Betrayal is worse than loss during cooperation

By integrating computational modeling and event-related potential (ERP) techniques, this study reveals that the emotional aversion to betrayal exerts a stronger influence on reducing cooperation than the financial loss itself, with neural data indicating that betrayal is processed earlier in decision-making than loss.

The Gist

The Big Question: Is it the Money or the Heartbreak?

Imagine you are playing a game where you have to decide whether to trust a partner.

• Scenario A (The Risk): You put your money in a slot machine. There's a chance you double it, and a chance you lose half. The machine is just a machine; it doesn't hate you.
• Scenario B (The Betrayal): You put your money in a "trust jar" with a friend. If they are nice, you both double your money. If they are mean, they take half your money and keep it.

The big question researchers asked is: When people refuse to play, is it because they are scared of losing the money (Loss Aversion), or because they are terrified of being stabbed in the back by a friend (Betrayal Aversion)?

The short answer? The heartbreak hurts way more than the empty wallet.

---

The Experiment: The "Trust vs. Luck" Game

The researchers set up a lab experiment with university students. They gave them two types of tasks that looked almost identical on the surface but felt very different in the gut.

1. The "Slot Machine" Task (Risk): Participants could gamble their money against a computer. The odds were clear.
2. The "Trust" Task (Cooperation): Participants could gamble their money against a "partner" (who was actually a computer pretending to be a person). The odds were the same, but the source of the risk was a human.

The Results:
People were much more cautious in the "Trust" task. Even when the odds of winning were great, they hesitated more to play with a "person" than with a "machine." They felt the potential for betrayal was a heavier weight than the potential for financial loss.

The Secret Weapon: Reading Minds with "Brain Waves"

To understand why this happens, the researchers didn't just ask people what they thought; they looked at their brains using EEG (brain waves). Think of this as a high-speed camera for thoughts. They were looking for two specific "flashes" of brain activity:

1. The P3 Flash (The "Gut Check"): This happens very fast (about 300-500 milliseconds). It's like your brain's initial "Whoa, that's interesting!" reaction.
2. The LPP Flash (The "Deep Dive"): This happens a bit later (600-800 milliseconds). It's like your brain sitting down to think, "Okay, let's really analyze the pros and cons."

What they found:

• Betrayal hits the "Gut Check" first: When the brain sensed the possibility of being betrayed by a human, the P3 flash changed immediately. The brain flagged the social risk instantly. It's like your alarm system going off the moment you see a stranger walking toward your car.
• Loss hits the "Deep Dive" later: The fear of losing money (financial loss) didn't show up in the fast "Gut Check." It only appeared in the slower LPP flash. This means the brain processes the emotional sting of betrayal before it even finishes calculating the math of the financial loss.

The Computer Model: The "Emotion vs. Math" Equation

The researchers built a computer model to simulate how people make these choices. They tried to write a formula for human behavior.

• Old Idea: People just calculate: Is the money worth the risk?
• New Discovery: People calculate: Is the money worth the risk? PLUS, how much does it hurt my feelings if I get betrayed?

The model showed that the "Emotional Hurt" factor (Betrayal Aversion) was a much stronger predictor of whether someone would say "No" than the "Financial Loss" factor. In fact, the emotional cost of betrayal was so heavy that it suppressed cooperation much more than the fear of losing money did.

The Takeaway: Why We Are Human

This study tells us that humans are wired to prioritize social safety over financial safety.

• The Analogy: Imagine you are walking down a dark alley.
  • Loss Aversion is worrying you might trip and scrape your knee (financial loss).
  • Betrayal Aversion is worrying a stranger might jump out and punch you (social betrayal).
  • Your brain screams "STOP!" at the thought of the punch long before it even thinks about the scraped knee.

In Conclusion:
When we decide whether to cooperate, we aren't just doing math. We are doing "emotional math." The fear of being betrayed by a friend is a primal, fast, and powerful force that stops us from cooperating much faster and harder than the fear of losing a few dollars. Our brains treat a broken promise as a much bigger emergency than a broken wallet.

Technical Summary

1. Problem Statement

Cooperative behavior is fundamental to human society, yet the mechanisms driving it remain complex. While loss aversion (the tendency to prefer avoiding losses over acquiring equivalent gains) and betrayal aversion (the emotional cost of being betrayed by another person) are known to influence cooperation, their relative potency and distinct neural mechanisms are not fully understood.

• The Gap: Previous research often treated betrayal and financial loss as separate phenomena or examined them in distinct contexts (social vs. non-social risks). It remains unclear whether the reluctance to cooperate stems primarily from the psychological impact of betrayal itself or the direct financial loss it causes.
• The Question: Does betrayal aversion exert a stronger suppressive effect on cooperation than loss aversion, and how are these two processes temporally and neurally dissociated?

2. Methodology

The study employed a multimodal approach combining computational modeling and event-related potentials (ERPs) to dissect the cognitive and neural dynamics of cooperation.

