Picture a young investor in Jakarta scrolling through Instagram late at night. A favorite content creator posts an excited prediction about a coin that is “about to explode.” Friends in a group chat are already buying. Within minutes, our investor moves money into the trade, skipping the slow work of research. What pushed that decision? A team of researchers set out to map the chain of psychological and social forces that sits behind moments exactly like this one.
In a study published in the Journal of Risk and Financial Management, Bambang Leo Handoko of Bina Nusantara University in Jakarta and three colleagues examined how personality traits, social pressure, and the credibility of online influencers connect to the choices people make in cryptocurrency markets. Their central finding is that these outside forces rarely act directly. Instead, they tend to work through mental shortcuts and crowd-following behavior, which in turn shape how much risk a person is willing to accept.
The puzzle the researchers wanted to solve
Cryptocurrency markets are famous for wild swings. The authors point to the November 2022 collapse of the FTX exchange, which helped drive Bitcoin’s price down to around $17,200, and to the U.S. approval of Bitcoin exchange-traded funds in January 2024, which pushed prices above $49,000. With prices this jumpy and regulation this thin, the researchers argue that purely rational financial models struggle to explain what investors actually do.
Earlier behavioral finance research often treated two patterns as direct causes of investment choices. The first is heuristic bias, which means relying on mental shortcuts or rules of thumb rather than careful analysis. The second is herding behavior, the tendency to copy what other people are doing instead of evaluating an asset on its own merits. The authors wanted to look a step further back and ask what produces these patterns in the first place, and how they ultimately translate into action.
To organize the investigation, they borrowed a framework from psychology called Stimulus-Organism-Response. The idea is simple: an outside trigger (the stimulus) sparks an internal mental state (the organism), which then produces a behavior (the response). In this study, personality traits and social cues serve as the stimulus. Heuristic bias and herding serve as the internal state. And the final cryptocurrency investment decision is the response. Sitting in the middle is risk tolerance, a person’s willingness to accept uncertainty and possible loss.
How the study was built
The team surveyed 367 retail cryptocurrency investors in Indonesia through an online questionnaire. To qualify, participants had to be active crypto investors, follow at least one crypto influencer, and have seen influencer content within the previous three months. The sample skewed young and relatively new to investing: about 35 percent were aged 21 to 30, students made up the largest occupational group, and most had been investing for two years or less. Nearly two-thirds put less than 10 percent of their monthly income into crypto.
The survey measured several personality traits drawn from established psychology research. Openness reflects curiosity and a taste for new ideas. Extraversion captures sociability and excitement-seeking. Conscientiousness describes self-discipline and organized, goal-directed behavior. The questionnaire also gauged influencer credibility (built from expertise, trustworthiness, attractiveness, and similarity to the viewer) and social influence from peers and family. Everything was rated on a five-point agreement scale.
To analyze the responses, the researchers used a statistical method called structural equation modeling, which estimates how a web of variables relate to one another at the same time. They also ran checks to confirm their survey items were measuring distinct concepts and were not simply picking up the same idea under different labels.
What the analysis revealed
All twelve of the study’s proposed relationships found statistical support, though their strengths varied considerably.
On the personality side, the three traits were all linked to higher heuristic bias, but not equally. Extraversion showed the strongest connection. The authors suggest that socially active, excitement-seeking people are exposed to more market chatter and influencer content, which may make them quicker to lean on mental shortcuts. Conscientiousness showed a more moderate link, which the researchers interpret as a steadying influence even when disciplined investors face time pressure. Openness, by contrast, was statistically significant but so small that the authors describe its practical effect as negligible. In their words, openness “does not meaningfully translate into biased decision-making in cryptocurrency investment.”
The social side told a clearer story. Both influencer credibility and social influence were linked to more herding behavior. When investors saw an influencer as believable, they were more likely to imitate that person’s moves. Pressure from friends, family, and online communities pulled in the same direction, nudging people to follow the crowd rather than run their own analysis.
From there, the study traced how these internal states connected to action. Heuristic bias had a strong link to risk tolerance and a moderate link to investment decisions. In other words, people who lean heavily on mental shortcuts tended to feel more comfortable taking risks and were more likely to invest. Herding behavior was also linked to both risk tolerance and investment decisions, but its direct effect on decisions was small. The authors read this as a sign that individual cognitive shortcuts shape crypto choices more powerfully than social copying alone.
Risk tolerance turned out to be a key connector. It was linked to investment decisions on its own, and it also served as the channel through which heuristic bias and herding flowed into action. The researchers describe risk tolerance as a “serial mediating mechanism,” meaning the biases appear to raise a person’s willingness to accept risk, and that raised willingness then shows up in the decision to invest. Together, the model accounted for about 62 percent of the variation in cryptocurrency investment decisions.
One detail stands out. The authors note that traditional thinking often treats risk tolerance as a fixed personality trait. Their results suggest it may be more changeable than that, shifting in response to mental shortcuts and social herding, especially under uncertainty. To check this, they tested an alternative version of the model in which risk tolerance was treated as something that strengthens or weakens other relationships, rather than as a channel. That version did not hold up statistically, which they take as support for their interpretation of risk tolerance as a bridge rather than a fixed dial.
Practical takeaways and important limits
The researchers draw out several suggestions. Because mental shortcuts showed the strongest pull, they argue that exchanges and fintech platforms could build in decision-support tools such as risk warnings, volatility alerts, and prompts that encourage investors to pause before impulsive trades. Given the influence of social media personalities, they suggest that regulators, including Indonesia’s Bappebti and Financial Services Authority, could require clearer disclosure of paid endorsements and standardized risk statements. For investors themselves, the authors recommend setting risk limits in advance and avoiding heavy reliance on social cues.
Several caveats deserve attention. This was a cross-sectional survey, meaning it captured a single snapshot in time. The authors are explicit that the findings should be read as associations rather than proof that one factor causes another. The sample was concentrated among young, less-experienced Indonesian investors, so the patterns may not extend to seasoned or institutional traders or to other countries. The researchers also note that data collection happened around the buzz of the Bitcoin ETF approval, a period of strong optimism that may have amplified socially driven behavior. Market conditions like bull and bear cycles were not built into the model, which they flag as a target for future work.
