This article was downloaded by: [Field, Nigel P.] On: 18 March 2009 Access details: Access Details: [subscription number 909678069] Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Death Studies Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713657620 Role of Attachment in Response to Pet Loss Nigel P Field a; Lisa Orsini a; Roni Gavish a; Wendy Packman a Pacific Graduate School of Psychology, Redwood City, California, USA a Online Publication Date: 01 April 2009 To cite this Article Field, Nigel P., Orsini, Lisa, Gavish, Roni and Packman, Wendy(2009)'Role of Attachment in Response to Pet Loss',Death Studies,33:4,334 — 355 To link to this Article: DOI: 10.1080/07481180802705783 URL: http://dx.doi.org/10.1080/07481180802705783 PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material Death Studies, 33: 334–355, 2009 Copyright # Taylor & Francis Group, LLC ISSN: 0748-1187 print=1091-7683 online DOI: 10.1080/07481180802705783 ROLE OF ATTACHMENT IN RESPONSE TO PET LOSS NIGEL P FIELD, LISA ORSINI, RONI GAVISH, and WENDY PACKMAN Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Pacific Graduate School of Psychology, Redwood City, California, USA This study examined the impact of attachment on grief severity following the death of a pet Seventy-one participants who had lost a dog or cat within the past year completed a set of measures that included an attachment measure assessing individual differences in attachment anxiety and avoidance, strength of the past attachment to the pet, the continuing bond with the deceased pet, social support, and complicated grief symptoms Attachment anxiety and strength of the past attachment to the pet were each uniquely predictive of more severe grief Furthermore, the continuing bond to the deceased pet partially mediated the impact of strength of the past attachment to the pet on grief severity No significant mediators of the effect of attachment anxiety on grief were found, however The results highlight the importance of distinguishing strength of attachment from attachment security in examining the effect of attachment on response to pet loss Inter-species attachment bonding is assumed to develop in much the same manner as bonding between humans in providing security and protection or stimulate a natural instinct to offer security and protection or reciprocal attachment security (Noonan, 1998; Voith, 1985) It therefore should follow that response to the loss of a pet attachment relationship will be similar to that found for loss of a human attachment bond Consistent with this, previous research has shown the strength of the attachment bond to a pet to be a significant predictor of grief severity following pet death (Brown, Richards, & Wilson, 1996; Gosse, 1989; Gosse & Barnes, 1994) The present study extends Received 20 October 2007; accepted 25 May 2008 Address correspondence to Nigel P Field, Pacific Graduate School of Psychology, 405 Broadway St., Redwood City, CA 94063 E-mail: nfield@pgsp.edu 334 Role of Attachment in Response to Pet Loss 335 prior work on pet loss in examining the role of attachment style on extent of grief following loss of a pet Furthermore, it attempts to identify mediators through which the attachment style might impact grief—specifically, social support, the strength of the pet attachment bond, and the continuing bond to the deceased pet To our knowledge, this is the first attempt to examine attachment style in pet loss Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 The Human–Pet Bond Pets occupy traditionally acknowledged human roles because they evoke similarly known patterns of emotion and behavior observed within human attachment relationships In addition, pets are endowed with human qualities, making them capable of filling roles such as ‘‘attachment figure, transitional object [and] therapist’’ (Noonan, 1998, p 17) or child (Albert & Bulcroft, 1987; Archer, 1997) Archer (1997) provided numerous examples of the similarities between pets and babies that may appear unmistakably accurate to an owner of a pet, such as their ability to elicit emotional responses without a commensurate capacity to verbalize meanings (see also Lorenz, 1970), or the uncritical acceptance that can be achieved in relationships with pets, known in psychological language as ‘‘unconditional positive regard ’’ (italics in original) or in layman’s language as ‘‘unconditional love and acceptance’’ (p 253) Similar to humans, pets carry the potential to provide an emotional attachment bond that promotes a sense of security and well-being (Sable, 1995) The attachment relationship between humans and pets is reinforced by the fact that pets are often viewed as part of the family (Raupp, 1999; Raupp, Barlow, & Oliver, 1997) People become emotionally close and committed to pets, accept responsibility for pets, share activities with pets and grieve when their pet dies Thus, the human–pet attachment bond is similar to that between humans Pet Bonding and Grief To the extent that the human–pet bond constitutes an attachment bond, a similar response to separation and loss of a pet should be found following loss of a human attachment bond This has been addressed in a number of studies Quackenbush (1985) found that Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 336 N P Field et al pet death is experienced in a manner similar to human death in terms of sleep loss, days missed from work, and other psychological and social difficulties Gerwolls and Labott (1994) found that loss of a pet after two weeks and eight weeks revealed grief scores that are similar to those reported after losing a human relationship In an interview study of adults grieving the loss of a pet, Cowles (1985) found reports of memories of their deceased pet that are commonly observed in the grief reaction to losing a human attachment relationships such as searching behavior, obsessive rumination of the events leading up to the pet’s death, thoughts of previous significant losses both animal and human, and fears of losing control or ‘‘going crazy.’’ Thus, losing a pet is very similar to losing a human attachment relationship In a study examining strength of attachment and grief, Gosse (1989) found degree of attachment to a pet to be a strong predictor of grief response following pet death Archer and Winchester (1994) similarly found a significant positive correlation between degree of attachment to the pet and grief Thus, the strength of the attachment bond to the pet is an important factor in extent of distress following pet loss Attachment Style and Response to Interpersonal Loss Beyond the significance of the strength of the attachment bond for understanding characteristic behavioral responses observed in grief reactions to loss, individual differences in attachment style has been shown to play a role in adjustment to loss (Archer, 1999; Mikulincer & Shaver, 2003) Attachment style pertains to the quality of relationships and not to strength of the attachment bond per se For example, abusive relationships may involve strong albeit maladaptive attachment ties that are more likely to be found among those who are insecurely attached (Henderson, Bartholomew, Trinke, & Kwong, 2005) Specifically, attachment style is an indicator of one’s cognitive and emotional capacity to negotiate care in human relationships Two different types of insecure attachment styles have been identified in the adult attachment literature—anxious and avoidant attachment Anxious attachment is characterized by excessive concern over abandonment or rejection in a person’s adult human relationships with significant others (Mikulincer & Shaver, 2003) Avoidant attachment is identified by Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Role of Attachment in Response to Pet Loss 337 a tendency to defensively dismiss the need for emotional connection and support in one’s adult human relationships with significant others (Mikulincer & Shaver, 2003) Although the nature of the influence of attachment style on grief has been investigated in regard to human relationships, it has not been explored in regard to human–pet bonds We address this in the present study in examining how the quality of a person’s human relationships, reflected in his or her attachment style, has bearing on how he or she responds to the death of a pet Attachment style has direct bearing on response to separation or loss and is an important predictor of variability in the grief response (Archer, 2001; Field & Sundin, 2001; Parkes, 2003; Stroebe, Schut, & Stroebe, 2005) Bowlby (1980) identified symptoms of chronic grief, such as pronounced distress and helplessness that not abate over time, to be associated with an anxious style of attachment and delayed grief with an avoidant style of attachment Support for anxious attachment as a predictor of chronic grief has been shown in a number of studies (e.g., Field & Sundin, 2001; Fraley & Bonanno, 2004; Wijingaards-de Meij, Stroebe, & Schut, 2007) Other anxious attachment manifestations of disordered mourning identified in the empirical literature involve negative appraisals of one’s ability to cope following loss of an attachment figure (Mikulincer & Florian, 1998) A perceived inability to cope following loss parallels the pre-morbid anxious attachment features such as compulsive care-seeking and care-giving, both behaviorally (e.g., proximity seeking) and cognitively (e.g., thinking about the attachment figure), which are aimed at maintaining a feeling of safety provided by the attachment relationship According to Field and Sundin (2001), compulsive careseeking prior to loss mirrors the post-loss behaviors of individuals with anxious attachment who make inordinate albeit futile attempts to recover the lost attachment However, over time, the continual activation of distress as a result of the failure to reestablish proximity with their loved one may heighten feelings of hopelessness and despair in the anxiously attached With respect to avoidant attachment, the results are mixed Field and Sundin (2001) did not find support for an association between compulsive self-reliance (i.e., avoidant attachment style) and bereavement-related adjustment in conjugal bereavement Fraley and Bonanno (2004) similarly found no effect of avoidant 338 N P Field et al attachment on bereavement-related symptomatology However, they found that a fearful type of avoidance (i.e., those who were high in both attachment avoidance and anxiety) was predictive of slower recovery following death of a spouse or child Most recently, Wijngaards-de Meij, Stroebe, and Schut (2007) found that avoidant attachment was predictive of more severe bereavementrelated symptomatology Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Mediators of Attachment Style on Grief Response Given that different forms of insecure attachment can interfere with successful adaptation to bereavement, it is important to identify the underlying mechanisms through which attachment style might affect adjustment Each of the following are possible mechanisms for this Social Support A way by which attachment style impacts poorer adjustment to loss of a pet could be through lack of social support According to attachment theory (Bowlby, 1973), internal working models of attachment are guided by prior experiences in close relationships with others and function to shape beliefs about available support in the environment Thus, past negative experiences of support will promote a sense of distrust or attachment insecurity regarding the availability of others while positive experiences will promote a sense of trust or attachment security in that availability In social support research, attachment style differences have come under study as potential predictors of variability in giving, using and benefiting from social support (Mikulincer et al., 2001) Davis, Morris, and Kraus (1998) found a negative correlation between insecure attachment and social support with avoidant attached individuals reporting having less social support than other insecure non-avoidant individuals on a