Parental Bonding Instrument (PBI)

1. Abstract

The Parental Bonding Instrument (PBI), developed in 1979 by Gordon Parker, Hilary Tupling, and L. B. Brown at the University of New South Wales, is one of the most widely utilized and thoroughly validated psychometric self-report instruments in psychiatric epidemiology, developmental psychopathology, and clinical psychology. Designed to retrospectively assess an individual’s perceptions of the parenting styles and interpersonal behaviors exhibited by their mother and father during their first sixteen years of life, the PBI operationalizes parental behavior along two fundamental, orthogonal dimensions: Parental Care (ranging from warm, affectionate, empathic, and responsive parenting to emotional coldness, indifference, and rejection) and Parental Overprotection or Control (ranging from intrusive control, infantilization, psychological over-involvement, and prevention of independent behavior to the encouragement of autonomy and self-reliance).

The instrument consists of 25 self-administered items completed separately for each parental figure. Responses are recorded on a four-point Likert-type scale spanning “Very like”, “Moderately like”, “Moderately unlike”, and “Very unlike”. Twelve items evaluate the Care dimension (yielding scores from 0 to 36), while thirteen items measure Overprotection/Control (yielding scores from 0 to 39). By applying empirical cut-off scores or median splits to both dimensions, respondents can be categorized into four distinct parenting quadrants: Optimal Parenting (high care, low overprotection), Affectionate Constraint (high care, high overprotection), Neglectful Parenting / Absent Bonding (low care, low overprotection), and Affectionless Control (low care, high overprotection).

Psychometrically, the PBI exhibits excellent internal consistency (Cronbach’s alpha coefficients typically ranging from .88 to .93 for Care and .74 to .85 for Overprotection) and exceptional long-term temporal stability, demonstrated prospectively over intervals of up to two decades without significant distortion by adult psychiatric episodes or mood states. The instrument has been translated into more than twenty languages and adapted globally, providing foundational empirical evidence connecting early developmental adversity—particularly “affectionless control”—to heightened adult vulnerability for major depressive disorder, anxiety disorders, personality pathologies, eating disorders, and suicide risk.

2. Keywords

Parental Bonding Instrument, PBI, parental care, parental overprotection, affectionless control, attachment theory, developmental psychopathology, retrospective assessment, psychometrics, Gordon Parker

3. Authors

The Parental Bonding Instrument was authored by a team of clinical psychiatric and psychological researchers based at the University of New South Wales (UNSW) in Sydney, Australia:

• Gordon Parker, AO, MD, PhD, DSc, FRANZCP: Scientia Professor of Psychiatry at the University of New South Wales and Founder of the Black Dog Institute. Professor Parker is a globally recognized authority on mood disorders, affective neuroscience, and developmental clinical psychology.
• Hilary Tupling, BA, DipPsych: Research Psychologist and clinical associate at the School of Psychiatry, University of New South Wales, who contributed significantly to the original item selection, psychometric reduction, and factor analytic validation of the scale.
• L. B. Brown, MA, PhD, DipEd, FBPSS: Professor of Psychology at the School of Applied Psychology, University of New South Wales, specializing in social psychology, psychometrics, and measurement theory.

Correspondence regarding the original development and licensing of the instrument is historically directed to Scientia Professor Gordon Parker at the Black Dog Institute, Hospital Road, Prince of Wales Hospital, Randwick, NSW 2031, Australia.

4. Purpose

The primary purpose of the Parental Bonding Instrument is to provide a standardized, objective, and retrospectively reliable assessment of fundamental parental attitudes and behaviors as experienced by a child during their formative developmental years (specifically up to the age of 16). Prior to the publication of the PBI in 1979, the psychiatric literature lacked a standardized, empirically derived dimensional measure capable of quantifying early relational dynamics between children and their primary caregivers. Existing parent-child interaction paradigms were either labor-intensive observational rubrics unsuitable for adult epidemiological investigations or psychodynamically oriented interview schedules characterized by low inter-rater reliability and ambiguous scoring.

