Brain state and cortical layer-specific mechanisms underlying perception at threshold
- Department of Neuroscience, Yale University, United States
- Interdepartmental Neuroscience Program, Yale University, United States
- Systems Neurobiology Laboratories, The Salk Institute for Biological Studies, United States
- Department of Psychiatry, Yale University, United States
- Wu Tsai Institute, Yale University, United States
- Department of Psychology, Yale University, United States
- Kavli Institute for Neuroscience, Yale University, United States
Figures
Figure 1 with 3 supplements
Orientation change detection task at perceptual threshold.
(A) Schematic of task structure. The monkey initiated a trial by fixating on the center of the screen. Two Gabor stimuli (represented by oriented lines) were presented for 200 ms and then turned off …
Figure 1—figure supplement 1
Laminar recordings in V4 (modified from Nandy et al., 2017).
(A) An artificial dura (AD) chamber is shown over dorsal V4 in the right hemisphere of Monkey A. The native dura mater was resected and replaced with a silicone-based AD, thereby providing an …
Figure 1—figure supplement 2
Additional psychometric function examples.
Example behavioral psychometric functions from nine recording sessions in the attend-in (red) and attend-away (blue) conditions. Behavioral performance (circles) is presented as a function of …
Figure 1—figure supplement 3
Psychometric function parameters.
(A) Distribution of the orientation threshold of the fitted psychometric functions in the attend-in (red solid) and attend-away (blue hatch) conditions for each recording session. The threshold of …
Figure 2 with 2 supplements
Hit trials have larger pupil diameter whereas microsaccades more often precede misses.
(A) Normalized pupil diameter for hit and miss trials in the threshold condition. 0 ms corresponds to non-target and target stimulus onset. Mean ± s.e.m. (B) Distribution of pupil diameter values …
Figure 2—figure supplement 1
Microsaccades are preferentially directed toward the target in correct trials and have a slight correlation with pupil diameter.
Data is presented for all trials, regardless of orientation change (not just the threshold condition). (A) The histograms represent the direction of microsaccades relative to the attended stimulus …
Figure 2—figure supplement 2
Single monkey pupil diameter and microsaccade data.
(A) Normalized pupil diameter for hit and miss trails in the threshold condition separated by monkey. 0 ms corresponds to non-target and target stimulus onset. Mean ± s.e.m. (B) Distribution of …
Figure 3 with 3 supplements
Target stimuli evoke higher firing rates in hit trials.
Rows correspond to different layers (top = superficial, middle = input, bottom = deep). (A) Performance for decoding targets from non-targets from single units and multi-units in each layer. Points …
Figure 3—figure supplement 1
Single monkey decoding performance.
(A) Performance for decoding targets from non-targets from single units and multi-units in the superficial layer for individual monkeys. Points in the left section of each plot show the decoding …
Figure 3—figure supplement 2
Single monkey firing rate data.
(A) Single monkey non-target population (single- and multi-unit) PSTH of visually responsive neurons for the hit (orange) and miss (dark-gray) trials in the threshold condition (mean ± s.e.m.). (B) …
Figure 3—figure supplement 3
Firing rates for individual neurons and reaction time in threshold condition.
(A) Target stimulus-evoked normalized firing rates in hit and miss trials for each recorded single- and multi-unit cluster in hit and miss trials. Clusters are divided by layer: left = superficial, …
Figure 4 with 2 supplements
Broad-spiking neurons in the superficial layer have decreased variability in hit trials.
(A) Rows correspond to different layers (top = superficial, middle = input, bottom = deep). The Fano factor of broad-spiking putative excitatory neurons for the hit and miss trials in the threshold …
Figure 4—figure supplement 1
Single monkey Fano factor data.
Rows correspond to different layers (top = superficial, middle = input, bottom = deep) and columns correspond to monkeys. The Fano factor of broad-spiking putative excitatory neurons for the hit and …
Figure 4—figure supplement 2
Narrow-spiking neurons do not have decreased variability in hit trials.
Top: Fano factor modulation index for each narrow-spiking neurons recorded in each layer, averaged in the 60 ms preceding non-target stimulus onset. Bottom: Bootstrapped estimation of the mean …
Figure 5 with 2 supplements
Deep-layer neurons are phase-locked to low-frequency rhythms in miss trials.
(A) Pairwise phase consistency (PPC) of single and multi-units in each layer to the local field potential (LFP) signal recorded from the same channel in hit and miss trials at threshold. PPC was …
Figure 5—figure supplement 1
Additional pairwise phase consistency (PPC) data.
(A–B) Top: Raw PPC values calculated for clusters recorded in the superficial (A) and (B) input layers in hit and miss trials, averaged into three frequency bands, 3–12 Hz, 15–25 Hz, and 30–80 Hz. …
Figure 5—figure supplement 2
Single monkey pairwise phase consistency (PPC) data.
Rows correspond to different layers (top = superficial, middle = input, bottom = deep) and columns correspond to monkeys. PPC of single and multi-units in each layer to the local field potential …
Figure 6
Hit trials are characterized by stronger feed-forward interlaminar population correlations.
(A) Canonical correlation analysis (CCA)-based population correlation as a function of time and interlaminar delay during the pre-stimulus and stimulus-evoked periods in hit and miss trials in an …
Figure 7 with 1 supplement
Greater interlaminar coherence in hit trials in the pre-stimulus and non-target stimulus-evoked periods.
Rows correspond to different pairs of layers (top = superficial-input, middle = superficial-deep, bottom = input-deep). (A) Multi-unit interlaminar spike-spike coherence (SSC) calculated in the 200 …
Figure 7—figure supplement 1
Single animal interlaminar coherence.
Rows correspond to different pairs of layers (top = superficial-input, middle = superficial-deep, bottom = input-deep). (A) Single monkey multi-unit interlaminar spike-spike coherence (SSC) …
Figure 8
Conceptual model for stimulus processing at perceptual threshold.
(A) Hit trials have a larger pupil diameter and fewer pre-target microsaccades, reflecting a state of increased arousal and greater eye position stability. Conversely, miss trials show decreased …
Author response image 1
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Other | Monkey | This paper | Species (Macaca mulatta) | |
| Software, algorithm | MATLAB | MathWorks | R2019a | |
| Software, algorithm | Cortex | NIMH | http://www.cortex.salk.edu/ |
Additional files
-
Supplementary file 1
Supplementary tables.
(a) Corresponding null hypothesis testing results. Null hypothesis testing results corresponding to the estimation statistics-based analyses. (b) Generalized linear model (GLM) coefficient values. Coefficients and significance values of the variables used in the GLM analysis. (c) GLM summary. Additional summary statistics of the GLM analysis.
- https://cdn.elifesciences.org/articles/91722/elife-91722-supp1-v1.docx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/91722/elife-91722-mdarchecklist1-v1.docx