measure of global support (Davis et al., 1998) In the context of bereavement, Wijngaards-de-Meij et al (2007) found that anxious attachment was associated with greater grief among parents who had lost a child through death and that this relationship was partially mediated by relationship satisfaction (i.e., spousal support) Role of Attachment in Response to Pet Loss 339 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 To the degree that attachment insecurity impacts negatively on the ability of the person to obtain emotional support following loss of a pet, such that those who are more insecure are more likely to lack this resource to help buffer their distress, social support should mediate the impact of insecure attachment on grief In effect, social support can provide the bereaved with a sense of felt security in being able to appeal to others for comfort to help him or her process the loss On the other hand, lack of social support may increase the likelihood that the bereaved will avoid processing the loss because the emotions, aroused by reminders of the loss, are too overwhelming to bear alone Pet Bond Pets may have served a compensatory function among those with insecure attachment whose relationships with other people are of lower quality In effect, such individuals may have developed a stronger bond to their pet as a result Consistent with this, Stallones, Marx, Garridy, and Johnson (1990) found that attachment to pets, but not simply pet ownership, was inversely associated to human social support with stronger attachment to pets reported by those who had less human social support In light of this more exclusive focus of attachment on their pet, the loss may have much greater implications for someone who is insecurely attached than for someone who is more capable of satisfying their attachment needs through human relationships These individuals therefore are likely to experience more severe grief following the loss of their pet This may be compounded by lack of social support to help them cope with the loss Continuing Bonds According to an attachment theory perspective, successful adaptation to bereavement requires relinquishing the goal of regaining physical proximity to the deceased (Field, Gao, & Paderna, 2005) The tendency of anxiously attached individuals to experience chronic grief following the loss implies such failure to so This may be reflected in an attempt to maintain a bond with the deceased pet that is indicative of avoidant coping If a continuing bond to the pet is a type of avoidant coping, it may interfere with processing the loss There is controversy in the bereavement Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 340 N P Field et al literature as to whether a continuing bond to the deceased is maladaptive or, in fact, is integral to successful adaptation to bereavement, however (see Klass, Silverman, & Nickman, 1996; Stroebe & Schut, 2005) If indeed it is a maladaptive form of avoidant coping, it might then follow that a continuing bond with the deceased pet would mediate the relationship between attachment anxiety and more severe grief On the other hand, a continuing bond may constitute an important resource in serving as an internalized secure base in helping to offset the pain of the loss that is fully compatible with acknowledging the permanence of the separation (Field et al., 2005) In this case, it should follow that securely attached individuals are best equipped to make use of a continuing bond as an internalized secure base to help them cope with the loss; it thus should mediate more successful adjustment reflected in lower grief This study addresses the possible mediating role of a continuing bond to the deceased pet in hindering or in facilitating successful adaptation to bereavement This study extends beyond the direct effects of the strength of the attachment bond to a pet on the adjustment following pet death, by highlighting the importance of looking at attachment style as a significant individual difference factor predictive of grief severity Moreover, it attempts to identify mediators through which attachment style might impact grief through its influence on social support, the strength of the past attachment to the pet, and the continuing bond to the deceased pet It is hypothesized that anxious and avoidant attachment will be predictive of more severe grief as mediated through their impact on poorer social support and as mediated through their impact on a stronger attachment bond to their pet to compensate for the lower quality of their attachment to people The continuing bond to the deceased as a mediator of the effect of attachment style on grief severity will also be examined although no clear hypothesis is offered as to its directionality as a mediator Method Participants A sample of pet loss survivors was solicited through an internet flyer sent to pet loss support counselors and other pet loss establishments Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Role of Attachment in Response to Pet Loss 341 (e.g., humane society, crematoriums, hospitals), through a computergenerated file of pet loss survivors that included contact information and an internet classified ad listing Two criteria were used to select research participants: (a) 18 years-of-age or older, and (b) loss of a dog or cat in death or separation (e.g., missing, stolen) within 12 months from the date of data collection Four hundred and ninety-six ground mail letters were mailed Out these, were returned as ‘‘not deliverable’’ and 401 did not respond The 71 people who agreed to participate in the study were sent an informed consent form to fill out prior to their participation and asked to return it along with a survey packet Demographic information on the study sample is summarized in Table The mean age of participants was 44.48 (SD ¼ 14.05) The majority were Caucasian (91%), female (84%), married (65%), and living with at least one other person (84%) TABLE Demographics Category Age (M ¼ 44.48, SD ¼ 14.05) Gender Female