The Role of Infrastructural and Psychological Factors in Sustainable Transportation Mode Choices

Abstract

Individual mobility behavior continues to pose a challenge to achieving climate goals, as motorized individual transportation is still favored over public transportation. The present study examines five possible drivers of more sustainable transportation mode choices: two infrastructural factors, specifically city center accessibility and railway accessibility, and three psychological variables: adaptability, climate change perception, and car orientation. A sample of N = 187 participants was collected in a German city in the Lower Rhine region. Our findings, based on ordinal logistic regression models, indicate that railway accessibility and car orientation are associated with both the use of motorized and public transportation. While center accessibility and adaptability predicted the use of motorized individual transportation, these variables did not significantly relate to the use of public transportation. Also, our results indicate that climate change perception does not relate to transportation use. This surprising finding is discussed in detail. On a more general level, the study’s insights reinforce previous findings and stress the importance of considering not only infrastructural factors in urban spaces but also the characteristics and attitudes of their inhabitants.

1. Introduction

The prevalence of motorized individual transport (MIT) in Germany poses a significant challenge to achieving national and global climate goals. With 47% of trips and 65% of passenger kilometers driven by cars [1], the transport sector is a major contributor to greenhouse gas emissions. This reliance on private vehicles not only degrades air quality and exacerbates urban congestion but also hinders progress toward the CO₂ reduction targets established by national and international climate agreements [2]. Despite growing awareness of climate change, a transition to more sustainable mobility remains critical [3].

To address these challenges, policymakers and researchers have introduced several strategies, including the widely recognized ‘avoid-shift-improve’ framework [4]. This framework aims to reduce travel demand, shift trips to sustainable modes of transport, and improve vehicle and fuel efficiency, as well as traffic management. The discourse has since expanded to encompass a more extensive range of concepts, including the term “Verkehrswende” (traffic transition), which is further subdivided into two constituent parts: “Antriebswende” (drive transition) and “Mobilitätswende” (mobility transition). These terms reflect a more holistic understanding of the necessary changes, integrating technological advancements in vehicle propulsion, systemic shifts in transport infrastructure and services, as well as fundamental changes in mobility behaviors. Manderscheid [5] highlights that these transitions are deeply embedded in social, cultural, and political contexts, underscoring the need for a multifaceted approach to sustainable mobility that extends beyond mere technological solutions.

Recent research has highlighted the critical role of psychological factors in transportation decisions [6,7,8,9] while also underscoring the complex interplay between individual behaviors, societal influences, and infrastructural factors [10] that shape these choices. Transport mode choice has been viewed as a conscious decision-making process, often modeled through frameworks such as extended versions of the theory of planned behavior [7,11,12,13]. Understanding the individual psychological drivers of behavior change, particularly in response to environmental challenges, is crucial for promoting sustainable travel behavior [7]. In this context, our study aims to explore the psychological drivers of adaptability, climate change perception (CCP), and car orientation, in conjunction with the infrastructural factors of the city center and railway accessibility from the residence. By adopting an interdisciplinary approach, we seek to enhance the understanding of individual mobility decisions and their underlying determinants. Although active travel modes such as walking and cycling are acknowledged as essential components of sustainable mobility, the present study focuses on the modal choice between MIT and PT, as this conflict represents a key leverage point for emission reduction and system-level transitions.

When explaining behavior change, adaptability as a personality trait is an important factor [14]. It captures the individual’s capacity to make appropriate responses to changing situations and circumstances [15]. Climate change and the associated growing need for sustainable travel behavior pose such circumstances, which make the concept of adaptability relevant to travel mode choice. Previous findings show that higher adaptability was found to facilitate adaptive performance in changing environments [16].

Another key factor, frequently discussed in the context of sustainability, is climate change perception (CCP), which encompasses “people’s perceptions of the reality and causes of climate change, and the perceived valence, spatial distance, and temporal distance of its consequences” [17]. CCP has been utilized to explain support for adaptation policies [18] and behaviors like information-seeking related to climate change [19]. While CCP is believed to predict adaptation intentions, such as greening gardens or installing green roofs, the evidence of its predictive power for such actions has been mixed [19]. In the travel context, CCP has been linked to an increased willingness to pay for on-demand transport services as a solution for the last-mile problem [20] and was shown to correlate with envisioning more environmentally friendly future travel behaviors [21].

