Veloped by (Mikulincer et al., 1990; Mikulincer and Erev, 1991). In completing thisFrontiers in Psychology | www.frontiersin.orgApril 2015 | Volume 6 | ArticleEin-Dor et al.order AMI-1 avoidant attachment and glucosequestionnaire, participants rated the extent to which each item was descriptive of their ECR on a 7-point scale ranging from not 
at all (1) to very much (7). Eight items assessed avoidant attachment (e.g., “I am uncomfortable when other people get too close to me”) and 7 assessed anxious attachment (e.g., “I worry about being abandoned”). In the present study, Cronbach s were 0.67 for the anxiety items and 0.76 for the avoidance items. Previous research has shown high concordance between this brief measure and the buy PTK/ZK 
36-item ECR measure (Brennan et al., 1998). Mean scores were computed for each scale, and the two scores were significantly correlated, r(280) = 0.55, p < 0.001. General anxiety level was assessed with a Hebrew version of the Brief Symptom Inventory (BSI; Derogatis and Melisaratos, 1983)--a 53-item self-report inventory in which participants rate the extent to which they have been aggravated (0 = "not at all" to 4 = "extremely") in the past week by various symptoms. The BSI anxiety subscale comprised six items (e.g., "Feeling tense or keyed up"), and its reliability and validity have been repeatedly demonstrated (Boulet and Boss, 1991). In the present study, Cronbach was 0.81, and thus self-rated anxiety was calculated by averaging the item ratings. Social support was assessed with the multidimensional scale of perceived social support (MSPSS; Zimet et al., 1990). It includes items to assess perceived support from friends, family and an intimate partner (e.g., "There is a special person who is around when I am in need"). In the present study, Cronbach was 0.93, and thus a total score of perceived social support was calculated by averaging the item ratings. Upon completion of the first session, participants were given a referral for a medical examination at the Mor Institute for Medical Data Ltd, and the medical center administrator was given the contact details of the prospective patient. Participants were invited to the medical center in the morning after a fast of 12 h. We controlled for time of awakening, morning activity, caffeine consumption and smoking, factors that can affect morning cortisol levels. All participants were instructed not to exercise before coming to the examination. The medical examination was conducted independently of all other study variables, by medical staff blind to the study goals and hypotheses. Upon participants' arrival, a nurse drew blood samples into serum tubes containing aprotinin (500 kallikrein-inhibiting units, or KIU, per ml of blood). The samples were centrifuged at 1,600 ?g for 15 min at 4 , and then transferred to plastic tubes and stored at ?0 . Cortisol was measured by the TKCO1-Coat-A-Count kit (Diagnostic Products Corporation, Los Angeles, CA, USA), and DHEA was assessed with the DHEA-DSL-9000-ActiveTM DHEA coated tube radioimmunoassay kit (Diagnostic System Laboratories, Webster, TX, USA). Basal glucose levels were measured with the Roche Diagnostics Serum Work Area Modular Analytics P-800 auto-analyzer (Roche Diagnostics, Basel, Switzerland). After the blood tests, an expert physician examined the participants to establish the occurrence of various medical conditions, including clinical hypertension, diabetes, and obesity. Test results were sent to the last author's (AHZ) res.Veloped by (Mikulincer et al., 1990; Mikulincer and Erev, 1991). In completing thisFrontiers in Psychology | www.frontiersin.orgApril 2015 | Volume 6 | ArticleEin-Dor et al.Avoidant attachment and glucosequestionnaire, participants rated the extent to which each item was descriptive of their ECR on a 7-point scale ranging from not at all (1) to very much (7). Eight items assessed avoidant attachment (e.g., "I am uncomfortable when other people get too close to me") and 7 assessed anxious attachment (e.g., "I worry about being abandoned"). In the present study, Cronbach s were 0.67 for the anxiety items and 0.76 for the avoidance items. Previous research has shown high concordance between this brief measure and the 36-item ECR measure (Brennan et al., 1998). Mean scores were computed for each scale, and the two scores were significantly correlated, r(280) = 0.55, p < 0.001. General anxiety level was assessed with a Hebrew version of the Brief Symptom Inventory (BSI; Derogatis and Melisaratos, 1983)--a 53-item self-report inventory in which participants rate the extent to which they have been aggravated (0 = "not at all" to 4 = "extremely") in the past week by various symptoms. The BSI anxiety subscale comprised six items (e.g., "Feeling tense or keyed up"), and its reliability and validity have been repeatedly demonstrated (Boulet and Boss, 1991). In the present study, Cronbach was 0.81, and thus self-rated anxiety was calculated by averaging the item ratings. Social support was assessed with the multidimensional scale of perceived social support (MSPSS; Zimet et al., 1990). It includes items to assess perceived support from friends, family and an intimate partner (e.g., "There is a special person who is around when I am in need"). In the present study, Cronbach was 0.93, and thus a total score of perceived social support was calculated by averaging the item ratings. Upon completion of the first session, participants were given a referral for a medical examination at the Mor Institute for Medical Data Ltd, and the medical center administrator was given the contact details of the prospective patient. Participants were invited to the medical center in the morning after a fast of 12 h. We controlled for time of awakening, morning activity, caffeine consumption and smoking, factors that can affect morning cortisol levels. All participants were instructed not to exercise before coming to the examination. The medical examination was conducted independently of all other study variables, by medical staff blind to the study goals and hypotheses. Upon participants' arrival, a nurse drew blood samples into serum tubes containing aprotinin (500 kallikrein-inhibiting units, or KIU, per ml of blood). The samples were centrifuged at 1,600 ?g for 15 min at 4 , and then transferred to plastic tubes and stored at ?0 . Cortisol was measured by the TKCO1-Coat-A-Count kit (Diagnostic Products Corporation, Los Angeles, CA, USA), and DHEA was assessed with the DHEA-DSL-9000-ActiveTM DHEA coated tube radioimmunoassay kit (Diagnostic System Laboratories, Webster, TX, USA). Basal glucose levels were measured with the Roche Diagnostics Serum Work Area Modular Analytics P-800 auto-analyzer (Roche Diagnostics, Basel, Switzerland). After the blood tests, an expert physician examined the participants to establish the occurrence of various medical conditions, including clinical hypertension, diabetes, and obesity. Test results were sent to the last author's (AHZ) res.
