Vulnerability to “Breadcrumbing” in a Sample of Adults in the United Kingdom: A Cross-Sectional Survey Study

Abstract

Breadcrumbing is a psychologically harmful dating behavior that involves sending non-committal signals to another person and periodically feigning interest in them, despite having no intention of taking the relationship forward. This is the first empirical study to examine the correlates of breadcrumbing experiences in the United Kingdom. A sample of 544 adults in the United Kingdom indicated their age, sex, sexual orientation, and relationship status and completed measures of social support and frequency of exposure to ghosting, gaslighting, and breadcrumbing. Hierarchical multiple regression analyses showed that age and social support were negatively correlated with breadcrumbing, and that ghosting and gaslighting were positively associated with breadcrumbing. When ghosting and gaslighting were entered in the model as predictors, the effects of social support were no longer statistically significant. The findings suggest that there is a cumulative effect of exposure to victimization that can increase vulnerability to breadcrumbing and that, under these circumstances, social support may cease to operate as a protective factor. Coping may be impaired, heightening the risk of revictimization. Younger adults appear to be at higher risk of breadcrumbing.

1. Introduction

The advent of social media, instant messaging, and geospatial social networking applications has engendered greater opportunity for interpersonal connections, including dating. Indeed, in 2024, 1 in 10 online adults in the United Kingdom visited a dating application, with 1 in 20 using it on a daily basis, and 1 in 5 online adults in the United Kingdom aged 18–24 years did so (Ofcom, 2024). Yet, online dating also presents some risks, such as increased exposure to victimization behaviors, including ghosting, gaslighting, and breadcrumbing.

Ghosting refers to the act of ending a relationship with somebody suddenly by severing all contact with them without explanation, often leading the victim to question what went wrong (Freedman et al., 2024). Gaslighting refers to the act of manipulating someone into self-doubt and thus into questioning their own perceptions, reality, and, in some cases, sanity, while denying that one is actually doing this (Akdeniz & Cihan, 2024). Breadcrumbing refers to the act of sending non-committal text or social media messages or signals to a past, present, or potential dating partner to lure them into a relationship, periodically feigning interest in them, despite having no serious intention of taking the relationship forward (Navarro et al., 2020). These messages and signals may include an occasional greeting by instant messaging, a sudden suggestion to meet up after a hiatus (but subsequently not following through), or an uncharacteristic “like” on social media, all of which are intended disingenuously to lead the victim to believe that the relationship may progress. These behaviors are qualitatively distinct but share some core characteristics: they generate uncertainty about the situation; precipitate self-doubt; and undermine self-esteem (Hailes & Goodman, 2025; LeFebvre & Fan, 2020; Lopes & Jaspal, 2025).

This study focuses on vulnerability to breadcrumbing specifically. As a complex behavior characterized by high levels of uncertainty, breadcrumbing may be especially challenging to detect and aversive for mental health. While there has been some work on other forms of victimization in relationships (e.g., ghosting, gaslighting, and coercion) (Jaspal & Lopes, 2025; Lopes & Jaspal, 2025), there is limited empirical research into the social psychological underpinnings of breadcrumbing and none that has been conducted in the United Kingdom. Drawing upon the literature on revictimization and using cross-sectional survey data from a sample of 544 adults in the United Kingdom, this study examines the following:

• The potential protective role of social support against exposure to breadcrumbing since social support has been found to be protective against other harmful interpersonal behaviors, such as intimate partner violence (Plazaola-Castaño et al., 2008);
• The potential risk factors of ghosting and gaslighting (as cumulative victimization experiences) in exposure to breadcrumbing in view of evidence that prior experiences of victimization can increase the risk of subsequent revictimization (Fereidooni et al., 2024).

Furthermore, since demographic differences in exposure to victimization behaviors have been observed empirically (Elias-Lambert et al., 2014; Jaspal & Lopes, 2025; Sweet, 2019), the effects of demographic variables are also controlled for.