Male Marital status Married Divorced Widowed Single Other Household status Living alone Not living alone Education High school College Graduate school Religious affiliations Buddhist Catholic Jewish Protestant Non-denominational Frequency 61 10 48 14 32 39 15 34 22 21 15 25 342 N P Field et al Measures DEMOGRAPHICS Standard demographic information was obtained on the study sample that included age, gender, ethnicity, religion, education, marital status, and whether living alone Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 ASPECTS OF THE LOSS Contextual factors and other aspects associated with the loss that could potentially affect the course and intensity of bereavement were assessed This included the nature of the death, length of time since the death, whether the participant obtained a new pet since the loss, type of pet, age of the pet, length of pet ownership, and number of prior pet losses PET ATTACHMENT SURVEY (PAS) The PAS Scale (Gosse, 1989) is an 11-item measure assessing strength of attachment in the past relationship with a deceased pet (e.g., ‘‘You had your pet near you when you studied, read, or watched TV’’) Each item is rated on a 5-point scale ranging from (almost never) to (almost always) The measure possessed satisfactory internal consistency (a ¼ 79) in the present study sample RELATIONSHIP SCALES QUESTIONNAIRE (RSQ) The 30-item RSQ is a self-report measure of individual differences in adult attachment (Griffin & Bartholomew, 1994) Each item is rated on a 7-point Likert scale ranging from (strongly disagree) to (strongly agree) Scaled scores for attachment anxiety (e.g., ‘‘I often worry that romantic partners don’t really love me’’) and attachment avoidance (e.g., ‘‘I am comfortable without close emotional relationships’’) were derived based on a procedure used by Fraley and Bonanno (2004) The two scales were moderately correlated (r ¼ 48) The internal consistency of the anxiety (a ¼ 80) and avoidance (a ¼ 81) subscales were in the satisfactory range INVENTORY OF COMPLICATED GRIEF SCREEN Complicated grief symptoms were measured by a 9item shortened version of the original 19-item Inventory of Role of Attachment in Response to Pet Loss 343 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Complicated Grief (Prigerson et al., 1995) Use of this shortened version was necessitated by the full measure including items that were descriptive of continuing bonds thus confounding complicated grief symptoms with continuing bonds expressions In a previous study involving human loss, this modified version was shown to be valid in terms of its internal consistency, relation to time since the loss, and perceived health (Filanosky, 2004) Items describe emotional, cognitive, and behavioral states associated with complicated grief Respondents rated the frequency with which they experienced each item during the past month on a 5-point scale, ranging from (never) to (always) The internal consistency of this shortened version was high (a ¼ 90) MULTIDIMENSIONAL SCALE OF PERCEIVED SOCIAL SUPPORT (MSPSS) The MSPSS (Zimet, Dahlem, Zimet, & Farley, 1988) consists of 12 items designed to measure perceived social support in relation to family, friends, and a significant other Each of the items are answered on a 7-point Likert scale ranging from (very strongly disagree) to (very strongly agree) A total score based on summing the 12 items was derived for the present study purposes This was based on the high internal consistency of this total scale score (a ¼ 94) CONTINUING BONDS WITH THE DECEASED A 4-item measure of continuing bonds with the deceased pet assessing extent of use of positive continuing bonds expressions involving fond memories, legacy of the deceased, and use of photographs for maintaining a bond Participants were instructed to rate how often they made use of each of these continuing bonds expressions during the past month on a 4-point scale ranging from (not at all) to (often) Its internal consistency was satisfactory (a ¼ 78) Results Grief and the Continuing Bond with the Pet The average time since the death for the sample as a whole was 5.40 (SD ¼ 3.78) months The mean score of 2.15 (SD ¼ 82) for complicated grief symptoms is somewhat lower than findings for human loss within the first year of bereavement using the same measure (M ¼ 2.55; SD ¼ 1.07), t ¼ À.2.55; p < 05 (Filanosky, 2004) 344 N P Field et al The item means and standard deviations for each of the continuing bonds (CB) measure items are shown in Table Table also includes the means and standard deviations for the same set of items from a database involving human loss within the first year of bereavement for comparative purposes (Filanosky, 2004) The mean scores for each of the pet loss CB items are not statistically significantly different from their human counterparts indicating that attachment to the pet continues after the death in a similar fashion to that found in humans Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Demographic Factors and Circumstances Surrounding Loss and Grief A series of correlational and t-test analyses were conducted to determine whether any of the demographic factors were predictive of grief severity An independent t-test analysis indicated a significant relationship between whether or not the participant was living alone and grief, such that those who were living alone reported greater grief (M ¼ 2.63, SD ¼ 1.00) than those not living alone (M ¼ 2.07, SD ¼ 76), t(69) ¼ 2.13, p < 05 No significant relationships were found for any of the other demographic factors with grief, however No differences were found in grief severity between those who procured a new pet since the death of their pet (M ¼ 2.18, SD ¼ 89) and those who did not so (M ¼ 2.14, SD ¼ 79) t(69) ¼ 0.22, ns Also, no significant relationships were found for grief severity with time since the death (r ¼ À.20, ns), length of pet ownership (r ¼ À.21, ns), or number of previous pet losses (r ¼ 10, ns) Furthermore, an analysis of variance (ANOVA) indicated that type of death