Relationship Burnout:
The Integrated Relationship Trajectory Scale, Version 1 (IRTS-1)

Regarding relationships, living in a constant state of anxiety may cross a point of no return leading to burnout, its anticipation, and separation. Relationships don’t just span marriage. While separation invokes adjacent categories of divorce and separation, separation also applies to cohabitating individuals. Cohabitation is important in relations, and its prevalence had dramatically increased in society, though there is difficulty in its measure (Bumpass & Sweet, 1989; Cohen, 2021; Vespa, 2014; c.f., Manning et al., 2019). Cohabitation offers the utility of marriage without its expectancies (e.g., financial stability, situational stability, employment stability). Yet relationships span more than just marriage and cohabitation. Relationships also include non-married and non-cohabitating individuals. Therefore, measuring burnout in social relationships calls for a more encompassing view of relationship beyond its range restricted state sanctioned measures of relationship (i.e., marriage), and its narrowly extended academic measure (e.g., marriage, cohabitation). 

Relationships are important for health and well-being. Relationships increase survival rate by 50% irrespective of age, sex, and/or health status (Holt-Lunstad et al., 2010). For example, married individuals experience less depression, increased happiness, increased life satisfaction, increased well-being, and lower mortality to suicide; co-habitation demonstrates equal benefits (Amato, 2015). Unfortunately, as people age, relationships decrease, and a significant portion (20–43%) of adults over 60 suffer frequent/intense loneliness (Perissinotto, 2012). Early identification of relationship burnout such that individuals may consider therapy or additional support may go a long way for individuals, mental health professionals, clinicians, councilors, therapists, employers, insurers, and governments managing the health and well-being of the public.

Unfortunately, literature on burnout is a vast topic, and while some aspects are studied in matters of marriage (e.g., Nejatian et al., 2022), there are well-funded existing measures of burnout across business domains. A relationship burnout scale may be founded on the Burnout Measure (Pines & Aronson, 1988), Maslach Burnout Inventory (Maslach, 1996), and its shorter version, the Burnout Measure, Short Version (BMSV; Malach-Pines, 2005). This shorter version examines for feelings of resent, tiredness, disappointment, hopelessness, entrapment, helplessness, depression, physical weakness/sickness, worthlessness/failure, insomnia, and quitting. 

In sociological studies, relationships today are more companionate than utilitarian, and come with expectations of shared housework, support, and shared efforts to protect from social stigma (Cohen, 2021). Burnout’s response may evidence a kind of deviant adaptation responding to disconnects between cultural goals of relations and institutional means in support of them (Merton, 1938). 

Turning toward the psychological, relationship burnout may inversely converge with a well-known popularized psychological three-part typology of commitment, intimacy, and passion in the triangular theory of love (Sternberg, 1986). More accessible to individuals are common items related to popular anecdotes describing burnout (e.g., argumentation, changed feelings, disappointment, boredom, disconnection, being liked, exclusion etc.). Another model identifies critical areas leading to separation as criticism, contempt, defensiveness, and stonewalling which comprise the four-horseman model along with antidotes of a gentle startup, a culture of appreciation, taking responsibility, and self-soothing (Gottman, 1993; Gottman & Krokoff, 1989; Gottman & Silver, 2015).

Of more recent research, individuals are driven by a universal need for relatedness which is clustered with additional universal needs of autonomy and competency as identified by self-determination theory (SDT; Ryan & Deci, 2000). In convergence with autonomy and competency, flow state (Nakamura & Csikszentmihalyi, 2002) has demonstrated protective effects against burnout (Aust et al., 2022). Existing burnout measures (e.g., Burnout Measure, BMSV) and well-known factors in relations crossing sociological and psychological studies are valuable candidates to consider in construct validity of a more generalized relationship burnout scale, aptly proposed as the integrated relationship trajectory scale (IRTS; pronounced like “hurts”), and its first version, herein referred to as the IRTS-1.

Evidence of criterion validity in a measure’s predicted consequences is critical (Kaplan & Saccuzo, 2018). An improved relationship burnout measure is expected to offer prediction of real-world separation, yet must be balanced with prediction of the opposite, relationship sustainability, for greater face-value validity. Borrowing from Newton’s (1687) two-body problem, relationships may be represented by a simple three-part typology. First, entry, as two bodies entering relatively stable orbits around each other, regardless of inertial differences. Second, orbit, where two bodies maintain said orbits, and third, departure, where one or both bodies reach escape velocity and separate from said orbits. These three types cover a continuous measure from entry, orbit, to departure and can be represented as a simple gauge for ease of accessibility while helping to advert maladaptive cognitions (e.g., catastrophizing) due to potential rumination on a scale focused on burnout alone.

While a relatively simple model, a generalized relationship burnout scale is expected to account for practices whereby a body reaching escape velocity may reconsider or may be beholden to inter-body efforts developing protective effects to prevent burnout as well as repair effects to reduce burnout in process. The integrated burnout scale is expected to normalize relative relations of not one body moving toward/away from another body, but both (i.e., a two-body problem). Having considered repair effects, protective effects, and relativity in a relationship, there may be individual and cultural differences (i.e., errors) determining what qualifies as items measuring burnout (e.g., BMSV’s “calling it quits” [Malach-Pines, 2005]).

