Frustration is a complex human emotion (Broom, 1998) that significantly influences cognitive and behavioral processes (Jeon, 2015). Negative emotions such as frustration have been extensively studied in relation to traffic safety as contributing factors to aggressive driving behaviors. Drivers experiencing frustration are prone to demonstrating increased risk-taking and non-compliant behaviors, both of which contribute to heightened accident risk (Jeon, 2015; Bjureberg et al., 2023; Jeon et al., 2014; Shinar, 1998), and aggressive driving is a frequently associated factor in road accidents (Dula and Ballard, 2003; James and Nahl, 2000). Despite these associations, research into the specific triggers and mechanisms underlying frustration in driving contexts remains limited, partly due to the multifaceted nature of frustration and its overlapping dimensions with other negative emotions. Addressing these gaps is essential for developing targeted interventions to reduce aggressive driving and improve traffic safety.

Frustration is a common negative emotional response that arises when individuals encounter obstacles or hindrances while striving to achieve their goals (Dollard et al., 1939; Lazarus, 1994). It encompasses a spectrum of emotional responses, ranging from mild irritation to intense anger, and is influenced by both situational and individual factors (Dollard et al., 1939; Miller, 1941; Berkowitz, 1990). Frustration is also recognized as a multifaceted emotion (Bruun et al., 2016), with numerous factors contributing to its manifestatio, including personal traits (Costa and McCrae, 1999), cognitive appraisals (Lazarus and Folkman, 1984), self-regulation, and expectations (Carver and Scheier, 2001), with numerous factors contributing to its manifestatio. As a first step, it is necessary to classify clusters of factors influencing frustration that are specifically relevant to traffic safety.

Typically, driving is perceived as a straightforward journey to a destination, often idealized in advertisements as a pleasant experience. However, the reality frequently deviates from this ideal. According to Reason et al. (1990), driving is a goal-oriented activity that is essential for daily tasks such as commuting, taking children to school, and shopping, but it is often disrupted by various traffic obstructions. For instance, consider a situation of needing to reach work or attend an important doctor’s appointment. Despite leaving home with adequate time, unforeseen events—such as accidents, road construction, or navigation system malfunctions—can delay arrival. These disruptions create a gap between drivers’ expectations for the journey and their actual experiences, often leading to frustration.

Previous research indicates that frustration during driving is commonly attributed to factors such as traffic jams, congestion, accidents, construction, and delays at red lights during peak hours or slow-moving vehicles on rural roads. Frustration in driving, however, extends beyond such traffic-related issues. For example, Bosch et al. (2020) identified three main categories of driving situations that can lead to frustration: (a) traffic-related situations, such as construction sites, traffic jams, and parking limitations; (b) in-vehicle situations, including difficulties during driving preparation, challenges interfacing with the vehicle, and the social atmosphere within the car; and (c) self-inflicted factors, such as stress, impatience, or distraction; and (d) adverse weather conditions.

Notably, frustration differs from anger, as anger is directed at someone perceived as responsible for an undesirable event (Wirtz and Strohmer, 2017) and varies in both intensity and arousal (Grimm et al., 2007). However, persistent frustration may escalate into anger and lead to aggressive behavior (Shinar, 1998; Berkowitz, 1989). Aggressive driving, in turn, can be a symptom of frustrated (Shinar, 1998).

Furthermore, heightened stress levels are commonly observed in frustrated drivers (Jeon and Zhang, 2013), and frustration has been shown to impair the cognitive skills required for safe driving (Jeon, 2015; Lee, 2010). In the context of driving-related frustration, previous studies have emphasized that frustration can significantly impact both road safety and the user experience (Löcken et al., 2017; Oehl et al., 2019).

Furthermore, frustration can be experienced and expressed in various ways depending on individual differences. Personal traits and coping mechanisms significantly influence how frustration is perceived and managed, in addition to the specific nature of the driving-related events. Gross (2008) emphasized that individuals differ in their emotional, behavioral, and physiological responses, as well as in their subsequent emotional regulation. These differences can also be linked to other emotional responses to frustration (e.g., anger, irritation, disappointment), physiological responses (e.g., increased heart rate, elevated cortisol levels), and behavioral expressions (e.g., aggressive behavior, verbal outbursts, task disengagement).

Understanding the triggers of frustration and identifying driver profiles prone to frustration constitute essential knowledge for developing strategies and interventions to mitigate or prevent frustration in traffic. For example, the link between aberrant driving behaviors and frustration suggests that interventions aimed at curbing such behaviors may reduce frustration. Harris and Nass (2011) conducted a driving simulator study in which frustration was deliberately induced. Their findings indicated that a voice assistant, which reassured drivers by explaining that other drivers’ reactions were not intended to offend, led to improved driving performance and fewer negative emotions.

Similarly, using an empathetic voice assistant that responds to detected anger with messages like, “Hey, are you alright? I can understand that you are a bit angry, sometimes I feel the same way. How about some music to take your mind off things?” has been demonstrated to reduce negative emotions (Braun et al., 2019).

