Directional Influence between the Human Amygdala and Orbitofrontal Cortex at the Time of Decision-Making
(A) Task summary. On every trial subjects were shown an image of a snack food for 1 s, at which time they were prompted to indicate whether or not they would be willing to eat the food using a four-item response scale (Strong-Yes, Yes, No, Strong-No). At the end of the experiment one trial was selected at random and the subject's choice was implemented using the actual food. Snacks could be appetitive or aversive, as measured by independent continuous liking-ratings provided by each subject. (B) Scatter plots (jittered) showing the association between prior continuous liking-ratings and choices for each food and subject. Lines correspond to least square fits. Correlation coefficients and p-values were: PT258: 0.73, p<10−27; PT206: 0.91, p<10−63, PT180: 0.58, p<10−15 (C) Estimated cumulative transition probabilities from an ordinal multinomial GLM that conditions choices in trial t on the response on the previous trial t -1. Dotted lines correspond to the best estimates from a restricted model without the autoregressive (i.e., history independent) component.