Table 2 Correlations between attention measures and the angle (representing direction) and magnitude (representing speed) of the neural trajectory across different conditions, compared to the respective chance distribution

From: Geometry of neural dynamics along the cortical attractor landscape reflects changes in attention

 

Movie

Task

 

Sitcom ep1

Sitcom ep2

GradCPT face

GradCPT scene

Angle (\(\vec{WD}\), attractor)

z = 3.570, q = 0.002*

z = 4.047, q < 0.0001*

z = −3.451, q = 0.002*

z = −3.277, q = 0.002*

Magnitude (\(\vec{WD}\))

z = −4.779, q < 0.0001*

z = −5.776, q < 0.0001*

z = 2.147, q = 0.048*

z = 3.096, q = 0.005*

Angle (\(\vec{B}\), attractor)

z = −2.275, q = 0.040*

z = 0.773, q = 0.440

z = 1.817, q = 0.093

z = 1.815, q = 0.093

Magnitude (\(\vec{B}\))

z = −4.778, q < 0.0001*

z = −1.377, q = 0.207

z = 1.137, q = 0.293

z = 0.825, q = 0.432

  1. Analyses were conducted on four runs of the SONG dataset: two sitcom episodes watching and gradual-onset continuous performance task (gradCPT) with either face or scene images. False discovery rate (FDR) correction was applied to control for multiple comparisons across 16 significance tests. Asterisks indicate FDR-corrected p < 0.05.