Parker and colleagues engineered the PBI to bridge the gap between empirical attachment paradigms and clinical epidemiology. Specifically, the scale serves several interconnected clinical and investigative functions:

• Etiological Investigation in Developmental Psychopathology: The PBI provides researchers with continuous parametric data to examine how early adverse relational environments correlate with subsequent biological and psychological vulnerabilities. Extensive epidemiological research has demonstrated that adverse bonding configurations act as distal, non-specific vulnerability factors for affective disorders, social anxiety disorder, generalized anxiety disorder, obsessive-compulsive disorder, borderline personality disorder, and substance abuse.
• Clinical Case Formulation and Psychotherapy: In individual psychotherapy, schema therapy, and cognitive-behavioral interventions, the PBI acts as a diagnostic and psychoeducational tool. It illuminates maladaptive core schemas, internalized relational scripts, and developmental deficits that stem from parental coldness, neglect, or enmeshment. Identifying a history of “affectionless control” helps clinicians anticipate therapeutic alliance ruptures, transference dynamics, and deep-seated fears of abandonment or engulfment.
• Genetic and Environmental Deconstruction: In behavioral genetics and twin studies (such as the landmark investigations conducted by Kenneth Kendler and colleagues), the PBI has been employed to separate shared environmental factors from non-shared environmental and genetic influences on personality development and psychiatric vulnerability.
• Evaluation of Retrospective Memory Integrity: Because the PBI asks adult patients to recall their childhood up to age 16, extensive methodological research has validated whether current emotional state or state depression creates “mood-congruent recall bias.” Evidence consistently demonstrates that PBI scores remain remarkably stable and reflect true historical perceptions rather than acute psychiatric state distortions.

5. Psychological Construct

The Parental Bonding Instrument measures parenting behavior across two independent, bipolar dimensional axes derived from extensive factor analytic research in parent-child interactions:

1. The Care Dimension

The Care dimension assesses the degree of emotional warmth, affection, empathy, availability, and loving responsiveness provided by the parent versus emotional distance, rejection, neglect, and coldness. The dimension is represented by 12 items (six positively worded and six negatively worded/reverse-coded).

• High Care: Characterized by verbal and non-verbal warmth, unconditional positive regard, active listening, comfort in times of distress, and validation of the child’s emotional state. Examples include speaking in a warm and friendly voice, being physically and verbally affectionate, smiling frequently, and demonstrating genuine empathy toward the child’s worries.
• Low Care (Coldness/Rejection): Characterized by affective detachment, harsh criticism, absence of praise, relational neglect, and emotional unavailability. The child experiences an atmosphere where they feel unwanted, misunderstood, ignored, or actively rejected.

2. The Overprotection / Control Dimension

The Overprotection (or Control) dimension evaluates the degree to which a parent exerts intrusive control, restricts psychological and behavioral autonomy, encourages dependency, and discourages independent exploration versus fostering autonomy, self-determination, and personal responsibility. The dimension is measured by 13 items (six positively worded denoting control/overprotection and seven negatively worded denoting encouragement of independence).

• High Overprotection: Characterized by infantalization (“tending to baby”), boundary violations, invasion of privacy, overbearing behavioral control, prevention of socialization with peers, and fostering systemic helplessness (e.g., communicating that the child cannot look after themselves without the parent).
• Low Overprotection (Autonomy Granting): Characterized by granting age-appropriate freedom, encouraging independent decision-making, respecting personal boundaries, and supporting the child’s natural process of individuation.

The Four Parenting Quadrants

By mapping an individual’s Care score on the vertical axis and Overprotection score on the horizontal axis using established population cut-off scores or sample medians, Parker established four distinct parental typologies:

6. Theoretical Framework

The Parental Bonding Instrument is grounded in a convergence of classical developmental theories, behavioral taxonomy, and psychodynamic attachment models:

1. Bowlby’s Attachment Theory

The primary conceptual foundation of the PBI is John Bowlby’s Attachment Theory (1969, 1973, 1980). Bowlby posited that infants and developing children rely on primary caregivers as a “secure base” from which to explore the environment and a “safe haven” to which they can return for comfort and regulation during times of threat. The Care dimension directly captures the caregiver’s capacity to serve as a warm, responsive safe haven. Conversely, the Overprotection dimension reflects an inhibition of the secure base function: by restricting exploration and infantilizing the child, overprotective parents undermine the child’s development of mastery, agency, and self-efficacy.

Furthermore, early parent-child interactions coalesce into Internal Working Models (IWMs)—cognitive-affective representational templates of self and other. Children exposed to low care and high overprotection develop internal working models depicting the self as inherently unlovable, incompetent, and vulnerable, while viewing the relational world as rejecting, unpredictable, or suffocating.

