Attachment theory proposes that early childhood experiences with caregivers lead individuals to form internal working models about relationships that guide perceptions and behaviors in close relationships throughout life (Bowlby, 1969; Fraley, 2019).
On the one hand, the assumption that working models function as trait-like representations implies that attachment styles should be fairly resistant to change, reverting back toward initial “set points” even after disruptions (Fraley & Brumbaugh, 2004).
On the other hand, formulations of working models as continually updated based on relational experiences suggest that major events could potentially alter attachment representations permanently if the experiences are impactful enough (Davila & Sargent, 2003).
Fraley, R. C., Gillath, O., & Deboeck, P. R. (2021). Do life events lead to enduring changes in adult attachment styles? A naturalistic longitudinal investigation. Journal of Personality and Social Psychology, 120(6), 1567–1606. https://doi.org/10.1037/pspi0000326
Key Points
- The study examined whether 25 different life events lead to immediate and/or enduring changes in adult attachment styles.
- About half of the events were associated with immediate changes in attachment, but only a quarter were associated with enduring changes.
- There were significant individual differences, with some people changing more than others following events.
- How people construed events was related to whether they showed enduring change in attachment.
Rationale
The study explains that attachment theory proposes mechanisms for both stability and change in attachment styles, but does not clearly specify whether experiences tend to lead to transient or enduring changes (Fraley et al., 2011).
Most prior longitudinal research on life events and adult attachment has simply focused on comparing individuals who did or did not experience a given event at one point in time (Gillath et al., 2016).
This between-persons approach cannot address whether events shape within-person changes in attachment security enduringly versus transiently.
Other studies have examined test-retest correlations in attachment security over time. But high stability estimates can mask enduring changes for subgroups of individuals (Karantzas et al., 2019).
This study uses a longitudinal interrupted time-series design with a within-person analysis to assess trajectories before and after events, overcoming limitations in previous research.
The design allows for examining both short-term and long-term changes in response to major life events.
Method
Attachment was measured using the ECR-RS scale on four dimensions: global/romantic avoidance/anxiety.
Life events were measured with a checklist of 25 events spanning domains like:
- Health (‘I was ill or sick’),
- Work (‘I started a new job’),
- Relationships (‘I started dating someone new’),
- Family (‘I found out that I or my partner was pregnant’),
- Location (‘I moved to a new location’),
- Leisure (‘I took a vacation or went on holiday’).
For each event experienced, participants rated whether it was positive/rewarding or stressful.
Sample
The data come from the yourPersonality longitudinal project that assessed adult attachment and life experiences approximately monthly over periods from 6 months to 3 years.
The sample included 4,920 participants with 3-24 waves of data.
In terms of demographics, the sample was predominantly white (74%) women (82%). The average age was 35 years. In terms of relationship status at the initial assessment, 39% were single, 35% dating, and 25% married. Education levels ranged from some high school (1%) to graduate degrees (28%).
For the analyses examining change in response to specific life events, subset samples were created for each of the 25 events. These “event-based samples” included participants who had experienced the given event at least once and completed a minimum of 3 assessment waves before and after the event.
The size of these subset samples varied widely across events based on how many participants met those criteria. For example, the subset for the event “I was ill or sick” included 832 participants, while the subset for the event “I retired” only included 14 participants.
This sample was used to analyze general patterns of stability and change in attachment styles over time.
Statistical Analysis
The key analysis involved fitting longitudinal mixed effects models to estimate attachment trajectories before and after events.
Separate models were run for each of the 25 events and for each attachment outcome (global/romantic avoidance/anxiety).
Results
Both immediate and enduring changes in attachment were observed in response to major life events, clarifying important attachment dynamics. But enduring change was less common on average than transient change.
Around half of events were associated with short-term/immediate changes in attachment, but only a quarter were related to enduring changes.
For example, starting a new relationship immediately decreased insecurity but no enduring changes. Having a conflict with a partner increases insecurity immediately and enduringly.
Immediate Changes
- About half of the 25 life events studied were associated with immediate changes in attachment styles. For example, starting a new relationship led to decreases in attachment insecurity.
