ADHD Atten Def Hyp Disord (2018) 10:151–160 https://doi.org/10.1007/s12402-017-0241-x ORIGINAL ARTICLE ADHD symptoms in healthy adults are associated with stressful life events and negative memory bias 1,2 1 3,4 3,4 Janna N. Vrijsen · Indira Tendolkar · Marten Onnink · Martine Hoogman · 1 1,4 1 1,3,4 Aart H. Schene · Guillén Fernández · Iris van Oostrom · Barbara Franke Received: 15 May 2017 / Accepted: 9 October 2017 / Published online: 28 October 2017 © The Author(s) 2017. This article is an open access publication Abstract Stressful life events, especially Childhood Introduction Trauma, predict ADHD symptoms. Childhood Trauma and negatively biased memory are risk factors for affective dis - Attention-deficit/hyperactivity disorder (ADHD) is prevalent orders. The association of life events and bias with ADHD in children (Polanczyk and Rohde 2007) and persists into symptoms may inform about the etiology of ADHD. Mem- adulthood in up to 60% of these young patients, resulting ory bias was tested using a computer task in N = 675 healthy in adult prevalence rates of 2.5–5% (Kessler et al. 2005; adults. Life events and ADHD symptoms were assessed Simon et al. 2009). ADHD is characterized by sustained and using questionnaires. The mediation of the association debilitating symptoms of inattention and/or hyperactivity/ between life events and ADHD symptoms by memory bias impulsivity (Faraone et al. 2015). The diagnostic and statisti- was examined. We explored the roles of different types of cal manual of mental disorders (DSM; American Psychiatric life events and of ADHD symptom clusters. Life events and Association 2013) classification ADHD is considered to rep - memory bias were associated with overall ADHD symptoms resent the extreme of a continuum of traits in the population as well as inattention and hyperactivity/impulsivity symptom in children and adults (Chen et al. 2008; Franke et al. 2012; clusters. Memory bias mediated the association of Lifetime Hoogman et al. 2012; Lubke et al. 2009). ADHD frequently Life Events, specifically Childhood Trauma, with ADHD occurs co-morbid with emotional problems, i.e. anxiety dis- symptoms. Negatively biased memory may be a cognitive orders and major depression (Herrmann et al. 2010; Ueker- marker of the effects of Childhood Trauma on the develop - mann et al. 2010), and this co-occurrence is higher for the ment and/or persistence of ADHD symptoms. inattentive than for the hyperactive/impulsive presentation (Friedrichs et al. 2012). Keywords ADHD · Childhood Trauma · Stressful The aetiology of ADHD is very heterogeneous and there events · Memory bias · Persistence are several developmental pathways leading to the same outcome. A growing body of research links ADHD onset and severity to psychosocial and neurocognitive factors, such as the experience of stressful life events and memory * Janna N. Vrijsen Janna.Vrijsen@radboudumc.nl processes. The exposure to stressful life events, and—more specifically—Childhood Trauma, has been shown to pre - Department of Psychiatry, Radboud University dict ADHD onset as well as persistence of the disorder into Medical Center, Donders Institute for Brain, Cognition adulthood (Biederman et al. 1995; Friedrichs et al. 2012; and Behaviour, PO Box 9101, 6500 HB Nijmegen, The Netherlands Sugaya et al. 2012), as well as the onset of other psychiat- ric disorders e.g. depression and anxiety disorders (Hovens Pro Persona Mental Health Care, Depression Expertise Center, Nijmegen, The Netherlands et al. 2010; Kessler et al. 1997). In total, some 20–50% of children with a history of Childhood Trauma have clinical Department of Human Genetics, Radboud University Medical Center, Nijmegen, The Netherlands levels of ADHD (Glod and Teicher 1996; McLeer et al. 1994). There also is some evidence that ADHD symptom Department of Cognitive Neurosciences, Radboud University Medical Center, Nijmegen, The Netherlands expression may differ between children with and without Vol.:(0123456789) 1 3 152 J. N. Vrijsen et al. Childhood Trauma: traumatized children have been found to To explore the relevance of the cognitive model (Beck be less hyperactive (Glod and Teicher 1996). Besides Child- 2008; Beck and Haigh 2014) for ADHD and improve our hood Trauma, recent stressful events, such as conflicts at knowledge about the relation between stressful life events, work, divorce, and monetary problems, are also associated memory bias, and ADHD, we here performed association with levels of ADHD severity (Able et al. 2007; Biederman studies in a large sample of self-reported healthy adults, and Faraone 2006; Sobanski et al. 2007). which provided information over a large range of ADHD Childhood Trauma may result in altered cognitive pro- symptom severity (Hoogman et al. 2012). In line with the cessing (Perry 2008). Specifically, the cognitive model by cognitive model (Beck 2008; Beck and Haigh 2014) and fol- Beck (2008) and Beck and Haigh (2014) assumes that the lowing association studies of the different individual compo- experience of traumatic events during childhood can result in nents, we assessed the possible mediation of the association dysfunctional basic assumptions about the self and the world. between life events and self-reported inattention and hyper- Stressful events in turn trigger these assumptions. Informa- activity/impulsivity symptoms by memory bias. Given the tion is processed in accordance with these assumptions, set- broad array of stressful life events—ranging from childhood ting the stage for cognitive biases and increasing the risk for sexual abuse to financial problems—found associated with the development of psychiatric problems such as depression. adult ADHD (symptom severity) in previous studies, we first One frequently studied cognitive bias is negative