Early Memories of Warmth and Safeness Scale (EMWSS)

Abstract

The Early Memories of Warmth and Safeness Scale (EMWSS) is a 21-item self-report psychometric instrument developed by Anja Richter, Paul Gilbert, and Kirsten McEwan in 2009. Designed within the theoretical architecture of evolutionary psychology, attachment theory, and Compassion-Focused Therapy (CFT), the EMWSS assesses adult subjective recall of positive emotional experiences in childhood—specifically experiences of feeling secure, loved, protected, soothed, and socially connected. While traditional developmental and psychiatric inventories have historically focused on adverse childhood experiences (such as physical, emotional, or sexual abuse, neglect, and parental antipathy), the EMWSS addresses the crucial, independent dimension of early affiliative memories. Respondents evaluate 21 items using a five-point Likert scale ranging from 0 (No, never) to 4 (Yes, most of the time), yielding a unidimensional global score ranging from 0 to 84. Psychometric investigations across clinical and non-clinical cohorts consistently demonstrate excellent internal consistency (Cronbach’s α typically ranging between .95 and .97), robust test-retest reliability, and a solid single-factor structure. Convergent validity is evidenced by significant negative correlations with self-criticism, external and internal shame, anxiety, and depression, alongside strong positive associations with current experiences of social safeness, secure attachment, and self-compassion. This article delivers a comprehensive analysis of the EMWSS, detailing its theoretical foundation, psychometric architecture, clinical and research applications, structural validity, and administrative standards.

Keywords

Early Memories of Warmth and Safeness Scale, EMWSS, Compassion-Focused Therapy, affiliative emotion, attachment memories, social safeness, soothing system, childhood emotional climate, self-criticism, psychometrics

Authors

The EMWSS was authored by researchers based at the Mental Health Research Unit, Kingsway Hospital, and the University of Derby, United Kingdom:

• Anja Richter, Ph.D.: Clinical Psychologist and psychometric researcher affiliated with the Mental Health Research Unit, Kingsway Hospital, Derby, UK. Dr. Richter has conducted foundational empirical investigations into the interface between developmental memories, attachment dynamics, and clinical vulnerability.
• Paul Gilbert, Ph.D., FBPsS, OBE: Professor of Clinical Psychology at the University of Derby and founder of Compassion-Focused Therapy (CFT) and the Compassionate Mind Foundation. Professor Gilbert is internationally renowned for his pioneering work on evolutionary models of affect regulation, shame, self-criticism, social rank theory, and the biological substrates of compassion.
• Kirsten McEwan, Ph.D.: Associate Professor and researcher in mental health, psychometrics, and nature-based health interventions at the University of Derby. Dr. McEwan has co-authored numerous seminal scales and psychometric papers examining social safeness, compassion, and emotional processing.

Purpose

The primary purpose of the Early Memories of Warmth and Safeness Scale (EMWSS) is to quantify the subjective recall of feeling safe, soothed, loved, and socially accepted within one’s family and immediate relational milieu during childhood. For decades, developmental psychology and psychiatric epidemiology operated under an asymmetrical threat-centric paradigm. Inventories such as the Childhood Trauma Questionnaire (CTQ) or the Adverse Childhood Experiences (ACE) study systematically measured trauma, hostility, overt abuse, and punitive parenting. However, evolutionary psychology and affective neuroscience suggest that the human brain possesses distinct neurobiological systems for processing threat versus safeness.

Consequently, the simple absence of abuse does not imply the presence of emotional warmth, safeness, or soothing. An individual may grow up in an environment devoid of physical violence or overt hostility, yet experience profound emotional barrenness, coldness, or low affiliative connection. Conversely, some individuals exposed to socio-environmental adversity may retain vivid emotional memories of feeling intensely cherished, comforted, and protected by at least one primary caregiver. The EMWSS was specifically designed to capture this vital, often neglected dimension of early relational experience.

