Monkeys exhibit a paradoxical decrease in performance in high-stakes scenarios

Adam L. Smoulder, Nicolas Pavlovsky, Patrick J. Marino, Alan D. Degenhart, Nicole T. McClain, Aaron P. Batista, Steven M. ChaseProceedings of the National Academy of Sciences 2021 · DOI

TL;DR

Researchers have demonstrated that rhesus monkeys, like humans, exhibit "choking under pressure" – performing worse when potential rewards are exceptionally high. This finding suggests that the neural mechanisms underlying this paradoxical behavior might be conserved across species, opening doors for neurobiological investigations.

What Problem Does This Paper Address?

The phenomenon of "choking under pressure," where performance declines in high-stakes situations despite increased incentives, is well-documented in humans. While various psychological explanations exist, the underlying neural basis remains elusive, partly due to the lack of a suitable animal model. This paper aims to establish whether choking is a uniquely human trait or if it's conserved in other species, thereby enabling future neurobiological research into its mechanisms.

Methods

The researchers trained three rhesus monkeys to perform a "speed + accuracy" reaching task. In this task, monkeys controlled a cursor to reach a target, with the size of the potential reward (Small, Medium, Large, or Jackpot) cued at the start of each trial. The Jackpot reward was significantly larger and much rarer (5% of trials) than the others. To investigate the behavioral underpinnings of choking, they analyzed reach kinematics, specifically the "ballistic reach endpoint prediction" (where the reach would have landed without error correction) and "homing time" (the time spent approaching the target in the final stages of the movement). They also conducted control experiments to isolate the effects of reward rarity and magnitude.

Datasets

The study utilized data collected from three rhesus monkeys performing the described tasks. No external datasets were used for training or evaluation. The paper also makes the kinematic data and analysis code publicly available.

Key Insights and Ideas

  • Choking is not uniquely human: The most significant finding is the clear demonstration that rhesus monkeys exhibit choking under pressure, mirroring human behavior.
  • Rarity and magnitude are key: The paradoxical performance decrease was only induced when the reward was both rare and exceptionally large. Neither rarity nor magnitude alone was sufficient to cause choking.
  • Overcaution as a driver: Monkeys choked due to overly cautious reaching behavior in high-stakes (Jackpot) trials. This manifested as a shorter initial ballistic component of the reach and increased homing time, leading to more "undershoot" failures.
  • Trade-off between errors: When rewards were small, monkeys showed a tendency towards "overshoot" failures, suggesting less precise reaching when stakes were low.
  • Potential for conserved neural mechanisms: The cross-species demonstration of choking strongly implies that the underlying neural circuitry might be conserved, making monkeys a valuable model for future neurobiological studies.

Key Results

  • All three monkeys showed a significant decrease in success rate when transitioning from "Large" to "Jackpot" rewards. For example, Monkey N's success rate dropped by 25.2% for Jackpot rewards.
  • The "ballistic reach endpoint prediction" became shorter with increasing reward size, indicating less distance covered in the initial phase of the reach.
  • "Homing time" increased progressively with reward size, meaning monkeys spent more time fine-tuning their approach to the target.
  • The proportion of "undershoot" failures increased significantly from Large to Jackpot rewards, while "overshoot" failures were more common for Small rewards.
  • Control experiments showed that a rare but not exceptionally large reward ("Rare-Large") did not induce choking, nor did a large but common reward ("Common-Jackpot").

Limitations and Future Work

The authors acknowledge that while the "speed + accuracy" task showed consistent choking across all animals, other epochs of the task (delay, target-hold) exhibited more idiosyncratic behavioral changes across individuals. This suggests that while overcaution during the reach is a common factor, other psychological explanations for choking (e.g., impulsivity, distraction) might also contribute, varying between individuals.

Future work is clearly directed towards leveraging this new animal model to investigate the neural underpinnings of choking. The authors speculate that neuromodulatory systems, such as dopamine and norepinephrine, might play a crucial role, given their known involvement in reward processing and cognitive control, and their inverted-U relationships with performance. Invasive neural recordings in monkeys performing this task could provide critical insights into these mechanisms. Furthermore, understanding the factors that contribute to choking could inform the design of training regimens to help individuals mitigate this phenomenon.

Summary updated

Abstract

In high-stakes situations, people sometimes exhibit a frustrating phenomenon known as "choking under pressure." Usually, we perform better when the potential payoff is larger. However, once potential rewards get too high, performance paradoxically decreases-we "choke." Why do we choke under pressure? An animal model of choking would facilitate the investigation of its neural basis. However, it could be that choking is a uniquely human occurrence. To determine whether animals also choke, we trained three rhesus monkeys to perform a difficult reaching task in which they knew in advance the amount of reward to be given upon successful completion. Like humans, monkeys performed worse when potential rewards were exceptionally valuable. Failures that occurred at the highest level of reward were due to overly cautious reaching, in line with the psychological theory that explicit monitoring of behavior leads to choking. Our results demonstrate that choking under pressure is not unique to humans, and thus, its neural basis might be conserved across species.
BibTeX
@inproceedings{smoulder2021monkeys,
  author = {Adam L. Smoulder and Nicolas Pavlovsky and Patrick J. Marino and Alan D. Degenhart and Nicole T. McClain and Aaron P. Batista and Steven M. Chase},
  title = {{Monkeys exhibit a paradoxical decrease in performance in high-stakes scenarios}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2021},
  doi = {10.1073/pnas.2109643118},
}
Figure