Beyond individual differences, mental health disorders may confound burnout measures (e.g., depression, bipolar disorder; Ahola, 2014; Beeney et al., 2019). Turning to a general relationship burnout scale’s potential content validity, attachment styles are covered in measures of the four horseman (Fowler & Dillow, 2011). Thankfully items from the Burnout Measure, Short Version may cover these confounds. Casting a wide “nomological net” (Cronbach & Meehl, 1955) across the prior constructs of entry, orbit, and departure in relations is expected to assemble a rough item pool, where development of a scale shall iterate in increasing reliability along with substantive, structural, and external validities (Clark & Watson, 1995; Kaplan & Saccuzzo, 2018; Loevinger, 1957). 

To validate the construction of the proposed IRTS-1, it is hypothesized (Hypothesis 1) that internal validity concerning popular descriptions of burnout leading to relationship departure should be correlated with more rigorous measures where matters of inverse correlated aspects of relationship entry and orbit. It is further hypothesized (Hypothesis 2) that the proportion of individuals in sustained relationship entry, orbit, and final departure shall be correlated to known studies of relationship longevity. In testing these hypotheses, several subscales are considered: (a) popular description of burnout characteristics; (b) the Burnout Measure, Short Version (Malach-Pines, 2005); (c) the triangular theory of love (Sternberg, 1986); (d) the four horsemen (Gottman, 1993; Gottman & Krokoff, 1989; Gottman & Silver, 2015); (e) family relations (Cohen, 2021); (f) strain adaptations (Merton, 1938); and (g) SDT’s needs of relatedness, autonomy, and competency (Ryan & Deci, 2000). 

Method

Participants

This study’s sample had been convenience sampled from students enrolled in Washington State University’s Fall 2023 Psychological Testing & Assessment course (PSYCH-412; N = 30). Sample demographic characteristics included student id (last four digits), gender, age, and type of relationship the participant had selected for burnout evaluation (see Table 1 for sample characteristics). The sample was comprised mainly of emerging adults (M = 28.4; SD = 9.065; range: 19–47), where males comprised 63% of the sample, and where individuals not reporting an ethnicity comprised 63% of the sample. Additional demographic data on the type of relationship had been used to examine different aspects of the IRTS with respect to relationship burnout. Participants had been compensated with course credit, as determined by the last four digits of a participant’s Washington State University student identification, which had been removed from these data prior to analysis.

Materials

Demographic measures had been recorded for age, gender, ethnicity, and type of relationship. Initial IRTS draft items had been developed through a combination of inductive approaches based on empirical experience and deductive approaches based on literature review to cover a relationship burnout construct. An initial item bank of 62 draft scale items had been assembled which included a subset of items from the Burnout Measure Short Version (BMS; Malach-Pines, 2005) for testing criterion-related (concurrent) validity, 4 question Perceived Stress Scale (PSS-4; Cohen et al., 1983; Cohen & Williamson, 1998) for testing construct related validity, and additional items based on the triangular theory of love (Sternberg, 1986), self-determination theory (SDT; Ryan & Deci, 2000), and sociological studies of relations (Cohen, 2021).

IRTS draft items had been reviewed for face validity by the author’s peers in the same Washington State University Fall 2023 Psychological Testing & Assessment course (PSYCH-412). Additional face validity review had been provided by three subject matter experts crossing psychological and medical fields from UW Medicine (e.g., psychiatric nurse practitioner, medical professional, psychological professional). Final face validity review had been provided by the instructor of the same PSYCH-412 course. Following draft item review, the item bank had been culled based on face validity feedback to develop the final IRTS scale.

The final IRTS consists of 24 items on a 5-point Likert scale ranging from 1 (never feel this way) to 5 (always feel this way). Items included 17 IRTS-1 items, 3 burnout subscale items based on the BMS for testing criterion-related (concurrent) validity, and 4 stress subscale items based on the PSS-4 for testing construct related validity. Some questions had been re-ordered to avoid response fatigue related to BMS and the more cognitively loaded PSS-4 questions.

Procedure

Survey items had been imported and collected using an online self-report survey implemented in Washington State University’s instance of Qualtrics XM between October 16–22, 2023. Data had been exported to Microsoft Excel and corrected for reverse scoring, and summation for IRTS-1, BMS, and PSS-4 subscales. Data had then been imported into the University of Amsterdam’s Jeffrey’s Amazing Statistics Program (JASP) for analysis (e.g., descriptive statistics, frequency tables, unidimensional reliability, ANOVA, correlation). Excel had been used to manually calculate frequency tables for age demographics. An ad-hoc one-way ANOVA analysis in JASP had been executed for evaluating statistical difference between groups of participant evaluated relationship types (e.g., spouse, friend, coworker, family member, cohabitant) in addition to an ad-hoc Pearson correlation coefficient to analyze the relationship between the BMS and PSS-4 subscales.