In clinical contexts, particularly within Compassion-Focused Therapy (CFT) and Schema Therapy, the EMWSS functions as a pivotal assessment tool. It helps clinicians discern why certain patients struggle to generate warm, soothing mental imagery or exhibit intense resistance (“fears of compassion”) when attempting self-reassurance. Patients who present with high self-criticism and severe shame often possess extremely low scores on the EMWSS. This profile reveals a neuroaffective system that was never adequately calibrated through early intersubjective soothing, making the somatic experience of warmth feel unfamiliar, frightening, or alien.

In research contexts, the EMWSS enables investigators to operationalize early positive emotional memories as a developmental antecedent, mediator, or moderator in longitudinal psychopathology studies. It facilitates empirical exploration of how affiliative memory representations buffer against psychological distress, regulate epigenetic stress-response mechanisms, and cultivate adult resilience, secure attachment, and socio-emotional competence.

Psychological Construct

The EMWSS assesses a multidimensional psychological construct consolidated into a single, cohesive affective-relational domain: the recalled subjective experience of affiliative warmth, safeness, and emotional attunement in childhood. Rather than measuring objective, historical parental behaviors (such as specific disciplinary strategies or chronological events), the instrument measures the enduring somatic and emotional resonance left by those early interactions. The core construct encompasses several interwoven psychological dimensions:

1. Social Safeness and Inner Calm

This dimension reflects feelings of tranquility, peace, and ease within one’s relational environment. Items such as item 1 (“I felt safe and secure”), item 8 (“I felt peaceful and calm”), and item 21 (“I felt at ease”) measure the baseline physiological state of low threat and high comfort. In developmental literature, this state signifies the down-regulation of the sympathetic nervous system and the activation of the parasympathetic “rest and digest” and ventral vagal social engagement system.

2. Relational Attunement and Epistemic Trust

Attunement refers to the child’s felt sense that their internal emotional world is recognized, validated, and accurately understood by caregivers. Captured by item 2 (“I felt appreciated the way I was”), item 3 (“I felt understood”), and item 5 (“I felt comfortable sharing my feelings and thoughts with those around me”), this dimension gauges the foundation of epistemic trust. It reflects the subjective confidence that one’s inner thoughts, vulnerabilities, and emotions can be disclosed safely without inviting ridicule, rejection, or emotional abandonment.

3. Intersubjective Soothing and Dyadic Regulation

Before human infants and children develop autonomous emotional self-regulation, they rely entirely on dyadic co-regulation with sensitive attachment figures. This dimension is measured through item 7 (“I knew that I could count on empathy and understanding…”), item 10 (“I could easily be soothed by people close to me when I was unhappy”), item 17 (“I knew I could rely on people close to me to console me when I was upset”), and item 20. These items assess the availability of a primary caregiver who served as a reliable “safe haven” capable of absorbing, metabolizing, and soothing distress.

4. Unconditional Acceptance and Belonging

A crucial factor in psychological resilience is the belief that one belongs and is intrinsically worthy of love, independent of performance or mistakes. Captured by item 9 (“I felt that I was a cherished member of my family”), item 14 (“I felt loved even when people were upset about something I had done”), item 16 (“I had feelings of connectedness”), and item 19 (“I had a sense of belonging”), this facet addresses the evolutionary drive to feel embedded within an affectionate social group. Unconditional acceptance buffers directly against shame, which evolutionary psychologists conceptualize as an acute fear of social devaluation and exclusion.

Theoretical Framework

The EMWSS is deeply anchored in Paul Gilbert’s evolutionary model of affect regulation, attachment theory as formulated by John Bowlby, and modern affective neuroscience, including Stephen Porges’ Polyvagal Theory.

