Article Figures & Data
Figures
- Figure 1.
Neuronal correlates of perceptual decision making in the mouse cortex. A, Top, Schematic of the experimental configuration of the audiovisual change detection paradigm for head-fixed mice. Bottom, Timeline of the audiovisual change detection task, indicating task contingencies for a visual change trial. Modified from Oude Lohuis et al. (2022b,c). B, Schematic representation of the relevant cortical areas represented on a flattened cortical surface. Acronyms are used for the major subdivision of the dorsal cortex following standard nomenclature (Wang and Burkhalter, 2007; Wang et al., 2012; Steinmetz et al., 2019). Highlighted in color are the areas from which data was analyzed. V1, primary visual cortex; PPC, posterior parietal cortex; ACC, anterior cingulate cortex. MOs, Supplementary motor cortex. C, Baseline-corrected average PSTHs recorded in (from top to bottom) V1, PPC, and ACC following a change in the orientation of the presented drifting grating. Red, hits; green, misses. Dark colors indicate max visual change (highest saliency), while light colors indicate threshold visual change (low saliency). Shaded areas indicate the standard error of the mean. Color bars on top of individual panels indicate time bins in which significant differences (p < 0.05, permutation-based test, FDR-corrected) were found between responses to, respectively, sensory stimuli with a difference salience (blue) or hit/miss trials (orange). See also Extended Data Figure 1-1. D, Outline (top) and timeline (bottom) of the contrast discrimination task, in which mice had to rotate a wheel to bring the Gabor patch with the highest contrast toward the center of the field of view. CW, clockwise; ACW, anticlockwise. Modified from Steinmetz et al. (2019). E, Same as C, but computed as a function of the difference in contrast between the stimulus presented in the contralateral field of view with respect to the recorded hemisphere (which was always the highest-contrast stimulus) and the stimulus presented ipsilaterally. The color darkness indicates the contrast difference. Statistical differences were computed as in panel C. Note that no difference between responses to sensory stimuli with different contrasts was observed. See also Extended Data Figure 1-1.
- Figure 2.
Network architecture and activity in the nominal setup. A, Network schematic. We developed a minimal network model with an excitatory (E) and an inhibitory (I) node in three cortical areas: V1, PPC, and PFC. Orange (blue) arrows indicate feedforward (feedback) connections, whose values were determined based on anatomical connectivity—roughly indicated by arrow thickness; see also panel B. Connections between excitatory and inhibitory nodes (black arrows) were calibrated to match experimental results. External input was applied to V1 (red arrow) to simulate visual stimuli. B, Synaptic weights between nodes. The top left 3 × 3 block corresponds to nonlocal connections (matrix W, see STAR methods), while the other three blocks correspond to local couplings γi. Values in highlighted cells (red lines) were experimentally derived. All other values were calibrated. C–E, Example firing rate traces in the three regions for two different values of applied current: C, low current Imax = 0.8 pA; D, E, medium current Imax = 2 pA. At medium currents a feedback bump may (D) or may not (E) appear depending on small changes in initial conditions. In C–E, row 1 reflects the input, rows 2–4 the activity of excitatory nodes, and rows 5–7 the activity of inhibitory nodes.
- Figure 3.
Network activity as a function of initial conditions and input currents. A, Examples of activity trajectories (random initial conditions) for different input currents, in the nominal setup of parameters. B, Distributions of integral quantity S for the trajectories in A. A late-activity bump is detected when the integral of V1 firing rate S lies in the interval [0.2, 0.35], highlighted in green. C, Frequency distribution of S for different values of applied current Iapp. Red columns highlight the values displayed in B. The statistics is obtained over 100 different realizations for every value of Imax, with initial conditions sampled from a uniform, random distribution ui(t = 0) ∈ [0, 0.05] spikes/s.
- Figure 4.
Probability of observing a late-activity bump, in the plane of parameters (α, Imax), for the sets of connections considered. In each panel, heatmaps show the probability of observing both an early and late activity bumps P2b as a function of applied input current Imax and morphing parameter α, applied to a different set of connections. The green and orange lines represent isolines of P1b = 99% and Pov = 99%, respectively, which border regions dominated by single bumps/inactivity and overshooting (see gray labels on the top-left panel). A white, dotted line marks the nominal setup (α = 1). For each panel, morphed connections are colored in red on the respective network scheme. For each couple of (α, Imax) values, we considered 50 random initial conditions (see Materials and Methods).
- Figure 5.
Effects of instantaneous link inactivation on feedback bumps. A, We selected initial states and nominal conditions leading to late-latency activity (dashed line, T* = none). We ran this network up to a chosen time T* and then instantaneously set the PFC→V1 link to zero. For Imax = 1.8 pA (A.1) the instantaneous inactivation prevented (T* = 50–300 ms) or suppressed (T* > 300 ms) the feedback bump in V1. Conversely, when a higher value of Imax is used (A.2, Imax = 3.0 pA), feedback activity can be restored in V1 despite the inactivation of the PFC→V1 link. B.1–2, Same as A.1–2 but for the isolation of PFC from the network. Note how the feedback bump is always prevented or suppressed irrespective of the value of Imax (compare Fig. 4). Similar results were obtained for all other set of connections (compare Fig. 4).
Tables
V1 PPC PFC V1 11.22 1.29 PPC 4.57 10.57 PFC 0.72 9.78 V1 PCC PFC 1 1 1 2.3 1.8 1.9 2 2 2 0.5 0.5 0.5 V1 PPC PFC 3 2 2 2 2 2 2 4 2 0.3 0.3 0.3 V1 PCC PFC 30 200 38 10 10 10 0.8 0.9 3.8 0.07 0.1 0.07
Figure 1-1
(A) Same as Fig. 1C, but including neuronal responses during false alarms (grey). Neuronal responses during error trials (licks toward the wrong side) are not shown in view of their rare occurrence. For visualization purposes, hit and miss trials are only shown for max visual change trials. (B) Same as Fig. 1E, but including neuronal responses during false alarms (grey) and during errors (black), the latter indicating responses towards the lowest-contrast visual stimulus. For visualization purposes, hit and miss trials are only shown for max visual change trials. Download Figure 1-1, TIF file.
Extended Data 1
The file code.zip includes all code used to generate the figures and perform the analyses presented in this manuscript. The folder "./ephys analysis/" contains code and instructions concerning experimental data and Figure 1. The files "Figure2CDE.m", "Figure3AB.m" and "Figure3C.m" simulate the model in the nominal setup, and generate the corresponding figures. The file "Morphing.m" allows changing the connectivity strengths with respect to the nominal setup, and plotting the resulting trajectories. The folder "./Figure4/" contains instructions, code and data concerning Figure 4. Download Extended Data 1, ZIP file.
In this issue