Attitudes as subjective evaluations of an object, a person, or behavior make an important contribution to explaining individual travel behavior [22,23] and, in some studies, have been found to possess even greater explanatory power than infrastructural variables [24]. However, the presence of value-action gaps indicates that attitudes do not always translate into behavior [25]. These gaps can arise from conflicting subjective norms, perceived behavioral control [26], or infrastructural barriers, such as a lack of viable travel options [22]. Given the pressing need for behavior change in response to environmental challenges, attitudes are particularly advantageous due to their malleability. Although the impact of changed attitudes on travel mode choice has shown mixed results [27], attitudes remain significant determinants of travel behavior [28].

Travel behavior and mode choice are influenced not only by psychological factors but also by spatial and geographical conditions. Two key predictors in this context are center accessibility and railway accessibility. These factors align with the extended “5D” framework of mobility research, which, in addition to density, diversity, and design, also includes destination accessibility and distance to transit [29]. Center accessibility refers to the walking travel time from a residence to the nearest city center. City centers are characterized by high densities of employment, services, and infrastructure, shaping the spatial organization of activities and transportation choices. The accessibility of centers impacts the spatial arrangement of activities and transportation choices, which in turn can shape travel patterns and mode preferences. Individuals living closer to the nearest center enjoy greater accessibility to various destinations and may rely less on personal vehicles due to the possibility of walking [30,31]. This relationship is also central to the “15-min city” concept [32], which aims to create urban environments where essential daily needs can be met without reliance on motorized individual transport. PT services ensure the daily mobility of large segments of the population in local areas. Accessibility of PT can generally be defined as the ease with which individuals can use these services and subsequently reach their desired destination. The accessibility of transit stops is determined by the concept of walkability measures, which assess the physical proximity of residences or other origins to transit stops or stations. This proximity is typically expressed in terms of walking distance or time. The importance of distance to transit is based on the understanding that the willingness to use PT decreases as the distance to access it increases [29,33,34,35].

Attractive PT is about more than just physical proximity and accessibility. Travel time—including waiting and transfer time—as well as the quality of the walking environment, such as perceived safety, are important considerations. It also includes factors such as service frequency, reliability, consistency of PT service, and network connectivity, which refers to the ease of reaching different destinations using a public transport network [36,37].

Land use patterns and urban density also play an important role. The size of a city significantly influences the availability and coverage of PT. Larger cities tend to have denser and more frequent transit networks compared to smaller cities [36]. Furthermore, the concept of transit accessibility is linked to broader urban planning paradigms such as “Transit-Oriented Development”. These approaches aim to create urban environments where essential services and amenities, including PT, are within easy reach of residents [32,38].

We chose railway accessibility as a predictor of transit accessibility. First, walking time to the station has a high elasticity of demand for intercity rail travel [39], and thus for longer trips. Second, railway stations serve as important intermodal hubs that enhance connectivity with other sustainable modes [40]. This high level of service often results in a more attractive alternative to car travel for longer distances [41].

The complex interplay between MIT reliance, psychological factors, and urban planning plays an important role in developing effective policies and interventions to reduce car use and promote sustainable mobility in Germany. By considering these multifaceted aspects, including the important role of proximity to train stations, policymakers and urban planners can work towards a more balanced and environmentally friendly mobility system that meets the diverse needs of the population while being consistent with climate goals.

Our study investigates the influence of psychological variables—specifically adaptability, CCP, and car orientation—on travel mode choice. This approach complements traditional research, which predominantly focuses on socio-demographic and economic attributes. By incorporating these psychological factors into regression models alongside infrastructural variables, we extend state-of-the-art research and aim to deepen the understanding of sustainable travel mode choice, moving beyond purely logistical explanations.

2. Methods

2.1. Survey Description

The survey was conducted in a multi-method format. The participants were able to take part via an online form on the LimeSurvey platform, via a paper questionnaire sent by mail, or via telephone. During the telephone interview, the answers were entered into the online form by a project employee. The participants were recruited within Krefeld, a German city in the lower Rhine region. The data collection was conducted in two waves: the first wave (October 2022–April 2023) used an opportunity sample, including flyers, posters, newspaper articles, and advertisements on the information screens within PT. Participants had the chance to participate in a raffle for two iPads and 20 PT tickets. The second wave (August 2023–October 2023) used a random sample based on address data from the local registration office. The data set was refined through the removal of participants who reported being underage (age < 18) or missing data in the variable age, and participants starting their first travel of the day outside the researched city. All analyses are based on the combined and cleaned sample (N = 187) consisting of 78 women, 108 men, and 1 person who did not indicate any gender. Participants in the sample were between 18 and 89 years old, with a mean age of M = 42.78 (SD = 16.89). The study was conducted following the ethical standards laid down in the Declaration of Helsinki and approved by the local ethics committee (ID 2308APGE4720).