Experimental Design

• Participants: 49 right-handed university students (final sample after exclusions).
• Tasks: Two decision-making tasks were administered with identical probabilities (10%–90%) and principal amounts (¥0.2–¥1.0):
  1. Risk Task (Financial Risk): Participants chose between a safe option (keeping the principal) and a risky option (doubling the principal with probability $p$ or losing half with probability $1-p$). Outcomes were determined by chance.
  2. Cooperation Task (Social Risk): Participants chose to cooperate with an anonymous partner. If successful, both doubled their money. If the partner "defected" (betrayed), the participant lost half their principal, which was transferred to the partner.
• Procedure: Participants performed both tasks while EEG was recorded. Post-task ratings of happiness/unhappiness were collected for positive and negative outcomes.

Computational Modeling

• Framework: Based on Cumulative Prospect Theory (CPT).
• Model Comparison: Seven models (M0–M6) were tested to determine the best fit for decision-making data:
  • M1: Standard CPT (Gain/Loss utility).
  • M2: CPT + Betrayal Aversion ($\kappa$) as an additive emotional cost.
  • M3: CPT + Betrayal Aversion + Prosocial Utility ($\omega$) (valuing the partner's gain).
  • M4–M6: Variants of M1–M3 with task-specific parameters.
• Selection: Model 3 (incorporating CPT, betrayal aversion, and prosocial utility) was selected as the winning model based on the Akaike Information Criterion (AIC).

Neural Recording (ERP)

• Equipment: 64-channel EEG system.
• Components Analyzed:
  • P3 (350–500 ms): Associated with rapid encoding of motivational significance and intuitive processing.
  • LPP (600–800 ms): Associated with sustained, elaborative emotional and motivational processing.
• Analysis: Linear mixed-effects regression was used to correlate the model-derived parameters (betrayal aversion $\kappa$ and loss aversion $\lambda$) with P3 and LPP amplitudes.

3. Key Contributions

1. Dissociation of Aversions: The study successfully separated the effects of betrayal aversion from loss aversion within the same cooperative context using computational modeling, demonstrating that betrayal has a unique, stronger inhibitory effect on cooperation than financial loss alone.
2. Temporal Sequencing: It provides the first evidence that betrayal and loss are processed sequentially in the brain: betrayal is encoded early (intuitive stage), while loss is integrated later (evaluative stage).
3. Neural Markers: Identified specific ERP signatures for these aversions:
   • P3: Sensitive to betrayal aversion (intuitive prosocial motivation).
   • LPP: Sensitive to both betrayal and loss aversion (elaborative evaluation).

4. Key Results

Behavioral Findings

• Risk Aversion: Participants were significantly less willing to take risks in the cooperation task compared to the risk task, even when probabilities and payoffs were identical.
• Emotional Impact: Cooperation tasks elicited greater happiness for gains and greater unhappiness for losses compared to the risk task, confirming the added emotional cost of social interaction.
• Decision Time: Decisions were faster in the cooperation task but were driven more by principal magnitude than probability, suggesting a cautious, heuristic-based approach to social risk.

Computational Modeling Results

• Betrayal vs. Loss: Both loss aversion ($\lambda$) and betrayal aversion ($\kappa$) negatively predicted cooperation willingness. However, betrayal aversion had a significantly stronger predictive power ($z = -6.97, p < 0.001$) than loss aversion.
• Individual Differences: Higher prosocial preference (valuing others' benefits) correlated with higher betrayal aversion but did not correlate with loss aversion.

Neural (ERP) Findings

• P3 Component (Early Stage):
  • Amplitude decreased as betrayal aversion increased (betrayal inhibition effect).
  • Amplitude was unaffected by loss aversion.
  • Interpretation: Betrayal is detected early as a threat to prosocial motivation, suppressing the intuitive drive to cooperate.
• LPP Component (Late Stage):
  • Amplitude decreased as both betrayal aversion and loss aversion increased.
  • Interpretation: During the later, deliberative phase, the brain integrates both the emotional cost of betrayal and the financial cost of loss to make a final decision.

5. Significance

• Theoretical Advancement: The findings challenge the view that financial loss is the primary driver of risk aversion in social contexts. Instead, the emotional cost of betrayal is the dominant factor suppressing cooperation.
• Mechanistic Insight: The study supports a dual-process framework in social decision-making:
  1. Fast/Intuitive: Betrayal is flagged immediately (P3), acting as a "stop signal" for cooperation.
  2. Slow/Reflective: Financial loss is weighed later (LPP) alongside the emotional cost.
• Practical Implications: Understanding that betrayal triggers a distinct, early neural response suggests that interventions to promote cooperation (e.g., in economics or organizational behavior) must address the trust and intention aspects of interaction, not just the financial incentives.
• Methodological Rigor: By combining computational modeling with high-temporal-resolution EEG, the study overcomes the limitations of fMRI in distinguishing closely timed cognitive processes in social decision-making.

In conclusion, the paper establishes that betrayal is psychologically and neurally "worse" than loss, acting as a primary, early-occurring barrier to human cooperation.