香京julia种子在线播放

A nationwide record linkage of the Danish Twin Registry (48) and the Danish Psychiatric Central Research Register (DPCRR) (49) identified eligible monozygotic twins. In addition to monozygosity, eligibility criteria were i) a personal or co-twin history of an affective spectrum diagnosis (i.e. International Classification for Diseases ICD-10 codes F30-34.0 and F38.0) (17) or for low-risk twins neither a personal nor a co-twin history affective spectrum diagnosis from January 1995 to June 2014, and ii) age 18–50 years.

Exclusion criteria for all groups were: birth weight under 1.3 kg, history of brain injury; current severe somatic illness, current substance abuse; current mood episode defined by Hamilton Depression Rating Scale (HDRS-17) (50) or Young Mania Rating Scale (51) (YMRS; scores >14), current pregnancy, or being dizygotic. The low-risk twins were also excluded if they reported other first-degree relatives with organic mental disorder, schizophrenia spectrum, or affective disorders.

Participants provided their written and informed consent to the study in accordance with the Helsinki declaration. The study was approved by the ethics committee for the Capital Region of Denmark (H-3-2014-003) and the Danish data protection agency (2014–331–0751).

Recruitment took place from December 2014 until January 2017. From an initial sample of 215 participants, 11 participants were excluded due to missing diagnoses (n = 6), affective disorder (n = 4), or alternative high-risk affective disorder (n=1). The final sample for this paper included 204 participants, with each classified as affected (n = 115; BD = 31 and UD = 84), high-risk (n = 49; BD = 11 and UD = 38), or low-risk (n = 40). There were 25 concordant affected twin-pairs (BD/BD: n = 5; UD/UD: n = 11; BD/UD: n = 9), 45 discordant twin-pairs (high-risk/UD: n = 36; high-risk/BD: n = 9), 19 low-risk twin pairs, and 26 single twins.

Participants attended a 1-day assessment at the Danish Research Centre for Magnetic Resonance at Copenhagen University Hospital Hvidovre. Further data from this sample have been reported elsewhere, including neurocognitive (52–55), clinical/psychological (56), and biological outcomes (57). Here we report data on assessments related to mental imagery for the first time.

Outliers above or below three standard deviations (SDs) of the mean were excluded. Groups were compared on baseline and clinical characteristics and on key outcomes from each imagery assessment based on previous literature (32). Each continuous dependent variable was examined with mixed-model analysis of variance with groups as fixed factors and modeling random effects for twin pairs to account for dependence within these. Categorical variable (sex only) was examined with logistic regression with groups as predictors and within twin-pair dependence adjusted for using generalized estimating equations estimates of the standard errors.

For our primary analyses, we included our three key groups, i.e., affected in remission (combining BD and UD), high-risk and low-risk, followed by pairwise comparisons when relevant. Comparing the affected (remitted) group with the low-risk group would help determine whether and which imagery abnormalities persist in remission; comparing the high-risk group with the other two groups would help determine whether and which imagery abnormalities are present in individuals who have not developed BD/UD despite genetic liability (consistent with the notion of an endophenotype for affective disorders). Effect sizes as Cohen d (67) were included to aid interpretations (0.30 = small; 0.50 = medium; 0.80 = large).

We conducted three additional sets of secondary (exploratory) analyses. First, we repeated the above analyses with only BD or UD on selected imagery measures for which previous literature indicates discrepancy between disorders (and hence combining both groups may not be always appropriate): i) emotional imagery of positive valence (PIT and MII) may be enhanced in BD only (35, 68, 69) and ii) aspects of mental rotation performance may be affected in BD (32) and UD (41). Second, we directly contrasted the BD and UD affected (remitted) groups, as previous studies have contrasted imagery measures between BD and UD groups predominantly during depression illness at the time of assessment (32). Finally, we conducted a series of multiple linear regressions with baseline clinical characteristics (HDRS and STAI-Y) as predictors, and key imagery outcomes (emerging as significant from our primary analyses) as dependent variables, based on our previous research demonstrating associations between emotional imagery and symptoms transdiagnostically (32). Manic symptoms (YMRS) were not included as these were matched between groups (see Results). Initially both predictors were entered simultaneously, and then non-significant predictors were removed from the model stepwise until only significant predictors remained.

Significance level was set to alpha = .05 for two-sided hypothesis-testing (unless otherwise stated for directional hypothesis-testing). As our primary analyses involved multiple comparisons, we also applied the Benjamini-Hochberg procedure (70) to control for family-wise false discovery rate (q = .10), and we reported these when they changed the pattern of results (for one pairwise comparison only). Data analyses were conducted using SPSS (71).

