Adult plasticity of spatiotemporal receptive fields of multisensory superior colliculus neurons following early visual deprivation

Abstract

Purpose: Previous work has established that the integrative capacity of multisensory neurons in the superior colliculus (SC) matures over a protracted period of postnatal life (Wallace and Stein, 1997), and that the development of normal patterns of multisensory integration depends critically on early sensory experience (Wallace et al., 2004). Although these studies demonstrated the importance of early sensory experience in the creation of mature multisensory circuits, it remains unknown whether the reestablishment of sensory experience in adulthood can reverse these effects and restore integrative capacity.

Methods: The current study tested this hypothesis in cats that were reared in absolute darkness until adulthood and then returned to a normal housing environment for an equivalent period of time. Single unit extracellular recordings targeted multisensory neurons in the deep layers of the SC, and analyses were focused on both conventional measures of multisensory integration and on more recently developed methods designed to characterize spatiotemporal receptive fields (STRF).

Results: Analysis of the STRF structure and integrative capacity of multisensory SC neurons revealed significant modifications in the temporal response dynamics of multisensory responses (e.g., discharge durations, peak firing rates, and mean firing rates), as well as significant changes in rates of spontaneous activation and degrees of multisensory integration.

Conclusions: These results emphasize the importance of early sensory experience in the establishment of normal multisensory processing architecture and highlight the limited plastic potential of adult multisensory circuits.

Fig. 1

Construction of a spatiotemporal receptive field (STRF) for an individual SC neuron. On the left is shown the visual (dotted ellipse) and auditory (solid ellipse) receptive fields of a representative neuron and the stimulus locations (dots) used for creating the STRF. On the right is shown the STRF created for a single spatial plane (i.e., 20° elevation, shown on the left in the gray shading). In this plot the normalized evoked response (scaled to the maximal response) as a function of time (x-axis) and spatial location (y-axis) is represented in a pseudocolor format in which lighter shades of grey represent higher firing rates.

Fig. 2

Representative example of spike density functions (SDFs), spatiotemporal receptive fields (STRFs), and spatial receptive fields (SRFs) for an SC neuron from a normally-reared (NR) animal. A: Rasters and SDFs illustrate this neuron’s response to the visual, auditory and combined visual-auditory stimulus conditions. Triangles represent the onset of the visual (black) and auditory (grey) stimuli. B: The STRFs for the three stimulus conditions for the 20° elevation plane (see fig. 1 for a description of the construction of this representation). C: Top panel shows the stimulus locations (in two dimensions – azimuth and elevation) used for the construction of the SRF plots. Below this are shown the visual, auditory and multisensory SRFs for this neuron, along with that predicted from a simple summation of the visual and auditory SRFs (V+ A) and a contrast plot showing the difference between the actual multisensory response and this prediction (M− [V+ A]). Note the areas of superadditive (darker shading) and subadditive (lighter shading) interactions in this contrast SRF.

Fig. 3

Representative example of spike density functions (SDFs), spatiotemporal receptive fields (STRFs), and spatial receptive fields (SRFs) for an SC neuron from a DR+ animal. Conventions are the same as for figure 2. Note in the SRF contrast plot (bottom right) that there is very little evidence for superadditive or subadditive multisensory interactions in this neuron.

Fig. 4

Comparison of multisensory stimulus evoked response dynamics between NR and DR+ populations. A: Mean evoked response duration. B: Mean peak firing rate. C: Mean evoked firing rate. D: Spontaneous firing rate.

Fig. 5

Comparison of temporal patterns of multisensory response and multisensory integration in relation to the onset and offset of the predicted multisensory response for the population of SC neurons from NR (A) and DR+ (B) animals. Whereas the top panels show changes in integrative capacity (%) as a function of time, the bottom panels depict changes in firing rate for the actual (dotted trace) and predicted (solid trace) multisensory conditions. Note that these responses are aligned to the predicted multisensory response’s onset (left) and offset (right). This analysis emphasizes the lack of significant non-linearities in the responses of neurons from DR+ animals, in striking contrast to the condition in normal animals.