Results

Measures of reliability of the IRTS-1 scale had been tested via Cronbach’s (1951) coefficient alpha (α) for internal consistency with the burnout subscale and stress subscales. Measures of validity of the IRTS-1 measure had been tested via Pearson correlations between the IRTS-1 scale with the BMS and PSS-4 informed subscales. The mean relationship burnout score obtained with the IRTS-1 in the student sample was 48 (SD = 13.75) evidencing a mostly centered leptokurtic distribution shifted slightly left on a range of possible scores from 24–120 where a score of 58 is centered. A slight floor effect is evidenced in some of the means reported across items in the scale.

Reliability

Rough reliability measures of item means and standard deviations had been calculated (see Appendix 1, Table 2). Overall reliability had tested for internal consistency across the IRTS-1 scale, and BMS and PSS-4 subscales using Cronbach’s (1951) coefficient alpha (α). The α is a function of the quantity of test items and the proportion of total variance of the scale corresponding to the covariance between test items. The α is ranged from 0–1 and represents the “lowest bound for reliability” (Kaplan & Saccuzzo, 2018, p. 110). Values for α may be less reliable (0.0–0.20), rather reliable (>0.20–0.40), quite reliable (>0.40–0.60), reliable (>0.60–0.80), and very reliable (>0.80–1.00; Hair et al., 2010). The IRTS-1 had somewhat high/good internal consistency (17 items; α = .86; see Appendix 1, Table 3).  Items from the BMS had an excellent internal consistency (3 items; α = .91; see Appendix 1, Table 5). Items from PSS-4 had poor internal consistency (4 items; α = .72; see Appendix 1, Table 7).

Beyond α, item-rest correlations (Lord & Novick, 1968) had been calculated to measure the association of a scale’s item with all other scale items (Zijlmans et al., 2018). These item-rest correlations had been calculated for the IRTS-1 and BMS/PSS-4 subscales (see Appendix 1, Tables 4, 6, 8). IRTS-1 questions Q1 (“I feel argumentative.”), Q2 (“I feel irritated”), Q4 (“I feel bored”), Q5R (“I feel like I want to go on a date.”), and Q6 (“I feel disappointed”) demonstrated low item-rest correlations of .186, .277, 0.264, and 0.246 respectively (see Appendix 1; Table 4). Item-rest correlations for the BMS subscale were > 0.819 (see Appendix 1; Table 6). Item-rest correlations for the PSS-4 subscale were > .353 (see Appendix 1; Table 8). 

Validity

A Pearson correlation coefficient (r) was calculated to assess the linear relationship between the IRTS-1 and BMS. There was a significant and positive correlation between these scales (r = 0.784, p < .001; see Appendix 2, Figure 1). This indicates that higher levels of relationship burnout as measured by the IRTS-1 were associated with higher levels of burnout as measured by the items selected for criterion validity from the BMS.

A Pearson correlation coefficient was calculated to assess the linear relationship between the IRTS-1 and PSS-4. There was a significant and moderate correlation between these scales (r = 0.402, p < .028; see Appendix 2, Figure 2). This indicates that higher levels of relationship burnout as measured by the IRTS-1 were associated with higher levels of perceived stress as measured by the stress subscale based on the PSS-4.

Group Differences

The results of the one-way ANOVA showed that the five relationship type groups did not significantly differ from each other on mean burnout scores, F(4,25) = 2.17, p = .10. 

Interpretation

Based on these analyses, overall reliability of the IRTS-1 is good, but not excellent (α  = .86). In addition, validity in criterion validity with the BMS (r = 0.784, p < .001) is not super strong. Including more items from the BMS may drive the relation up. Though the IRTS-1 against the existing BMS items demonstrated excellent internal consistency (α = .91), poor internal consistency against the PSS-4 (α = .784) invites an opportunity to also include the entire subscale of the PSS (Cohen et al., 1983) or the more reliable PSS-10 (Cohen & Williamson, 1988), rather than the PSS-4. 

Alternatively, it is possible that iterating on IRTS-1 items might offer better improvements. For example, argumentativeness, irritation, boredom, wanting to go on dates, and disappointment may be less relevant to relationship burnout and perceived stress. Considering this, it is possible that evaluating construct validity with the PSS-4 may be tangential to, but less related to relationship burnout, where alternate established measures more related to relationship burnout may be more meaningful. That said, relationship burnout may be confounded by perceived stress, as there is the possibility that functioning relationships better within zones of shared Yerkes & Dodson’s (1908) findings on optimal arousal which could imply floor/ceiling effects in measures related to perceived stress. Improvements to the IRTS-1 with respect to items related to stress measures may benefit from additional iteration following these findings.

Considering issues between the IRTS-1 and PSS scales, a post-hoc Pearson coefficient correlation was calculated between the BMS and PSS-4 subscales. There was a non-significant and non-existent correlation between these scales (r= .030, p < 0.876; see Appendix 2, Figure 3). This indicates that perceived stress and existing measures of burnout are unrelated. This finding further informs future iteration on IRTS-1 items by suggesting minimal representation of items related to perceived stress or alternate forms of its measurement.