was not predictive of grief severity, F(4, 69) ¼ 121, ns TABLE Continuing Bonds Items Pet Items Focused on fond memories of deceased Shared fond memories with others of deceased Positive influence of deceased on who I am today Looking at photographs or pictures of deceased Human M SD M 3.19 3.10 2.57 2.94 0.89 0.88 1.08 1.04 3.18 2.87 2.85 2.64 SD T 0.98 0.09 0.96 1.58 1.04 À1.60 1.13 1.73 345 Role of Attachment in Response to Pet Loss TABLE Correlations for Main Study Measures Measure Anxiety – Avoidance 48ÃÃà Alone 22 Past bond to pet 19 Continuing bond 22 Social support À.51ÃÃà Grief 53ÃÃà à Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 p < 05 Ãà p < 01 ÃÃà – 23 16 19 À.41ÃÃà 25à – 31 31Ãà À.14 29à – 66ÃÃà À.30à 58ÃÃà – À.19 67ÃÃà – À.32Ãà – p < 001 Simple correlations among the main set of variables are shown in Table There are a number of noteworthy features in the pattern of correlations here Prominent relationships are found among strength of past attachment to the pet, pet CB, and grief; this highlights the importance of strength of bond on both a tendency to maintain attachment following the death and grief severity The strong relationship between pet CB and grief is particularly noteworthy here Attachment anxiety and avoidance are predictive of grief and social support, but not with strength of past attachment bond to the pet nor with pet CB The pattern of correlations thus suggests how strength of past attachment to the pet and attachment style are distinct predictors of grief severity Attachment Style and Grief To test the hypothesis that attachment style plays a role in adjustment to loss of a pet, such that higher anxiety and avoidance and=or their interaction would be predictive of more severe grief, a regression analysis was conducted in which grief symptoms were regressed on anxiety, avoidance, and their interaction Prior to conducting this analysis, the anxiety and avoidance scores were centered on their group means and the interaction was computed based on these centered scores (see Aiken & West, 1991) The results are summarized in Table The results indicate that attachment anxiety is a significant predictor of grief, such that higher anxiety scores are associated with greater grief However, a significant interaction effect shows that the strength of the effect of anxiety on grief is moderated by avoidance, such that a stronger 346 N P Field et al TABLE Attachment Styles Predicting Grief Effect Beta SE Beta b Anxiety Avoidance Anxiety  Avoidance 30 34 36 12 19 10 29à 22 39ÃÃà Note R2 change ¼ 24, p < 001 p < 05 ÃÃà p < 001 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 à relationship exists between anxiety and grief at higher levels of avoidance No significant effect is shown for avoidance Mediating Models A series of hierarchical regression analyses were conducted to determine whether social support, strength of past bond to the pet, and the continuing bond to the pet each served as a mediator of the effect of attachment styles on grief This involved entering anxiety, avoidance, and their interaction in Step of the regression model and the specified mediating variable in Step of the regression analysis as predictors of grief Support for a mediational model would be shown by a significant reduction in the amount of variance contributed by the combined attachment style measures from Step to Step in predicting grief in conjunction with a significant relationship shown for the mediator with grief in Step (Baron & Kenny, 1986) The Step results are summarized in Tables 5–7 The absence of a significant reduction from Step (see Table 4) to Step in the strength of the standardized beta TABLE Mediational Model Analysis for Social Support Effect Beta SE Beta b Anxiety Avoidance Anxiety  Avoidance Social support 21 16 35 À.09 11 11 10 10 25 19 38Ãà À.11 Note R2 change ¼ 01, ns p < 01 Ãà Role of Attachment in Response to Pet Loss 347 TABLE Mediational Model Analysis for Past Pet Bond Effect Anxiety Avoidance Anxiety  Avoidance Past pet bond Beta SE Beta b 18 15 36 66 08 08 08 11 21à 18 40ÃÃà 54ÃÃà Note R2 change ¼ 28, p < 001 p < 05 ÃÃà p < 001 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 à coefficients accounted for by the attachment style measures in predicting grief when the mediator was included in the regression model indicates that social support, attachment bond to the pet, and continuing bonds respectively not mediate the relationship between attachment style and grief Path Analysis For illustrative purposes, a path analysis was conducted to highlight the broader configuration of relationships among the set of relevant variables, using the AMOS statistical software (Arbuckle, 2003) The structure of the path model was informed by the pattern of significant correlations shown in Table This involved a recursive path model in which attachment anxiety and strength of attachment were modeled as distal predictors, and social support and CB as mediators, or proximal predictors, of grief Attachment avoidance was not included in the path model because it was not a significant predictor of grief in the regression analysis (see Table 4) A model trimming process was used in TABLE Mediational Model Analysis for Continuing Pet Bond Effect Anxiety Avoidance Anixety  Avoidance Continuing bond to pet Note R2 change ¼ 29, p < 001 à p < 05 ÃÃÃp < 001 Beta SE Beta b 19 13 32 61 08 08 08 09 23à 15 36ÃÃà 56ÃÃà Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 348 N P Field et al FIGURE Path model of the effect of strength of past bond to the pet, the continuing bond to the pet, and attachment anxiety on grief symptoms The values next to each of the paths represent standardized parameter estimates à p < 05 ÃÃà p < 001 which all non-statistically significant paths in the just-identified model were deleted from the path model in arriving at the final model (for more detail on model trimming, see Kline, 1998) The final model is shown in Figure This model had adequate fit based on v2 ¼ 4.14, ns, CFI ¼ 1.00, NFI ¼ 0.96, and RMSEA ¼ 