The Tripartite Model of Affect Regulation

Gilbert’s Compassion-Focused Therapy delineates three evolved, functionally distinct emotion regulation systems:

• The Threat and Self-Protection System: Focused on identifying perils, maintaining safety, and avoiding harm. It is biologically linked to the amygdala, the hypothalamic-pituitary-adrenal (HPA) axis, and neurochemicals such as adrenaline and cortisol. Emotions associated with this system include fear, anger, disgust, and shame.
• The Drive, Seeking, and Resource-Acquisition System: Geared toward pursuing essential resources, status, mating opportunities, and goals. It is governed by the nucleus accumbens, dopaminergic pathways, and the sympathetic nervous system, eliciting excitement, exhilaration, and ambition.
• The Soothing, Affiliative, and Contentment System: Designed to promote rest, recuperation, social bonding, and cooperative attachment. It is neurochemically mediated by oxytocin and endogenous opioids (endorphins), and physiologically expressed via the myelinated ventral vagus nerve (the parasympathetic system). This system produces subjective feelings of peace, safeness, warmth, and interpersonal connection.

The EMWSS specifically measures the degree to which an individual’s early developmental landscape provided the interpersonal interactions necessary to stimulate and establish the soothing system. When children receive continuous affectionate contact, warm vocal tones, and reassuring facial expressions, their endogenous opioid and oxytocin architectures are repeatedly activated. This neurochemical conditioning provides an internal psychological template for self-soothing and social safeness later in life.

In contrast, individuals who lack early memories of warmth and safeness often present with an underdeveloped soothing system alongside a chronically hyperactive threat system. In adult life, when faced with setbacks, criticism, or relational loss, they cannot easily access internal self-comfort or soothing memories. Instead, their threat system responds with fierce internal hostility (self-criticism, self-disgust) or emotional collapse, because soothing mechanisms were never calibrated through intersubjective warmth.

Validity

Extensive psychometric investigations have established robust evidence for the construct, convergent, discriminant, and predictive validity of the EMWSS across both non-clinical and clinical populations.

Construct and Convergent Validity

In their seminal validation paper, Richter, Gilbert, and McEwan (2009) examined the relationship between the EMWSS and multiple established psychological constructs. As theoretically predicted:

• Self-Criticism and Self-Reassurance: The EMWSS demonstrated powerful, statistically significant negative correlations with inadequate self-criticism (r = -.48, p < .001) and hated self-criticism (r = -.45, p < .001) as measured by the Forms of Self-Criticising/Attacking & Self-Reassuring Scale (FSCRS). Conversely, it correlated strongly and positively with the ability to be self-reassuring (r = .50, p < .001).
• Shame: Strong inverse associations emerged between EMWSS scores and both external shame (Other As Shamer Scale; r = -.44, p < .001) and internal shame (Experience of Shame Scale; r = -.46, p < .001). Individuals lacking early warmth memories tend to view themselves as intrinsically flawed and expect others to judge or reject them.
• Attachment and Social Safeness: EMWSS scores correlate positively and robustly with adult secure attachment styles and the Social Safeness and Pleasure Scale (SSPS; r > .60, p < .001), validating that recalled childhood warmth serves as a critical foundation for adult relational security and current somatic ease.
• Parental Bonding: Validation studies assessing the Parental Bonding Instrument (PBI) demonstrate strong positive correlations between the EMWSS and parental care dimensions (maternal and paternal care, r = .65 to .75), alongside robust negative correlations with parental overprotection, coldness, and neglect.

Discriminant Validity

Discriminant validity analyses confirm that the EMWSS does not simply measure the inverse of childhood abuse or general negative affectivity. Studies utilizing both the Childhood Trauma Questionnaire (CTQ) and the EMWSS demonstrate that while childhood abuse correlates negatively with EMWSS scores, the two constructs exhibit distinct variance. Hierarchical multiple regression models reveal that the EMWSS accounts for unique, incremental variance in depression, anxiety, and self-compassion even after controlling for objective markers of physical, sexual, or emotional trauma. This confirms that the presence of early warmth is an independent protective factor, rather than merely the mirror image of trauma.