In conclusion of analysis, the IRTS-1 demonstrates utility in measure of relationship burnout across relationship types but needs improvements. Hypothesis 1 must be accepted as there is a positive correlation between the IRTS-1 and BMS subscale, though there is doubt about the use of the PSS-4 subscale, at least as it is incorporated into IRTS-1’s item set. Due to an oversight in item bank construction, review, and survey implementation, Hypothesis 2 had not been evaluated due to demographic details not being entered to collect information on relationship duration. Irrespective of Hypothesis 2 omission, results demonstrate that IRTS-1 has potential for measurement of general relationship burnout, though improvements to both the IRTS-1 scale and the way PSS is used in for criterion validity is called for in a future iteration (i.e., IRTS-2).

Discussion

The purpose of this research had been to generalize theoretical conception of relationship burnout induced from popular description and deduced from existing literature focused on specific types of relationship burnout (e.g., marriage and dating etc.). There are many kinds of relationships in society including relationships with cohabitants, friends, family, coworkers, bosses, subordinates etc. Anywhere where there are two people (i.e., dyad), there is a relationship, and thus there is potential for burnout. Results demonstrated that the IRTS-1’s ability to measure general relationship burnout has high/good internal consistency (α  = .86), significant criterion validity (r = 0.784, p < .001), yet moderate construct validity (r = 0.402, p < .028). An ANOVA analysis of results between relationship types demonstrated no statistical difference which supports generalizability regardless of relationship type. These results support the hypothesis that it is possible to develop a generalized measure of relationship burnout.

Item Set Improvements. Though the first hypothesis is supported, it is possible that iterating on scale items to remove weaker items may increase internal consistency further. Following modification to scale items, including more items from the BMS and PSS-10 rather than the PSS-4 may yield increases in criterion and construct validity. However, results cast some doubt about PSS’s use as it relates to the core construct of relationship burnout, at least in the way perceived stress is measured and recorded. Running an adhoc correlation between the BMS and PSS-4 subscales had given weight to this doubt, however existing literature may offer promise of a corrective method.

Optimal Arousal Over Perceived Stress

Considering the non-significant and non-existent correlation between the BMS and PSS-10 scales (r = .030, p < 0.876) and significant criterion validity when using the BMS, perceived stress might seem less related to the construct of relationship burnout. Yet, perhaps relationship stress is subject to optimal arousal, which demonstrates a curvilinear relationship between performance and stress. For example, the classic Yerkes-Dodson law (1908) may be at play in relationships where too little stress in the relationship leads to boredom, and too much stress leads to anxiety. Perceptions of boredom in dyads is a factor in studies of relationship maintenance and outcomes of relationship satisfaction (Dobson et al., 2023). Considering optimal arousal, there is also convergent validity with research on flow, which demonstrates increased feelings of joy with optimal experiences within a narrow lane between anxiety and boredom where individuals experience challenge and meet it with increases in skill (Csikszentmihalyi, 1975). 

Considering optimal arousal and flow, a possible improvement is to retest with a linear transformation of PSS scales mapped to a curvilinear form (i.e., low and high stress is high, middling stress is low). It is possible that this stress mapping may yield better results and increase construct validity, bringing increased convergent validity with studies of optimal experience which are core to positive psychology. Literature review of optimal arousal and flow in dyads as these relate to burnout is extremely limited; OpenAI ResearchGPT, which currently indexes 200 million research articles reports the same limitation (see Appendix 3), and Google Scholar searches report irrelevant results. Regardless, utilizing a transformed PSS scale may still yet be more valuable than measures of optimal arousal and flow. For example, existing measurements of optimal arousal are dependent on the much more invasive experience sampling method (ESM; Farnworth et al., 1996; Larson & Csikszentmihalyi, 1983), which would transform the IRTS-1 into a longitudinal measure which is not ideal for the benefits of cross-sectional measures allowing for immediate feedback, at least on the first measure.

While utilizing stress in testing construct validity of the IRTS-1 seems meaningful and valid at face validity, existing literature on burnout and performance seems to implicate exhaustion rather than stress, though earlier studies found inconclusive evidence (Corbeaneu et al., 2023; cf. Taris, 2007). Considering these results, it may be possible that linear measures of stress vs. performance might be less valid without consideration of a more sophisticated relationship between stress and performance (i.e., stress’s curvilinear relationship to performance). However, while tempting to “double down” on the relationship between stress and relationship burnout due to its strong face validity, a homeostatic measure of stress’s curvilinear relation seems more ideal, fits with well accepted theories of motivation, and converges with studies on optimal arousal and flow.

Longitudinal Demographics Not Reported

Unfortunately, due to an oversight, relationship duration had not been included in item bank construction, review, and survey implementation. The absence of relationship duration made the second hypothesis impossible to evaluate. Though this is unfortunate, it is fortuitous, for it had been more valuable to narrowly focus efforts on evaluating reliability and validity of IRTS-1 in evaluation of the first hypothesis, namely its generalizability in evaluation of overall relationship burnout regardless of relationship type. The additional time not implementing items and analyzing results allowed a shifting in research scope to run adhoc analyses for further investigation of the first hypothesis.

Why Use the IRTS?