2.2. Survey Content

The participants answered a questionnaire on demographics and general questions on mobility, a trip diary, and scales assessing psychological variables. In the trip diary of everyday mobility, participants reported behavioral data for trips taken on a Tuesday, Wednesday, or Thursday. This included information on trip origins and destinations, start and end times, travel purposes, and the modes of transportation used.

2.2.1. General Transportation Mode Use

In examining the transportation use patterns, the survey question “How often have you used these modes of transportation in the past 12 months?” was strategically recorded to create two composite dependent variables for subsequent analyses. The first variable, General PT Use, was constructed by combining responses for local and regional PT and long-distance PT. The second variable, General MIT Use, was created by aggregating responses for the car as a passenger, car as a driver, motorcycle and scooter, and taxi. These recoded variables serve as dependent variables in the analyses, allowing for a comprehensive examination of PT use and MIT choice.

2.2.2. Infrastructural Variables

Further analyses are based on two infrastructural variables: walking travel time in minutes from each residence to the nearest train station and the nearest city center. To calculate walking travel time to the destinations, the address information and the associated coordinates of residential locations were used. These coordinates served as starting points for subsequent calculations using QGIS (Quantum Geographic Information System, version 3.28.13-Firenze) in conjunction with the ORS (OpenRouteService, version 1.10.0) tool. The walking time calculation within the ORS tool was based on an assumed walking speed of 5 km/h. The city’s centers were also digitized as polygons in QGIS according to the city’s central concept [42]. The center accessibility of the studied city revolves around a primary main center and three district centers. The demarcation of the centers includes the functionally dense population of retail and service facilities. The centroids of the polygons were used as destination points for subsequent calculations.

The predictor of PT captures the continuous nature of travel times rather than categorizing them as acceptable or unacceptable. The granularity allows for statistical analysis, as highlighted by [43], facilitating deeper insights into the relationship between accessibility and urban factors. Moreover, as noted by [44], continuous measures enhance our ability to identify specific areas needing intervention, leading to more effective PT planning and policymaking. Figure 1 illustrates the spatial distribution of participants, railway stations, and the city center, providing an overview of their relative locations within the study area.

2.2.3. Psychological Variables

We investigated three psychological variables: adaptability as a trait variable, CCP, and the attitudinal variable car orientation.

Adaptability was assessed using a validated scale by [45]. It consists of ten items representing three facets of adaptability: cognitive, affective, and behavioral. The scale measures an overarching adaptability factor on a five-point Likert scale from 1 (=strongly disagree) to 5 (=strongly agree) and showed good validity in our sample (Cronbach’s Alpha (α) = 0.90).

CCP was assessed using the climate change perceptions scale by [17]. The applied short scale consists of five items, each aiming at one type of CCP: reality, causes, valence of consequences, spatial, as well as temporal distance of consequences. Answers were given on a seven-point Likert scale ranging from 1 (=strongly disagree) to 7 (=strongly agree), and the fifth item was reverse-coded. The scale showed good validity in our sample (α = 0.78).

Car orientation was captured using the eponymous subscale from the validated questionnaire on psychological factors influencing the use of cars, public transport, and bicycles (PsyVKN) by [23]. Participants answered the five items on a five-point Likert scale ranging from 1 (=does not apply) to 5 (=applies). Unlike in the original scale, we did not reverse-code the items, thereby making care of an easier interpretation of results. Higher values of our variable car orientation stand for a stronger (positive) tendency towards cars. The scale showed good validity in our sample (α = 0.84).

Due to an administrative error, the dataset was missing some values for the psychological variables. For car orientation, 49 values were missing; in the cases of adaptability and CCP, 54 values were missing. To reach a better-powered outcome in later analysis, we imputed the missing values with the arithmetic mean.

3. Results

Descriptive statistics and ordinal logistic regressions were computed using IBM SPSS Statistics, Version 29.0.0.0. A proportional odds ordinal logistic regression model was used to examine the relationship between the predictors (infrastructural and psychological variables) and General PT Use and General MIT Use as dependent variables, both ordinal variables representing the frequency of use for the respective transportation modes. Confidence intervals were estimated using bootstrapping with approximately 10,000 samples, as the data did not meet the assumptions of normality. The analysis followed standard procedures for ordinal logistic regression.

