In the information age, people are facing an increasing demand to concentrate and perform a variety of daily tasks and challenges that require attention (Buschman and Miller, 2007). However, research shows that attentional resources are limited (Kahneman and Tversky, 1973; Falkinger, 2008). Economics researchers have examined the concept of limited attention in several contexts, such as Friday earnings announcements (DellaVigna and Pollet, 2009), marital events (Lu et al., 2016), and large jackpot lotteries (Huang et al., 2019). In present times, the excessive and problematic use of mobile phones has been widely documented.¹ It is worth noting that mobile phones compete for limited attention through the extensive mobile applications (apps) of social networks, videos, games, and other entertainment types. Beneito and Vicente-Chirivella (2022) discovered that banning mobile phones in schools had a significant positive impact on student performance in Spain. The use of mobile phones in the classroom, despite its potential to facilitate structured learning, can also lead to distractions that negatively impact academic performance. Brown et al. (2020) used the Blackberry Internet Service outage and discovered that a higher level of information gathering and increased trading activity were associated with greater views or downloads of disclosure filings; these findings suggest that the mobile Internet has the potential to divert the attention of stock market activities. However, the extent to which entertainment apps distract individuals and hinder their decision-making abilities remains unclear. In this study, we addressed these questions by examining the number of entertainment apps installed by investors and their online peer-to-peer (P2P) investment performance.

A growing body of evidence from psychology and health research shows that the use of mobile phones has a detrimental impact on the cognitive attention required to focus on daily tasks effectively (Anderson et al., 2017; Grewal et al., 2018). Moreover, entertainment apps are among the most concerning sources of distraction (Brooks, 2015; Swar and Hameed, 2017; Chu et al., 2021). Wu et al. (2018) found that the fear of missing out drives people to constantly speculate about what their families and friends are doing on social networks and the latest popular short videos; additionally, a sense of excitement and achievement linked to passing higher levels encourages people to play mobile games. Thornton et al. (2014) postulated that simply having a visually noticeable mobile phone can evoke a sense of being excluded from a “broad social and informational network... that one is not part of at the moment,” even when people are not using their phones. Wilmer et al. (2017) claimed that smartphones can negatively impact focused and sustained attention and diminish the capacity for achieving a state of flow (Csikszentmihalyi and Larson, 2014). Extending these findings to the investment context, we suggest that the attention-diverting effects of entertainment apps may result in investors making suboptimal investment decisions.

By using the number of installed mobile entertainment apps of P2P investors to measure attention distraction, our study found that investors with a greater number of such apps have a significantly higher likelihood of default and a lower internal rate of return, even after controlling for loans, borrowers, and investor-level characteristics. We were fully aware that there could be potential omitted control variables and that a reverse causality relationship may exist when downloading more entertainment apps for leisure or in response to poor investment outcomes. To address these concerns, we used a time-varying difference-in-difference analysis based on the entertainment company's seven large-scale national Internet service outages. We found that diminishing attentional distraction shock resulted in improved performance. We also used the Baidu search index of entertainment-related app keywords in the investor's city and the average number of entertainment apps downloaded among people of the same age and gender as the investor as instrumental variables (IV), and we found that the results remained robust.

To further assess the attention-distraction effect of entertainment apps, we differentiated between Friday and other working days and high- and low-speed Internet regions. DellaVigna and Pollet (2009) discovered that Friday distracted investors from job-related tasks. Since entertainment-related apps facilitate chatting with families and friends to schedule upcoming weekends and searching for new movies and videos to enjoy at weekends, they could cause a more pronounced attention-limitation effect on Friday compared with Monday to Thursday. In addition, a smooth mobile network is essential for a better user experience, especially for entertainment-related apps that comprise a large number of videos, pictures, and real-time communication. We discovered that the impact of entertainment-related apps on limited attention is more pronounced in high-speed Internet regions because of the improved user experience. We also found consistent evidence to support the identification validity of using the number of entertainment apps as an indicator of different levels of attention distraction among P2P investors.

Regarding the underlying mechanism of poor performance as a result of entertainment apps, research on economics and finance has shown that, if attention is limited, some information has to be ignored (Peng and Xiong, 2006; Choi and Choi, 2019; Wang and Song, 2022; Zhong, 2022). Moreover, people may rely on simple mental shortcuts to process information when their attention is limited. Barber and Odean (2008) discovered that attention-grabbing stocks tend to be bought in, and Lacetera et al. (2012) showed the existence of a “left-digit” bias. Iscenko (2020) found that borrowers exposed to higher levels of distraction tend to direct their attention toward lenders with whom they have an existing personal relationship. Using data from a P2P lending market, we explored the potential mechanisms underlying investment behavior from the perspectives of information-processing ability and familiarity heuristics. The results showed that investors with higher levels of distraction, indicated by more entertainment apps, exhibit worse investment performance when faced with more readily available information and are familiar with the borrowers.

