The Effect of Inclusion Criteria on the Functional Properties Reported in Mouse Visual Cortex

Abstract

Neurophysiology studies require the use of inclusion criteria to identify neurons responsive to the experimental stimuli. Five recent studies used calcium imaging to measure the preferred tuning properties of layer 2/3 pyramidal neurons in mouse visual areas. These five studies employed different inclusion criteria and reported different, sometimes conflicting results. Here, we examine how different inclusion criteria can impact reported tuning properties, modifying inclusion criteria to select different subpopulations from the same dataset of almost 17,000 layer 2/3 neurons from the Allen Brain Observatory. The choice of inclusion criteria greatly affected the mean tuning properties of the resulting subpopulations; indeed, the differences in mean tuning because of inclusion criteria were often of comparable magnitude to the differences between studies. In particular, the mean preferred temporal frequencies (TFs) of visual areas changed markedly with inclusion criteria, such that the rank ordering of visual areas based on their TF preferences changed with the percentage of neurons included. It has been suggested that differences in TF tuning support a hierarchy of mouse visual areas. These results demonstrate that our understanding of the functional organization of the mouse visual cortex obtained from previous experiments critically depends on the inclusion criteria used.

Keywords: calcium imaging; data analysis; inclusion criteria; neurophysiology; visual cortex.

Figure 1.

Tuning characteristics in published studies. A, Mean preferred TF tuning of seven visual areas reported in five published studies. B, C, Same as in A but reporting the OSI and the DSI.

Figure 2.

Example cells that pass inclusion criteria exclusively. A, All DF/F responses to the preferred stimulus condition (TF and direction) of a cell that passes all published inclusion criteria. B, Heatmap of mean %DF/F responses to each stimulus condition (TF × direction). C, Mean %DF/F responses (± SEM) to stimuli of different grating directions in the same example cell. D–F, Same as in A but with a cell that passes most criteria, but not Study 2. G–I, Same as in A but with a cell that only passes Study 1 criteria.

Figure 3.

Most studies select for neurons along similar axes of the data. A, Six density plots of the mean response at the preferred stimulus condition (%DF/F) against the SD of the responses at the preferred stimulus condition where each point represents a single neuron. For each study, colored neurons are those selected for by inclusion criteria. Heatmap represents the density of neurons. B–D, Tuning characteristics after inclusion criteria are applied to Allen Brain Observatory. B, Mean TF tuning of six visual areas when different inclusion criteria are applied. C, D, show the mean OSI and DSI of six visual areas, respectively. E, Venn Diagram of neurons that were selected for by each inclusion criteria. Area of circles represents the number of neurons. Letters indicate example neurons from Figure 2.

Figure 4.

Tuning characteristics of neurons based on robustness. A, E, I, Mean TF, OSI, and DSI tuning of neurons in V1, AL, and PM based on what percentage of most robust cells (cells with low CV) are included in the analysis. Shaded regions indicate SEM. The minimum percentage most robust cells displayed is 5%. B, Distribution of TF tuning of 10% least robust cells. C, Distribution of TF tuning of 10% most robust cells. F, G, J, K, Same as in B, C but with OSI and DSI. D, H, L, Mean TF, OSI, and DSI tuning of neurons in all visual areas comparing the 10% most robust neurons to the entire population of neurons. M, Heat map displaying p values for Mann–Whitney U test comparing the 10% most robust neurons and the entire population of neurons. The color scale is centered at p = 0.05/6 to account for Bonferroni correction. N, Mean DSI calculated for neurons selected to match the mean CV for each the neurons selected by each criterion, for each area, compared with the mean DSI for the neurons selected by that criteria and area. O, P, Same as in M but for OSI and TF.

Figure 5.

Tuning characteristics of neurons based on robustness with cross-validated metrics. A, E, I, Mean TF, OSI, and DSI tuning of neurons in V1, AL, and PM based on what percentage of neurons are included in the analysis, starting with the most robust neurons. Shaded regions indicate SEM. The minimum percentage most robust cells displayed is 5%. B, Distribution of TF tuning of 10% least robust neurons. C, Distribution of TF tuning of 10% most robust neurons. F, G, J, K, Same as in B, C but with OSI and DSI. D, H, L, Mean TF, OSI, and DSI tuning of neurons in all visual areas in 10% most robust neurons versus the entire population of neurons.

Figure 6.

How trial number changes tuning metrics and CV. A, Mean CV calculated at the preferred condition using different numbers of trials and the cross-validation method. B, Mean peak response at the preferred condition versus SD at the preferred condition using only four trials and the cross validation method. C, Same as in B but using 14 trials. D–F, OSI, TF, and DSI calculated using the cross-validation method as a function of the number of trials used in the analysis.