Skip to main content
. 2013 Sep;49(8):2067–2079. doi: 10.1016/j.cortex.2012.11.010

Fig. 5.

Fig. 5

FIR analysis was used to investigate the time-course of responses in the HC (A) and PHC. (B). In each case the increase in activity for trials where BE occurred compared to those where it did not are plotted, with standard error bars. The different FIR time-windows are displayed on the x axis, and percent change in fMRI BOLD response on the y axis. For both regions a significant increase in activation as early as 2–4 sec following the presentation of the first scene was apparent (which was before the presentation of the second scene). Furthermore, given that the BOLD response lags behind stimulus presentation with a peak of 6 sec, this reflected a rapid response to the first scene. (C) The results of the DCM model comparison analysis. This plot displays the exceedance probability on the y axis, which describes how likely each model is compared to any other model. This is shown for the three possible models. The ‘back’ model was the winner in both hemispheres, suggesting that the HC is the driving force behind the BE effect, and influences PHC.