ویژگی های خاطرات مزاحم در یک نمونه از جامعه افسردگی، بهبود افراد افسرده و هرگز افسرده

Abstract An accumulating body of evidence has revealed that intrusive autobiographical memories of negative events play a role in depression. Despite increasing understanding of the phenomenological experience of these memories, previous research in this area has been conducted in either nonclinical samples, or in clinical samples without an adequate control group. This study aimed to replicate and extend findings with dysphoric samples by comparing the content and characteristics of intrusive memories in clinically depressed (n = 25), recovered (n = 30) and never-depressed (n = 30) participants. Participants completed mood measures, and a battery of self-report questionnaires that indexed intrusive memory frequency, avoidance and characteristics. Intrusive memories were common and shared strikingly similar characteristics across the three groups. The key finding was that depressed participants reported higher levels of intrusion-related distress, associated emotions (especially sadness and helplessness), interference as a result of the memories and memory vividness compared to the never-depressed group. Despite similar levels of intrusion, there were group differences in avoidance such that depressed participants reported higher levels of avoidance than the never-depressed group. These results provide further support for the proposal that clinical interventions for depression could usefully incorporate components that aim to reduce intrusive autobiographical memories and target avoidance strategies.

Results Sample characteristics Demographic characteristics The total sample was comprised of 50 female (58.8%) and 35 male (41.2%) participants with a mean age of 24.26 years (range = 18–54, SD = 6.05). There were no significant differences between groups in terms of age (F(2,82) = 2.21, p > .05) or gender (χ2 (2, N = 85) = 1.85, p > .05). Table 1 presents demographic and sample characteristics for each group. Table 1. Sample and demographic characteristics. Currently depressed n = 25 Recovered depressed n = 30 Never-depressed n = 30 Gender (% female) 68 60 50 Age (years) 25.48 (7.22) 25.07 (6.55) 22.43 (3.80) Marital status (% single) 68.0 86.7 90.0 Educational history (% completed at minimum year 12 (high school) or equivalent) 100 100 100 Employment status (% unemployed) 52.0 63.3 63.3 Ethnicity (% of sample) Caucasian 36.0 26.7 23.3 Asian 36.0 50.0 70.0 Other 28.0 23.3 6.7 BDI-II 28.60 (8.61) 12.23 (7.06) 6.03 (3.72) BAI 21.20 (9.17) 11.40 (6.79) 7.13 (7.08) Current episode (months) 20.84 (37.11) – – Age of first episode 17.41 (5.19) 17.81 (5.18) – Number of previous episodes 2.35 (2.08) 2.07 (1.36) – Number of Axis I diagnosesa 1.20 (1.35) 0.60 (1.00) 0.27 (0.58) Note. BDI-II = Beck Depression Inventory, BAI = Beck Anxiety Inventory. Except where noted, values refer to mean and (standard deviation) scores. a Value refers to number of Axis I diagnoses excluding MDD. Table options DSM-IV Axis I diagnoses All participants in the currently depressed group met criteria for a current MDE. Seven participants were experiencing their first MDE; the remainder met criteria for MDD (recurrent). Of the MDD (recurrent) subgroup, ten participants reported one to three previous episodes, three participants reported four to five previous episodes, three participants reported that their previous episodes were ‘too numerous to count’, and two were uncertain as to the number of past episodes. The total number of Axis I diagnoses (not including MDD) ranged between 0 and 5 for this group. Nine participants from the currently depressed group met criteria for MDD only (36%). The remaining 16 participants (64%) had at least one comorbid Axis I diagnosis. Axis I comorbidity for the depressed group was as follows; Panic Disorder without Agoraphobia (n = 1), Generalised Anxiety Disorder (GAD, n = 10), Obsessive Compulsive Disorder (OCD, n = 4), Social Phobia (n = 9), Specific Phobia (n = 3), and Eating Disorder Not Otherwise Specified (EDNOS, n = 3). In addition, five participants in the currently depressed group were taking antidepressant medications (Sertraline, n = 2, Fluoxetine, n = 1, Escitalopram, n = 1, and Citalopram, n = 1). Six (24%) participants