Figure 4. Response modulation by spatial attention and running is independent and different in reliability.

(A) Example cells showing different response modulation patterns. Upper traces, visual responses when mice attend in or out of the RF location. Lower traces, visual responses when mice are running or stationary. The values under the traces are modulation indices.

(B) Scatterplot showing no correlation between attentional and running modulation indices.

(C) Top, an example cell showing stable running-dependent modulation across days. Corresponding cell images and regions of interest (red) are shown for each day. Values below images are modulation indices. Bottom, scatterplots comparing running modulation index between day N and consecutive days. Data from more than 4 days after the first day were pooled together.

(D) Correlation of running modulation index across days. Error bars, 90% CI of bootstrapped mean. The numbers of session pairs are indicated at the bottom.

(E) As in (C), attentional modulation.

(F) As in (D), attentional modulation.

(G) Within-block reliability of attentional modulation. The scatterplot compares the modulation indices between the first and second half of the same blocks.

(H) Between-block reliability of attentional modulation. Attentional modulation indices were calculated separately for the first and third blocks and compared in the scatterplot. Same neurons as in (G). p values in (G) and (H) were determined by a randomization test.

(I) Summary. Two types of state-dependent response modulation are imposed on single neurons independently. After repeated state changes, single neurons are either variably or stably modulated by spatial attention and running, respectively.

p values of all example neurons were determined by bootstrap test. All calcium response traces are mean response to all sparse noise stimuli within RFs. See also Figures S7–S10.