Figures and data
Task timeline.
Subjects chose between two snack food lotteries on each trial. Subjects learned about the lotteries through random food draws. Every 4-8 seconds, subjects sampled a new draw from each lottery. They were allowed to sample as many times as they wanted but were incentivized to sample approximately 7 draws per trial. Sampled food draws were presented for 2 seconds, followed by a fixation cross appearing for 2-6 seconds with random jitter. The trial ended when the subject chose the left or right lottery, using the respective index finger. Upon making their choice, subjects were presented with a food drawn from their chosen lottery.
Example trial with the sampled value and accumulated value.
The sampled value (SV; red) and accumulated value (AV; black) are plotted for this example trial. For the first draw, the SV and AV are identical. However, as the trial proceeds, the two signals diverge. In the model, a choice is made when the AV reaches a pre-specified decision boundary.
Choice data.
(a) The probability of choosing left based on the left – right value difference for both SV and AV. As the value difference becomes greater in favor of one option, the probability of choosing that option increases, for both SV and AV. (b) The effect of gaze on choice. The longer that subjects looked at one lottery over the other, over the course of the whole trial, the more likely they were to choose that lottery.
Regions responding to sampled and accumulated value
(a) vmPFC showed a significantly positive correlation with |ΔSV|, but did not respond to |ΔAV|. (b) Both pre-SMA and IPS (as well as the dlPFC, not pictured) showed a significantly positive correlation with |ΔAV|, but no correlation with |ΔSV|. Voxels thresholded at p < .05.
Beta plots from the vmPFC, striatum, pre-SMA, IPS, and dlPFC (GLM1).
Displayed are regression coefficients from each region for absolute sampled value difference (|ΔSV|) and absolute accumulated value difference (|ΔAV|). Both vmPFC and striatum show a similar pattern of BOLD activity that scales positively with |ΔSV|, but does not respond to |ΔAV|. The opposite pattern can be seen in the pre-SMA, IPS, and dlPFC which both show a strong positive correlation between BOLD activity and |ΔAV|, but no relationship to |ΔSV|.
Regions responding to sampled gaze-weighted value (SGWV) and accumulated gaze-weighted value (AGWV).
SGWV correlates with activity in (a) vmPFC, and (b) striatum, while AGWV correlates with activity in (c) pre-SMA. Voxels thresholded at p < .05.
Representation of gaze-weighted evidence.
The vmPFC shows a positive interaction between sample value (SV) and gaze location, while the pre-SMA shows a negative interaction between accumulated value (AV) and gaze location. Both results are consistent with gaze enhancing the value of fixated items.
Non-uniform temporal weighting.
In both gaze-weighted and non-gaze-weighted models, participants showed stronger recency than primacy effects, both in terms of (A) the model parameters, and (B) the resulting temporal weighting functions averaged across all trials. Error bars are standard errors clustered by participant.