0.02 (0.00 to 0.18) fit indices There are a few noteworthy features of this model First, the effect of attachment anxiety on grief is exclusively a direct effect Although it is significantly linked with social support, the path for social support with grief is not significant; therefore attachment anxiety does not exert an indirect effect through social support Also, neither is there an indirect link through CB since attachment anxiety is not significantly associated with CB On the other hand, strength of attachment has both significant indirect and direct effects on grief The indirect effect of strength of attachment on grief through CB Role of Attachment in Response to Pet Loss 349 suggests that a stronger bond with the pet when alive has an impact of the strength of CB that, in turn, results in more severe grief Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Discussion The primary objective of this study has been to extend beyond the direct effects of the strength of the attachment bond to a pet on the severity of grief by examining attachment style as a significant individual difference factor predictive of adjustment following pet death Furthermore, it attempted to identify mediators through which attachment style might impact grief—principally social support, strength of past attachment to the pet, and the continuing bond to the deceased pet The results indicated that although the effects of attachment anxiety and the anxiety x avoidance interaction were shown to be predictive of grief, this effect was not mediated by each of the above factors This finding on the effect of attachment anxiety on grief is consistent with Bowlby’s (1973, 1980) theory of loss and attachment, which speculates that anxiously attached adults have difficulty regulating emotions during security threat experiences such as loss of an attachment figure, which leads to prolonged feelings of helplessness indicative of complicated grief This result replicates the findings for the effect of attachment anxiety in human loss (e.g., Field & Sundin, 2001; Fraley & Bonanno, 2004) The significant Anxiety  Avoidance interaction effect on grief is consistent with the findings of Fraley and Bonanno (2004) with respect to human loss This finding highlights the importance of distinguishing fearful versus dismissive variants of avoidant attachment (see Bartholomew & Shaver, 1998); this significant interaction effect suggests that fearful individuals (i.e., those who are high in both anxiety and avoidance) are prone to be most affected by the loss of their pet The non-significant results for attachment avoidance suggest that defensive deactivation of attachment characteristic of this attachment style does not interfere with adjustment to bereavement—at least as it is manifested in grief-specific symptomatology This is in keeping with the results for human loss in which attachment avoidance was not shown to have an effect on adjustment ollowing the death of a significant human relationship (e.g., Field & Sundin, 2001; Fraley & Bonanno, 2004) It was also Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 350 N P Field et al noteworthy that avoidance was not linked with a stronger past bond to the pet despite its negative impact on social support Results of the path analysis revealed how attachment anxiety and strength of the past bond to the pet both served as unique predictors of grief This highlights the importance of distinguishing strength of attachment from type or quality of attachment in terms of attachment style in examining the role of attachment in adjustment to loss (see Shaver & Tancredy, 2001) The indirect effect found for strength of pet bond on grief as mediated through continuing bonds suggested that those with a stronger past bond with their pet were also more likely to maintain a stronger bond with the pet following the death that in turn incited greater grief Because the items contained in the continuing bonds measures are positively phrased whereas the items in the complicated grief are negatively valenced, the strong association between the two cannot be explained in terms of simple matching of valence On the other hand, because attachment anxiety was not associated with continuing bonds, despite its significant positive relationship with grief, this suggests that continuing bonds represent a more adaptive component of grief as distinguished from complicated grief symptoms An important focus in future research would be to examine more closely the relationship between continuing bonds and grief through a longitudinal design The non-significant correlation between attachment anxiety and strength of the attachment to the pet further highlights this important distinction between strength of attachment and quality of attachment as identified in terms of attachment style It also offers no support for the position that those who are insecurely attached (i.e., anxiously attached) are more likely to form a stronger attachment to a pet to compensate for failure to meet their attachment needs in human relationships The significant negative relationship found between social support and grief symptoms in the simple correlational analysis is consistent with previous grief studies (e.g., Stroebe & Stroebe, 1987) A similar significant negative correlation found for attachment anxiety with social support is also in keeping with past research (e.g., Wijngaards-de Meij et al., 2007) Despite this, the results of the hierarchical regression analysis as well as the path analysis results provided no support for social support as a mediator of the relationship between attachment anxiety and grief Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Role of Attachment in Response to Pet Loss 351 symptoms In fact, in the regression and path analysis, social support was no longer a significant predictor of grief when considered alongside other predictors of grief It suggests that social support is a less important factor in pet loss than in human loss It also may be that the social support