It could be asked, why even use a scale such as the IRTS? Why not use the BMS? Unfortunately, the BMS had been developed to measure overall burnout of an individual and is not specific to relationships. Offering tools for individuals and clinicians to narrow the source of burnout may be of significant clinical value and may inform therapies as it may offer a way to identify relationships correlated/causal to burnout early and often. Clarifying burnout in relationships rather than burnout in general could allow individuals to identify maladaptive cognitions that had been based on inaccurate perceptions more specifically (e.g., “it’s the job”, “it’s the money”, “it’s the people”) for use in cognitive behavioral therapy (CBT; Beck, 2020). The same specificity may also improve history taking steps of eye movement desensitization and reprocessing (EMDR; Shapiro, 2017). For example, attributing burnout to workplaces may be misplaced and be more related to relationships with one’s cohabitant, manager, co-worker, friend, or aggregated across multiple measures of generalized relationship burnout within and between groups.

Limitations

Limitations of this study may further hurt its validity as the sample had been from university students with a small sample size (n = 30). Beyond sample size issues, it is also possible that some items may be poorly worded, and that improvements to wording may lead to less priming, which could impact subsequent responses. IRTS-1’s length of items in self-report survey could also induce response fatigue. As administered, IRTS-1 demographic questions had been placed at the beginning of the self-report survey rather than the end which may have inadvertently induced stereotype threat (see Steele & Aronson, 1995) which may evidence perceived stereotypes and difficulties in relationships (e.g., adversarial relationships with management, cultural assumptions) which may then bias item responses. Changing the order of items also may exhibit different responses due to response fatigue and differences in response due to sequence/order effects.

Improving the IRTS

 There are improvements to be made to IRTS-1 as a general measure of relationship burnout. Test items that seem irrelevant could be replaced or removed. It is also possible that the IRTS-1’s construct related to stress may be misplaced or be insufficiently measured. It is possible with increased iteration that the IRTS-1 may provide a useful tool in the clinical toolkit, and provide individuals, and dyads ways to more frequently measure their burnout between each other such that if there is a mutual interest, both parties may engage in efforts to reduce it.

Conclusion

The IRTS-1 is a possible candidate for initial measures of generalized relationship burnout, regardless of relation type. Beyond utilization of intra-dyad measures of burnout such as the IRTS-1 for personal and therapeutic use, it may be useful in industrial/organizational psychology applications (e.g., human resources, leadership, team, and peer evaluations). The ability to more frequently measure relationship burnout between team members could provide a more meaningful indicator of a core contributor of performance than more broad measures than the BMS. This would be a highly valuable tool used in candid conversations. The use of IRTS-1 offers not only a platform for discussing ways to improve relations, but also may ease the burden of preparing for relationship separation. Tools for human resources departments such as IRTS-1 may make difficult conversations easier.

 There is promise in the IRTS-1 and tools to measure generalized relationship burnout, however more work is needed to increase its reliability and validity. In terms of overall recommendations for future research, removing irrelevant test items, increasing BMS items, and solving the stress measures as it relates to optimal arousal seems the best next step. Though new scales are developed continuously in the field of psychology, the IRTS-1 may open the door to a first-generation method to measure general relationship burnout in specific relationships, with further improvements in a proposed second version as an IRTS-2. This tool is anticipated to provide a quick diagnostic for peoples to do something about intra-relationship burnout sooner, rather than later and thus may contribute toward the well-being, health, and productivity of social organization—further iteration is welcome.

References

Ahola, K., Hakanen, J., Perhoniemi, R., & Mutanen, P. (2014). Relationship between burnout and depressive symptoms: A study using the person-centered approach. Burnout Research, 1(1), 29–37. https://doi.org/10.1016/j.burn.2014.03.003

Amato. (2015). Marriage, cohabitation and mental health. Family Matters (Melbourne, Vic.), 96, 5–13.

Aust, F., Beneke, T., Peifer, C., & Wekenborg, M. (2022). The relationship between flow experience and burnout symptoms: A systematic review. International Journal of Environmental Research and Public Health, 19(7), 3865. https://doi.org/10.3390/ijerph19073865

Beck, J. S. (2020). Cognitive Behavior Therapy (3rd ed.). Guilford Press.

Beeney, J. E., Hallquist, M. N., Scott, L. N., Ringwald, W. R., Stepp, S. D., Lazarus, S. A., Mattia, A. A., & Pilkonis, P. A. (2019). The emotional bank account and the four horsemen of the apocalypse in romantic relationships of people with borderline personality disorder: A dyadic observational study. Clinical Psychological Science, 7(5), 1063–1077. https://doi.org/10.1177/2167702619830647

Bumpass, L. L., & Sweet, J. A. (1989). National estimates of cohabitation. Demography, 26(4), 615–625. https://doi.org/10.2307/2061261

Clark, L. A., & Watson, D. (2016). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. https://doi.org/10.1037/1040-3590.7.3.309

Cohen, P. H. (2021). Marriage and cohabitation. In P. H. Cohen, The family: Diversity, inequality, and social change (3rd ed., pp. 272–317). W. W. Norton & Company.

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 385-396. https://doi.org/10.2307/2136404

Cohen, S., & Williamson, G. M. (1988). Perceived stress in a probability sample of the United States. In Spacapan S. & Oskamp S. (Eds.), The social psychology of health: Claremont symposium on applied social psychology (pp. 31–67). Sage.