3.1. Descriptive Statistics

Table 1. Descriptive Statistics.

Variable	M	SD	Range
Center Accessibility	23.98	14.78	1.54–77.14
Railway Accessibility	30.48	21.64	4.00–106.92
Adaptability	3.22	0.64	1.10–5.00
Climate Change Perception	6.00	0.85	2.20–7.00
Car Orientation	2.95	0.91	1.00–5.00
General PT Use	4.30	2.14	1.00–7.00
General MIT Use	4.82	1.99	1.00–7.00

Note. Lower values reflect better center and railway accessibility; PT = Public Transportation; MIT = Motorized Individual Transport.

reports the mean values (M), standard deviations (SD), and the range of our variables. Furthermore, the respective intercorrelations are shown in

Table 2. Correlation Matrix for the variables of interest.

Variable	Center Accessibility	Railway Accessibility	Adaptability	Climate Change Perception	Car Orientation	General PT Use	General MIT Use
Center Accessibility	-
Railway Accessibility	ρ = 0.120	-
Adaptability	ρ = 0.125	ρ = 0.206 **	-
Climate Change Perception	ρ = 0.025	ρ = 0.145 *	ρ = 0.064	-
Car Orientation	ρ = −0.008	ρ = 0.114	ρ = 0.035	ρ = −0.162 *	-
General PT Use	ρ = −0.107	ρ = −0.169 *	ρ = 0.027	ρ = 0.030	ρ = −0.181 *	-
General MIT Use	ρ = 0.225 **	ρ = 0.210 **	ρ = −0.098	ρ = 0.010	ρ = 0.359 **	ρ = −0.581 **	-

Note. N = 187, presented are values of Spearman’s Rho (ρ) and their significances (p). PT = Public Transportation; MIT = Motorized Individual Transportation. * p < 0.05, ** p < 0.01.

.

3.2. Effects of Spatial Factors and Psychological Variables on Mobility Behavior

Two ordinal logistic regression analyses were conducted to investigate the general use of transportation modalities: one analysis for General PT Use and another for General MIT Usage. In both ordinal logistic regressions, the predictor variables center accessibility, railway accessibility, adaptability, climate change perception, and car orientation were considered. The predictor variables were tested a priori to verify that there was no violation of the assumption of no multicollinearity. Correlation coefficients between the predictors did not exceed 0.80, which would have indicated strong correlations and suggested multicollinearity (Table 2). Results are presented in

Table 3. Ordinal logistic regression analysis to predict general public transportation use.

Predictors	Estimate	SE	p	95% CI Bootstrap (10,000 Samples)
Constant	681.61		0.004
Center Accessibility	−0.018	0.009	0.048	[−0.037, 0.002]
Railway Accessibility	−0.016	0.006	0.012	[−0.026, −0.006]
Adaptability	0.136	0.206	0.510	[−0.357, 0.656]
Climate Change Perception	−0.043	0.157	0.784	[−0.332, 0.300]
Car Orientation	−0.375	0.148	0.011	[−0.725, −0.037]

Note. Significant influences are marked in bold.

for the dependent variable PT usage and in

Table 4. Ordinal logistic regression analysis to predict general motorized individual transport use.

Predictors	Estimate	SE	p	95% CI Bootstrap (9797 Samples)
Constant	561.40		<0.001
Center Accessibility	0.035	0.010	<0.001	[0.017, 0.055]
Railway Accessibility	0.020	0.007	0.002	[0.009, 0.032]
Adaptability	−0.551	0.221	0.013	[−0.997, −0.128]
Climate Change Perception	0.090	0.171	0.598	[−0.270, 0.489]
Car Orientation	0.971	0.167	<0.001	[0.639, 1.391]

Note. Significant influences are marked in bold.

for MIT usage.

When focusing on General PT Use (Table 3) as the dependent variable, the model fit was statistically significant, χ² = 17.36, p = 0.004. The Pseudo R-Square values (Nagelkerke = 0.09) suggest a weak relationship between the predictors and General PT Use [46]. Higher car orientation significantly reduces the odds of PT usage (β = −0.375, p = 0.011), while higher railway accessibility is associated with increased PT usage (β = −0.016, p = 0.012 due to reverse-coded variable). Center accessibility does not hold up as a predictor when applying bootstrapping; therefore, no significant relation can be reported. Also, the psychological variables CCP show no predictive value within our model.