1.1. Social Support

Social support is defined as the belief that one can rely on others for help in times of need. This includes providing emotional support, feelings of self-esteem, a sense of belonging, and material support (S. Cohen & Hoberman, 1983). Social psychological theories, including social identity theory (Scheepers & Ellemers, 2019), stress and coping theory (Lazarus & Folkman, 1984), and the social cure perspective (Avanzi et al., 2021), highlight the significance of social support in psychological well-being. Moreover, identity process theory (Breakwell, 2021) considers the derivation of social support to be an adaptive, effective, and sustainable coping strategy when one is faced with a potential “hazard” to identity (see also Jaspal, 2025).

Breadcrumbing tends to be an ongoing experience with the perpetrator appearing in the victim’s life, disengaging and, periodically, re-engaging with them. The experience is characterized by uncertainty since the victim is deliberately kept “on standby,” and left wondering whether or not the breadcrumber is genuinely interested in them (Navarro et al., 2020). In this sense, breadcrumbing is conceptually distinct from ghosting, which, conversely, refers to the complete cessation of communication with the victim.

Social support can enable the individual to exchange confidences with a confidant, to seek advice and guidance from them, and to append meaning to events and experiences. According to the buffering hypothesis (S. Cohen & Wills, 1985), social support reduces the risk of stress-induced psychopathology. Here it is argued that social support may also reduce the risk of exposure to behaviors that the individual comes to recognize as potentially stress-inducing. When exposed to potential breadcrumbing, the victim may be supported by a confidant to recognize the behavior as such. Accordingly, social support may prevent exposure to breadcrumbing. Indeed, social support has been found to be inversely associated with other harmful dating behaviors, such as intimate partner violence (Plazaola-Castaño et al., 2008). Therefore, it is hypothesized that social support will be associated with less frequent exposure to breadcrumbing from dating partners.

1.2. (Re)victimization

Victimization is defined as the process of becoming a victim of maltreatment or aggression that causes psychological, emotional, or physical harm (Muratore, 2023). As ghosting, gaslighting, and breadcrumbing are all associated with psychological distress in victims (Lopes & Jaspal, 2025; Navarro et al., 2020), they can be regarded as acts of victimization. The concept of revictimization is defined as any subsequent victimization after an initial experience (Graham-Kevan et al., 2015), such as ghosting, gaslighting or breadcrumbing. While this study does not empirically examine revictimization per se, literature in this area is useful for understanding the potential cumulative effects of exposure to victimization behaviors.

Much research into revictimization has focused upon the impact of childhood sexual abuse upon subsequent experiences of abuse in adulthood (Fereidooni et al., 2024). Research by Graham-Kevan et al. (2015) found that exposure to a greater number of types of trauma was associated with a higher likelihood of revictimization, that is, of experiencing subsequent trauma events. Moreover, Jaspal et al. (2017) found that ethnic minority gay men who reported a childhood sexual abuse experience were more likely to report subsequent victimization based upon both their sexuality and ethnicity in adulthood (see also Evans et al., 2014).

Overall, research suggests that victims of childhood abuse are more likely to suffer from subsequent abuse during adulthood, but there is limited evidence regarding the cumulative effects of victimization experienced exclusively in adulthood. This study addresses this gap.

1.3. Cumulative Effects of Victimization

In this article, the cumulative effects of exposure to more than one form of victimization, such as ghosting, gaslighting, for breadcrumbing are examined. The study assumes no particular chronology to experiences of victimization but rather focuses on their co-occurrence.

Ghosting refers to the complete cessation of contact with an individual, while breadcrumbing refers to intermittent contact with an individual after an initial cessation of contact. Gaslighting essentially calls into question the individual’s interpretation of their experiences. Though qualitatively distinct, these types of victimization are related, e.g., the breadcrumber essentially ghosts the victim before re-appearing and they may gaslight the victim to justify their reappearance after a hiatus. As such, these victimization experiences are expected to be correlated.

There are many reasons why an individual who experiences ghosting or gaslighting may be at increased risk of breadcrumbing. For instance, social learning may precipitate in the victim a tolerance of abusive behaviors from significant others (Cochran et al., 2011). Moreover, object relations theory postulates that individuals’ (early) interactions with others provide a psychological template for anticipating and interpreting future interactions (Greenberg & Mitchell, 1983; Kernberg, 1976). The low self-esteem provoked by exposure to one form of victimization may engender the perception that subsequent experiences of victimization are to be expected or even deserved (Gilbert, 2011).