from this group were receiving psychological or counselling treatment at the time of interview. For the recovered depressed group, twelve participants reported having only one previous episode (40%). Of the remainder of participants in this group, 13 reported having two to three previous episodes, three reported having four to five past episodes, one participant reported having too many episodes to count, and one participant was uncertain as to the number of past episodes. For the recovered group, the total number of Axis I diagnoses per participant ranged between 0 and 4. Twelve participants from the recovered depressed group met criteria for at least one Axis I diagnosis. Axis I diagnoses for this group was as follows; Agoraphobia without Panic Disorder (n = 1), GAD (n = 4), OCD (n = 2), Social Phobia (n = 7), Specific Phobia (n = 4), and EDNOS (n = 1). Seven participants (23.3%) from the recovered depressed group were receiving psychological or counselling treatment at the time of interview. Three participants in the recovered group were taking antidepressant medications (Effexor, n = 1, Meclobemide, n = 1, Sertraline, n = 1). For the never-depressed control group, the total number of Axis I diagnoses per participant ranged between 0 and 2. Six individuals in the never-depressed group met criteria for a diagnosis of at least one DSM-IV Axis I disorder based on responses to the SCID-I/NP (First et al., 1996). Axis I diagnoses for the never-depressed group was as follows; GAD (n = 1), OCD (n = 2), Social Phobia (n = 1) and Specific Phobia (n = 3). Four participants (13.3%) from this group were receiving psychological or counselling treatment at the time of interview. Analyses One-way analyses of variance (ANOVAs) were carried out to compare the currently depressed, recovered depressed and never-depressed groups on mood measures, as well as intrusive memory variables. Significant ANOVA’s were followed up with Tukey pairwise between-group comparisons. Mood The groups differed on both BDI-II and BAI measures (see Table 1 for means): BDI-II (F(2,82) = 82.13, p < .01), BAI (F(2,82) = 23.75, p < .01). The depressed group reported higher BDI-II and BAI scores than the other two groups (ps < .01). The recovered group reported higher BDI-II scores than the never-depressed group (p > .05). No other group differences were found on these measures (ps > .05). General characteristics of negative intrusive memories The majority of participants reported a specific intrusive memory of a negative life event. Twenty-four participants (96%) in the currently depressed group, 24 participants (80%) in the recovered depressed group, and 22 participants (73.3%) in the never-depressed group reported specific negative intrusive memories. The percentage of participants that reported intrusive memories did not differ according to group (χ2 (2, N = 85) = 5.00, p = .082). Data from six participants (two from each group) were excluded from the remaining analyses as their memories were reported to have last occurred greater than 12 weeks prior to interview. Thus, this left 22, 22 and 20 memories for the currently depressed, recovered depressed and never-depressed groups respectively for remaining analyses. Negative interpersonal events accounted for approximately 70% of memories reported across the three groups. Interestingly, participants’ subjective ratings of distress did not differ according to intrusive memory content (F(4,59) = 2.13, p = 0.088) (see Table 2 for the distribution of memory content across groups). Table 2. Content, impact of event scale and general characteristics of negative intrusive memories. Currently depressed n = 22 Recovered depressed n = 22 Never-depressed n = 20 Content (frequency/%) Interpersonal event 16 (72.7) 15 (68.2) 15 (75.0) Death/illness involving other 3 (13.6) 2 (9.1) 1 (5.0) Personal assault/abuse 1 (4.5) 2 (9.1) 1 (5.0) Illness/injury involving self 0 (0.0) 1 (4.5) 0 (0.0) Other 2 (9.1) 2 (9.1) 3 (15.0) Impact of event scale IES intrusion 17.82 (7.02) 17.27 (9.28) 12.90 (6.78) IES avoidancea 20.23 (10.25) 15.95 (8.10) 11.60 (8.24) General characteristics Time since memory occurred 1.45 (0.86) 3.27 (4.10) 1.80 (1.44) Frequency past week 3.75 (2.57) 3.00 (2.41) 2.79 (2.22) Frequency within daya 