Predictive and Clinical Validity

Cross-sectional and longitudinal clinical research shows that low EMWSS scores predict elevated vulnerability to major depressive disorder, generalized anxiety, borderline personality organization, and complex post-traumatic stress disorder (CPTSD). In psychotherapy research, EMWSS baseline scores predict treatment trajectory in Compassion-Focused Therapy. Individuals entering treatment with critically depressed EMWSS scores (< 25) frequently display intense early therapeutic resistance termed “fears of compassion.” For these individuals, activating affiliative affects initially triggers grief, longing, or acute threat responses, requiring clinicians to proceed slowly with targeted somatic down-regulation.

Reliability

The EMWSS exhibits exceptional reliability metrics across diverse demographic groups, cultural translations, and clinical samples:

• Internal Consistency: In the original validation study by Richter et al. (2009) comprising non-clinical adult samples, the scale yielded an exceptional Cronbach’s alpha coefficient of α = .97. Subsequent cross-cultural validation studies have replicated these findings:
  • Portuguese adult clinical and non-clinical validations (Matos et al., 2013) reported Cronbach’s alphas of α = .96 and α = .97.
  • Portuguese adolescent adaptations (Marta-Simões et al., 2017) demonstrated internal consistency of α = .95.
  • Spanish, Japanese, and Turkish adaptations have consistently observed Cronbach’s alpha coefficients between .94 and .97, and McDonald’s omega (ω) values exceeding .95, indicating that all 21 items reliably index the shared latent construct.
• Test-Retest Reliability: Longitudinal reliability assessments across temporal intervals ranging from four to eight weeks show strong temporal stability. In a sample of non-clinical adults re-tested after six weeks, the intraclass correlation coefficient (ICC) was r = .84 (p < .001), indicating that recalled emotional memories of early warmth remain relatively stable cognitive-affective representations in adulthood when measured outside of major therapeutic interventions.
• Item-Total Correlations: Corrected item-to-total correlation coefficients for the 21 items consistently exceed .60, with the majority falling between .68 and .85. These metrics confirm high item homogeneity with no problematic or poorly discriminating items.

Factor Analysis

Structural evaluations of the EMWSS through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) provide robust evidence for a clear unidimensional structure.

Exploratory Factor Analysis (EFA)

In the original scale development study (Richter et al., 2009), principal axis factoring conducted on the initial item pool revealed a clear, single dominant factor. The scree plot demonstrated a pronounced inflection after the first factor, with the primary factor accounting for approximately 60.5% to 64.2% of the total variance. Factor loadings for all 21 retained items were uniformly high, ranging from .62 to .88, with no substantial cross-loadings or fragmentation into secondary dimensions. Items tapping feelings of being loved (e.g., item 11, “I felt loved”) and feelings of safety (e.g., item 1, “I felt safe and secure”) yielded the strongest primary loadings (.82 to .88).

Confirmatory Factor Analysis (CFA)

Subsequent psychometric examinations using CFA have repeatedly validated the single-factor model across diverse populations. Standard structural equation modeling indicates that a one-factor solution provides a good-to-excellent fit to empirical data, provided error covariances between syntactically adjacent items with minor semantic overlap (e.g., item 10 and item 17, both relating to consolation/soothing when unhappy) are appropriately specified:

• Comparative Fit Index (CFI): Values consistently range between .93 and .97 across validation studies, well above the conventional .90 threshold for acceptable model fit.
• Tucker-Lewis Index (TLI): Typically ranges between .92 and .96.
• Root Mean Square Error of Approximation (RMSEA): Generally ranges from .048 to .068 (with 90% confidence intervals spanning .041 to .075), confirming minimal residual error.
• Standardized Root Mean Square Residual (SRMR): Commonly observed below .045.

Measurement invariance testing across gender and age cohorts supports full metric and scalar invariance. This indicates that men and women interpret the 21 items in a mathematically equivalent manner, permitting unbiased cross-group comparisons of latent means.