Means and SDs are presented in Table 1 . There were no significant group differences in sex, age, education, or IQ. As expected, the affected twins scored higher (than high-risk and low-risk groups) on baseline symptoms of depression and anxiety, but there were no significant differences between high-risk and low-risk twins. There were no significant group differences in symptoms of mania. A high proportion of affected twins in remission reported current pharmacological medications ( Table 1 ). Restricting our primary analyses to twins without medication did not change the pattern of results, hence we report results including all twins.

This large study of 204 monozygotic twins investigated—for the first time—whether mental imagery abnormalities are present in i) affected twins with affective disorders in remission and ii) twins with high familial risk for affective disorders. Regarding the first aim, we found that remitted twins with a history of affective disorders reported greater emotional impact of intrusive prospective imagery, and greater vividness, likelihood, and subjective experience of (deliberately-generated) negative prospective imagery compared with unaffected twins at either low-risk or high-risk of affective disorders. That is, affected twins are more prone to imagining the future (whether a desired holiday or a feared examination) and when they did so it felt more real and emotional. The affected twins in remission also reported less likelihood of (deliberately-generated) positive prospective imagery compared to both high-risk and low-risk twins (however, the latter result did not survive correction for multiple comparisons). In contrast, the twin groups did not differ on measures of cognitive stages (non-emotional) of mental imagery.

In relation to the second aim, we found no evidence of mental imagery abnormalities in twins with high familial risk for affective disorders but who were nevertheless not affected. Finally, mood and anxiety symptoms were significantly associated with greater emotional impact of prospective imagery across our entire sample, replicating findings that support a dimensional and transdiagnostic role of imagery abnormalities in psychopathology (32).

Our findings extend previous work identifying abnormalities in prospective emotional mental imagery across affective disorders in the presence of significant depressive symptoms (7, 32, 33, 37). Interestingly, our pattern of results were the same for analyses conducted using a sub-sample restricting the affected group in remission to only those individuals with no residual subclinical symptomatology (HDRS scores <8; n = 89). This suggests that imagery abnormalities are not just related to residual affective symptomatology—these critically may represent a trait-related phenomenon in individuals with affective disorders. The distinction between “state” or “trait” needs to be further delineated, for example using direct comparisons of participants during remission versus during illness episodes (while acutely depressed and/or in manic/mixed episodes). While imagery-based abnormalities that are equivalent during both illness and remission are consistent with the notion of “trait” markers, those abnormalities that more pronounced during illness episodes may additionally represent “state” markers.

The association between biases in prospection and psychopathology remains understudied, although recent research suggests its potential relevance for depression (72) and anxiety (73). Our study indicates, for the first time, that individuals with a history of affective disorders continue to experience vivid intense images of negative future events even during long-term remission, similar to other cognitive-emotional biases including imagery of the past (27). Instead, abnormalities in positive prospective imagery appeared limited to likelihood ratings across affective disorders (although these results did not survive more stringent corrections for multiple comparisons), and did not differentiate BD versus UD (both in remitted states). Future replications need to confirm if the inability to experience vivid positive images of future events as previously reported is only prominent during illness/depressive states (63, 74).

Our findings also highlight that unlike emotional imagery, cognitive (non-emotional) imagery processes appear intact across affective disorders in remission. While we previously found some evidence of both better and worse performance on cognitive stages of imagery in a sample of mixed euthymic and depressed BD (32), the present study confirms that cognitive stages of imagery remain largely preserved in affective disorders (32). These cognitive processes rely on general executive function as well as on more specific visuospatial abilities (1). How these processes—as measured by cognitive imagery tasks—compare to other components and measures of cognition remains unclear. This is relevant as executive function in particular is often compromised in recovered individuals with affective disorders (75, 76). However, affective disorders are largely heterogeneous with regards to cognition (77, 78); for instance, affected twins in our sample did not show any cognitive deficits compared to low-risk twins (54). Answering the above question may be relevant with a view to personalizing therapeutic interventions. As examples, common cognitive therapy techniques such as reappraisal may rely on well-functioning cognition and hence be less efficacious if executive function is compromised, and likewise, imagery-based techniques could be advantageous in those with intact cognitive stages of imagery.