memory examined the association of lifetime stressful events (life- bias, the preferential and more frequent recall of negative time life stress) with ADHD symptoms as well as mediation compared to positive material (Mathews and MacLeod of this association by memory bias. Our second aim was to 2005). This type of bias is most pronounced for self-relevant explore the specificity of type of life events on ADHD symp- material (Steinberg et al. 2003) and is a stable risk factor toms. Hence, Childhood Trauma (early life stress) and recent for subclinical and clinical levels of emotional problems (De stressful events (recent life stress) were both association with Raedt and Koster 2010; Gotlib and Joormann 2010). ADHD symptom severity. Because both the presence as well Recent evidence from a large naturalistic psychiatric as the diversity of stressful events has been related to psy- cohort suggests that negative memory bias may be a cogni- chopathology (Hovens et al. 2010; Vrijsen et al. 2017), we tive marker for a broad range of mental disorders, including used variables representing the number of events. We also ADHD (Vrijsen et al. 2017). Importantly, the association of explored the association between life events and inattention Childhood Trauma, negative memory bias, and psychiatric and hyperactivity/impulsivity symptom clusters separately. problems remained when excluding depressed patients from The results will provide a first indication of the relevance of the analyses, for which the link with bias had earlier been Childhood Trauma and memory bias for ADHD symptoms shown. This supports the current proposition that psychiatric and may instigate further research. disorders share underlying behavioral and neurobiological dimensions (Insel et al. 2010; Insel 2014), one of which is negative memory bias. Regardless of the preliminary evi- Methods and materials dence for stressful life events in concordance with negative memory bias as global risk factors for psychopathology, Participants biased memory processing of emotional information has only sparsely been directly addressed in ADHD research. The study was performed using the Cognomics Initiative A first study reported an association of both inattention and Resource, the Brain Imaging Genetics (BIG) sample (http:// hyperactivity/impulsivity symptoms with memory bias for www.cognomics.nl), which consists of self-reported healthy angry faces compared to happy faces (d’Acremont and Van mainly young adults. Subsamples of the current sample have der Linden 2007). Preliminary evidence for memory bias previously been described elsewhere (e.g. Hoogman et al. in ADHD also came from a small study by Krauel et al. 2012; van Oostrom et al. 2012; Vogel et al. 2014; Vrijsen (2009), who found that adolescents with ADHD and comor- et al. 2015a). Participants were screened using a self-report bid externalizing problems showed less positive memory questionnaire for the following exclusion criteria: history bias compared to healthy controls and ADHD-only partici- of somatic disease potentially affecting the brain, current or pants. Examining memory bias in relation to ADHD symp- past psychiatric or neurological disorder, use of medication toms may inform about the aetiology of ADHD. Moreover, (except hormonal contraceptives) or illicit drug(s) during cognitive markers—i.e. memory biases—offer a relatively the past 6 months, history of substance abuse, current or easy target for (computerized) treatment (Hertel et al. 2017; past alcohol dependence, current pregnancy or lactation, and Koster and Bernstein 2015; Mathews and MacLeod 2002; menopause. All participants were fluent in Dutch. A total of Vrijsen et al. 2016). Hence, examining the role of memory 785 participants completed an ADHD symptom question- bias in ADHD may also aid in the development of treatment naire (see below; Kooij et al. 2005). Of those participants, tools. memory bias data was available for 675 individuals. The 1 3 ADHD symptoms in healthy adults are associated with stressful life events and negative memory… 153 BIG study was approved by the regional medical ethics com- studies (e.g. van Oostrom et al. 2012; Vogel et al. 2014; mittee. All participants gave written informed consent and Vrijsen et al. 2014a, b). Participants were asked to indicate, were financially compensated for participation. whether they had experienced a set of 21 life events before the age of 16 years, after the age of 16, and/or within the ADHD symptoms last year. In line with the study by Vogel et al. (2014), a ‘Lifetime Life Events’ variable was calculated, indicating The ADHD DSM-IV-TR Rating Scale for use in adults was the total number of experienced life events. Based on previ- used to assess current ADHD symptoms (Kooij et al. 2005). ous studies (van Oostrom et al. 2012; Vrijsen et al. 2014a, b, This instrument has shown internal and external validity in 2015b), a ‘Childhood Trauma’ variable and a ‘Recent Stress’ a large population-based adult sample (Kooij et al. 2005). variable were also calculated. ‘Childhood Trauma’ indicated Symptoms in the last 6 months were reported on a 4-point the number of different traumatic events (aggression, sexual, scale. A symptom was considered to be present if partici- and/or physical abuse) the participant experienced within or pants answered ‘often’ or ‘very often’. The scores on the outside the family before the age of 16 years. ‘Recent Stress’ 23 items were recalculated to the original 18 DSM-IV-TR reflected the number of different stressful events (health ADHD criteria, of which nine criteria are related to the inat- problems, health problems of a close one, death of a family tention (IA) symptom domain and nine to the hyperactiv- member, problems within the romantic relationship, divorce, ity/impulsivity (HI) symptom domain. The variables ‘Total a conflict at work, monetary problems, or legal issues) the ADHD’ symptoms (possible range 0–18),’ IA-symptoms’ participant experienced within the last year. (range 0–9), and ‘HI-symptoms’ (range 0–9) were derived from the data (for more details see Hoogman et al. 2012). Memory bias The Total ADHD variable, as well as the IA-symptoms and the HI-symptoms subscales had acceptable reliability in the Memory bias was assessed using a web-based version of current sample, α = .78, α = .72, and α = .64, respectively. the self-referent encoding/evaluation task (SRET; Hammen and Zupan 1984). The SRET has been used in previous Positive and negative affective state publications from BIG (Gerritsen et al. 2012; van Oostrom et al. 2012; Vogel et al. 2014; Vrijsen et al. 2015). The task The Positive and Negative Affect Schedule (PANAS; Watson had explicit instructions. During encoding, 12 negative and et al. 1988) was used to assess affective state. The PANAS 12 positive trait adjectives were presented one by one on a including its subscales have been shown to be reliable and screen in fixed random order. Participants were instructed to valid for measuring positive and negative affect in a large indicate for each word, whether it was self-referent or not by non-clinical sample (Crawford and Henry 2004). This instru- pressing one of two buttons. Following a 2.5 min distraction ment comprises two mood scales, one for the assessment of task (mental arithmetic), participants were asked to type in positive affect and one for negative affect. Ten descriptors as many of the words they could remember from the encod- are used for each scale to define their meanings, resulting ing phase within 3 min. Responses to the first and last two in a 20-item questionnaire using a 5-point scale that ranges words of the word list were excluded from analysis in order from ‘very slightly or not at all’ to ‘extremely’. Because of to avoid primacy and recency effects. Incorrect responses the relevance for memory bias and risk for psychopathology were checked manually, and spelling errors as well as plu- (e.g. Bower 1981), we selected the negative affect subscale rals if the original word was singular (and vice versa) were (possible range 10–50) for this study. This subscale has been permitted. found to correlate highly with depressive symptom levels In line with previous the studies using this task, two out- (Tarlow and Haaga 1996). This subscale was found to be come variables were calculated: proportion of self-referent highly reliable in the current sample (10 items; α = .88). negative recall (negative memory bias) and proportion of self-referent positive recall (positive memory bias). The pos- Stressful life events itive and negative memory bias variables were calculated by dividing the number of adjectives endorsed as self-referent Stressful life events were assessed with an adapted version and recalled in a given valence category by the total number of the List of Threatening Evens Questionnaire (Brugha and of self-endorsed adjectives. Cragg 1990). The original instrument has good test–retest reliability, high agreement between participant and inform- Statistical analyses ant ratings, as well as good agreement with interview-based ratings specificity (Brugha and Cragg 1990). The adapted Age and gender were entered as covariates in the media- instrument has not been validated, but has previously been tional model analyses based on the associations of these var- used in studies on biased processing including patient iables with the independent variables: Lifetime Life Events 1 3 154 J. N. Vrijsen et al. and memory bias. The PANAS negative affect scale total Table 1 Sample descriptives including means (standard deviations) or percentages and range and absolute numbers for the variables: age, score was added as covariate in all analyses to correct for sex, negative affect, life events, and ADHD symptoms (N = 675) variation due to differences in negative ‘depressotypic’ state. Variable Mean (SD) and range and Prior to analyses, log transformation was applied in case of absolute numbers; or % non-normally distributed scores. After transformation, the of total values for asymmetry all lay between − 2 and + 2, and were hence acceptable in order to prove normal univariate distri- Age in years 22.7 (3.8), range 18–39 bution (George and Mallery 2010). One-tailed bivariate cor- Sex (% female) 62% relations were calculated between the life events variables, PANAS negative affect score 13.5 (4.8), range 10–47 ADHD symptoms, and positive and negative memory bias. Experienced Lifetime Life Events 99% For these correlations, Bonferroni multiple testing correc- Number of different Lifetime Life Events 5.1 (3.3), range 0–23 tion was based on the number (6) of comparisons among Experienced Childhood Trauma 21% life events, positive bias, negative bias, and ADHD symp- Number of different Childhood Trauma 0.3 (0.6), range 0–3 tom variables, resulting in a significance level of p < .0083. 