In conclusion, frustration encountered during driving can undermine road safety (Berkowitz, 1989; Lee, 2010). Addressing and reducing frustration is, therefore, crucial. However, the triggers of frustration are diverse and vary among individuals (Bosch et al., 2020; Ferreri and Mayhorn, 2020). Accordingly, this study investigates the multifaceted nature of frustration in various driving situations, employing Latent Profile Analysis (LPA) to explore the interaction between individual characteristics and frustration.

The LPA method is particularly advantageous for identifying and categorizing distinct subgroups within a population based on their responses to frustration. This approach facilitates a nuanced understanding of how individuals experience and react to frustrating scenarios. Our primary objective is to identify the factors that trigger drivers’ frustration and determine which subgroups of drivers are particularly susceptible. We examine three critical dimensions: (1) personality traits, (2) group use categories, and (3) emotional responses to frustrating scenarios. Accordingly, we aim to answer the following Research Questions (RQs):

We apply a person-centered analytical approach to categorize drivers into distinct profiles based on their responses to various frustration triggers and their associated emotional reactions. By focusing on drivers in Sweden, this study provide a comprehensive understanding of the factors contributing to frustration.

The results are presented sequentially to represent this study’s four primary LPA modeling steps. Prior to discussing the associated results, an explanation of the analytic process at each step is included.

The primary objective of this study was to identify and analyze latent profiles of drivers experiencing varying levels of frustration across different driving situations. Through Latent Profile Analysis (LPA), we identified four distinct profiles of frustration: Minimal, Low, Moderate, and Severe, representing different likelihood of frustration. The Severe profile, which comprises 11% of the sample, demonstrated the highest frustration levels across all driving situations, while the Minimal profile, representing 12%, showed consistently low frustration levels. Individuals in the Severe Profile were more likely to be older drivers (>45 years), more likely to be women, engaged in shared driving services, and highly associated with DBQ lapses. However, we observed similar patterns in both frustration situations and emotional responses to frustration across the four identified profiles, although with varying intensity. These findings support the idea that triggers and responses to frustration can be similar. However, differences in frustration experiences among individuals within each cluster may be mediated by demographic and behavioral factors. In this study, these factors included age, gender, behavioral dimensions, and user group categories, which emerged as key determinants of variations in frustration intensity across the profiles.

All identified profiles reveal both similar patterns and distinct levels of emotional responses in drivers’ frustration experiences across profiles. Boredom and unmotivated responses exhibited minimal variation across profiles, maintaining consistent intensity. However, individuals in the Severe profile experienced significantly higher levels of high-arousal emotional responses to frustration, with higher mean scores than other profiles. Conversely, the Minimal Profile exhibited resilience, with lower mean scores across all emotional responses. This supports the notion that higher likelihood of frustration arousal is linked to more intense emotional responses (Lazarus and Folkman, 1984). Interestingly, Shorkey and Crocker (1981) note that individuals respond to frustration in either adaptive or maladaptive ways. Adaptive responses are constructive, aiming to resolve the issue preventing goal achievement — either by avoiding the problem beforehand or employing problem — solving strategies once it arises. Based on the higher frustration levels observed in the Severe Profile, it is plausible to propose that the predominantly older drivers in this group may exhibit maladaptive responses to frustrating events. This pattern is consistent with theories of emotion regulation and coping. Managing emotions by controlling environmental triggers is a lifelong strategy that becomes more self-directed with age. As individuals gain control over their surroundings and emotional challenges, their ability to regulate emotions improves, especially in older adults (Thompson, 1991; Charles et al., 2001; Charles and Piazza, 2009). This finding is in contrast with our observation as the older drivers are more likely to be in the Severe profile, suggesting they are less able to manage frustrating situations.

Frustration levels were highest in scenarios involving Goal Achievement Obstructed (e.g., running late or encountering unexpected obstacles) and Unpleasant Experiences (e.g., noisy passengers or unpleasant appointments). Conversely, Limited Control (e.g., malfunctioning car systems or poor weather) and Road Obstruction (e.g., traffic jams) were associated with a significantly lower likelihood of frustration. Additionally, demographic variables and the DBQ lapses dimension were significant predictors of profile membership.

The higher likelihood of experiencing frustration in situations triggered by Goal Achievement Obstructed, such as “Driving behind a big truck and not being able to have full vision,” “Driving in a construction zone,” and “Being in a traffic jam,” aligns with previous studies indicating that the importance of the goal, along with the intensity of the desire to achieve it, have a strong connection to frustration (Dollard et al., 1939; Franz et al., 2020; Gross et al., 1997).