Schema therapy, which combines cognitive-behavioral techniques with interpersonal, experiential, and psychodynamic tenets (Martin & Young, 2010), was developed on the premise that revictimization following childhood abuse can be attributed to maladaptive schemas. These schemas are characterized by pervasive, dysfunctional beliefs about the self (e.g., “I am unworthy of love”) and others (“others will reject me”) and develop during childhood due to unmet psychological and emotional needs, such as the need for acceptance and recognition from significant others (Martin & Young, 2010). The schemas subsequently influence relationships in adulthood. Moreover, due to unmet needs from parents, family, and peers, the individual may develop unhealthy coping styles, e.g., self-blaming, and maladaptive beliefs about the self and others (e.g., “I am to blame for what is happening to me”), and emotions (e.g., shame and guilt). Self-blame and shame are powerful emotions that are often accompanied by self-deprecating thoughts and feelings of punishment, self-criticism, and self-deservedness, which indicate loss of social attractiveness, thus promoting revictimization (Gilbert, 2011). These emotions can preclude help-seeking and the derivation of social support in victims (Fereidooni et al., 2024). All this in turn increases vulnerability to revictimization in adulthood.

Indeed, past research has found that victims of childhood abuse are not only more prone to abuse perpetrated by their dating partners during adulthood, but they also show early maladaptive schemas, such as emotional inhibition schemas (Crawford & Wright, 2007) and disconnection/rejection schemas (Gay et al., 2013), as well as dysfunctional coping strategies, such as self-blame, anxious coping, and disengagement (Arata, 1999; Gibson & Leitenberg, 2001). A recent systematic review found dysfunctional coping strategies and negative emotions (e.g., guilt and shame) to be robust predictors of revictimization (specifically by intimate partner violence) in adulthood (Fereidooni et al., 2024). Furthermore, recent research has found support for early maladaptive schemas predicting victimization by cyberdating, which includes behaviors such as ghosting and breadcrumbing, in both men and women (Celsi et al., 2021).

Based on theory and research focusing on revictimization, it is hypothesized that individuals who have frequently been ghosted and/or gaslighted may subsequently find themselves in a situation where they are also more susceptible to breadcrumbing.

1.4. Hypotheses

• Social support should be negatively associated with breadcrumbing, while controlling for the effects of the demographic variables.
• The harmful dating behaviors of ghosting and gaslighting should be positively associated with breadcrumbing, independently of the effects of social support.

2. Materials and Methods

2.1. Design, Procedure, and Participants

This study used a cross-sectional correlational survey design. A convenience sample of 544 adults in the United Kingdom was recruited on Prolific (https://www.prolific.com/, accessed on 18 June 2025), a participant recruitment platform. The eligibility criteria were being aged 18 years or over, having had at least one dating experience, and residing in the United Kingdom. A post hoc G* power analysis was conducted (Faul et al., 2007) to evaluate statistical power. The alpha was set at 0.05; the sample size was set at N = 544; the number of predictors was set at 5; and the assumed effect size was set at 0.9, consistent with existing literature (e.g., J. Cohen, 1962, 1968, 1969; Thompson, 2007). Results showed a statistical power of 0.95. This suggests that we can safely accept the study’s hypotheses and avoid Type II errors (Dziak et al., 2020). Participants indicated their age, sex, sexual orientation, and relationship status and completed measures of social support, ghosting, gaslighting, and breadcrumbing.

The mean age of participants was 30.71 (SD = 5.70). There were 279 (51.3%) men and 265 (48.7%) women, and 287 (52.8%) self-identified as heterosexual and 257 (47.2%) as non-heterosexual. In total, 407 (74.8%) participants were in a romantic relationship and 137 (25.2%) were single.

2.2. Measures

The 12-item Interpersonal Support Evaluation List (S. Cohen & Hoberman, 1983) was used to measure social support on a 4-point Likert scale (1 = definitely false to 4 = definitely true). The scale included items such as “I feel that there is no one I can share my most private worries and fears with” (reverse-scored) and “When I need suggestions on how to deal with a personal problem, I know someone I can turn to.” A higher mean score indicated higher social support (α = 0.90).

