Table 1
Demographics and clinical characteristics of the sample.
| M (SD) | |
|---|---|
| N = 726 n (%) | |
| Gender | |
| Cisgender female | 476 (66%) |
| Cisgender male | 225 (31%) |
| Non-binary, transgender, gender-fluid, or not reported | 23 (3.2%) |
| Ethnicity | |
| Hispanic or Latino/Latinx/Latine | 122 (17%) |
| Non-Hispanic and non-Latino/Latinx/Latine | 602 (83%) |
| Not reported | 2 (0.28%) |
| Race | |
| African American | 20 (2.8%) |
| American Indian/Alaskan native | 2 (0.3%) |
| Asian | 125 (18%) |
| Native Hawaiian/other Pacific Islander | 1 (0.1%) |
| White | 432 (62%) |
| More than one race | 56 (8.1%) |
| Other or not reported | 62 (8.9%) |
| Age | 19 (2) |
| Mood Symptoms | |
| MASQ-LOI | 16.71 (6.79) |
| N = 422 n (%) | |
| Current Mood Diagnoses | |
| (Unipolar) Depressive Disorder | 121 (29%) |
| Major depressive disorder | 73 (17%) |
| Persistent depressive disorder | 48 (11%) |
| Bipolar Disorder | 26 (6.2%) |
| Bipolar I disorder | 11 (2.6%) |
| Bipolar II disorder | 10 (2.4%) |
| Bipolar disorder not otherwise specified | 5 (1.2%)) |
| Non-Psychiatric Control | 275 (65%) |
[i] Notes: MASQ-LOI = Mood and Anxiety Symptom Questionnaire, Loss of Interest Anhedonia subscale.
Figure 1
Probabilistic Reward Task and Graphical Illustration of Action-DDM. (A) On each trial, participants saw a face with a long or short mouth and responded by pressing a button to indicate which mouth length was shown. Rewards were delivered three times more often for correct identifications of the rich vs. lean stimulus; assignment of short/long mouths to the rich/lean conditions was counterbalanced. (B) Graphical illustration of Action-DDM. Shaded nodes represent observed data and unshaded nodes represent parameter estimations. Double-bordered nodes represent trial-wise computed variables. Parameters include alpha: learning rate; Bv: the degree to which value differences influenced drift rate; Bz: the degree to which value differences influenced starting point bias; vintercept: baseline stimulus processing efficiency or drift rate; t: non-decision time; a: decision threshold.
Table 2
Summary of PRT performance statistics.
| VARIABLE | MEAN | SD | |
|---|---|---|---|
| Block 1 | response bias | 0.05 | 0.21 |
| discriminability | 0.67 | 0.27 | |
| rich acc | 0.82 | 0.10 | |
| lean acc | 0.79 | 0.12 | |
| rich RT | 529.64 | 91.94 | |
| lean RT | 530.00 | 92.43 | |
| rich correct RT | 530.67 | 89.50 | |
| lean correct RT | 529.14 | 89.91 | |
| rich error RT | 540.25 | 143.50 | |
| lean error RT | 555.30 | 144.90 | |
| Block 2 | response bias | 0.08 | 0.23 |
| discriminability | 0.64 | 0.27 | |
| rich acc | 0.82 | 0.11 | |
| lean acc | 0.76 | 0.14 | |
| rich RT | 532.88 | 94.20 | |
| lean RT | 541.15 | 95.56 | |
| rich correct RT | 532.62 | 91.94 | |
| lean correct RT | 543.75 | 93.81 | |
| rich error RT | 550.44 | 140.57 | |
| lean error RT | 554.92 | 135.46 | |
| Average | response bias | 0.07 | 0.19 |
| discriminability | 0.66 | 0.25 | |
| rich acc | 0.82 | 0.09 | |
| lean acc | 0.77 | 0.12 | |
| rich RT | 531.26 | 88.70 | |
| lean RT | 535.57 | 88.98 | |
| rich correct RT | 531.65 | 86.19 | |
| lean correct RT | 536.45 | 86.99 | |
| rich error RT | 544.77 | 128.20 | |
| lean error RT | 554.31 | 128.11 |
Figure 2
Posterior Predictive Performance. (A) Observed and simulated overall accuracy by stimulus type and response time (fast RT < .1 RT quantile; slow RTs > .9 RT quantile). (B) Observed and simulated changes in response bias and discriminability. The task trials were binned into eight timepoints with 25 trials each and response bias and discriminability were calculated independently for each timepoint. (C) Observed and simulated changes in response time distributions across four timepoints with 50 trials each. For a stimulus (rich or lean), we plotted the RT distributions when participants responded correctly (positively-valued RTs) and incorrectly (flipped to be negatively-valued RTs for illustrations).
Figure 3
Comparisons of PRT behavioral parameters between non-psychiatric control group (NC) and unipolar depression group (UNI).
Note: PRT summary statistics and Action-DDM parameters are indicated in gold and red respectively.
Figure 4
Correlations between anhedonic symptoms and PRT parameters.
Note: PRT summary statistics and Action-DDM parameters are indicated in gold and red respectively.