Effects of acetylcholine on response profiles and orientation tuning. Upper graphs show raster plots and histograms when ACh was not applied (black curves) and when it was applied (grey curves, width of the curves denotes standard error) for four example cells (A–D). Row 2 (ACh not applied) and 3 (ACh applied) show the activity as a function of time (horizontal axis in ms relative to stimulus onset, time 0) and bar orientation (vertical axis, preferred orientation [deg] corresponds to second tick mark from the base). Row 4 shows the difference plot between those shown in row 2 and 3. Cell A was affected by ACh at all orientations and over the entire response period. Cell B was mostly affected at orientations close to the preferred. It was one of the few cells where ACh caused a change in latency. Cell C was affected by ACh application over a broad range of orientations, but most strongly at those flanking the preferred, and during late response periods. Cell D was inhibited by ACh application. The inhibition was strongest when the preferred orientation was presented and during the early response period. The histograms are based on 20 trials (top 20 rows of rasters) without and 20 trials with ACh application each.

Firing rate of neurons as a function ACh application. (A) Effect of ACh on spontaneous activity. Most neurons increased their spontaneous activity upon ACh application. (B–F) Effects of ACh application on stimulus driven activity (preferred bar orientation) during the different response periods. horizontal axis, activity when ACh was not applied, vertical axis, activity when ACh was applied. Black symbols, cells facilitated by ACh, grey symbols, cells inhibited by ACh.

Median modulation index for the spontaneous activity and firing rate in the presence of a bar with preferred orientation. (A) Modulation index for cells facilitated by ACh during spontaneous activity and during different response periods following response onset. (B) Modulation index for cells inhibited by ACh during spontaneous activity and during different response periods following response onset. Error bars denote 25, 75 percentiles.

Signal-to-noise ratios (SNR). (A) SNR during the whole response period (0–500 ms) when ACh was not applied (horizontal axis) and when it was applied (vertical axis) for cells facilitated (filled symbols) and cells inhibited (open symbols). (B) SNR during the first 0–100 ms following response onset. (C) SNR during the period from 300 to 400 ms following response onset. Symbols denote the median SNR obtained from bootstrapping. Error bars denote 25, 75 percentiles obtained from the bootstrap distribution.

Interspike interval distributions (ISI) and coefficient of variation (CV). (A) Raster plot and histogram of a cell response to the preferred bar orientation when ACh was not applied (upper graph) and when ACH was applied (lower graph). Dashed vertical lines show the period from which ISI values were obtained. (B) ISI (black histograms) obtained from the raster plots shown under A. Grey curve shows a gamma function fitted to the ISI distributions. The CV decreased upon ACh application for this cell. (C) Average ISI as a function of time following response onset (black curve, left vertical axis) measured within a sliding window of 25 ms widths. Average coefficient of variation (grey curve, right vertical axis) as a function of time following response onset measured within a sliding window of 25 ms. Dots represent individual ISI durations.

Cell examples of orientation tuning with ACh not applied and ACh applied. The upper row shows examples cells demonstrating a sharpening of the tuning curve (A), an increase of the amplitude (B), and a broadening of the tuning curve (C). Box plots underneath the tuning curves show the median (and 25, 75 percentiles) of the tuning parameters amplitude, FWHM, and offset derived from bootstrapping. P-values indicated whether the parameters of interest were significantly affected by ACh application. Grey symbols and fitted line, orientation tuning when ACh was applied. Black symbols and fitted line, orientation tuning when ACh was not applied.

Population data of orientation tuning. (A) Amplitude of the Gaussian fitted to the orientation tuning data during the three different response periods. (B) Tuning widths, FWHM of the Gaussian fitted to the orientation tuning data during the three different response periods. (C) Tuning offset of the Gaussian fitted to the orientation tuning data during the three different response periods. horizontal axis, parameters of interest when ACh was not applied, vertical axis: parameters of interest when ACh was applied. Filled symbols median (and 25, 75 percentiles) of the parameter of interest for cells facilitated by ACh determined by bootstrapping; open symbols median (and 25, 75 percentiles) of parameter of interest for cells inhibited by ACh determined by bootstrapping.

Schematic representation of the V1 cellular network and its modulation by ACh. Cell bodies are symbolized by filled circles or triangles, dendrites by bars, axonal terminal zones by transparent ellipsoids. Lateral intracortical connections are symbolized by purple horizontal lines (on which presynaptic muscarinic receptors are located (grey boxes). In addition the axonal terminal zones could also be populated by these receptors (not shown for clarity). Thalamocortical terminals are symbolized by vertical blue lines. Nicotinic presynaptic receptors located on these terminals are shown as black triangles. Postsynaptic receptors can be located on the soma or the dendrites, but are shown on somas exclusively for simplicity. The omission of receptors on some cell bodies was also performed for clarity. Most of the recordings in our study were from supragranular layers (electrode position). ACh applied to layer 1 interneurons (l1) would increase the excitability of pyramidal cells (p2/3) by inhibiting layer 2/3 (b2/3) interneurons (). In layer 2/3 ACh activates muscarinic receptors on pyramidal cells causing cell depolarization, increased excitability, and reduced spike frequency adaptation (; ). A simultaneous activation of muscarinic receptors located on interneurons would counteract the depolarization of the pyramidal cell, due to increased GABAergic inhibition (; but note that hyperpolarization of interneurons can also occur ). Postsynaptic M2 muscarinic receptors on layer 4 cells decrease the excitability of layer 4 neurons (), thereby possibly reducing the excitatory feed forward drive to layer 2/3 pyramidal cells. Presynaptic muscarinic receptors in most layers reduce the efficacy of lateral excitatory (M2, M4) and inhibitory (M1) connections resulting in decreased intracortical integration (; ; ; ; ). Nicotinic receptors in layer 4 of primate V1 boost the synaptic efficacy of thalamocortical connections, increasing the relative efficacy of feed-forward signals (; ; ). Postsynaptic nicotinic receptors are mostly confined to regular and irregular spiking interneurons in layer 2/3 (; ; ; ; but see ), where they increase the inhibitory drive within the network, resulting in inhibition of pyramidal cells, as well as disinhibition (). The schema highlights the diversity of receptors at different locations, their effects, and their specificity dependent on localization. As a result application of ACh has a variety of effects, depending on whether application is local, or whether the whole network is affected. l1, interneuron from layer 1; db, double bouquet cell; b, basket cell; p, pyramidal cell; ss, spiny stellate cell; LTS, low threshold spiking interneuron; numbers refer to the location of the cell body.