3.60 (3.68) 1.92 (1.12) 1.43 (0.94) Time since event (weeks) 94.32 (191.25) 136.05 (219.39) 149.85 (258.23) Age at event 24.09 (7.85) 21.59 (7.47) 19.30 (6.37) Vantage perspective −1.95 (1.32) −1.30 (2.08) −1.04 (2.21) MEQ Vividness 24.39 (4.88) 25.78 (4.41) 23.81 (6.04) Coherence 32.91 (5.90) 31.62 (6.38) 34.00 (5.18) Accessibility 19.00 (4.45) 20.62 (3.36) 20.33 (3.79) Sensory detail 28.56 (7.02) 26.56 (7.15) 26.57 (7.55) Time perspective 21.70 (6.44) 20.96 (6.80) 21.14 (6.67) Visual perspective 25.00 (5.55) 22.46 (6.74) 22.48 (7.84) Emotional intensitya 24.65 (3.96) 22.87 (5.94) 18.90 (6.75) Note. IES = Impact of event scale, MEQ = Memory experiences questionnaire. Except where noted, values refer to means (standard deviations). a Tukey test revealed the depressed group reported higher ratings on these variables than never-depressed group. Table options There was a trend towards a significant difference in the time (in weeks) since the memory last occurred prior to interview (F(2,61) = 3.05, p = .055). There were no significant differences between groups in terms of the time elapsed since the event in the memory occurred or participants’ age at the time when the remembered event occurred (ps > .05) (see Table 2 for means). Length and frequency of negative intrusive memories Participants’ estimates of the duration of their memories (i.e., the amount of time that intrusive memories remained in mind) ranged from lasting less than one minute to over an hour. The majority of participants in all groups reported intrusive memories that lasted five minutes or less (72%, 77%, and 70% for depressed, recovered and never-depressed groups respectively). Four participants in the currently depressed group reported memories that lasted over an hour, whereas one participant from the recovered depressed group (and none from control group) reported a memory of this length. The participants who reported that they had experienced their intrusive memory in the week prior to interview were asked to estimate how many days they had experienced the intrusive memory in the preceding week, and how frequently their memory occurred during the day. The groups did not differ on how frequently they experienced the memory during the week (p > .05). However, significant differences were found in the frequency within a given day (F(2,39) = 3.40, p < .05). The never-depressed group experienced their memories less frequently during the day compared to depressed participants (p < .05) (no other group differences were found, ps > .05) (see Table 2 for means). Intrusion and avoidance of intrusive memories The groups did not differ on IES Intrusion scores (F(2,61) = 2.46, p = 0.094). However, they differed on IES Avoidance scores (F(2,61) = 4.88, p < .05) (see Table 2 for means). The depressed group reported higher levels of avoidance than never-depressed group (p < .01). There were no other group differences found for avoidance scores (ps > .05). Specific characteristics of intrusive memories and associated emotions Intrusive memory interview results Sensory features For all three groups, the majority of memories reported were described as having a sensory quality. The predominant sensory modality reported was that the memory was a visual image. Across groups, 32%, 23%, and 25% reported visual image only (for depressed, recovered and never-depressed groups, respectively), 9%, 32% and 20% reported thoughts only (for depressed, recovered and never-depressed groups respectively), and 49%, 45% and 50% reported that their memory consisted of a visual image in combination with other elements (for depressed, recovered and never-depressed groups, respectively). There were no differences across groups in the percentages of thoughts (χ2 (2, N = 64) = 3.51, p > .05), percentages of visual images (χ2 (2, N = 64) = .50, p > .05), or percentages of visual + other elements (χ2 (2, N = 64) = .74, p > .05) that were reported. General memory characteristics For the IMI variables, there were significant between-group differences for ratings of vividness (F(2,62) = 3.92, p < .05), distress (F(2,61) = 4.02, p < .05) and interference caused by the memory (F(2,61) = 4.67, p < .05). The depressed group reported higher overall ratings of memory vividness, distress and