Our study adds to the literature in many ways. First, it extends the examination of the attention-distraction effect at the individual investor level. Economics researchers have used various methods to identify limited attention and have incorporated potential distractions to understand how attention is allocated to a particular event (Lu et al., 2016; Campbell et al., 2019). However, existing research mainly focuses on market-wide attention distraction and professional institutional investors. Ben-Rephael et al. (2017) pointed out that the decision to focus on retail investors, institutional investors, or both should be empirically determined because of their different roles in the finance market. Bajo et al. (2021) studied the distraction effect of birthdays and found a decrease in trading activity in a three-day window. In this study, we showed that mobile entertainment apps can cause investor attention to shift away from investment, leading to a decrease in investment performance.

Second, our study also contributes to the understanding of the underlying mechanism of inferior decision-making caused by limited attention. Previous theoretical models suggested that limited attention can explain category learning (Peng and Xiong, 2006) and style investing (Barberis and Shleifer, 2003; Wahal and Yavuz, 2013) and have important implications for other asset pricing. In this study, we attempted to empirically test whether investors neglect partial information and use heuristics when their attention is limited. Our study echoes previous studies on how limited attention affects the incorporation of information into the decision-making process. There is substantial evidence that people tend to use simple heuristics when processing information in tasks such as categorization (Mullainathan, 2002; Barberis and Shleifer, 2003), simple forecasting (Hong et al., 2007), and others (Busse et al., 2013; DeHaan et al., 2015).

Third, the study contributes to this literature by demonstrating that mobile entertainment apps can negatively impact productivity. Previous studies showed that the use of mobile phones makes it difficult to perform daily routine activities (Liebherr et al., 2020; Chu et al., 2021). In contrast, some studies showed that smartphone apps are positively used for education and health monitoring.

This study focused on entertainment apps, which are widely acknowledged as attention-diverting and widely used (Bayer et al., 2016; Long et al., 2016). In this study, we expanded the discussion to investment decisions and proposed that the attention distraction caused by entertainment apps results in poor investment decisions. To the best of our knowledge, the potential adverse effect of entertainment apps has not been empirically tested in real-life decision-making with real financial consequences.

We conducted several robustness tests (see Table 8). We first examined the window period of half a year and 3 months after the investors registered on the platform. The app information was recorded at the time of registration in 2015 and was not updated. This resulted in measurement noise for the recorded data for a substantial amount of time after 2015. Therefore, to mitigate the effect of changes in the number of mobile apps, we restricted the sample to half a year and 3 months after registration. Despite the potential measurement noise being excluded, the effect remained significant (columns 1–4). We then examined any possible influences resulting from the empirical methods used. We used a linear probability model to test the relationship between Ln (#Enter App) and Default; the coefficient remained positive and significant (column 5). To determine if the effect was due to the total number of installed apps, we also included the natural logarithmic value of the total number of apps as a control in the baseline regression Ln (#Total App). The coefficients for the primary variable showed little change. In contrast, the Ln (#Total App) coefficient was not significant, suggesting that attention is mainly affected by specific types of apps (e.g., entertainment apps) instead of all kinds (Columns 6 and 7).

Psychologists and behaviorists have recently become interested in the widespread possession of mobile phones, demonstrating that it is crucial in understanding people's emotions, behaviors, and psychological wellbeing. Our study delves into this area by investigating the impact of attention-distracting entertainment apps on financial market investment performance. By analyzing real financial market investment data, we discovered that limited attention resulting from the use of entertainment-related apps is a key determinant of poor investment performance. We found that limited attention increases the likelihood of defaults and lower returns. To further prove the impact of limited attention, we used an exogenous natural experiment of entertainment service unavailability and found that outage can enhance investment performance. Moreover, results from the instrumental variable were also consistent. The mechanism analysis showed that lower-performance bidding behaviors were subject to information neglect and familiarity heuristics. The evidence is consistent with previous findings that limited attention has adverse effects and leads to cognitive shortcuts, thereby uncovering the potential underlying processes behind inattentive thinking.

Our study provides insights into economic agents' cognitive behaviors when they process information, especially when their cognitive resources are constrained. Although we do not delve into all the mental processes involved in economic decision-making under limited attention, future research should further explore this area thoroughly. Incorporating the cognitive effects of mobile phones into both theoretical and empirical analyses of financial markets would be a promising direction for future research. In addition, comprehensive data on individual mobile phone usage will greatly improve the research framework.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

YZ and YL contributed to conception and design of the study. YD organized the database. YZ and YD performed the statistical analysis. YZ wrote the first draft and revised draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.