measure used in the present study was not a sensitive enough measure in that it did not specifically focus on social support in the context of pet loss but instead assessed the quality of social support in general without reference to a particular event or concern In future research it may be useful to include a social support measure that more specifically targets social support in the context of a specific stressful life event such as pet loss Although attachment anxiety and continuing bonds were not related, such a relationship might have been found if CB expressions that were more clearly indicative of failure to relinquish the goal to re-establish physical proximity to the deceased were considered The CB items used in the present study may not have been indicative of such failure, such as a focus on fond memories, which is fully compatible with accepting the loss If indeed CB expressions more indicative of denial of the death were included, such as in keeping the pet’s supplies on hand as though needed as if the pet were still alive, a relationship between attachment anxiety and CB might have been found There are a number of limitations in this study One limitation has to with the use of a retrospective measure of the strength of the past attachment to the pet It could be argued that an individual’s level of current grief may influence his or her memory for the strength of the past attachment to the pet such that those with more severe grief are likely to judge this attachment as stronger The effect of current affective state on memory and judgment is well-established in the social cognition literature (e.g., Bower, 1981; Field, Thompson, & Gallagher-Thompson, 2006) This may partly explain the relatively strong correlation between the strength of the past attachment to the pet and grief However, because the pet bond measure used in the present study assesses strength of the pet bond based largely on behavioral indices of strength of attachment, this may be less of a problem than in the case of a less behavior-oriented measure A second limitation involves the cross-sectional nature of the design It is not possible to ascertain the directionality of the relationships among variables in the same way that it is possible with Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 352 N P Field et al longitudinal data Furthermore, proper assessment of mediational models requires longitudinal analyses modeled on temporally ordered events Although the results of the path analysis are compelling in providing a sensible model of the broader configuration of arrangement among the variables, such causal models must be interpreted with great caution A third limitation involves the relatively low response rate for participation in the study Consequently, those who participated may not be representative of the larger population of bereaved pet owners with implications regarding generalizability of the present study findings Finally, no information was obtained on psychiatric illness existing prior to the loss It is likely that those with pre-existing psychiatric illness would have reacted more strongly to the death of their pet A number of implications follow from the findings One implication is that indeed the loss of a pet is comparable to that shown for human loss in terms of its psychological impact The extent of grief and continuing bond expressions item mean scores are supportive of the viewpoint that the relationship with the pet constitutes an attachment bond In effect, distress in response to separation and an attempt to maintain proximity, expressed in grief and continuing bonds respectively, are important criteria in defining an attachment bond This highlights the clinical implications of pet loss in showing that grief counselors need to extend their involvement beyond an exclusive focus on human loss Arguing for the pet bond as an attachment bond is supported further by the findings for the effect of attachment anxiety on grief The findings for this attachment style as an individual difference factor in the context of pet loss that is similar to that found in human loss adds further credence to the pet bond as an attachment bond This is true both in terms of normative response to loss as well as individual difference that moderate this References Aiken, L S & West, S G (1991) Multiple regression: Testing and interpreting interactions London: Sage Albert, A & Bulcroft, K (1987) Pets and urban life Anthrozooăs, 1(1), 925 Arbuckle, J A (2003) AMOS users guide 5.0 Chicago: Small Waters Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Role of Attachment in Response to Pet Loss 353 Archer, J (1997) Why people love their pets? Evolution and Human behavior, 18, 237–259 Archer, J (1999) The nature of grief: The evolution and psychology of reactions to loss New York: Routledge Archer, J 2001 Grief from an evolutionary perspective In R O Stroebe Hansson, W Stroebe, & H Schut (Eds.), Handbook of bereavement research: Consequences, coping and care (pp 263–283) Washington, DC: American Psychological Association Archer, J & Winchester, G (1994) Bereavement following death of a pet British Journal of Psychology, 85, 259–271 Baron, R M & Kenny, D A (1986) The moderator–mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations Journal of Personality and Social Psychology, 51, 1173–1182 Bartholomew, K & Shaver, P (1998) Methods of assessing adult attachment: Do they converge? In J A Simpson & W S Rholes (Eds.), Attachment theory and close relationships (pp 25–45) New York: Guilford Bower, G H (1981) Mood and memory American Psychologist, 36, 129–148 Bowlby, J (1973) Attachment and loss: Vol Separation: Anxiety and anger New York: Basic Books Bowlby, J (1980) Attachment and loss: Vol Loss: Sadness and depression New York: Basic Books Brown, B H., Richards, H C., & Wilson, C A (1996) Pet bonding and pet bereavement among adolescents Journal of Counseling and Development, 74, 505 Cowles, K V (1985) The death of a pet: Human responses to the breaking of the bond