Concerning the ordinal logistic regression analysis using General MIT Usage as the dependent variable, the model fit was also significant, χ² = 57.43, p < 0.001, with a strong relationship between the predictors and the outcome, Nagelkerke = 0.27 [46]. The spatial variables center and railway accessibility prove to be significant positive predictors, but with very small parameters (Table 4). Longer travel durations to the center and the railway therefore predict higher General MIT Use. When considering the psychological variables, adaptability and car orientation emerge as significant predictors with rather high estimates, while CCP remains a non-significant predictor. Adaptability has a negative and car orientation has a positive coefficient, indicating that more adaptable individuals will use MIT less often, and a higher car orientation will lead to more frequent usage of MIT.

Overall, the travel durations between the residential location, the center, and the next station seem to influence modality usage, even though the effect of walking travel time between home and the center could not be found for PT. If the travel time is shorter, the choice falls on PT more often. The psychological variables provide further insights into this modality choice: general car orientation predicts both modality usages significantly, with a consistent directionality. If participants are more car-oriented, they will rather use MIT than PT. Meanwhile, CCP contributes only a small effect to any of our two models. Last, adaptability predicts MIT usage, but the prediction of PT use is limited.

4. Discussion

Our study shed light on the interrelation between five potential drivers and the general use of MIT as well as PT use. Interesting insights were found for infrastructural and psychological drivers. Travel durations between the residential location and the next station show a connection to the use of MIT and PT. Shorter durations thereby contributed to the choice of a sustainable transport mode. Then again, travel durations between the residence and the city center were found to be valid predictors of MIT, but not of PT use. Longer travel durations led to a more frequent choice of MIT. We also investigated the role of three psychological variables, which have only partially been subject to prior investigations about travel mode choice. While trait adaptability was found to be a predictor for the general use of MIT (but not of PT), the CCP did not show any significant value within our models. However, the explanatory power of our models was limited to a moderate amount of variance, which should be considered when interpreting these findings. Our study confirmed findings on the attitude of car orientation, which reliably affected the MIT use as well as the use of PT. Lower values on adaptability and a higher car orientation predicted a higher use of MIT, and lower car orientation predicted a higher use of PT. When looking into the matter of sustainable transportation, the missing predictive value of CCP is also an interesting finding.

The findings on the infrastructural factors are consistent with previous studies that have demonstrated the important role of proximity to PT facilities in promoting sustainable transportation choices [29,33]. The finding that shorter walking times to stations predicted higher PT use aligns with the concept of transit-oriented development, which emphasizes the need for dense, mixed-use areas near transit hubs to promote ridership. However, the effect of proximity to stations on PT use is likely mediated by additional urban design factors, such as the pedestrian quality of access routes, land-use diversity in the vicinity of stations, and last-mile connectivity through feeder systems or micromobility options and can influence whether a short distance is perceived convenient and usable in practice [44].

The present results suggest that a reduction in walking time to major destinations is associated with a reduction in MIT use. This supports the hypothesis that individuals who live closer to centers are less likely to rely on their cars for daily activities. This finding is consistent with previous research that has demonstrated a reduction in MIT use when walking or biking access to a destination is improved [29]. In contrast to the reduction in MIT use, walking time to the nearest center did not play a significant role in predicting PT use. Residents living closer to centers may prefer alternative modes of transportation, such as bicycling, that offer similar flexibility to MIT but may not affect PT use. This finding can be better understood by examining the implications of the compact city (green, healthy, resilient, smart, safer, etc.) and 15-min or 20-min city concepts that improve people’s quality of life. These concepts prioritize reducing travel times and increasing walkability by creating environments where destinations for daily activities are within walking or at least cycling distance. By promoting such accessibility, these concepts could reduce car reliance. Thus, while these concepts are potentially successful in reducing MIT use, they do not necessarily increase PT use, as the immediate need for PT is reduced when active modes are feasible and convenient.

