Differential effects of real versus hypothetical monetary reward magnitude on risk-taking behavior and brain activity

Abstract

Human decisions are more easily affected by a larger amount of money than a smaller one. Although numerous studies have used hypothetical money as incentives to motivate human behavior, the validity of hypothetical versus real monetary rewards remains controversial. In the present study, we used event-related potential (ERP) with the balloon analogue risk task to investigate how magnitudes of real and hypothetical monetary rewards modulate risk-taking behavior and feedback-related negativity (FRN). Behavioral data showed that participants were more risk averse after negative feedback with increased magnitude of real monetary rewards, while no behavior differences were observed between large and small hypothetical monetary rewards. Similarly, ERP data showed a larger FRN in response to negative feedback during risk taking with large compared to small real monetary rewards, while no FRN differences were observed between large and small hypothetical monetary rewards. Moreover, FRN amplitude differences correlated with risk-taking behavior changes from small to large real monetary rewards, while such correlation was not observed for hypothetical monetary rewards. These findings suggest that the magnitudes of real and hypothetical monetary rewards have differential effects on risk-taking behavior and brain activity. Real and hypothetical money incentives may have different validity for modulating human decisions.

Paying out one versus paying out all trials and the decrease in behavioral and brain activity in the Balloon Analogue Risk Task

“…We used real losses after the balloon explosion instead of the standard “no gain” to improve the ecological validity of decision‐making during the BART. We used this type of modification successfully in our previous ERP studies of decision‐making under uncertainty (Xu et al, , ). The experiment lasted about 20–25 min for each condition.…”

“…Quantification of the FRN is still a controversial issue. In numerous studies, the FRN was measured by the mean amplitudes within the fixed time windows (e.g., Tian et al, ; Xu et al, , ). However, because the FRN overlaps with the P200 and P300 (e.g., Foti, Weinberg, Dien, & Hajcak, ; Yeung, Holroyd, & Cohen, ), any variance in these two components might be attributed to the FRN, especially when these two components are influenced by the same experimental factors as the FRN.…”

“…Similarly, Yau et al () observed an attenuated FRN and P300 amplitude in problematic internet‐using adolescents, which was thought to reflect their decreased sensitivity to external feedback and their impaired ability to assign attention to motivationally salient outcomes. Our recent ERP studies demonstrated a larger FRN in response to money loss during decision‐making with real rewards compared to those with hypothetical rewards—suggesting that real and hypothetical monetary rewards have differential effects on decision‐making and brain activity (Xu et al, , ).…”

“…Therefore, although each additional pump increases the reward associated with a particular balloon, it also increases the risk that the balloon will explode, and all accumulated money will be lost. Uncertainty aversion in the BART is quantified by the average number of pumps excluding balloons that popped (Lejuez et al, ), and the effect of the previous trial's outcome on the current decision‐making behavior is assessed by comparing the average number of pumps of the unexploded balloon when the previous decision was successful or failed (Xu et al, , ; Xu, Xiao, & Rao, ).…”

“…The BART has become one of the most widely used performance measures of decision‐making behavior under uncertainty (Harrison, Young, Butow, Salkeld, & Solomon, ); it has also been shown to relate to real‐world risk‐taking behaviors (e.g., smoking, drug use, and unprotected sex) across a broad range of studies (Aklin, Lejuez, Zvolensky, Kahler, & Gwadz, ; Bornovalova, Gwadz, Kahler, Aklin, & Lejuez, ; Buelow & Blaine, ; Lejuez et al, ). Researchers have also recently explored ERPs during the completion of the BART to evaluate the neural activity underlying risk‐taking and decision‐making in various at‐risk groups (Euser, van Meel, Snelleman, & Franken, ; Fein & Chang, ; Xu et al, , ; Yau, Potenza, Mayes, & Crowley, ). For example, Fein and Chang reported smaller FRN amplitudes and larger P300 amplitudes in response to negative feedback during the BART in treatment‐naive alcoholics with a greater family history density of alcohol problems.…”