Abnormalities in prospective imagery were not present in high-risk twins compared to low-risk twins, indicating that these do not fulfill the criteria for an illness “endophenotype” (28). Our data are more consistent with the notion that imagery-based abnormalities reflect (neuro)cognitive markers of the disorder itself—abnormalities that are likely to play a role in increasing future relapse given their persistence in remitted states. We note that previous research has shown altered emotional mental imagery in individuals with phenotypic (rather than genetic/familial) risk for BD, mainly based on the presence of hypomanic-like experiences (32, 43, 69, 79). However, our study was the first to focus instead on genetic-based risk using a twin design enabled by the unique Danish registers (49). As a potential explanation for this discrepancy, maladaptive prospective emotional imagery may be associated with phenotypic characteristics of affective disorders that lie on the same dimension of clinical symptoms, such as the actual presence of dysphoric mood (47, 62) or hypomanic-like experiences (32, 43, 69, 79), but we did not measure such features in our high-risk twins. Another potential explanation is that our high-risk twins had an average higher age compared to at-risk groups from previous research. Thus it is also possible that our sample included individuals who had a familial risk but were actually “resilient” to developing affective disorders, at least in terms of mental imagery characteristics. We have previously argued that higher extroversion and lower neuroticism may mediate sensitivity to adverse events and better functioning in this high-risk sample (56), although only future longitudinal studies will be able to clarify factors determining vulnerability or resilience.

Notably, we found no evidence of differences between affective disorders categories of BD vs. UD. The only exception was that remitted BD twins retrospectively described their imagery at times of depressed mood as more compelling compared to remitted UD twins. This is consistent with previous findings on suicidal imagery (37), suggesting that when asking about past mental images during depressive episodes it may be important to distinguish for specific suicidal content to clarify potential phenomenological differences. We also found that greater emotional impact of prospective imagery was associated with anxiety and low mood symptoms across our whole sample. If replicated in samples with a larger number of individuals with BD, overall these results will indicate that individuals with affective disorders may share a trait-like common profile of mental imagery characteristics, in line with a most recent dimensional and transdiagnostic model of the role of emotional imagery in psychopathology (32).

If prospective emotional imagery remains abnormally enhanced after recovery from affective disorders episodes, this could represent a vulnerability factor for future relapses through their impact on emotion and mood instability. We have previously proposed that vivid intense mental images may act as an emotional amplifier and maintain mood instability (7, 29). These novel findings highlight the importance of future longitudinal studies investigating whether imagery abnormalities during remission indeed predict subsequent illness recurrence rates.

Prospection has a key functional role in individuals’ daily life: we imagine future scenarios as a way to plan action, anticipate potential events and direct decision-making, manage uncertainty, and strengthen motivation toward goals (30, 31). Future imagery can also support emotion regulation, e.g., by anticipating future rewards to overcome present difficulties (80, 81). If prospection is biased in those with past affective episodes, remitted individuals will experience negative future scenarios more vividly, more likely to happen, and such intrusive images of the future will carry maladaptive emotional impact.

Finally, our findings indicate that imagery-based treatment innovation could be guided toward tertiary rather than primary prevention—at least in relation to familial risk. Building on previous therapy protocols (82–84) and recent experimental studies (85), targeting prospective imagery abnormalities using imagery-based cognitive therapy techniques has potential as a brief relapse prevention intervention, in particular in patients with greater mood instability (86). Emerging work in cognitive science also suggests that simple competing tasks can dampen recurrent “flashforward” imagery by taxing working memory while holding the flashforward image in mind (15, 87). Future studies should measure change in prospective imagery characteristics after imagery-based interventions to develop novel interventions based on mechanistic hypotheses (16).

Several limitations should be noted in our study. First, we did not include a dizygotic twins group; hence, we could not assess the interaction between environmental and genetic influence on variance, which is preferable when investigating potential endophenotypes. Second, we did not directly compared twins at high-risk for UD versus at high-risk for BD. Such comparison could establish whether familial risk in imagery abnormalities is more evident in high-risk twins given a co-twin history of BD rather than UD, because of the higher heritability of BD (19) relative to UD (20). Unfortunately we had low power to run those analyses because the high-risk for BD group was very small (n = 11). Third, we did not include a full comprehensive assessment of mental imagery (40), so it is possible that biases in other imagery domains—such as trauma memory imagery (5, 14, 88) and/or reward/motivational imagery (89, 90)—are also present in remitted affective disorders and/or high-risk twins. This might be relevant to the prevalence of trauma in affective disorders (91) and of persistent anhedonic symptomatology (92), which are both associated with relapse and worse prognosis. Finally, results from our analyses comparing patients affected with UD versus BD (n = 84 versus n = 31, respectively) and from our secondary analysis should be interpreted with caution, given the small sample size of the separate BD group (n = 31 affected and n = 11 high risk, respectively).

Major strengths of our study are the large and population-based twin sample (enabling a unique design to test hypotheses regarding familial risk), twin groups well-matched on demographic variables, and the use of validated scales and tasks that allow building on previous research and future replication. Further, participants had been in long-term remission providing greater confidence that findings were not state-related (although it remains possible that imagery is also linked to states). Importantly, our findings are unlikely to be attributed to other cognitive confounders, as our groups did not differ in general cognitive function (54). Finally, to our knowledge, this was also the second study ever to combine emotional in tandem with cognitive (non-emotional) testing of imagery abnormalities in a clinical sample.