2. Dimensional Models of Parenting Style

Parker and colleagues drew extensively upon the early factor-analytic child psychology literature, notably the foundational work of Earl S. Schaefer (1959, 1965) and Wesley C. Becker (1964). Schaefer analyzed maternal behavior and established that the vast spectrum of child-rearing practices could be parsimoniously mapped onto a circumplex organized by two primary orthogonal axes: Acceptance vs. Rejection (corresponding to the PBI Care dimension) and Psychological Autonomy vs. Psychological Control (corresponding to the PBI Overprotection dimension). Becker synthesized cross-cultural behavioral data confirming that these two dimensions account for the vast majority of observed variance in parent-child interactions across diverse demographic cohorts.

3. Cognitive Vulnerability and Schema Theories

The PBI integrates smoothly with modern cognitive paradigms, such as Aaron T. Beck’s cognitive theory of depression and Jeffrey Young’s Schema Therapy framework. In Young’s model, early toxic relational experiences directly generate Early Maladaptive Schemas (EMS). A low-care maternal or paternal environment contributes directly to schemas in the Disconnection and Rejection domain (e.g., Emotional Deprivation, Abandonment/Instability, Defectiveness/Shame). High overprotection fosters schemas within the Impaired Autonomy and Performance domain (e.g., Dependence/Incompetence, Vulnerability to Harm, Enmeshment/Undeveloped Self).

7. Validity

The psychometric validity of the Parental Bonding Instrument has been scrutinized and confirmed across hundreds of empirical investigations over more than four decades:

1. Construct and Structural Validity

Construct validity was initially established by Parker et al. (1979) by correlating PBI scores with independent, concurrent ratings made by subjects’ parents, as well as blinded psychiatric interview assessments. Parental self-ratings on identical dimensional constructs demonstrated moderate-to-high correlations with their adult offspring’s PBI scores ($r = .55$ to $.70$), indicating that adult offspring reports reflect objective parental styles rather than pure internal projection. Cross-informant concordances between adult siblings rating the same parents independently have yielded robust correlations (typically $r = .50$ to $.65$), confirming that the instrument captures enduring family dynamics and shared environmental realities.

2. Convergent and Discriminant Validity

The PBI demonstrates high convergent validity with other standardized relational and attachment measures:

• It correlates strongly with the Adult Attachment Interview (AAI): individuals with insecure, preoccupied, or unresolved attachment classifications consistently report significantly lower parental care and higher parental overprotection compared to secure counterparts.
• It shows robust convergence with the EMBU (Egna Minnen Beträffande Uppfostran; My Memories of Upbringing), with the PBI Care dimension aligning tightly with the EMBU Emotional Warmth factor, and Overprotection aligning with Rejection and Overinvolvement.
• Discriminant validity is supported by low-to-negligible correlations with unrelated psychological constructs such as intelligence, socioeconomic background, and general cognitive abilities.

3. Predictive Validity in Psychiatric Epidemiology

The predictive validity of the PBI is exceptionally well documented across psychiatric disorders:

• Major Depressive Disorder (MDD): Numerous longitudinal and case-control studies have identified maternal and paternal “affectionless control” as a strong risk factor for unipolar depression, early-onset dysthymia, and treatment-resistant depressive states. Parker (1983) and subsequent meta-analyses demonstrated that individuals in the affectionless control quadrant carry an odds ratio (OR) between 2.5 and 4.0 for developing major depression compared to those experiencing optimal parenting.
• Anxiety and Panic Disorders: Overprotection has emerged as a pronounced specific risk factor for agoraphobia, social anxiety disorder, and obsessive-compulsive disorder. Longitudinal cohort studies demonstrate that children raised by highly overprotective, intrusive parents exhibit compromised habituation to fear stimuli and heightened behavioral inhibition.
• Eating Disorders and Personality Pathology: In populations with anorexia nervosa and bulimia nervosa, elevated maternal and paternal overprotection scores are ubiquitous, reflecting systemic family enmeshment and suppressed developmental autonomy. Furthermore, borderline and avoidant personality disorders correlate systematically with extreme scores on low care and high control.

8. Reliability

The reliability of the Parental Bonding Instrument has been established across multiple operational parameters, confirming its robustness as an enduring psychometric measure:

1. Internal Consistency

Across diverse international clinical and non-clinical samples, the PBI exhibits strong internal consistency:

• Care Subscale: Cronbach’s alpha coefficients routinely fall between $\alpha = .88$ and $.93$ for maternal care, and $\alpha = .89$ and $.94$ for paternal care. Split-half reliability coefficients for the Care scale consistently exceed $.88$.
• Overprotection / Control Subscale: Cronbach’s alpha values typically range from $\alpha = .74$ to $.85$ for both maternal and paternal ratings. The slightly lower alpha relative to the Care dimension reflects the multidimensional nature of control, which captures both behavioral constraint and psychological/emotional infantilization.