0 536 Based on the correlations (see Table 2) and to assess the 1 104 overall relevance of life events for ADHD symptom levels, 2 28 the mediation of the association of Lifetime Life Events 3 7 with total ADHD symptoms by memory bias was tested. Experienced recent stressful events 30% Associations as well as mediation were also tested sepa- Number of different recent stressful 0.4 (0.1), range 0-4 events rately for the IA and HI symptom clusters to get an indica- 0 474 tion of potential specificities of findings. Subsequent model 1 137 were constructed testing the mediation of the association 2 53 of Childhood Trauma with total ADHD symptoms as well 3 9 as the symptom clusters by memory bias. Mediation was 4 2 tested using the PROCESS macro for SPSS (Hayes 2013). A Total ADHD symptoms 2.9 (2.9), range 0–16 bootstrapping method was used to assess the indirect effect ADHD inattention symptoms 1.2 (1.7), range 0–9 based on 1000 bootstrapped samples using bias-corrected ADHD hyperactivity/impulsivity symp- 1.7 (1.7), range 0–8 and accelerated 95% confidence intervals (BCa CI). toms Results negative memory bias, but not positive memory bias, was Sample descriptives and correlational analyses significantly positively correlated with total ADHD symp- tom level (p < .001); both inattention and hyperactivity/ Complete data were available for 675 participants. Descrip- impulsivity symptoms contributed to this correlation. A tives of the sample are shown in Table 1. Age was signifi- significant positive correlation was also seen between the cantly correlated to the number of Lifetime Life Events, Lifetime Life Events and Recent Stress variables, and total r(673) = .22, p < .001, but not to the memory bias variables, ADHD score (both p < .001). Both inattention and hyper- all p > .07. When comparing men and women on the number activity/impulsivity symptoms contributed to the correla- of Lifetime Life Events and memory bias, we saw gender tion with Lifetime Life Events. Hyperactivity/impulsivity differences on both the positive and negative memory bias symptoms but not inattention symptoms were significantly scores, t(673) = 3.33, p = .001 and t(673) = 2.02, p = .044, correlated with Recent Stress (p = .001 and p = .210, respec- respectively. tively). Furthermore, positive correlations were found for The correlational structure of the independent and negative memory bias and Lifetime Life Events as well dependent variable data is given in Table 2. As apparent, as Childhood Trauma (p < .005 for both), but not Recent Stress. None of the three life events variables correlated with positive memory bias. Therefore, mediation analyses were The PANAS negative affect scale was significantly positively cor - only run for negative memory bias. Also, because of the related with the number of Lifetime Life Events (r = .08), Recent absence of correlations between Recent Stress and nega- Events (r = .13), total ADHD symptom levels (r = .34) as well as tive memory bias variable, mediation of the association of both IA (r = .27) and HI subscales (r = .31), and the negative mem- ory bias score (r = .22), all p < .05. Negative affect was not signifi- cantly correlated with Childhood Trauma variable, r(673) = .07, p = .079. 1 3 ADHD symptoms in healthy adults are associated with stressful life events and negative memory… 155 Table 2 Bivariate correlations (including p-values) between stressful life events variables, positive and negative memory bias, and ADHD symptom level (including subscales) variables (N = 675) 1. 1.1 1.2 2. 3. 4. 4.1 4.2 1. LT Life Events – 1.1 CH Trauma .46, p < .001* – 1.2 RC Stress .47, p < .001* .13, p < .001* – 2. Positive bias − .03, − .06, .02, p = .322 – p = .254 p = .055 3. Negative bias .16, p < .001* .12, p = .001* .05, p = .095 − .09, – p = .007* 4. Total ADHD .15, p < .001* .11, p = .002* .11, p = .002* − .04, .23, p < .001* – symp. p = .165 4.1 IA symp. .12, p = .001* .08, p = .019 .03, p = .210 − .06, .19, p < .001* .80, p < .001* – p = .073 4.2 HI symp. .14, p = .001* .11, p = .003* .02, p = .001* .01, p = .367 .20, p < .001* .87, p < .001* .44, p < .001* – IA inattention symptom score, HI hyperactivity/impulsivity symptom score, LT Life time, CH childhood, RC recent * Significant at the p < .0083 level (p-level threshold after correction for multiple testing) Negative memory bias ab Life Events ADHDsymptoms c’ * Fig. 1 Model for the relationship between life events and ADHD of the association between Life Events and Negative memory bias, symptoms as mediated by negative memory bias. *The regression b = coefficient of the association between Negative memory bias and coefficient between life events and ADHD symptoms when nega- ADHD symptoms, c′ = estimate of the direct effect of Life Events on tive memory bias is included in the model. a = regression coefficient ADHD symptoms Recent Stress and ADHD symptoms by negative memory indirect effect ‘ab’ was small (0.005), but statistically signifi- bias was not examined. cant as the bootstrapped unstandardized indirect effect 95% confidence interval (BCa CI) ranged from .002 to .01 (i.e. Life events and ADHD symptoms—mediation not including zero). The ratio of the indirect effect (referred by negative memory bias to as P ) to the direct effect was .18. The P value provides M M an effect size measure, and in this case it indicates that 18% In the following, negative memory bias was examined as a of the effect of life events on ADHD symptoms operates mediator of the association between life events and ADHD indirectly through negative memory bias. symptoms according to the model presented in Fig. 1. Indeed, the relationship between Lifetime Life Events Model specification for IA and HI ADHD symptom and total ADHD symptoms was found to be mediated by clusters negative memory bias: the regression coefficients between Lifetime Life Events and negative memory bias as well as In separating the ADHD total scores into scores for the two between negative memory bias and total ADHD symptoms different symptom domains, we found both the associations were statistically significant. In addition, the association of Lifetime Life Events with inattention (IA) symptom score between Lifetime Life Events and total ADHD symptoms as well as with hyperactivity/impulsivity (HI) symptom stayed significant, when allowing for mediation by negative score to be mediated by negative memory bias (see Table 3: memory bias, with p < .001 in the total effect model. The Model 2 and 3, respectively with BCa CI [.001, − .01] for 1 3 156 J. N. Vrijsen et