- On average, immediate effects were small to moderate in size (about 0.05 SDs change for avoidance, 0.13 SD for anxiety).
- Some findings were counterintuitive – for example, being apart from one’s partner was related to decreases in anxiety on average.
Enduring Changes
- About a quarter of the events were related to enduring changes in attachment styles. For example, having a conflict with one’s partner was associated with enduring increases in romantic attachment insecurity.
- On average, enduring effects were small in size (under 0.10 SDs).
- There were significant individual differences in enduring change. The likelihood of changing by more than 1/3 SD ranged from 5% to 86% depending on the event.
- Perceiving events more positively was associated with greater secure shifts in attachment.
General Change Patterns
- On average, people tended to become less insecure (especially less anxious) over time regardless of events. This could be due to biases, effects of participating in the study, or other processes.
Selection Effects
- People higher in attachment insecurity were more likely to eventually experience certain “destabilizing” events like relationship breakups.
Insight
The study clarifies the important conceptual and empirical distinctions between short-term versus long-term changes in attachment.
It overcomes significant limitations in the existing literature concerning the impact of life events on attachment dynamics.
The intensive longitudinal within-person interrupted time series design allows stronger causal inferences about the effects of different life events.
Strengths
- Large sample size with very intensive repeated measurements over long time frames
- Naturalistic measurement of life events as they organically occurred
- Examination of broad range of events beyond typical attachment-relevant events
- Separation of short-term vs enduring changes
- Analysis of individual differences in change
Limitations
- Low base rates for some events like retirement, limiting conclusions
- Sample is not fully representative (mostly white women)
- Can’t make very long-term forecasts about enduring change
- There is no data on experiences like therapy that could impact attachment
- Limitations on causal inferences without controlled experiments
Conclusions
The study reveals how life events can perturb attachment trajectories, sometimes transiently and sometimes enduringly.
There appear to be complex dynamics between socialization processes that can change attachment and selection effects whereby attachment shapes experiences.
More research is needed to understand moderators of change, like event construals and individual differences. But the intensive longitudinal design provides a strong methodological foundation for studying attachment dynamics.
Fraley, R. C., Gillath, O., & Deboeck, P. R. (2021). Do life events lead to enduring changes in adult attachment styles? A naturalistic longitudinal investigation. Journal of Personality and Social Psychology, 120(6), 1567–1606. https://doi.org/10.1037/pspi0000326
References
Bowlby, J. (1969). Attachment. Attachment and loss: Vol. 1. Loss. New York: Basic Books.
Davila, J., & Sargent, E. (2003). The meaning of life (events) predicts changes in attachment security. Personality and Social Psychology Bulletin, 29(11), 1383-1395.
Fraley, R. C. (2019). Attachment in adulthood: Recent developments, emerging debates, and future directions. Annual Review of Psychology, 70, 401–422.
Fraley, R. C., & Brumbaugh, C. C. (2004). A dynamical systems approach to conceptualizing and studying stability and change in attachment security. In W. S. Rholes & J. A. Simpson (Eds.), Adult attachment: Theory, research, and clinical implications (pp. 86–132). New York: Guilford Press.
Fraley, R. C., Gillath, O., & Deboeck, P. R. (2021). Do life events lead to enduring changes in adult attachment styles? A naturalistic longitudinal investigation. Journal of Personality and Social Psychology, 120(6), 1567–1606. https://doi.org/10.1037/pspi0000326
Fraley, R. C., Roisman, G. I., & Haltigan, J. D. (2013). The legacy of early experiences in development: Formalizing alternative models of how early experiences are carried forward over time. Developmental Psychology, 49(1), 109–126. https://doi.org/10.1037/a0027
Gillath, O., Karantzas, G., & Fraley, R. C. (2016). Adult attachment: A concise guide to theory and research. Academic Press.
Karantzas, G. C., Deboeck, P. R., Gillath, O., & Fraley, R. C. (2019). Stability and change in attachment styles: A focus on the individual and life events and experiences that moderate change over time. Personality and Social Psychology Review. Manuscript submitted for publication.