2. Short- and Intermediate-Term Test-Retest Reliability

Initial reliability trials by Parker, Tupling, and Brown (1979) over a 3-week interval yielded Pearson test-retest correlation coefficients of $r = .76$ for Care and $r = .63$ for Overprotection. Subsequent evaluations by Wilhelm and Parker (1990) over intervals spanning several months to several years demonstrated retest correlations ranging between $r = .70$ and $.87$.

3. Long-Term Stability Across Two Decades

In a landmark 20-year prospective study published by Wilhelm, Niven, Parker, and Hadzi-Pavlovic (2005), a non-clinical cohort of subjects completed the PBI at baseline and again 20 years later. Pearson correlation coefficients over the two-decade interval remained remarkably high:

• Maternal Care: $r = .61$
• Maternal Overprotection: $r = .65$
• Paternal Care: $r = .74$
• Paternal Overprotection: $r = .65$

Crucially, structural equation modeling demonstrated that intercurrent major depressive episodes, life crises, or changes in personal marital status did not systematically distort retrospective scores, establishing that the PBI captures stable cognitive representations of childhood parenting rather than state-dependent mood artifacts.

9. Factor Analysis

The underlying factor structure of the Parental Bonding Instrument has been an area of active psychometric investigation and debate:

1. Original Exploratory Factor Analysis (Parker et al., 1979)

The original psychometric derivation began with an extensive pool of 114 parent-behavior items administered to 150 undergraduate students and clinical outpatients. Following initial item reduction, principal components factor analysis with varimax orthogonal rotation yielded a clear two-factor solution that accounted for the majority of common variance:

• Factor I (Care): Items reflecting warm communication, affection, empathy, understanding, and positive regard loaded positively on this factor (e.g., Item 1: .76; Item 6: .83; Item 12: .79), whereas items denoting coldness, indifference, and emotional neglect loaded negatively (e.g., Item 4: -.73; Item 16: -.78).
• Factor II (Overprotection / Control): Items capturing intrusion, restriction of freedom, infantilization, and control loaded positively on this factor (e.g., Item 9: .68; Item 10: .65; Item 23: .72), while items measuring the promotion of autonomy and behavioral independence loaded negatively (e.g., Item 21: -.70; Item 22: -.66; Item 25: -.62).

2. Alternative Factor Solutions (Three-Factor Models)

While the original two-factor model remains standard in international research, several subsequent confirmatory factor analyses (CFA) have suggested that a three-factor solution may provide a superior fit to empirical data in certain populations:

• Kendler’s Model (1996): In a large population-based sample of adult twins, Kenneth Kendler identified that the 13-item Overprotection scale cleanly bifurcates into two distinct sub-dimensions: Protectiveness (genuine parental concern and monitoring) and Authoritarianism / Intrusion (pathological invasion of boundaries and coercive control). The Care dimension remained completely intact as a robust unitary construct.
• Cubis et al. (1989) & Murphy et al. (1997): Structuring the instrument into three latent factors—Care, Encouragement of Autonomy, and Overprotection / Control—often yields superior model fit indices in structural equation models ($CFI > .94, RMSEA < .05$).

Despite these alternative configurations, Gordon Parker and colleagues have maintained that the classical two-factor orthogonal model possesses superior clinical utility, ease of interpretation, and epidemiological comparability across disparate historical cohorts.

10. Instrument / Measurement Tool

The structural characteristics, administration guidelines, and scoring criteria of the PBI are outlined below:

• Test Type: Retrospective dimensional self-report questionnaire.
• Target Domain: Perceived parental behaviors, attitudes, and bonding styles experienced during childhood and adolescence up to the age of 16 years.
• Administration Format: Self-administered paper-and-pencil questionnaire or computerized digital test form. The respondent completes the form twice: once for their Mother (or primary female caregiver) and once for their Father (or primary male caregiver).
• Target Population: Adolescents and adults aged 16 years and older.
• Item Count: 25 items per parental form (total of 50 responses if both parents are rated).
• Response Scale: Four-point Likert-type scale:
  • Very like
  • Moderately like
  • Moderately unlike
  • Very unlike
• Scoring System: Items are scored on a scale from 0 to 3 points:
  • Positively Scored Items (0 to 3):
    • Very like = 3
    • Moderately like = 2
    • Moderately unlike = 1
    • Very unlike = 0
  • Reverse Scored Items (R) (3 to 0):
    • Very like = 0
    • Moderately like = 1
    • Moderately unlike = 2
    • Very unlike = 3
• Subscale Composition:
  • Care Subscale (12 items): 1, 2(R), 4(R), 5, 6, 11, 12, 14(R), 16(R), 17, 18(R), 24(R). Theoretical Score Range: 0 to 36. Higher scores indicate greater warmth and emotional responsiveness.
  • Overprotection / Control Subscale (13 items): 3(R), 7(R), 8, 9, 10, 13, 15(R), 19, 20, 21(R), 22(R), 23, 25(R). Theoretical Score Range: 0 to 39. Higher scores indicate greater intrusion, infantilization, and control.
• Standard Cut-off Scores (Parker et al., 1979):
  • Maternal Form: Care cut-off = 27.0; Overprotection cut-off = 13.5.
  • Paternal Form: Care cut-off = 24.0; Overprotection cut-off = 12.5.

11. Permissions & Fee and Test Year

The Parental Bonding Instrument was formally introduced to the international scientific community in 1979 in the British Journal of Medical Psychology. Scientia Professor Gordon Parker and the Black Dog Institute have historically maintained an open-access philosophy for non-commercial academic research and clinical practice. The questionnaire items and scoring criteria are distributed freely for educational, non-profit clinical, and investigative applications, with the provision that formal attribution and citation be accorded to the original authors.

Commercial reproduction, large-scale commercial pharmaceutical trials, or integration into proprietary electronic health record (EHR) systems typically require written consultation and permission from Professor Gordon Parker via the Black Dog Institute, School of Psychiatry, University of New South Wales (Sydney, Australia).

12. References

• Becker, W. C. (1964). Consequences of different kinds of parental discipline. In M. L. Hoffman & L. W. Hoffman (Eds.), Review of Child Development Research (Vol. 1, pp. 169-208). Russell Sage Foundation.
• Bowlby, J. (1969). Attachment and Loss: Vol. 1. Attachment. Basic Books.
• Bowlby, J. (1973). Attachment and Loss: Vol. 2. Separation: Anxiety and Anger. Basic Books.
• Bowlby, J. (1980). Attachment and Loss: Vol. 3. Loss: Sadness and Depression. Basic Books.
• Cubis, J., Lewin, T., & Dawes, F. (1989). The Parental Bonding Instrument: Factorial structure, validity and reliability in an adolescent population. Journal of Child Psychology and Psychiatry, 30(2), 291-304. https://doi.org/10.1111/j.1469-7610.1989.tb00242.x
• Kendler, K. S. (1996). Parenting and the etiology of major depression in women: A twin study research approach. Psychological Medicine, 26(2), 245-259. https://doi.org/10.1017/s0033291700034654
• Murphy, E., Brewin, C. R., & Silka, L. (1997). The Parental Bonding Instrument: Reliability and stability over a 12-month period in a community sample of adult offspring. Psychological Medicine, 27(1), 209-214. https://doi.org/10.1017/s0033291796004318
• Parker, G. (1983). Parental Overprotection: A Risk Factor in Psychosocial Development. Grune & Stratton.
• Parker, G. (2010). The Parental Bonding Instrument (Annotated Bibliography). Black Dog Institute. https://www.blackdoginstitute.org.au/
• Parker, G., Tupling, H., & Brown, L. B. (1979). A parental bonding instrument. British Journal of Medical Psychology, 52(1), 1-10. https://doi.org/10.1111/j.2044-8341.1979.tb02487.x
• Schaefer, E. S. (1959). A circumplex model for maternal behavior. The Journal of Abnormal and Social Psychology, 59(2), 226-235. https://doi.org/10.1037/h0041114
• Wilhelm, K., & Parker, G. (1990). Reliability of the Parental Bonding Instrument and Intimate Bond Measure Scales. Australian and New Zealand Journal of Psychiatry, 24(2), 199-202. https://doi.org/10.3109/00048679009077683
• Wilhelm, K., Niven, H., Parker, G., & Hadzi-Pavlovic, D. (2005). The stability of the Parental Bonding Instrument over a 20-year period. Psychological Medicine, 35(3), 387-393. https://doi.org/10.1017/s0033291704003469

13. Items of the Scale