A closer examination of unpleasant experiences, such as “Driving when I am not happy about going to my destination (i.e., unpleasant appointment)” and “Driving when passengers are noisy,” suggests that baseline mood or other underlying factors may be associated with experiencing frustration. This finding is consistent with the idea that non-traffic-related factors, such as emotional state and interpersonal stress, play a role in driving frustration, especially for those in the Severe Profile. In these frustrating situations, neither the unpleasant appointment nor the noisy passengers directly induces frustration in the driver, as they are not related to the traffic situation or car interface, and we can consider them as the moderator factors in increasing the likelihood of frustration in drivers. Moreover, the lower level of association between Limited Control and Road Obstruction and experiencing frustration is consistent with theoretical models that emphasize frustration directly to a lack of control and decreased arousal (Donnerstein and Wilson, 1976) and identify traffic situations that are recognized as primary triggers for frustration in driving.

Additionally, car-sharing and leasing behaviors were positively associated with the Severe profile, suggesting that individuals engaging in these practices may experience a higher likelihood of frustration levels. This finding suggests that drivers who lease or share cars may have less tolerance for frustration, whereas the car owner user group is less strongly associated with a higher likelihood of frustration levels. While drivers who own their cars have a lower likelihood of experiencing frustration, this finding might support the notion of a direct association between the experience of frustration and contrast with expected outcomes (Lazarus, 1994), as the sharing group, which likely consists of more naive and less experienced drivers, might exhibit a lower tolerance for unexpected experiences, resulting in less resilience to frustrating situations. This suggests that helping drivers in the Severe profile develop strategies to handle unexpected outcomes more effectively could be a useful approach to minimizing frustration in traffic.

As for the DBQ dimensions, while prior studies suggest that violations and errors have the most pronounced impact on aggressive driving behavior and traffic accidents (Özkan and Lajunen, 2005), our findings indicate that Violations and Errors did not significantly differentiate the profiles. However, DBQ lapses was significantly lower in the Minimal and Low Profiles compared to the Severe Profile, suggesting that lapses in driving behavior are more strongly associated with higher frustration levels. Lapses, on the other hand, were significantly lower in the Minimal and Low profiles compared to the Severe profile. This finding suggests that DBQ lapses is the most important driver behavioral dimension to focus on when trying to prevent frustration. Lapses require dedicated attention in future research to better understand their role in frustration regulation.

This study contributes to applied social psychology by elucidating the complex interplay between driving situations, emotional responses to frustration, and individual differences in experiencing frustration. Identifying distinct frustration profiles provides a deeper understanding of how different drivers experience and react to frustration and can inform more tailored interventions. These insights can inform the development of targeted interventions to mitigate frustration and enhance driving experiences. For instance, tailored stress management programs could be designed for high-frustration profiles, while promoting car-sharing initiatives might alleviate frustration for specific demographic groups. Furthermore, the findings underscore the importance of considering demographic and behavioral factors in understanding driver frustration. Policymakers and transportation planners can leverage this knowledge to create more inclusive and supportive driving environments. For example, designing road infrastructures that minimize goal obstruction and unpleasant experiences could reduce frustration and improve traffic flow.

This study was conducted as an exploratory analysis of cross-sectional data, which introduces several limitations. One is that it was conducted using self-reported data, which can introduce biases such as social desirability bias, recall bias, and subjective interpretation of questions. Participants might underreport or overreport their experiences and emotions related to driving frustration as the experience might be close to the self-report time. We recommend that future studies focus on a longitudinal approach over time and use real-time data collection with a combination of objective measures, such as physiological stress indicators, to complement self-reported data and provide a more comprehensive understanding of driver frustration. Moreover, while the study categorizes various driving situations, it does not encompass all possible scenarios drivers can encounter. Hence, other significant driving situations that influence frustration levels might not have been included in the current analysis, potentially overlooking critical aspects of the driving experience. Also, while the LPA approach allowed for identifying distinct profiles, further research is needed to explore the underlying mechanisms that drive the profile differences and explore whether similar profiles can be found in other populations. Therefore, we proposed to include other potential predictors, such as socioeconomic status, personality traits, and cultural factors, that were not considered here in future studies. We chose to apply a person-centered approach, Latent Profile Analysis (LPA), to identify distinct driver profiles based on their frustration levels because this approach allows for a clearer understanding of how individual characteristics and driving experiences may create subgroups with unique patterns of frustration. While LPA offers valuable insights into how distinct subgroups experience frustration, we recognize the merit of comparing this approach to a variable-centered approach such as factor analysis, which might provide a more global understanding of frustration as a continuous factor. As suggested by Kam and Zhou (2016), comparing these two approaches could provide useful insights into the relative fit of each model and their ability to explain relationships with external variables. We acknowledge that conducting a factor analysis on the frustration scale would allow us to assess its psychometric properties and potentially enhance the validity of the subgroups identified in our study. While parcels served as an appropriate data reduction technique for conducting LPA in this study, their use assumes a formative measurement model. This assumption requires further investigation to validate the instrument’s structure and ensure replicability in future studies.

In conclusion, this study comprehensively analyzes individual differences frustration through LPA, providing a comprehensive analysis of drivers’ frustration and identifying subgroups more prone to experiencing it. This work opens new avenues for future studies to move toward a more personalized and human-centric approach to developing coping strategies and interventions for road safety.