al. Table 3 Regression coefficients, standard errors, and model sum- negative memory bias, and ADHD symptom level (including the IA mary information for the mediational models testing the association and HI subscales) controlled for the effect of age, gender, and nega- between the life events variables (lifetime and Childhood Trauma), tive affect (N = 675). The tested model is depicted in Fig. 1 Antecedent Consequent M (Negative memory bias) Y (ADHD symptoms) Coeff. SE p Coeff. SE p Model 1: Lifetime Life Events and total ADHD symptoms X (Lifetime Life Events) a 0.004 0.001 < .001 c′ 0.027 0.008 .002 M (Neg. memory bias) – – – b 1.254 0.313 < .001 Constant i − 0.024 0.023 0.289 i 0.586 0.128 .001 M Y 2 2 R = 0.076 R = 0.145 F(4670) = 13.699, p < .001 F(5669) = 22.673, p < .001 Model 2: Lifetime Life Events and IA ADHD symptoms X (Lifetime Life Events) a 0.004 0.001 < .001 c′ 0.018 0.007 .016 M (Neg. memory bias) – – – b .986 0.273 < .001 Constant i − 0.024 0.023 .289 i 0.276 0.159 .084 M Y 2 2 R = 0.076 R = 0.119 F(4670) = 13.699, p < .001 F(5699) = 18.023, p < .001 Model 3: Lifetime Life Events and HI ADHD symptoms X (Lifetime Life Events) a 0.004 0.001 < .001 c′ 0.019 0.007 .008 M (Neg. memory bias) – – – b .906 0.267 < .001 Constant i − 0.024 0.023 .289 i 0.301 0.156 .054 M Y 2 2 R = 0.076 R = 0.118 F(4670) = 13.699, p < .001 F(5699) = 18.898, p < .001 Model 4: Childhood Trauma and total ADHD symptoms X (Childhood Trauma) a 0.030 0.010 .002 c′ 0.148 0.078 .059 M (Neg. memory bias) – – – b 1.323 0.313 < .001 Constant i − 0.028 0.023 .221 i 0.567 0.184 .002 M Y 2 2 R = 0.070 R = 0.317 F(4670) = 12.607, p < .001 F(5669) = 21.157, p < .001 Model 5: Childhood Trauma and IA ADHD symptoms X (Childhood Trauma) a 0.030 0.010 .002 c′ 0.063 0.068 .353 M (Neg. memory bias) – – – b 1.049 0.273 < .001 Constant i − 0.028 0.023 .221 i 0.267 0.160 .096 M Y 2 2 R = 0.070 R = 0.112 F(4670) = 12.607, p < .001 F(5669) = 16.914, p < .001 Model 6: Childhood Trauma and HI ADHD symptoms X (Childhood Trauma) a 0.030 0.010 .002 c′ 0.135 0.067 .043 M (Neg. memory bias) – – – b 0.942 0.267 < .001 Constant i − 0.028 0.023 .221 i 0.284 0.156 0.070 M Y 2 2 R = 0.070 R = 0.114 F(4670) = 12.607, p < .001 F(5669) = 17.239, p < .001 All of the indirect effects of X on Y were statistically significant, which means evidence for mediation was found in all models. This was revealed by the bootstrapped Confidence Intervals not including zero in all models IA inattention symptom score, HI hyperactivity/impulsivity symptom score, X predictor life events variable, Y outcome variable, ADHD symp- tom variable, M mediator, negative memory bias; a = regression coefficient of the association between life events and negative memory bias, b coefficient of the association between negative memory bias and ADHD symptoms, c’ estimate of the direct effect of life events on ADHD symptoms, i constant coefficient both models). The strength of effect appeared similar for the Childhood Trauma and ADHD symptoms two symptom clusters (see coefficients in Table 3). P was . 21 for Model 2 and .18 for Model 3. When examining the associations with Childhood Trauma only (Model 4), we found that the direct effect of Childhood 1 3 ADHD symptoms in healthy adults are associated with stressful life events and negative memory… 157 Trauma on negative memory bias was significant, but the the development of ADHD symptoms (e.g. Stevens et al. direct effect of Childhood Trauma on total ADHD symptoms 2008). In addition and conversely, children showing high fell short of reaching statistical significance, with p = .059. ADHD symptom levels may be particularly prone to expe- Nevertheless, negative memory bias still seemed to mediate riencing Childhood Trauma such as aggression or abuse, some aspects of Childhood Trauma on total ADHD symp- because they tend to act out more and/or because of their toms with BCa CI [.01, − .08], as the indirect effect in the family members might show problems with impulse control total effect model was significant, and 27% of the effect of themselves (ADHD has a strong genetic basis; Franke et al. Childhood Trauma on ADHD symptoms seems to operate 2012; Harold et al. 2013). Although we had also hypoth- indirectly through negative memory bias. esized a possible association between Recent Stress and For the separate ADHD symptom domains, we observed ADHD symptoms, given earlier reports of more Recent an interesting pattern. As apparent from Table 3, Models 5 Stress on ADHD severity in a clinical sample (van der Meer and 6 (BCa CIs [.01, − .08] and [.01, − .06], respectively), et al. 2014), the observed nonsignificant correlations did not effects of Childhood Trauma on inattention severity oper - provide evidence for such a link in our population sample. A ate largely through memory bias (P = .49): early trauma potential explanation might be that the correction for nega- strongly affected negative memory bias, but no significant tive affect scores removed much of the variance related to direct effects on inattention symptoms were observed. For Recent Stress. We also employed a rather conservative sig- hyperactivity/impulsivity the pattern was different, with sig- nificance level for the correlations, and it may be important nificant direct and indirect effects, and mediating memory to note that the direct association between Recent Stress bias explaining only approximately 21% of the effect. and ADHD symptoms (total and hyperactivity/impulsiv- ity scores) did reach an uncorrected level of significance of p < .05. Although our results cannot address the direc- Discussion tion of the association, they may indicate that while recent stressors can still exacerbate current symptom severity The current study substantiates earlier suggestive findings somewhat, Childhood