Regarding the psychological variables, adaptability is by definition linked to an individual’s capacity to respond to change [15], such as a world critically influenced by climate change and the related need for adaptive behavior in the form of more sustainable mobility. Our findings that less adaptable persons would more likely choose MIT for their daily travels are therefore conforming with our expectations and prior research [16,24]. Surprisingly, we did not find this relationship regarding the use of PT. This might hint at a more complex relationship between trait adaptability and the use of sustainable transport modes. Possibly, there could be variables hindering more adaptable individuals from choosing PT, which was not considered in the present study. For instance, MIT may reflect proactive, self-directed behaviors requiring openness to innovation—domains, where adaptability plays a direct role—while PT use, might depend more on external factors like accessibility or cost, which are less tied to psychological traits. Similarly, the distinction between autonomy-driven MIT (e.g., cycling) and necessity-driven PT (e.g., bus/train use) could explain varying adaptability effects. Future studies should, therefore, take a closer look at this relationship and search for intervening variables in the interplay between personality and transport mode choice. Also, climate change as a situation that makes behavior change necessary could be too far away from the individual’s daily reality. For comparison, when a construction site is on the individual’s way to work, they will probably quickly search for an alternative route or mode of transportation. Meanwhile, the effects of climate change do not become evident in such an immediate manner. Our study took the first steps to explore the so-far understudied relationship between adaptability and travel mode choice, which can and should serve as a foundation for future research.

Similarly, we took a first step towards understanding the role of CCP in travel mode choice. Interestingly, it did not show a role in the use of PT or MIT. These findings support our previously explained presumption that climate change could be an alteration that is too far away from an individual’s daily mobility and reality of life. This would be in line with construal level theory, which explains how people perceive events like climate change as psychologically near or, in that case, more distant [47]. Prior adaptive behaviors successfully predicted by CCP were more closely related to climate change [19,21], giving this possible explanation more support. Still, it is a reportable finding that individuals highly aware of climate change will not, per se, use more sustainable transportation modes. Future research should validate and extend these insights, potentially exploring alternative psychological variables like environmental responsibility and social norms, which may more directly influence travel behavior. Refining the conceptualization and measurement of CCP—perhaps focusing on perceived behavioral control—could also yield more informative results and deepen our understanding of sustainable travel choices. Taken together, the findings point to specific factors that may serve as promising policy intervention points. Most notably, improving access to public transport infrastructure—particularly railway accessibility—emerged as a significant predictor of PT use. This supports urban development strategies that prioritize transit-oriented development and regional rail integration. In addition, car orientation, while relatively stable, showed a strong inverse association with PT uptake. This underlines the importance of long-term strategies aimed at shifting attitudes through targeted communication, education, or incentive-based interventions.

While the present study was conducted in a German urban context, the identified mechanism—particularly the interplay of infrastructural accessibility and car-oriented attitudes—may offer valuable insights for other cities facing similar sustainability challenges. Nevertheless, the generalizability of the results should be critically examined because of cultural differences in international urban settings.

Lastly, the attitude of car orientation successfully predicted the transport mode. More car-inclined individuals more often chose MIT and less often used PT in their daily travels. These findings are in line with prior research [23,28] and point towards a stable association between this attitudinal variable and mobility behavior. While car orientation appears to be a relatively stable attitudinal factor, it is not immutable. Studies highlight the potential of interventions such as targeted public awareness campaigns [48], behavioral nudges [49], or integrated mobility services such as Mobility-as-a-Service [50] to gradually shift user attitudes toward more sustainable transport options. These approaches work by increasing visibility, convenience, and the perceived normality of non-car modes.

Although not part of our analytical model, active modes such as walking and cycling are increasingly promoted in urban transport policies and share several behavioral determinants with PT and MIT use. As such, their theoretical integration into future research could enhance the understanding of full modal substitution patterns.

Like every empirical work, our study does not come without limitations. As shown, the findings are in line with prior work. However, the insights rely on correlation data, which does not allow conclusions on the directionality of interrelations. The assumption that people with higher car orientation give a lower priority to station access when moving might be just as valid and supported by prior research [51,52].

While we used innovative psychological variables to predict travel mode choice, established predictors like sociodemographic variables were not included in the analysis [53]. By sharpening our focus on five predictors, others were not considered, and this approach might reduce the informative value of our models. We decided to accept this limitation to apply a robust statistical method, given the limited sample size, and to contain innovative insights into the role of so far understudied psychological variables. In that way, we were able to create new insights that can now be integrated into future research.

The non-significant predictive value of climate change perception may be due to limitations in its measurement. Future research should explore alternative psychological variables, such as environmental responsibility and social norms, which may more strongly influence travel behavior. Refining the conceptualization and measurement of climate change perception itself—perhaps focusing on perceived behavioral control—could also yield more informative results and deepen our understanding of sustainable travel choices.

