area representing frequencies in a 0.4 octaves frequency band. Color codes for experimental groups are the same as in B. Note that only W1 animals had reduced representations of frequencies <4 kHz.

Figure 4.

Exposure to downward FM sounds in window 2 broadens frequency tuning. A, Representative cortical tuning bandwidth maps based on BWs measured at 70 dB SPL. Insets are representative receptive fields that had been recorded in the locations outlined in the bandwidth maps. The vertical axis of the receptive field plots depicts sound intensity from 0 to 70 dB SPL. The horizontal axis depicts frequencies from 1 to 32 kHz. B, Tuning bandwidth as a function of sound pressure level. Error bars depict SEM, the majority of which are occluded by the data symbols. Only window 2 animals showed broadened frequency tuning.

Figure 5.

Exposure to downward sweeps in all four windows alters CF distribution and sweep direction selectivity. A, The SDSI was down-shifted in animals that had been exposed to downward sweeps in windows 1 through 4, but not in animals exposed to the sweep in windows 1 and 2. B, χ² statistic of sweep direction selectivity. C, Characteristic frequency distribution along the tonotopic axis. Both window 1–2 and window 1–4 groups had fewer neurons tuned to frequencies <4 kHz. D, Tuning bandwidth as a function of sound pressure levels. The window 1–2 group, but not the window 1–4 group, showed broadened frequency tuning at 50–70 dB SPL. *p < 0.05.

Figure 6.

Development of response threshold, tuning bandwidth, and sweep response in different time windows. A, Frequency-dependent development of response threshold. Low frequency responses emerge after the age of P16, whereas the response thresholds for middle- and high-frequency neurons were not different among the four time windows. A 1 dB plotting offset was introduced between groups for better group distinction. B, Tuning bandwidth increases in window 2. Error bars depict SEM. C, Sweep response magnitude increases in window 3. The spike response to FM sweeps was counted in 30 ms windows. Some of the error bars are occluded by the data symbols. D, The strength of sweep direction selectivity did not change during development. The mean of the absolute values of the χ²-based SDSI is used so that selectivity for upsweeps and downsweeps are equally considered.

Figure 7.

Example cortical maps and receptive fields. A, Representative cortical CF, BW, and response threshold maps obtained in four different developmental windows. B, Representative receptive fields of the neurons whose recording sites are outlined in the maps. The vertical axis of the receptive field plots depicts sound intensity from 0 to 70 dB SPL. The horizontal axis depicts frequencies from 1 to 32 kHz. Neurons in the primary auditory cortex of the P14 animal had narrower frequency tuning than the older animals, while their response thresholds were comparable. The size and shape of the functionally determined primary auditory cortex were more variable in juvenile animals (<P40) than in adults (>P40). The difference in AI sizes seen here was not significant (p > 0.05).