Article Figures & Data
Figures
- Figure 1.
The fMRI paradigm timeline and symbol explanation for each of the three domains. Each subject completed 48 win, 48 loss, and 24 neutral trials, and the symbol–domain relationship was counterbalanced.
- Figure 2.
The eye-tracking experiment. A, Paradigm timeline. B, Exact display of the eye-tracking stimulus shown during the decision phase. C, The eight areas of interest used to extract the number of fixations.
- Figure 3.
Whole-brain activation during the different parts of reward and loss processing (whole-brain corrected p < 0.05, based on 3dClustSim correction [k > 33, p < 0.005], df = 47) and obtained using the fMRI paradigm. Color coding: magenta, RP; green, RPE; blue, LP; yellow, LPE. The first and the third row represent the positive contrasts vs baseline (>0), and the second and fourth row represent the negative contrasts vs baseline (<0) in the reward and loss domains, respectively. Respective t value color bars are shown on the right side.
- Figure 4.
Correlation of fMRI and eye-tracking data, showing individual data points and the respective ROIs used for the contrast value extraction. A, Contrast values of the left VMPFC during RPE processing correlate with higher values of Df win [i.e., increased attention toward values (f[v]) compared with probabilities (f[p]) in the win domain; r = 0.306, p = 0.046, df = 42]. B, Contrast values of the right VMPFC during RPE processing correlate with an increased attention toward values [f(v)] compared with probabilities [f(p)] in the win domain (r = 0.397, p = 0.008, df = 42). C, Contrast values of the PCC during LPE processing correlate with an increased attention toward probabilities [f(p)] compared with values [f(v)] in the loss domain (r = −0.284, p= 0.065, df = 42). D, Contrast values of the left VS during LPE processing correlate with an increased attention toward probabilities [f(p)] compared with values [f(v)] in the loss domain (r = −0.316, p = 0.039, df = 42).
Tables
Regressor Parametrical modulation Contrasts of interest Onset of choice, win domain Yes: RP RP > 0 Onset of choice, loss domain Yes: LP LP > 0 Onset of choice, neutral domain Yes: RP Onset of result, win domain Yes: RPE RPE > 0 Onset of result, loss domain Yes: LPE LPE > 0 Onset of result, neutral domain Yes: RPE Six movement regressors NA NA, not applicable.
- Table 2:
Overview of the parametric modulator calculations used for estimating brain activation
Number of symbols shown RP RPE in case of win (1): RPE = 1 − RP RPE in case of no win (0): RPE = 0 − RP Win domain 1 1 1 − 1 = 0 0 − 1 = −1 2 1/2 1 − 1/2 = 1/2 0 − 1/2 = −1/2 3 1/3 1 − 1/3 = 2/3 0 − 1/3 = −1/3 4 1/4 1 − 1/4 = 3/4 0 − 1/4 = −1/4 Number of symbols shown LP LPE in case of loss: LPE = −1 − LP LPE in case of no loss: LPE = 0 − LP Loss domain 1 0 −1 − 0 = −1 0 − 0 = 0 2 −1/2 −1 − (−1/2) = −1/2 0 − (−1/2) = 1/2 3 −2/3 −1 − (-2/3) = −1/3 0 − (−2/3) = 2/3 4 −3/4 −1 − (−3/4) = −1/4 0 − (−3/4) = 3/4 Contrast Region Laterality MNI coordinates Cluster size t Cluster P (FWE corrected) x y z RP > 0 MTG R 66 −49 2 1368 7.01 <0.001 oMFG L −57 29 −10 2017 6.70 <0.001 vmPFC L −9 38 −7 2017 5.64 <0.001 MTG L −63 −55 29 1337 6.17 <0.001 PCUN L 3 −52 20 671 6.07 <0.001 POG R 39 −19 56 175 5.05 0.011 RPE > 0 VS L −12 5 −10 1805 7.95 <0.001 VS R 12 5 −10 1805 7.84 <0.001 vmPFC L −6 44 −1 1094 6.87 <0.001 vmPFC R 3 47 −1 1094 6.35 <0.001 MTG L −60 −46 5 189 4.35 0.004 VIS L −3 −76 2 235 4.33 0.001 MCC L/R 0 −16 44 317 4.26 <0.001 Cluster size FWE-corrected, voxel threshold = 0.005; df = 47. L, Left; R, right.
Contrast Region Laterality MNI coordinates Cluster size t Cluster p (FWE corrected) x y z LP > 0 dmPFC R 3 50 35 392 6.06 <0.001 TPJ L −57 −67 29 647 5.76 <0.001 MTG R 60 −43 5 159 5.68 0.015 PCUN R 6 −55 35 139 5.19 0.028 MTG L −63 46 −1 174 4.90 0.001 TPJ R 57 −58 35 447 4.68 <0.001 vmPFC L/R 0 26 −16 158 4.49 0.015 LPE > 0 VIS R 9 −79 −1 4845 14.54 <0.001 AI R 42 20 −1 129 5.47 0.038 Cluster size FWE-corrected, voxel threshold = 0.005, df = 47. L, Left; R, right.