The two-stage sampling design comprising an opportunity, and a random sample is susceptible to self-selection bias and potential inconsistencies in participant characteristics. This and the mostly online questionnaire might have led to an over-representation of young individuals already interested in sustainable mobility. The generally high awareness and small variance in climate change perception might have impacted our results, diminishing existing effects in the population. Notably, the achieved sample size was relatively small, suggesting that the population may be difficult to access. Under the given circumstances, the use of this sampling method is acknowledged as a limitation, albeit a necessary compromise given the scarcity of alternative approaches.

Moreover, our study did not include active mobility options, such as walking and cycling, as dependent variables, despite their growing importance as sustainable modes of transport. Active mobility offers low-carbon alternatives to MIT and PT while also providing health benefits and enhancing the quality of public spaces through decreased noise and air pollution [54]. Incorporating walking and cycling could have provided a more comprehensive understanding of travel mode choices, especially within the context of the shift towards greener transportation options. When building on the presented insights, active mobility options should be considered.

Also, the limited sample size could be a subject of critique. The implemented questionnaire was rather long, thus bearing a bigger respondent burden [55]. At the same time, the extended questionnaire gave the possibility to integrate innovative psychological variables besides well-known predictors and made the present research possible. We counteracted the limited sample size by applying robust statistical methods (bootstrapping) and by analyzing a less complex regression model.

The study’s interdisciplinary approach offers valuable starting points for future research and practical guidance for policymakers and traffic planners. One key finding is that infrastructural, as well as psychological factors, play a role in the choice of MIT and PT, and both should be considered when studying decision-making processes. Therefore, decision-making on future mobility concepts should incorporate not only infrastructural considerations but also individual perspectives, such as attitudes, traits, and fundamentals of behavioral change. In this context, railway accessibility and car attitudes remain crucial. When exploring what drives PT use, both infrastructure improvements and the shift away from a car-oriented attitude should be emphasized. Additionally, MIT usage is predicted by the factors of center accessibility and adaptability—aspects that are relatively stable and difficult to modify, given the fixed nature of city centers and personality traits. To effectively shift behavior away from MIT, it is essential to identify and target other, more malleable factors. At the same time, adaptability is a variable relevant to adaptive performance [56] and, thereby, behavior change. Future research should consider the insights provided in this study while also considering change intentions and actual behavior change.

5. Conclusions

The interdisciplinary approach of this study offers new perspectives that are of interest to both researchers and practitioners in transportation planning. We found that the distance between the residential location and the railway station, and the attitude of car orientation, predicted the use of public and motorized transportation. For motorized transportation, center accessibility and adaptability emerged as additional predictors, however, this relation did not hold up for public transportation use. Climate change perception did not show predictive value in either of the models. The results emphasize the need to integrate infrastructural and psychological factors into planning and research. The dual association of these factors suggests that effective policy must consider not only the physical layout of urban spaces but also the attitudes and behaviors of people who use them. Improving public transport infrastructure, such as increasing railway accessibility, is essential to encourage the use of sustainable transport options. However, policymakers and planners also need to address the entrenched car-oriented mindset of many (urban) residents. By promoting PT as a convenient and flexible alternative and by implementing strategies to reduce car use, a more environmentally friendly and socially inclusive urban mobility system that supports reduced car dependence and increased public transport use can be achieved.

In addition, the study highlights the challenges posed by more fixed factors, such as center accessibility and individual adaptability, in influencing MIT use. As these factors are less amenable to change, it is important to identify other leverage points that can be targeted to encourage the adoption of more sustainable travel behaviors. At the same time, indirect strategies may help to offset the behavioral effects of such relatively fixed factors. For instance, limited center accessibility could be addressed by strengthening last-mile connectivity, enhancing the walkability of surrounding areas, or improving digital access to central services. In the case of lower adaptability, planners with simplified user interfaces or offering supportive infrastructures that reduce perceived complexity, such as the Deutschland-Ticket. Beyond the focus on public and motorized transport, the growing importance of active modes such as walking and cycling should also be acknowledged in the context of sustainable urban mobility. Finally, the study highlights that personality factors (particularly adaptability) play a complex role in influencing transportation choices that should be researched, also considering adaptive performance, adaptation theory, and possible intervening variables. This would pave the path towards personalized interventions, facilitating behavior change and promoting sustainable transportation.