Corbeanu, A., Illiescu, D., Ion, A., & Spînu, R. (2023). The link between burnout and job performance: A meta-analysis. European Journal of Work and Organizational Psychology, 32(4), 599–616. https://doi.org/10.1080/1359432X.2023.2209320

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297–334. https://doi.org/10.1007/BF02310555

Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281–302. https://doi.org/10.1037/h0040957

Csikszentmihalyi, M. (1975). Beyond boredom and anxiety. Jossey-Bass Publishers.

Dobson, K., Stanton, S. C. E., Balzarini, R. N., & Campbell, L. (2023). Are you tired of “us?” Accuracy and bias in couples’ perceptions of relational boredom. Journal of Social and Personal Relationships, 40(10), 3091-3120. https://doi.org/10.1177/02654075231168141

Farnworth, L., Mostert, E., Harrison, S., & Worrel, D. (1996). The experience sampling method: Its potential use in occupational therapy research. Occupational Therapy International, 3, 1–17. https://doi.org/10.1002/oti.23

Fowler, C., & Dillow, M. R. (2011) Attachment dimensions and the four horsemen of the apocalypse. Communication Research Reports, 28(1), 16-26. https://doi.org/10.1080/08824096.2010.518910

Gottman, J., & Silver, N. (2015). The seven principles for making marriage work: A practical guide from the country’s foremost relationship expert. Harmony.

Gottman, J. M. (1993). A theory of marital dissolution and stability. Journal of Family Psychology, 7(1), 57–75. https://doi.org/10.1037/0893-3200.7.1.57

Gottman, J. M., & Krokoff, L. J. (1989). Marital interaction and satisfaction: A longitudinal view. Journal of Consulting and Clinical Psychology, 57(1), 47–52. https://doi.org/10.1037/0022-006X.57.1.47

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Prentice Hall.

Holt-Lunstad J., Smith, T. B., & Layton J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLOS Medicine 7(7), e1000316. https://doi.org/10.1371/journal.pmed.1000316

Kaplan, R. M., & Saccuzzo, D. P. (2018). Psychological testing: Principles, applications, and issues (9th ed.). Cengage.

Larson, R., & Csikszentmihalyi, M. (1983). The Experience Sampling Method. New Directions for Methodology of Social & Behavioral Science, 15, 41–56.

Loevinger, J. (1957). Objective tests as instruments of psychological theory. Psychological Reports, 3(3), 635–694. https://doi.org/10.2466/pr0.1957.3.3.635

Lord, F. M., & Novick, M. R. (1968). Statistical theories of mental test scores. Addison-Wesley.

Malach-Pines, A. (2005). The burnout measure, short version. International Journal of Stress Management, 12(1), 78–88. https://doi.org/10.1037/1072-5245.12.1.78

Manning, W. D., Joyner, K., Hemez, P., & Cupka, C. (2019). Measuring Cohabitation in U.S. National Surveys. Demography, 56(4), 1195–1218. http://www.jstor.org/stable/45199466

Maslach, C. (1996). Maslach burnout inventory manual (3rd ed.). Mindgarden.com.

Merton, R. K. (1938). Social structure and Anomie. American Sociological Review, 3, 672–682. https://doi.org/10.2307/2084686

Nakamura, J., & Csikszentmihalyi, M. (2002). The concept of flow. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 89–105). Oxford University Press.

Nejatian, M., Alami, A., Momeniyan, V., Delshad, N. A., & Jafari, A. (2021). Investigating the status of marital burnout and related factors in married women referred to health centers. BMC Womens Health, 21(1), 25. https://doi.org/10.1186/s12905-021-01172-0

Newton, I. (1687). Philosophiæ naturalis principia mathematica. Jussu Societatis Regiae ac Typis Josephi Streater. https://www.loc.gov/item/2021667054/

Perissinotto, C. M., Cenzer, I. S., & Covinsky, K. E. (2012). Loneliness in older persons: A predictor of functional decline and death. Archives of Internal Medicine, 172(14), 1078–1084. https://doi.org/10.1001/archinternmed.2012.1993

Pines, A., & Aronson, E. (1988). Career burnout: Causes and cures. Free Press.

Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78. https://doi.org/10.1037/0003-066X.55.1.68

Shapiro, F. (2017). Eye Movement Desensitization and Reprocessing (EMDR) therapy (3rd ed.). Guilford Press.

Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality & Social Psychology, 69(5), 797–811. https://doi.org/10.1037//0022-3514.69.5.797

Sternberg, R. J. (1986). A triangular theory of love. Psychological Review, 93(2), 119–135. https://doi.org/10.1037/0033-295X.93.2.119

Taris, T. W. (2007). Is there a relationship between burnout and objective performance? A critical review of 16 studies. An International Journal of Work, Health & Organizations, 20(4), 316–334. https://doi.org/10.1080/02678370601065893

Vespa, J. (2014). Historical trends in the marital intentions of one-time and serial cohabitators. Journal of Marriage and Family, 76(1), 207-217. https://doi.org/10.1111/jomf.12083

Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative Neurology and Psychology, 18, 459-482. https://doi.org/10.1002/cne.920180503

Zijlmans, E. A. O., Tijmstra, J., van der Ark, L. A., & Sijtsma, K. (2018). Item-score reliability in empirical-data sets and its relationship with other item indices. Educational and psychological measurement, 78(6), 998–1020. https://doi.org/10.1177/0013164417728358

Appendix 1: Tables

Table 1

Sample Characteristics

Variable	n	%	M	SD
Age
10–19	2	6.7	19.000	0.000
20–29	18	60	23.278	2.244
30–39	6	20	34.500	2.881
40–49	4	13.3	47.000	0.000
Gender
Male	19	63.3
Female	9	30
Non-Binary / Third Gender	2	6.7
Ethnicity
Black / African American	6	20
Hispanic	1	3.3
Other	4	13.3
I prefer not to answer	19	63.3
Relationship Type
Spouse	7	23.3
Cohabitant	9	30
Family Member	2	6.7
Coworker	8	26.7
Friend	4	13.3

Note. N = 30. Participants were on average 28.4 years old (SD = 9.065).

Table 2

Item means and standard deviations.

Question	Item	M	SD
IRTS
Q1	I feel argumentative.	2.40	0.97
Q2	I feel irritated.	2.33	0.88
Q3	I feel disconnected.	2.73	1.20
Q4	I feel bored.	2.10	1.12
Q5R*	I feel like I want to go on a date.	3.83	1.44
Q6	I feel disappointed.	2.13	1.04
Q7R*	I feel my relationship is supported by friends and family.	2.67	1.47
Q8	I feel my relationship is stigmatized publicly.	1.63	1.30
Q9R*	I feel valued in my relationship.	2.23	1.19
Q10R*	I feel like I can make my own decisions in my relationship.	2.07	1.20
Q11	I feel useless.	1.87	1.33
Q12R*	I feel competent.	1.87	1.04
Q13R*	I feel passionate.	2.70	1.37
Q14R*	I feel intimate.	3.67	1.47
Q15	I feel distant.	2.63	1.30
Q16R*	I feel committed.	2.23	1.17
Q19R*	I feel committed.	2.27	1.14
BMS
Q17	I feel helpless.	2.27	1.20
Q18	I do things that I enjoy doing.	1.97	1.10
Q20	I feel trapped.	1.90	1.24
PSS-4
Q21	In the last month, how often have you felt that you were unable to control the important things in your life?	2.00	1.05
Q22R*	In the last month, how often have you felt confident about your ability to handle your personal problems?	2.93	1.17
Q23R*	In the last month, how often have you felt that things were going your way?	2.97	0.93
Q24	In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?	2.77	1.41

* Indicates item is reverse scored.

Table 3

IRTS-1 frequentist scale reliability statistics.

Estimate		Cronbach’s α
Point estimate		0.866
95% CI lower bound		0.782
95% CI upper bound		0.924

Table 4

IRTS-1 frequentist individual item reliability statistics (item’s contribution to scale).

Item		Item-rest correlation
Q1		0.186
Q2		0.277
Q3		0.673
Q4		0.264
Q5R		0.101
Q6		0.246
Q7R		0.717
Q8		0.352
Q9R		0.609
Q10R		0.753
Q11		0.562
Q12R		0.365
Q13R		0.626
Q14R		0.510
Q15		0.755
Q16R		0.542
Q19R		0.761

Table 5

BMS frequentist scale reliability statistics.

Estimate		Cronbach’s α
Point estimate		0.911
95% CI lower bound		0.830
95% CI upper bound		0.956

Table 6

BMS frequentist individual item reliability statistics.

Item		Item-rest correlation
Q17		0.819
Q18		0.898
Q20		0.758

Table 7

PSS-4 subscale frequentist scale reliability statistics.

Estimate		Cronbach’s α
Point estimate		0.715
95% CI lower bound		0.492
95% CI upper bound		0.851

Table 8

PSS-4 frequentist individual item reliability statistics.

Item		Item-rest correlation
Q21		0.548
Q22R		0.353
Q23R		0.655
Q24		0.529

Appendix 2: Figures

Figure 1

Correlation Plot of IRTS-1 Items and BMS Subscale Items

Figure 2

Correlation Plot of IRTS-1 Items and PSS-4 Subscale Items

Figure 3

Correlation Plot of BMS and PSS-4 Subscales

Appendix 3: Adhoc AI Literature Review

When searching for literature review on articles regarding relationship burnout, flow, and optimal arousal, the following prompt had been used: “Is there literature covering the concept of optimal arousal (Yerkes-Dodson, 1908) and flow (Csikszentmihalyi, 1975) in relationships?” This prompt had been input into a new session with OpenAI ResearchGPT and ChatGPT-4. The results of this query are found in Table 9.