Instrument / Measurement Tool

Below is the structural summary of the EMWSS as standardized in psychometric literature:

• Test Type: Retrospective self-report psychometric questionnaire.
• Target Respondent Group: Adults (aged 18 and older); validated adaptations also exist for adolescents (aged 12 to 18).
• Total Number of Items: 21 items.
• Response Format: 5-point Likert-type scale scored from 0 to 4:
  • 0 = No, never
  • 1 = Yes, but rarely
  • 2 = Yes, sometimes
  • 3 = Yes, often
  • 4 = Yes, most of the time
• Scoring Methodology: The scale is scored by simply summing the numerical values assigned to all 21 items. All items are positively phrased; there are no reverse-scored items.
• Score Range: Minimum possible score = 0; Maximum possible score = 84.
• Score Interpretation Guidelines:
  • Very Low Recalled Warmth (0 – 27): Indicates severe childhood emotional barrenness, lack of felt security, or absence of soothing figures. Strongly correlated with chronic shame, harsh self-criticism, fears of compassion, and clinical psychopathology.
  • Low to Moderate Recalled Warmth (28 – 48): Indicates inconsistent, conditional, or fragmented experiences of early safeness and attunement. Individuals may access self-soothing under low stress but easily collapse into self-doubt or threat responses under interpersonal strain.
  • Moderate to High Recalled Warmth (49 – 68): Reflects a generally solid, dependable developmental foundation characterized by emotional responsiveness, family belonging, and regular soothing.
  • Very High Recalled Warmth (69 – 84): Demonstrates an exceptionally rich reservoir of early affiliative memories. Highly predictive of secure adult attachment, robust emotional resilience, mature self-compassion, and low baseline vulnerability to affective disorders.
• Administration Duration: Approximately 3 to 5 minutes.

Permissions & Fee and Test Year

The Early Memories of Warmth and Safeness Scale (EMWSS) was published in 2009 by Anja Richter, Paul Gilbert, and Kirsten McEwan. In keeping with the scientific ethos of Professor Paul Gilbert and the Compassionate Mind Foundation, the EMWSS is an open-access psychometric instrument. It is freely available for clinical, educational, and academic research purposes without fee, provided that appropriate scholarly attribution is accorded to the original authors and publication. Commercial reproduction, inclusion in paid software suites, or corporate commercial deployment requires formal licensing permission from the copyright holders and the original publisher (The British Psychological Society / John Wiley & Sons, Ltd.). Researchers can access the official instrument and accompanying research materials via the University of Derby repository or the Compassionate Mind Foundation official portal.

References

The following peer-reviewed publications document the empirical validation and clinical application of the EMWSS:

• Bowlby, J. (1982). Attachment and loss: Vol. 1. Attachment (2nd ed.). Basic Books.
• Gilbert, P. (2009). Introducing compassion-focused therapy. Advances in Psychiatric Treatment, 15(3), 199–208. https://doi.org/10.1192/apt.bp.107.005264
• Gilbert, P. (2014). The origins and nature of compassion focused therapy. British Journal of Clinical Psychology, 53(1), 6–41. https://doi.org/10.1111/bjc.12043
• Marta-Simões, J., Ferreira, C., & Mendes, A. L. (2017). Exploring the psychometric properties of the Early Memories of Warmth and Safeness Scale in Portuguese adolescents. Journal of Child and Family Studies, 26(8), 2092–2101. https://doi.org/10.1007/s10826-017-0726-y
• Matos, M., Pinto-Gouveia, J., & Duarte, C. (2013). When I don’t feel safe in my own skin: The role of early memories of warmth and safeness on psychopathology. Psychology and Psychotherapy: Theory, Research and Practice, 86(3), 288–304. https://doi.org/10.1111/j.2044-8341.2012.02065.x
• Porges, S. W. (2011). The polyvagal theory: Neurophysiological foundations of emotions, attachment, communication, and self-regulation. W. W. Norton & Company.
• Richter, A., Gilbert, P., & McEwan, K. (2009). Development of an early memories of warmth and safeness scale and its relationship to psychopathology. Psychology and Psychotherapy: Theory, Research and Practice, 82(2), 171–184. https://doi.org/10.1348/147608308X395000

Items of the Scale