TY - JOUR
T1 - Outcome saliency modulates behavioral decision switching
AU - Sun, Sai
AU - Yu, Rongjun
AU - Wang, Shuo
N1 - Funding Information:
We thank Ping Zhang for collecting some of the data and Paula Webster for valuable comments. This research was supported by the China Scholarship Council (to S.S.), and an NSF CAREER Award (1945230), West Virginia University (WVU), WVU PSCoR Program, ORAU Ralph E. Powe Junior Faculty Enhancement Award, and the Dana Foundation (to S.W.). The funders had no role in study design, data collection, and analysis, decision to publish, or preparation of the manuscript.
PY - 2020/8/31
Y1 - 2020/8/31
N2 - Goal-directed decision making often requires evaluating the outcomes of our decisions, assessing any gains or losses, learning from performance-related feedback, and deciding whether to alter our future decisions. However, it is unclear how these processes can be influenced by the saliency of an outcome (e.g., when one aspect of the outcome is accentuated more than another). Here we investigated whether decision strategies changed when certain aspects of the task outcome (win/loss or correct/incorrect) became more salient and how our brain encoded such saliency signals. We employed a simple two-alternative forced choice gambling task and quantified the frequency at which participants switched decisions to an alternative option in the subsequent trial after receiving feedback on their current selection. We conducted three experiments. In Experiment 1, we established the baseline decision switching behavior: participants switched more frequently following incorrect trials than correct trials, but there was no significant difference between win and loss trials. In Experiment 2, we highlighted the utility (win or loss) or performance (correct or incorrect) dimension of the chosen outcome and we found that the difference in switching frequency was enlarged along the highlighted dimension. However, Experiment 3 showed that when using non-specific saliency emphasis of the outcome, the saliency effect was abolished. We further conducted simultaneous EEG recordings using specific saliency emphasis and found that the feedback-related negativity, P300, and late positive potential could collectively encode saliency modulation of behavioral switching. Lastly, both the frontal and parietal theta-band power encoded the outcome when it was made more salient. Together, our findings suggest that specific outcome saliency can modulate behavioral decision switching between choices and our results have further revealed the neural signatures underlying such saliency modulation. Altering the saliency of an outcome may change how information is weighed during outcome evaluation and thus influence future decisions.
AB - Goal-directed decision making often requires evaluating the outcomes of our decisions, assessing any gains or losses, learning from performance-related feedback, and deciding whether to alter our future decisions. However, it is unclear how these processes can be influenced by the saliency of an outcome (e.g., when one aspect of the outcome is accentuated more than another). Here we investigated whether decision strategies changed when certain aspects of the task outcome (win/loss or correct/incorrect) became more salient and how our brain encoded such saliency signals. We employed a simple two-alternative forced choice gambling task and quantified the frequency at which participants switched decisions to an alternative option in the subsequent trial after receiving feedback on their current selection. We conducted three experiments. In Experiment 1, we established the baseline decision switching behavior: participants switched more frequently following incorrect trials than correct trials, but there was no significant difference between win and loss trials. In Experiment 2, we highlighted the utility (win or loss) or performance (correct or incorrect) dimension of the chosen outcome and we found that the difference in switching frequency was enlarged along the highlighted dimension. However, Experiment 3 showed that when using non-specific saliency emphasis of the outcome, the saliency effect was abolished. We further conducted simultaneous EEG recordings using specific saliency emphasis and found that the feedback-related negativity, P300, and late positive potential could collectively encode saliency modulation of behavioral switching. Lastly, both the frontal and parietal theta-band power encoded the outcome when it was made more salient. Together, our findings suggest that specific outcome saliency can modulate behavioral decision switching between choices and our results have further revealed the neural signatures underlying such saliency modulation. Altering the saliency of an outcome may change how information is weighed during outcome evaluation and thus influence future decisions.
UR - http://www.scopus.com/inward/record.url?scp=85090014868&partnerID=8YFLogxK
U2 - 10.1038/s41598-020-71182-9
DO - 10.1038/s41598-020-71182-9
M3 - Journal article
C2 - 32868828
AN - SCOPUS:85090014868
SN - 2045-2322
VL - 10
JO - Scientific Reports
JF - Scientific Reports
IS - 1
M1 - 14288
ER -