Table 9

OpenAI ResearchGPT and ChatGPT-4 Results on Optimal Arousal and Flow in Relationships

OpenAI ResearchGPT	OpenAI ChatGPT-4
The concept of optimal arousal, as described in the Yerkes-Dodson Law (1908), and flow, as conceptualized by Csikszentmihalyi (1975), have been extensively studied in various contexts, including leisure activities, task performance, and creativity. However, there appears to be limited direct research connecting these concepts specifically within the context of relationships. Optimal Arousal and Flow: Optimal arousal is closely associated with the experience of flow. This connection is often used to explain why people engage in leisure activities and games (Xie, 2014).Yerkes-Dodson Law and Task Performance: The Yerkes-Dodson Law describes a non-linear relationship between arousal and performance. Performance improves linearly on simple tasks and is impaired at high levels of arousal (Diamond, 2005).Flow and Creativity: The concept of flow has been linked to creativity, as demonstrated in studies on musical composition, where flow was found to influence the quality of compositions produced in group tasks (MacDonald, Byrne, & Carlton, 2006).Flow in Work and Achievement: High skill and challenge at work, which are key components of flow, are associated with greater positive mood, task interest, and performance among achievement-oriented employees (Eisenberger, Jones, Stinglhamber, Shanock, & Randall, 2005). In conclusion, while there is significant research on optimal arousal and flow in various settings, the specific application of these concepts to relationships is not directly covered in the existing literature. The studies mentioned above, however, provide insights into how these concepts might play a role in interpersonal dynamics, particularly in activities involving leisure, creativity, and work.	I conducted a search for literature on the concept of optimal arousal (Yerkes-Dodson, 1908) and flow (Csikszentmihalyi, 1975) specifically in the context of relationships. However, the search results did not yield direct references to academic or scholarly literature that explicitly discusses these concepts within the specific framework of relationships. The Yerkes-Dodson Law of optimal arousal and Csikszentmihalyi’s concept of flow are well-studied in various contexts such as psychology, education, and work performance, but their direct application or study in the context of romantic or interpersonal relationships might be less prevalent or explicitly defined in the literature. If you are looking for a more general understanding of these concepts or their application in other areas, I can certainly provide more information. Alternatively, if you have a specific aspect of relationships or a particular type of relationship in mind (such as romantic, familial, or professional), I can try to find more targeted information.
References Diamond, D. (2005). Cognitive, endocrine and mechanistic perspectives on non-linear relationships between arousal and brain function. Dose-Response, 3(1),  https://doi.org/10.2201/nonlin.003.01.001 Eisenberger, R., Jones, J., Stinglhamber, F., Shanock, L., & Randall, A. (2005). Flow experiences at work: for high need achievers alone? Journal of Organizational Behavior, 26(7), 755-775. https://doi.org/10.1002/JOB.337 MacDonald, R., Byrne, C., & Carlton, L. (2006). Creativity and flow in musical composition: an empirical investigation. Psychology of Music, 34(3), 292–306. https://doi.org/10.1177/0305735606064838
Xie, P. (2014). Optimal arousal, tourism. In J. Jafari & H. Xiao (Eds.), Encyclopedia of Tourism (pp. 1-2). Springer.

Appendix 4: IRTS-1 Scale

Table 9

Integrated Relationship Trajectory Burnout Scale, Version 1 (IRTS-1) Final Scale

Item	1Never feel this way	2Sometimes but less than half the time	3 Half the time	4 Most of the time	5 Always feel this way
1. I feel argumentative.
2. I feel irritated.
3. I feel disconnected.
4. I feel bored.
5. I feel like I want to go on a date. *
6. I feel disappointed.
7. I feel my relationship is supported by friends and family. *
8. I feel my relationship is stigmatized publicly.
9. I feel valued in my relationship. *
10. I feel like I can make my own decisions in my relationship. *
11. I feel useless.
12. I feel competent. *
13. I feel passionate. *
14. I feel intimate. *
15. I feel distant.
16. I feel committed. *
17. I feel depressed. **
18. I feel helpless. **
19. I do things that I enjoy doing. *
20. I feel trapped. **
21. In the last month, how often have you felt that you were unable to control the important things in your life? ***
22. In the last month, how often have you felt confident about your ability to handle your personal problems? */***
23. In the last month, how often have you felt that things were going your way? */***
24. In the last month, how often have you felt difficulties were piling up so high that you could not overcome them? ***

 

^*   Reverse scored.

^**   For testing criterion-related (concurrent) validity (from Burnout Measure; Short Version; Malach-Pines, 2005).

^*** For testing construct related validity (from Perceived Stress Scale 4 [PSS-4]; Cohen et al., 1983; Cohen & Williamson, 1988).

Appendix 5: BMS Scale (Malach-Pines, 2005)

Item	1 Never	2 Almost Never	3 Rarely	4 Sometimes	5 Very Often	6 Very Often	7 Always
Tired Disappointed with people Hopeless Trapped Helpless Depressed Physically weak/Sickly Worthless/Like a Failure Difficulties sleeping “I’ve had it”

 

Appendix 6: PSS-4 Scale (Cohen et al., 1983)

Item	0 Never	1 Almost Never	2 Some-times	3 Fairly Often	4 Very Often
1. In the last month, how often have you felt that you were unable to control the important things in your life?
2. In the last month, how often have you felt confident about your ability to handle your personal problems? *
3. In the last month, how often have you felt that things were going your way? *
4. In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?

 

^*   Reverse scored.