Trauma—possibly through memory showing associations between life events and population processes—may be a risk factor for both the development ADHD symptoms, and shows that such associations might and persistence of ADHD symptoms into adulthood. The be particularly driven by childhood traumatic events. The latter is important, since we still lack a good understanding associations were robust, when controlling for negative of the factors that influence persistence of ADHD symptoms ‘depressotypic’ affect, and were observed for both inatten- into adulthood. Persistence has for example been associated tion and hyperactivity/impulsivity. We provide new evidence with childhood ADHD severity and whether or not patients for the mediation by negative memory bias of associations received treatment during childhood (Kessler et al. 2005), between life events and population ADHD symptom scores. as well as neural and cognitive processes (e.g. Alderson This first exploration of the relevance of memory bias as et al. 2013; Onnink et al. 2014). Understanding (and in the marker for ADHD symptoms will hopefully stimulate fur- future possibly predicting) outcome of ADHD in adulthood ther (sub)clinical research into cognitive biases in ADHD. is crucial to inform treatment options, clinical planning, and Associations between Lifetime Life Events and ADHD lifestyle choices of individuals at risk. symptoms seemed to be driven mainly by childhood trau- Though mediation of negative memory bias was seen matic events. Indeed, Childhood Trauma has a strong effect for the association between Childhood Trauma and total on the developing child, at the neural, cognitive, and emo- ADHD symptoms, it was surprisingly more specific for tional level (e.g. Biederman et al. 1995; Stein et al. 1997; hyperactivity/impulsivity levels in this population cohort. Sugaya et al. 2012) as well as being a known risk factor for Childhood Trauma affects brain areas vital for cognitive control and memory functioning (Bremner 2002; Bremner and Narayan 1998; Bremner et al. 2003; McGaugh 2004; Stein et al. 1997). This may result in less effective suppres- sion of unwanted negative material in memory, resulting in The pattern of results is similar when using the dichotomized memory bias (Nolen-Hoeksema et al. 2008). Besides nega- Childhood Trauma variable (yes vs. no trauma) in the mediational tive memories, the neural effects might—directly or through models. All of the indirect effects of Childhood Trauma on ADHD memory bias—be associated with less effective suppression symptoms (total, IA, and HI subscales) were statistically significant, of hyperactive and impulsive tendencies. To find out whether which means evidence for mediation was found in all models. This was revealed by the bootstrapped confidence intervals not including this interpretation is relevant for the clinical extreme of zero in all models. The only difference with the models including the hyperactivity/impulsivity, substantiation by studies in clini- continuous Childhood Trauma variable was that the direct effect on cal samples is required. HI symptoms did not reach statistical significance using the dichoto- mous Childhood Trauma variable (p = .068). 1 3 158 J. N. Vrijsen et al. Important to note is that, because of the cross-sectional that negative bias might have affected participants’ recall. design of our study, we could not study causation and cur- Because factual events during pre-determined periods of life rently cannot know the direction of the effects observed. Our were assessed, and because adults’ recall of childhood events model was built under the assumption that memory bias can has been shown to be fairly accurate (Brewin et al. 1993), be a marker for ADHD, but studies in e.g. major depression it seems unlikely that this would have significant effects on show that levels of memory bias can also be influenced by our study beyond potentially adding some random noise, Recent Stress (Gotlib and Joormann 2010). Memory bias limiting power. is thus partly state-dependent, but through controlling for In conclusion, biased processing of emotional informa- negative affect we seem to tap into trait-like characteristics tion may be a marker for adult ADHD symptom severity in of negative memory bias. As we also did not find effects of addition to being a risk factor for depression onset, mainte- recent life events, the model constructed and shown in Fig. 1 nance, and recurrence (De Raedt and Koster 2010). Memory is the most likely model explaining the relationship between bias association with ADHD symptoms was independent of life events, memory bias, and ADHD symptoms. The devel- current depressive mood state, and more negative memory opment and persistence of ADHD symptoms in the popula- bias was linked to increased levels of both inattention and tion may have an ‘emotional’ component, akin to anxiety and hyperactivity/impulsivity. The current results make the study depression. In fact, ADHD shares some symptoms with these of cognitive biases in neurodevelopmental disorders an inter- disorders, e.g. avoidance and cognitive impairment. Perhaps esting new research direction. negative cognitive schemata (Beck 2008), based on childhood Acknowledgements This work made use of the BIG (Brain Imaging experiences, thus play a role in the development of neurode- Genetics) database, first established in Nijmegen, The Netherlands, in velopmental disorders as well (see also recent evidence by 2007. This resource is now part of Cognomics (www.cognomics.nl), a Vrijsen et al. 2017). We present a first evaluation of the rel- joint initiative by researchers of the Donders Centre for Cognitive Neu- roimaging, the Human Genetics and Cognitive Neuroscience depart- evance of life stress and memory bias for ADHD symptoms ments of