Men are more prone than women to exhibit sunk cost bias—the tendency to persist with an investment despite its disadvantages—when exposed to romantic cues, according to new research published in the Journal of Consumer Research. Surprisingly this effect is not limited to romantic contexts but also extends to consumer behavior, suggesting that deep-seated evolutionary drives can subtly shape decision-making in various context.
The sunk cost bias refers to the tendency to persist with a decision or investment based on resources already spent, even when abandoning it might be the more rational choice. For example, someone might continue watching a movie they don’t enjoy simply because they’ve already invested an hour of their time. It is often viewed as irrational because the resources already invested (the “sunk costs”) cannot be recovered, and decisions should ideally be based on future outcomes rather than past expenditures.
Traditionally, explanations for this bias have focused on psychological and cognitive factors, such as the desire to avoid waste, fear of regret, or a need to justify prior decisions. However, some researchers suggest that sunk cost bias may not always be irrational. From an evolutionary perspective, committing to an investment might have been advantageous in contexts like mating and survival, where persistence and resource commitment signaled fitness and reliability to potential partners or ensured access to critical resources.
This evolutionary angle underpins the rationale for the study. The researchers proposed that sunk cost bias might serve an adaptive purpose in mating contexts, particularly for men, who historically adopted proactive and resource-intensive strategies to secure mates. They hypothesized that mating motives could trigger an implemental mindset—a focus on achieving specific goals—that heightens the tendency to stick with prior investments, even in unrelated domains like consumption.
To explore how mating motives influence sunk cost bias, the researchers conducted one field study and seven controlled experiments, six of which were pre-registered to ensure transparency.
The researchers began by exploring whether men and women differ in their susceptibility to sunk cost bias in a romantic context. They recruited 220 participants (114 women) from a Chinese online survey platform, with an average age of 30.7 years. Participants were randomly assigned to one of two conditions: a romantic relationship decision or a consumption decision.
In the romantic condition, participants imagined they had spent over a month chatting with a potential dating partner, sent virtual gifts, and arranged an in-person meeting. Before meeting, they learned of a more attractive and better-matched potential partner. They had to choose whether to stick with the original partner or meet the new one. In the consumption condition, participants imagined they had paid a non-refundable deposit for shoes but found the same shoes cheaper elsewhere. They decided whether to buy the shoes at the original store or switch to the cheaper option.
The goal was to see if men exhibited stronger sunk cost bias than women in romantic decisions but not in consumption decisions. This design controlled for potential confounding factors, like perceived morality, which was assessed using participant ratings. This study established that men were more likely than women to adhere to sunk costs in a mating scenario, setting the stage for broader investigations.
Building on Study 1, the researchers examined whether exposure to mating cues influenced sunk cost bias in unrelated domains, such as investment and consumption decisions. The sample consisted of 231 heterosexual participants (150 women), with an average age of 25.3 years, recruited from the United States via Prolific.
Participants were randomly assigned to one of two conditions: a mating cue condition or a control condition. In the mating cue condition, participants rated the attractiveness of images of opposite-sex individuals and imagined going on a romantic date with the most appealing person. In the control condition, participants rated images of apartments and imagined discussing rental details with the owner.
Afterward, participants were asked to make decisions in two scenarios. In the investment scenario, they had to decide whether to continue funding a plane project that was mostly complete but now outcompeted by a superior alternative. In the consumption scenario, they decided between attending a concert for which they had a $100 ticket or a preferred concert with a $50 ticket. Mating cues significantly heightened sunk cost bias among men in both scenarios but had no effect on women, reinforcing the role of mating motives.
The researchers designed Studies 3a and 3b to differentiate sunk cost bias, which involves prior investment, from status quo bias, which does not. Study 3a involved 420 Chinese university students (233 women, average age 23.1 years) recruited from a survey platform. Study 3b involved 561 participants (301 women, average age 31.1 years) from the Credamo platform.
In Study 3a, participants completed a task to earn a lottery opportunity (prior investment condition) or were given the lottery directly (no prior investment condition). Mating cues were manipulated similarly to Study 2. Participants then decided whether to switch to a lottery with better odds. Results showed that men’s sunk cost bias increased in the prior investment condition when exposed to mating cues, while no such effect was found in the no-investment condition.
Study 3b used a restaurant reservation scenario. In the prior investment condition, participants had paid a deposit; in the no-investment condition, they had made a free reservation. As in Study 3a, mating cues heightened sunk cost bias for men in the prior investment condition but had no effect in the no-investment condition, confirming the distinction between sunk cost and status quo biases.
To explore the mechanism behind mating cues and sunk cost bias, the researchers conducted Studies 4a and 4b, recruiting 539 participants (301 women, average age 30.5 years) and 551 participants (302 women, average age 30.6 years), respectively, from Credamo.
In Study 4a, participants described an upcoming event they were planning and completed a mating cue manipulation as in earlier studies. An implemental mindset was measured with items assessing participants’ focus on goal-directed behaviors (e.g., planning actions). Results showed that mating cues heightened an implemental mindset in men but not in women.
In Study 4b, participants’ implemental mindset was manipulated. Those in the strengthened condition listed steps to address a hypothetical problem, while those in the control condition described the problem. All participants then made a sunk cost decision about continuing a plane project. Results showed that an implemental mindset increased sunk cost bias for both men and women, supporting its role as a mediator.
Study 5 used a process-by-moderation approach to confirm the role of implemental mindset in linking mating cues and sunk cost bias. The sample included 572 participants (305 women, average age 31.8 years) recruited from Sojump. Participants were exposed to mating or control cues and then completed a task to manipulate their implemental mindset (e.g., choosing between two laptops to activate goal-oriented thinking).
Participants then made a sunk cost decision about continuing to watch a boring movie after investing an hour. In the control condition, mating cues increased men’s sunk cost bias. However, in the strengthened implemental mindset condition, mating cues had no effect, as all participants exhibited heightened sunk cost bias. This result confirmed that the implemental mindset drives the observed effects.
The final study applied these findings in a real-world context. The researchers collaborated with a university coffee shop, recruiting 240 Chinese students (120 women) over seven days. Participants were offered paid membership cards (5 RMB) with a 20% discount on coffee for two weeks. The cards featured either a romantic design (mating cue) or a neutral design (control).
The researchers tracked card usage and conducted follow-up surveys to measure participants’ motivations and behaviors. Results showed that men exposed to romantic cards used their memberships more frequently, demonstrating stronger sunk cost bias when mating cues were present. Women’s behavior was unaffected by the romantic design, highlighting the gender-specific effects of mating motives on investment persistence.
Together, the findings provide evidence that mating motives can significantly influence decision-making by activating an implemental mindset, particularly among men. This mindset drives a stronger commitment to prior investments, manifesting as sunk cost bias, even in contexts unrelated to mating.
The research highlights fundamental gender differences rooted in evolutionary mating strategies, where men are more likely to exhibit goal-directed persistence in response to mating cues. These effects extend beyond romantic scenarios into consumer behavior, showing how deep-seated psychological mechanisms shape everyday choices.
The findings also have real-world implications for marketing and consumer behavior, the researchers said. For instance, marketers could use romantic cues to boost men’s commitment to loyalty programs or products where upfront investments are required. However, the researchers added that policymakers and consumers should remain vigilant about how such tactics might exploit psychological biases, leading to irrational financial decisions.
The study, “He Loves the One He Has Invested In: The Effects of Mating Cues on Men’s and Women’s Sunk Cost Bias,” was authored by Rui Chen, Hao Sun, Zhaoyang Guo, and Haipeng (Allan) Chen.