interference with daily activities than the never-depressed group (ps < .05). There were no other group differences found for memory vividness and distress (ps > .05), suggesting that the recovered group rated intermediate levels of these variables. However in contrast, for the interference ratings, the recovered group reported lower ratings of interference than the depressed group (p < .05), whereas there were no differences on interference ratings between never-depressed and recovered groups (p > .05). There were no group differences on the remaining characteristics – that is, nowness/reliving the event, autonoetic awareness, lack of context, observer perspective, and ratings of uncontrollability ( Fig. 1 presents group means on IMI variables). Intrusive Memory Interview items. VAS=visual analogue scale, where 0=not at all, ... Fig. 1. Intrusive Memory Interview items. VAS = visual analogue scale, where 0 = not at all, and 100 = very much. Figure represents means, and error bars represent standard errors (SEM). SEM’s are presented rather than standard deviations due to constraints of figure. * = CD > ND, # = CD < RD. Figure options Vantage perspective The groups did not differ in terms of the vantage perspective from which they recalled their intrusive memory, as rated by the IMI (see Table 2 for means) (F(2,57) = 2.24, p > .05). Of the participants who reported memories that were included in the analysis, 14, 16 and 14 participants (from depressed, recovered and never-depressed groups, respectively) scored responses at the extreme ends of the continuum on the vantage perspective scale. Of these, 100%, 81%, and 71% of these memories were coded as first person perspective, respectively, across the three groups. There were no between-group differences in the percentage of first person perspective memories (χ2 (2, N = 44) = 4.42, p > .05) Intrusion-related emotions In terms of specific emotions on the IMI (see Fig. 2 for means), the groups differed in their ratings of helplessness (F(2,61) = 4.78, p < .05), sadness (F(2,61) = 3.69, p < .05) and anger (F(2,61) = 3.28, p < .05). Tukey tests revealed that the depressed group reported higher levels of sadness and helplessness than the never-depressed group (ps < .05). The depressed group also rated higher levels of helplessness than the recovered group (p < .05). There were no other between group differences for sadness and helplessness (ps > .05). However, for anger, there was a trend towards the depressed group reporting higher anger scores than the recovered group (p = .06). Interestingly, the groups did not differ in their ratings of memory-related detachment, numbness, guilt, surprise, happiness or shame (all ps > .05). However, there was a trend towards a difference for fear (F(2,61) = 2.69, p = .08) and disgust (F(2,61) = 2.81, p = .068). Although these differences did not reach significance, an inspection of the means suggests a trend towards higher ratings for the depressed group across most negative emotions (except for guilt) in comparison with the other groups. Intrusive Memory Interview items assessing emotional responses to intrusive ... Fig. 2. Intrusive Memory Interview items assessing emotional responses to intrusive memories. VAS = visual analogue scale, where 0 = not at all, and 100 = very much. Figure represents means and error bars represent standard errors (SEM). SEM’s are presented rather than standard deviations due to constraints of figure. * = CD > ND, # = CD < RD. For anger, there was a trend observed, where CD > RD (p = .06). Figure options Memory experiences questionnaire results The groups differed on the MEQ Emotional Intensity subscale (F(2,60) = 4.75, p < .05) (see Table 2 for means). The depressed group reported higher MEQ Emotional Intensity scores than the never-depressed group (p < .01). There were no other between-group differences on this item (ps > .05). There were no overall group differences found on MEQ ratings of MEQ Vividness, MEQ Coherence, MEQ Accessibility, MEQ Sensory Detail, MEQ Visual Perspective, and MEQ Time Perspective (all F’s < 2.4, ps > .05). The MEQ Accessibility results should be interpreted with caution given the low reliability estimate for this scale in our sample. ( Table 2 presents means for MEQ items). 4