the Radboud university medical center and the Max Planck in a healthy sample. Important to note is that the effect sizes Institute for Psycholinguistics in Nijmegen. The Cognomics Initiative of the correlations are small, which is not surprising given the is supported by the participating departments and centers and by exter- limited variance in ADHD symptoms in this healthy sample. nal grants, i.e. the Biobanking and Biomolecular Resources Research Infrastructure (Netherlands) (BBMRI-NL), the Hersenstichting Neder- Substantiation in a subclinical an eventually clinical sample land, and the Netherlands Organisation for Scientific Research (NWO). are future steps. If further substantiated, addressing schemata The research leading to these results also receives funding from the in cognitive behavioral therapy or computerized cognitive European Community’s Seventh Framework Programme (FP7/2007– trainings (i.e. Cognitive Bias Modification; Koster and Bern- 2013) under Grant Agreement No 602805 (Aggressotype), and from the Horizon2020 Marie-Curie ETN Grant No 643051 (MiND). B. stein 2015) for ADHD might be beneficial. Franke is supported by a Vici grant from NWO (Grant Number 016- Our study should be viewed in the context of some 130-669). We wish to thank all persons who kindly participated in the strengths and limitations. Strength of the study is the large BIG research. We would like to thank Maartje de Vries for her support sample of participants. This is especially important, when with data processing. examining novel markers for a given disorder. We used a Compliance with ethical standards population sample not containing diagnosed patients with ADHD for our study, which afforded us the possibility Conflict of interest Author BF received a speaker fee from Merck. to study the effect of symptoms severity along the entire The other authors do not have conflicts of interest to report. continuum observed in the population. We view this as a strength, but it also requires confirmation of findings in a Ethical standards All participants gave written informed consent and were financially compensated for participation. The study was clinical sample, before any we can start making inferences approved by the medical center’s regional medical ethics committee. for disease and treatment. We studied two classes of stressful The authors assert that all procedures contributing to this work comply life events (i.e. Recent Stress and Childhood Trauma), add- with the ethical standards of the relevant national and institutional com- ing to the understanding of the role of Childhood Trauma mittees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. in ADHD. However, the variance of recent stressful events and Childhood Trauma is limited in a healthy sample. Web- based data collection was used, which limits the control- Open Access This article is distributed under the terms of the lability of data collection. However, testing participants in Creative Commons Attribution 4.0 International License (http://crea- tivecommons.org/licenses/by/4.0/), which permits unrestricted use, their natural setting eliminates experimenter effects (Har - distribution, and reproduction in any medium, provided you give appro- ris and Rosenthal 1985), in turn increasing the ecological priate credit to the original author(s) and the source, provide a link to validity of the results. The measure of life events provides the Creative Commons license, and indicate if changes were made. a limitation, as trauma severity and subjective stress associ- ated to the traumas were not measured. Another possible limitation of the measurement method for the life events is 1 3 ADHD symptoms in healthy adults are associated with stressful life events and negative memory… 159 deficit/hyperactivity disorder in adults, a review. Mol Psychiatry References 17(10):960–987. doi:10.1038/mp.2011.138 Friedrichs B, Igl W, Larsson H, Larsson JO (2012) Coexisting psychi- Able SL, Johnston JA, Adler LA, Swindle RW (2007) Functional and atric problems and stressful life events in adults with symptoms of psychosocial impairment in adults with undiagnosed ADHD. Psy- ADHD–a large Swedish population-based study of twins. J Atten chol Med 37(1):97–107. doi:10.1017/S0033291706008713 Disord 16(1):13–22. doi:10.1177/1087054710376909 Alderson RM, Kasper LJ, Hudec KL, Patros CH (2013) Attention- George D, Mallery P (2010) SPSS for Windows step by step. A simple deficit/hyperactivity disorder (ADHD) and working memory in study guide and reference. Pearson, Boston adults: a meta-analytic review. Neuropsychology 27(3):287–302. Gerritsen L, Rijpkema M, van Oostrom I, Buitelaar J, Franke B, doi:10.1037/a0032371 Fernández G, Tendolkar I (2012) Amygdala to hippocam- Association AP (2013) Diagnostic and statistical manual of mental pal volume ratio is associated with negative memory bias in disorders, 5th edn. Author, Washington, DC healthy subjects. Psychol Med 42(2):335–343. doi:10.1017/ Beck AT (2008) The evolution of the cognitive model of depression S003329171100122X and its neurobiological correlates. Am J Psychiatry 165(8):969– Glod CA, Teicher MH (1996) Relationship between early abuse, post- 977. doi:10.1176/appi.ajp.2008.08050721 traumatic stress disorder, and activity levels in prepubertal chil- Beck AT, Haigh EA (2014) Advances in cognitive theory and therapy: dren. J Am Acad Child Adolesc Psychiatry 35(10):1384–1393. the generic cognitive model. Annu Rev Clin Psychol 10:1–24. doi:10.1097/00004583-199610000-00026 doi:10.1146/annurev-clinpsy-032813-153734 Gotlib IH, Joormann J (2010) Cognition and depression: current sta- Biederman J, Faraone SV (2006) The effects of attention-deficit/hyper - tus and future directions. Annu Rev Clin Psychol 6(1):285–312. activity disorder on employment and household