Pets and the Family, 8, 135–161 Davis, M H., Morris, M M., & Kraus, L A (1998) Relationship-specific and global perceptions of social support: Associations with well-being and attachment Journal of Personality and Social Psychology, 74, 468–481 Field, N P., Gao, B., & Paderna, L (2005) Continuing bonds in bereavement: An Attachment theory based perspective Death Studies, 29, 1–23 Field, N P & Sundin, E C (2001) Attachment style in adjustment to conjugal bereavement Journal of Social and Personal Relationships, 18, 347–361 Field, N P., Thompson, L W., & Gallagher-Thompson, D (2006) Impact of current grief on memory for past grief in spousal bereavement Memory, 14, 297–306 Filanosky, C A (2004) The nature of continuing bonds with the deceased and their effect on bereavement outcome Dissertation Abstracts International: Section B: The Sciences and Engineering, 65(2-B), 1027 Fraley, R C & Bonanno, G A (2004) Attachment and loss: A test of three competing models on the association between attachment-related avoidance and adaptation to bereavement Personality and Social Psychology Bulletin, 30, 878–890 Gerwolls, M K & Labott, S M (1994) Adjustments to the death of a companion animal Anthrozooăs, 7, 172187 Gosse, G H (1989) Factors associated with the human grief experience as a result of the death of a pet Dissertation Abstracts International: Section B: The Sciences and Engineering, 49(11-B), 5011 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 354 N P Field et al Gosse, G H & Barnes, M J (1994) Human grief resulting from the death of a pet Anthrozooăs, 7, 103–112 Griffin, D W & Bartholomew, K (1994) The metaphysics of measurement: The case of adult attachment Advances in Personal Relationships, 5, 17–52 Henderson, A., Bartholomew, K., Trinke, S., & Kwong, J (2005) When loving means hurting: An exploration of attachment and intimate abuse in a community sample Journal of Family Violence, 20, 219–230 Klass, D., Silverman, P., & Nickman, S L (Eds.) (1996) Continuing bonds: New understandings of grief Washington, DC: Taylor & Francis Kline, R B (1998) Principles and practice of structural equation modeling New York: Guilford Lorenz, K (1970) Studies in animal and human behavior (R Martin, Trans., Vol 1) Cambridge, MA: Harvard University Press (Originally published in 1970) Mikulincer, M & Florian, V (1998) The relationship between adult attachment styles and emotional and cognitive reactions to stressful events In J Simpson & S Rholes (Eds.) Attachment theory and close relationships (pp 143–165) New York: The Guilford Press Mikulincer, M., Gillath, O., Halevy, V., Avihou, N., Avidan, S., & Eshkoli, N (2001) Attachment theory and reactions to others’ needs: Evidence that activation of sense of attachment security promotes empathetic responses Journal of Personality and Social Psychology, 81, 1205–1224 Mikulincer, M & Shaver, P R (2003) Attachment theory and affect regulation: The dynamics, development, and cognitive consequences of attachmentrelated strategies Motivation & Emotion, 27(2), 77–102 Noonan, E (1998) People and pets Psychodynamic Counseling, 4(1), 17–31 Parkes, C M (2003) Acquainted with grief illness Crisis & Loss, 11(1), 37–46 Prigerson, H G., Frank, E., Stanislav, V K., Reynolds, C F., Anderson, B., Zubenko, G S., Houck, M S., George, C J., & Kupfer, D J (1995) Complicated grief and bereavement-related depression as distinct disorders: Preliminary empirical validation in elderly bereaved spouses American Journal of Psychiatry, 152(1), 22–30 Quackenbush, J (1985) The death of a pet—how it can affect owners Veterinary Clinics of North America: Small Animal Practice, 15(2), 395–402 Raupp, D C (1999) Treasuring, trashing or terrorizing: Adult outcomes of childhood socialization about companion animals Society & Animals, 7(2), 141–159 Raupp, D C., Barlow, M., & Oliver, J (1997) Perceptions of family violence: Are companion animals in the picture? Society and Animals, 5, 219–237 Sable, P (1995) Pets, attachment, and well-being across the life cycle Social Work, 40, 334–341 Shaver, P R & Tancredy, C M 2001 Emotion, attachment, and bereavement: A conceptual commentary In M S Stroebe & R O Hansson (Eds.) Handbook of bereavement research: Consequences, coping, and care (pp 63–88) Washington, DC: American Psychological Association Stallones, L., Marx, M B., Garrity, T F., & Johnson, T R (1990) Pet ownership and attachment in relation to the health of U.S adults, 21 to 64 years of age Anthrozooăs, 4(2), 100112 Role of Attachment in Response to Pet Loss 355 Downloaded By: [Field, Nigel P.] At: 23:47 18 March 2009 Stroebe, M & Schut, H (2005) To continue or relinquish bonds: A review of consequences for the bereaved Death Studies, 29, 477–494 Stroebe, M., Schut, H., & Stroebe, W (2005) Attachment in coping with bereavement: A theoretical integration Review of General Psychology, 9(1), 48–66 Stroebe, W & Stroebe, M (1987) Bereavement and health New York: Cambridge University Press Voith, V L (1985) Attachment of people to companion animals Veterinarian Clinics of North America: Small Animal Practice, 15(2), 289–295 Wijngaards-de Meij, L., Stroebe, M., & Schut, H (2007) Patterns of attachment and parents’ adjustment to the death of their child Personality and Social Psychology Bulletin, 33, 537–548 Zimet, G D., Dahlem, N W., Zimet, S G., & Farley, G K (1988) The Multidimensional Scale of Perceived Social Support Journal of Personality Assessment, 52, 30–41  ... of continuing bonds with the deceased pet assessing extent of use of positive continuing bonds expressions involving fond memories, legacy of the deceased, and use of photographs for maintaining... full measure including items that were descriptive of continuing bonds thus confounding complicated grief symptoms with continuing bonds expressions In a previous study involving human loss, this... attachment bond In effect, distress in response to separation and an attempt to maintain proximity, expressed in grief and continuing bonds respectively, are important criteria in defining an attachment