A single item developed by Navarro et al. (2020) was used to measure the frequency of ghosting experiences. Participants read the following paragraph that described ghosting: “’Ghosting’ refers to the act of a person cutting off all communication (temporarily or permanently) with someone with whom they have some kind of relationship (e.g., friendship or romantic relationship). It is a way to break up a relationship (sudden or gradual) in which all contact with that person is cut off, or their attempts to communicate with the person who started it are ignored. They may stop responding to your phone calls or WhatsApp messages, or stop following you or block you on social media.” They then indicated how often dating partners had ghosted them in the last year using a 5-point Likert scale (0 = never to 4 = more than 5 times). A higher score indicated more frequent experiences of ghosting.

The 14-item Victim Gaslighting Questionnaire (Bhatti et al., 2023) was used to measure gaslighting experiences on a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). The scale included items such as “You often find yourself questioning your own sanity because of their words” and “You often find them denying things even when there is proof.” Participants completed the scale while thinking about a present or past dating partner. A higher mean score indicated greater experience of gaslighting (α = 0.96).

A single item developed by Navarro et al. (2020) was used to measure the frequency of breadcrumbing experiences. Participants read the following paragraph that described breadcrumbing: “’Breadcrumbing’ literally refers to leaving breadcrumbs so that someone can follow the trail. Breadcrumbers do not stop talking on WhatsApp, sending random DMs or text messages, or giving an occasional like on a social network site in order to not discard the other person at all, but the relationship does not progress. Breadcrumbing can happen after a relationship has broken down, but the breadcrumber does not wish to let their partner go. It is also a way to maintain a date on ‘hold’ and can suggest that the person is not really attracted to you but is interested in remaining relevant/attractive to others.” They then indicated how often dating partners had breadcrumbed them in the last year using a 5-point Likert scale (0 = never to 4 = more than 5 times). A higher score indicated more frequent experiences of breadcrumbing.

2.3. Statistical Analyses

Analyses were conducted using IBM SPSS Version 29. First, means and standard deviations were calculated for the continuous variables and frequencies and percentages for the categorical variables. Second, Pearson correlations were conducted to examine associations between the continuous variables. Third, independent samples t-tests bootstrapped at 1000 samples were conducted to examine differences in the continuous variables. Finally, hierarchical multiple regression analyses were conducted to examine the effects of social support, ghosting, and gaslighting upon breadcrumbing, while controlling for the effects of age and relationship status. The demographic variables of age and relationship status were entered in the first block, followed by social support in the second block, and ghosting and gaslighting in the third block. Results were considered to be statistically significant if the p-value was below 0.05.

3. Results

3.1. Descriptive Statistics

Participants reported high social support (M = 3.47, SD = 0.66) and medium exposure to gaslighting (M = 2.89, SD = 1.21). Moreover, 324 (59.6%) individuals in the sample reported ever having experienced ghosting, and 272 (50%) reported ever having experienced breadcrumbing.

3.2. Correlations

Table 1. Correlation matrix.

Variables	1	2	3	4
1. Age
2. Social support	−0.04
3. Ghosting	−0.09 *	−0.19 **
4. Gaslighting	−0.02	−0.34 **	0.33 **
5. Breadcrumbing	−0.17 **	−0.18 **	0.30 **	0.30 **

* p < 0.05, ** p < 0.001.

provides an overview of the correlations between the continuous variables in this study. As age was statistically significantly negatively correlated with breadcrumbing, its effects were controlled for in the model.

3.3. Differences in Breadcrumbing by Sex, Sexual Orientation, and Relationship Status

Independent samples t-tests showed that there were no statistically significant differences in breadcrumbing by sex (men = 0 vs. women = 1) or sexual orientation (heterosexual = 0 vs. non-heterosexual = 1). However, there were differences by relationship status (0 = single vs. 1 = partnered), t(542) = 4.219, p < 0.001, Cohen’s d = 0.42, with single people (M = 1.07, SD = 1.11) reporting more frequent breadcrumbing than partnered people (M = 0.90, SD = 0.05). Therefore, the effects of relationship status were controlled for in the model.