Contrast k threshold* Contrast direction Region Laterality Peak MNI coordinates Cluster size (k) Peak t x y z RP > 0 34 Positive AI Left n.s. n.s. n.s. Right vmPFC Left −9 47 −13 556 4.60 Right 6 50 −10 556 4.80 VS Left n.s. n.s. n.s. Right Negative AI Left −33 23 5 105 5.26 Right n.s. n.s. n.s. vmPFC Left n.s. n.s. n.s. Right VS Left n.s. n.s. n.s. Right RPE > 0 32 Positive AI Left −33 17 −7 371 5.29 Right 30 20 −7 697 7.37 vmPFC Left −3 47 5 583 6.34 Right 3 47 −1 583 6.35 VS Left −12 5 −10 371 7.95 Right 12 5 −10 697 7.84 Negative AI Left n.s. n.s. n.s. Right vmPFC Left n.s. n.s. n.s. Right VS Left n.s. n.s. n.s. Right ↵*Whole-brain corrected, p < 0.05 (based on 3dClustSim correction: k > 33, p < 0.005). n.s., Not significant.
Contrast k threshold* Contrast direction Region Laterality Peak MNI coordinates Cluster size (k) Peak T x y z LP > 0 34 Positive AI Left n.s. n.s. n.s. Right vmPFC Left −6 29 −10 52 4.43 Right 0 26 −16 52 4.49 VS Left n.s. n.s. n.s. Right Negative AI Left −30 26 −4 114 3.77 Right n.s. n.s. n.s. vmPFC Left n.s. n.s. n.s. Right VS Left −18 11 −1 114 6.60 Right 12 8 5 95 5.26 LPE > 0 33 Positive AI Left −30 26 2 47 4.61 Right 33 29 −1 74 4.26 vmPFC Left n.s. n.s. n.s. Right VS Left n.s. n.s. n.s. Right Negative AI Left n.s. n.s. n.s. Right vmPFC Left −12 47 −4 390 3.86 Right 6 41 −13 390 5.58 VS Left −15 11 −10 733 7.95 Right 24 −1 −13 178 5.26 ↵*Whole-brain corrected p < 0.05 (based on 3dClustSim correction: k > 33, p < 0.005).
Variable N Minimum Maximum Mean SD Total fixations, win domain 44 7.90 65.23 29.32 12.726 Total fixations, loss domain 44 7.50 61.17 28.94 11.720 Df win domain: [f(v) − f(p)] 44 −9.13 12.20 2.51 4.959 Df loss domain: [f(v) − f(p)] 44 −7.77 11.07 2.73 4.521 Df high risk, win domain: [f(h) − f(l)] 44 −5.00 2.07 −0.35 1.395 Df high risk, loss domain: [f(h) − f(l)] 44 −3.97 4.37 0.34 1.495 Percentage of high-risk choices, win domain 44 26.67 100.00 66.59 20.466 Percentage of high-risk choices, loss domain 44 30.00 100.00 77.27 18.584 Average reaction time, all trials 44 3.02 17.99 8.51 3.544 Average reaction time, win domain 44 2.92 21.63 8.68 3.935 Average reaction time, win domain, high-risk choices 44 2.92 22.68 8.74 3.958 Average reaction time, win domain, low-risk choices 41 3.23 19.76 8.96 4.058 Average reaction time, loss domain 44 3.12 16.38 8.56 3.471 Average reaction time, loss domain, high-risk choices 44 3.12 15.59 8.37 3.363 Average reaction time, loss domain, low-risk choices 44 3.57 38.20 10.33 6.309 - Table 8:
Additional significant Pearson correlations (uncorrected) of the variables from the eye-tracking task
Variable 1 Variable 2 r P N Df win Df loss 0.94 <0.001 44 Average reaction time, win domain −0.39 0.009 44 Average reaction time, win domain, high-risk choices −0.38 0.011 44 Average reaction time, win domain, low-risk choices −0.38 0.016 41 Df loss Average reaction time, loss domain −0.46 0.002 44 Average reaction time, loss domain, high-risk choices −0.48 0.001 44 Average reaction time, loss domain, low-risk choices −0.36 0.023 44 Df high risk win Percentage of high-risk choices, win domain 0.31 0.039 44 - Table 9:
Main bootstrapped Pearson correlation results (uncorrected, seed set at 10 with 10,000 repetitions) of the fMRI and eye-tracking data (df = 42)
Eye-tracking variable ROI fMRI contrast variable RP > 0 RPE > 0 Df win Left AI −0.05 (0.169) 0.14 (0.142) Right AI 0.11 (0.122) −0.02 (0.145) Left OSS −0.03 (0.189) 0.01 (0.125) Right OSS −0.08 (0.210) 0.02 (0.148) PCC 0.11 (0.151) 0.10 (0.129) Left vmPFC 0.22 (0.156) 0.31 (0.150)* Right vmPFC 0.18 (0.150) 0.40 (0.138)* Left VS 0.09 (0.135) 0.23 (0.140) Right VS 0.17 (0.140) 0.02 (0.131) LP > 0 LPE > 0 Df loss Left AI 0.10 (0.144) −0.20 (0.127) Right AI 0.13 (0.116) −0.12 (0.151) Left OSS 0.17 (0.129) −0.12 (0.145) Right OSS 0.03 (0.157) −0.21 (0.142) PCC −0.02 (0.129) −0.28 (0.134)* Left vmPFC −0.16 (0.159) −0.17 (0.143) Right vmPFC 0.01 (0.156) −0.15 (0.152) Left VS −0.01 (0.152) −0.32 (0.152)* Right VS −0.08 (0.163) <0.01 (0.138) ↵*p < 0.05.
In this issue
