Progress and challenges for understanding the function of cortical microcircuits in auditory processing

Abstract

An important outstanding question in auditory neuroscience is to identify the mechanisms by which specific motifs within inter-connected neural circuits affect auditory processing and, ultimately, behavior. In the auditory cortex, a combination of large-scale electrophysiological recordings and concurrent optogenetic manipulations are improving our understanding of the role of inhibitory-excitatory interactions. At the same time, computational approaches have grown to incorporate diverse neuronal types and connectivity patterns. However, we are still far from understanding how cortical microcircuits encode and transmit information about complex acoustic scenes. In this review, we focus on recent results identifying the special function of different cortical neurons in the auditory cortex and discuss a computational framework for future work that incorporates ideas from network science and network dynamics toward the coding of complex auditory scenes.

Fig. 1

Simplified views of cortical circuits. (a) Diagram of excitatory–inhibitory circuit with recurrent connections. Theoretical and experimental studies demonstrate that inhibition stabilizes between excitatory and inhibitory neurons in the auditory cortex. (b) Inhibitory–excitatory network can be extended to include several interneuron subtypes. (c) Schematic diagram of connectivity between select neurons in the auditory cortex (note that layer-specific information is omitted here): Exc: Excitatory neurons; PV: parvalbumin-positive interneurons; SOM: somatostatin-positive interneurons; VIP: vasopressin-positive interneurons; TC: Thalamo-cortical projection neurons. All neuron types receive additional inputs from other brain areas, which were omitted from the diagram for simplicity. Open circles: excitatory synapses; closed circles: inhibitory synapses. Solid lines indicate dominant projections; dashed lines indicate occasional connections

Fig. 2

Inhibitory interneurons affect auditory cortical responses and behavior. Activating PVs with ChR2 (a) increases tone-evoked responses and improves behavioral frequency discrimination acuity, whereas suppressing PVs using Arch has the opposite effect (b). (c) Direct activation of excitatory neurons with ChR2 does not change tone-evoked responses or behavioral frequency discrimination acuity on average. (c e) Left: diagram of optogenetic manipulation. Center: mean tone-evoked response magnitude under light-off and light-on conditions based on neuronal recordings. Right: Behavioral response to a shift in frequency under light-off and light-on conditions. Adapted from ref.

Fig. 3

Specific inhibitory neuron type mediates auditory adaptation. (a) Top: The effect of SOM and PV inactivation on stimulus-specific adaptation to frequent tones was tested using an oddball stimulus, with two tones at 10–90 ratio, light every 5th tone. Bottom: The mean firing rate (FR) during repeated tones adapted with successive presentations of the standard tone. (b) SOMs provide stimulus-specific inhibition, as the effect of SOM suppression increased with repeated standard tones. PVs provided constant inhibition regardless of adaptation. a, b adapted from ref. . (c) Passive exposure to a tone stimulus lead to a decrease in excitatory and an increase in inhibitory activity over 5 days. Left: calcium activity was imaged using two-photon microscopy in populations of identified inhibitory and excitatory neurons before and after subjecting the mouse to prolonged exposure to tones. Neuronal activity was measured as spike counts inferred from the imaged fluorescence signal. Right: change index of the mean activity in response to the tone to which the mouse was exposed, averaged over populations of excitatory (red) or inhibitory (blue) neurons, over days since prolonged tone exposure onset. Mean excitatory activity decreased with exposure, whereas mean inhibitory activity increased. (d) Among the inhibitory neurons, the activity of SOMs increased following passive tone exposure, whereas the activity of PVs decreased. Mean z-scored time course of Calcium activity of SOMs or PVs in response to a tone at day 1 (black traces) and day 5 (blue traces). c, d adapted from ref.

Fig. 4

Progressively complex view of cortical dynamics. (a) Diagram of the time course of inhibition-stabilized recurrent dynamics. Adapted from ref. . (b) Reduced model of mutually coupled Excitatory (Exc)—PV—SOM network. Exc rate is a non-linear—linear function of excitatory synaptic inputs (filled circles) evoked by the tone, as well as inhibitory inputs (open circles) from PVs and SOMs. PV and SOM firing rate is a non-linear—linear function of excitatory synaptic inputs from excitatory cells and tone-evoked excitatory inputs. Optogenetic manipulation by Arch is modeled as an inhibitory synaptic input. Adapted from ref. . (c) Left: network modules identified based on the correlation network structure. Adapted from ref. . Right: diagram of the time course of transformation in brain network structure with learning: Nodes belonging to the same module are colored in the same color. Black lines refer to the edges of the network. Note that with learning, the connectivity within and between modules is transformed. Adapted from ref.