3.4. Hierarchical Multiple Regression Models

Hierarchical multiple regression analyses (see

Table 2. Hierarchical multiple regression models predicting breadcrumbing.

Variables	Adj. R2	F	p	β	t	p	95% CI
Model 1	0.047	14.528	<0.001
Age				−0.141	−0.306	0.001	[−0.038, −0.010]
Rel. status				−0.152	−3.563	<0.001	[−0.528, −0.153]
Model 2	0.065	13.683	<0.001
Age				−0.142	−3.355	<0.001	[−0.038, −0.010]
Rel. status				−0.115	2.510	0.008	[−0.451, −0.067]
Social support				−0.145	3.273	<0.001	[−0.337, −0.089]
Model 3	0.238	35.000	<0.001
Age				−0.129	−0.462	<0.001	[−0.035, −0.009]
Rel. status				−0.016	1.718	0.692	[−0.215, 0.143]
Social support				−0.049	2.185	0.234	[−0.189, 0.046]
Ghosting				0.370	−0.423	<0.001	[0.316, 0.493]
Gaslighting				0.156	8.735	<0.001	[0.060, 0.191]

Rel. status = relationship status.

) were conducted to examine the effects of social support, ghosting, and gaslighting upon breadcrumbing, while controlling for the effects of age and relationship status.

At Step 1, the demographic variables of age and relationship status were entered in the model. Age was statistically significantly negatively associated with breadcrumbing and being single was statistically significantly positively associated with breadcrumbing. Age and relationship status explained 5% of the variance in breadcrumbing.

At Step 2, social support was entered in the model and was statistically significantly negatively associated with breadcrumbing. Age and relationship status remained significant predictors. Social support explained an additional 2% of the variance in breadcrumbing.

At Step 3, ghosting and gaslighting were entered in the model and were both statistically significantly positively associated with breadcrumbing. Social support ceased to be a significant predictor, but age remained a significant predictor. Ghosting and gaslighting explained an additional 17% of the variance in breadcrumbing.

4. Discussion

This study set out to examine the cumulative effects of exposure to more than one form of victimization experienced in adulthood, building upon long-standing research into revictimization following childhood abuse. The hypotheses that social support would be negatively associated with breadcrumbing and that frequent exposure to ghosting and gaslighting would be positively associated with breadcrumbing were supported. However, the effects of social support were no longer significant when ghosting and gaslighting were entered in the model as predictors. These findings have implications for how victims of harmful dating behaviors, such as breadcrumbing, can be supported to cope more effectively.

The prevalence of ghosting and breadcrumbing was high in the sample, but the findings are especially relevant to younger adults, who reported more frequent exposure to breadcrumbing than older adults. This is unsurprising given the proclivity among younger adults to use dating applications (Ofcom, 2024), where such abusive behaviors are unfortunately pervasive (Timmermans et al., 2021), but concerning in view of the known link between exposure to abusive dating behaviors and poor mental health (Lopes & Jaspal, 2025) and the mental health crisis in younger adults (McGorry et al., 2025).

The results suggest that, though qualitatively distinct behaviors, ghosting, gaslighting, and breadcrumbing tend to co-occur. The cross-sectional design of the study did not provide insight into the chronology of exposure to these behaviors, i.e., whether exposure to ghosting and gaslighting precedes breadcrumbing. However, as a more complex victimization experience that incorporates elements of both ghosting and gaslighting, it is reasonable to hypothesize that breadcrumbing is predicted by ghosting and gaslighting. People who frequently experience ghosting and gaslighting appear to be at higher risk of breadcrumbing. The attacks on identity resulting from one form of victimization (e.g., ghosting or gaslighting) may diminish the individual’s resilience to withstand and resist another (e.g., breadcrumbing), increasing their vulnerability to it (see Breakwell, 2021).

