Embodied Decision-Making: Motor Simulation and Risk Preference

Introduction

Recent work in embodied cognition has established that abstract reasoning engages the sensorimotor cortex (Glenberg & Kaschak, 2002; Kintsch, 2008). However, the extent to which this simulation process influences high-level decision-making remains unclear. Risk and uncertainty, fundamental to judgment and choice, involve mental models of future outcomes—simulations that may recruit motor representations for action planning. In the domain of gambling and financial decisions, this motor engagement could plausibly influence the subjective weighting of outcomes and probabilities.

We hypothesized that constraining motor simulation during decision-making would shift risk preferences toward more conservative choices. Specifically, we predicted that participants unable to engage hand movement during a risky choice task would downweight uncertain gains and overweight potential losses, consistent with prospect theory (Tversky & Kahneman, 1992). This prediction rests on the assumption that motor simulation of approach (reaching toward gains) versus avoidance (withdrawal from losses) constitutes part of the computational basis for risk evaluation.

Method

Participants

We recruited 289 participants (M age = 24.3 years, 58% female) from Amazon Mechanical Turk and local university participant pools across three experiments. Participants were screened for color blindness and non-native English speakers were excluded. Each study offered $6–8 compensation plus performance-contingent bonuses up to $3. No significant differences in demographics or baseline gambling propensity were observed across studies (F's < 1.2, p's > .15).

Procedure

In all three studies, participants completed a 100-trial gambling task in which they selected between a safe option (e.g., certain $5) and a risky option (e.g., 50% chance of $12, 50% chance of $0). Payoff structures were counterbalanced across trials. In the motor-constraint condition, participants placed their hands flat on a table with fingers extended and were instructed to maintain this posture throughout the task, monitored via webcam. In the control condition, participants' hands rested naturally. Before the main task, all participants completed a 20-trial practice block to familiarize themselves with the response interface (keyboard-based, with corresponding finger constraints in the motor-constraint group). Trials presented choice matrices on-screen for 5 seconds, followed by outcome feedback.

Study 2 additionally included a working memory span task (Operation Span; Turner & Engle, 1989) administered before the gambling task to test whether effects were mediated by cognitive load. Study 3 included a neutral-prime condition in which participants gestured freely before (but not during) the gambling task, to test specificity of motor constraint effects.

Results

A logistic mixed-effects model predicting risky choice (with random intercepts for participant) revealed a significant main effect of motor constraint condition (β = −0.43, SE = 0.15, z = −2.87, p = .004), aggregated across all three studies. Participants in the motor-constraint condition selected the risky option on 38% of trials (95% CI [33%, 43%]) versus 48% of trials in the control condition (95% CI [44%, 52%]), corresponding to a medium effect size (Cohen's d = 0.62).

In Study 2, working memory span did not significantly mediate this effect (indirect effect = −0.06, 95% CI [−0.19, 0.08]), suggesting motor constraint influenced risk preference independent of general cognitive load. In Study 3, the neutral-prime condition showed equivalent risk preference to the control condition (47% risky choices), indicating that the effect required concurrent motor constraint, not merely prior motor activation. Effect sizes remained consistent across all three studies (d's = 0.58–0.68), and a random-effects meta-analysis showed a robust aggregate effect (g = 0.62, 95% CI [0.39, 0.85]).

Discussion

These findings extend embodied cognition into financial decision-making, suggesting that the sensorimotor simulation of approach and avoidance contributes to risk weighting. The robustness across three independent samples and the specificity of the effect to concurrent (not prior) motor constraint support a causal interpretation. The independence from working memory span argues against a pure cognitive-load explanation; instead, the data suggest that motor simulation per se provides information to the decision-making system.

Future work should examine whether this effect generalizes to real-world financial domains and whether individual differences in motor imagery ability predict natural variation in risk preference. The clinical implications for populations with motor dysfunction (e.g., Parkinson's disease, stroke) warrant investigation. More broadly, these findings underscore the importance of considering embodied mechanisms when designing interventions to promote financial literacy or reduce risky decision-making in vulnerable populations.

References

• Glenberg, A. M., & Kaschak, M. P. (2002). Grounding language in action. Psychonomic Bulletin & Review, 9(3), 558–565.
• Kintsch, W. (2008). How the mind computes the meaning of metaphor: Imagined actions and embodied language. Cognitive Neurodynamics, 2(3), 179–191.
• Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task dependent? Journal of Memory and Language, 28(2), 127–154.
• Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and Uncertainty, 5(4), 297–323.
• Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monographs, 9(2), 1–27.
• Slovic, P. (2007). "If I look at the mass I will never act": Psychic numbing and genocide. Judgment and Decision Making, 2(2), 79–95.
• Critchley, H. D., Wiens, S., Rotshtein, P., Öhman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature Neuroscience, 7(2), 189–195.