Identity process theory (Breakwell, 2021) outlines a series of strategies that individuals deploy when exposed to threats to identity, that is, when feelings of self-esteem, self-efficacy, continuity, and distinctiveness are undermined by events and experiences. Social support is viewed in identity process theory as an adaptive, effective, and sustainable coping strategy. This position is supported by much empirical evidence (see Gariépy et al., 2016), including evidence that social support increases engagement in self-protective behaviors when faced with a hazard (Gifford et al., 2025; Jaspal & Breakwell, 2022). Indeed, a significant determinant of recovery following a trauma event is the presence of a social support network and people who have access to social support have a decreased likelihood of trauma symptoms (Brewin et al., 2000; Evans et al., 2014). After all, social support can mitigate distress and maladaptive behaviors through the use of adaptive problem-solving strategies—often co-created with one’s supporter (Green & Pomeroy, 2007).

Yet, the theory also notes that there may be limits to coping and that, under some circumstances, even those coping strategies that are normally adaptive, effective, and sustainable may fail to provide self-protection. The findings indicate that, when the individual is exposed to multiple types of victimization (in this case, ghosting and gaslighting), the protective effects of social support against exposure to breadcrumbing may diminish. In real terms, this suggests that the victim may not recognize breadcrumbing as such and may not heed advice, guidance, and general support from others. Put simply, they are too overwhelmed by the multi-faceted victimization that they can no longer capitalize on the available social support in a way that prevents breadcrumbing. The individual concomitantly exposed to multiple forms of victimization may internalize maladaptive, dysfunctional beliefs and emotions that inhibit self-protective action against breadcrumbing, possibly believing that this is acceptable or perhaps even deserved. In their systematic review, Fereidooni et al. (2024) found that the quality of support offered by others was an important determinant of revictimization likelihood. It may be that support needs to be framed in a way that is cognizant of the individual’s personal circumstances and identity when they experience breadcrumbing and indeed other forms of victimization.

It is worth noting that research in this area, including our own, is generally reliant upon self-report data. People indicate the extent to which they believe that they have been ghosted, gaslighted, and breadcrumbed. One’s interpretation of others’ behavior is, of course subjective and one’s experience is viewed through the lens of one’s own identity (Jaspal, 2025). It is quite possible that some people misattribute others’ behavior to ghosting, gaslighting, and breadcrumbing. Indeed, it has been noted that victimization, especially early victimization, is associated with passive-dependency and greater sensitivity to rejection in intimate relationships, as well as with decreased capacity to make rational attributions of one’s own and others’ thoughts, feelings, and behaviors and to engage in sound social judgment and self-evaluation (Ornduff, 2000; Ornduff & Kelsey, 1996; Ornduff et al., 2001). Individuals may develop an affective bias to perceive and anticipate harm in relational contexts (Ornduff, 2000). This may increase the risk not only of actual victimization but also of misinterpreting and misattributing others’ actions, whether in thought or deed.

Limitations

This study used a cross-sectional survey design and does not provide insight into causation. It also provides no insight into the chronology of exposure to victimization behaviors. Future research should use experimental and longitudinal methods as a next step. Moreover, the study did not control for early maladaptive schemas, thoughts of unlovability, self-criticism or associated negative emotions, such as guilt and shame, that may also preclude help-seeking and heighten the risk of exposure to breadcrumbing (Gilbert, 2011; Martin & Young, 2010). In particular, the mediating role of personality traits between childhood victimization and revictimization in adulthood has been observed (e.g., Ornduff et al., 2001), suggesting that this would be a worthwhile next step in this area. This is the first published empirical study of breadcrumbing in the United Kingdom, complementing existing research conducted in Argentina, India, and Spain. Given that social norms (including in relation to dating and relationships) differ by cultural context, further research into breadcrumbing in other cultural contexts would be valuable. In particular, it would be advantageous to replicate this study cross-culturally.

5. Conclusions

Victimization experiences, such as ghosting, gaslighting, and breadcrumbing, tend to co-occur, reflecting their cumulative effects. Social support may have limited efficacy in protecting against victimization when the individual is overwhelmed by multiple forms of it. Social support must be tailored to the circumstances and identity of the individual. In some cases, social support should be supplemented by therapeutic support. More generally, there is a need for greater societal awareness of the harms associated with increasingly common but harmful behaviors, such as ghosting, gaslighting, and breadcrumbing.