income. MedGen- doi:10.1146/annurev.clinpsy.121208.131305 Med 8(3):12 Hammen C, Zupan BA (1984) Self-schemas, depression, and the pro- Biederman J, Milberger S, Faraone SV, Kiely K, Guite J, Mick E, cessing of personal information in children. J Exp Child Psychol Davis SG (1995) Impact of adversity on functioning and comor- 37:598–608 bidity in children with attention-deficit hyperactivity disor- Harris MJ, Rosenthal R (1985) Mediation of interpersonal expectancy der. J Am Acad Child Adolesc Psychiatry 34(11):1495–1503. effects: 31 meta-analyses. Psychol Bull, 97, 363–386 doi:10.1097/00004583-199511000-00017 Harold GT, Leve LD, Barrett D, Elam K, Neiderhiser JM, Natsuaki Bower GH (1981) Mood and memory. Am Psychol 36(2):129–148 MN, Thapar A (2013) Biological and rearing mother influences Bremner JD (2002) Neuroimaging of childhood trauma. Semin Clin on child ADHD symptoms: revisiting the developmental inter- Neuropsychiatry 7(2):104–112 face between nature and nurture. J Child Psychol Psychiatry Bremner JD, Narayan M (1998) The effects of stress on memory and 54(10):1038–1046. doi:10.1111/jcpp.12100 the hippocampus throughout the life cycle: implications for child- Hayes AF (2013) Introduction to mediation, moderation, and condi- hood development and aging. Dev Psychopathol 10(4):871–885 tional process analysis: a regression-based approach. Guilford Bremner JD, Vythilingam M, Vermetten E, Southwick SM, McGlashan Press, New York T, Staib LH, Charney DS (2003) Neural correlates of declarative Herrmann MJ, Biehl SC, Jacob C, Deckert J (2010) Neurobiological memory for emotionally valenced words in women with post- and psychophysiological correlates of emotional dysregulation traumatic stress disorder related to early childhood sexual abuse. in ADHD patients. Atten Defic Hyperact Disord 2(4):233–239. Biol Psychiatry 53(10):879–889 doi:10.1007/s12402-010-0047-6 Brewin CR, Andrews B, Gotlib IH (1993) Psychopathology and early Hertel PT, Maydon A, Cottle J, Vrijsen JN (2017) Cognitive Bias experience: a reappraisal of retrospective reports. Psychol Bull modification: retrieval practice to simulate and oppose rumina- 113(1):82–98 tive memory biases. Clin Psychol Sci 5(1):122–130 Brugha TS, Cragg D (1990) The list of threatening experiences: the Hoogman M, Rijpkema M, Janss L, Brunner H, Fernandez G, Buite- reliability and validity of a brief life events questionnaire. Acta laar J, Arias-Vásquez A (2012) Current self-reported symptoms Psychiatr Scand 82(1):77–81 of attention deficit/hyperactivity disorder are associated with Chen W, Zhou K, Sham P, Franke B, Kuntsi J, Campbell D, Asherson total brain volume in healthy adults. PLoS ONE 7(2):e31273. P (2008) DSM-IV combined type ADHD shows familial associa- doi:10.1371/journal.pone.0031273 tion with sibling trait scores: a sampling strategy for QTL linkage. Hovens JG, Wiersma JE, Giltay EJ, van Oppen P, Spinhoven P, Pen- Am J Med Genet B Neuropsychiatr Genet 147B(8):1450–1460. ninx BW, Zitman FG (2010) Childhood life events and childhood doi:10.1002/ajmg.b.30672 trauma in adult patients with depressive, anxiety and comor- Crawford JR, Henry JD (2004) The positive and negative affect sched- bid disorders vs. controls. Acta Psychiatr Scand 122(1):66–74. ule (PANAS): construct validity, measurement properties and doi:10.1111/j.1600-0447.2009.01491.x normative data in a large non-clinical sample. Br J Clin Psychol Insel TR (2014) The NIMH Research Domain Criteria (RDoC) project: 43(Pt 3):245–265. doi:10.1348/0144665031752934 precision medicine for psychiatry. Am J Psychiatry 171(4):395– d’Acremont M, Van der Linden M (2007) Memory for angry faces, 397. doi:10.1176/appi.ajp.2014.14020138 impulsivity, and problematic behavior in adolescence. J Abnorm Insel T, Cuthbert B, Garvey M, Heinssen R, Pine DS, Quinn K, Wang P Child Psychol 35(2):313–324. doi:10.1007/s10802-006-9092-1 (2010) Research domain criteria (RDoC): toward a new classifica- De Raedt R, Koster EHW (2010) Understanding vulnerability for tion framework for research on mental disorders. Am J Psychiatry depression from a cognitive neuroscience perspective: a reap- 167(7):748–751. doi:10.1176/appi.ajp.2010.09091379 praisal of attentional factors and a new conceptual framework. Kessler RC, Davis CG, Kendler KS (1997) Childhood adversity and Cognit Affect Behav Neurosci 10(1):50–70. doi:10.3758/ adult psychiatric disorder in the US National Comorbidity Survey. cabn.10.1.50 Psychol Med 27(5):1101–1119 Faraone SV, Asherson P, Banaschewski T, Biederman J, Buitelaar JK, Kessler RC, Adler LA, Barkley R, Biederman J, Conners CK, Faraone Ramos-Quiroga JA, Franke B (2015) Attention-deficit/hyper - SV, Zaslavsky AM (2005) Patterns and predictors of attention- activity disorder. Nat Rev Dis Primers 1:15020. doi:10.1038/ deficit/hyperactivity disorder persistence into adulthood: results nrdp.2015.20 from the national comorbidity survey replication. Biol Psychiatry Franke B, Faraone SV, Asherson P, Buitelaar J, Bau CH, Ramos-Qui- 57(11):1442–1451. doi:10.1016/j.biopsych.2005.04.001 roga JA, CollaboraTion IMPA (2012) The genetics of attention 1 3 160 J. N. Vrijsen et al. Kooij JJ, Buitelaar JK, van den Oord EJ, Furer JW, Rijnders CA, Hodia- early severe institutional deprivation: presentation and associa- mont PP (2005) Internal and external validity of attention-deficit tions in early adolescence. J Abnorm Child Psychol 36(3):385– hyperactivity disorder in a population-based sample of adults. 398. doi:10.1007/s10802-007-9185-5 Psychol Med 35(6):817–827 Sugaya L, Hasin DS, Olfson M, Lin KH, Grant BF, Blanco C (2012) Koster EH, Bernstein A (2015) Introduction to the special issue on Child physical abuse and adult mental health: a national study. J cognitive bias modification: taking a step back to move for- Trauma Stress 25(4):384–392. doi:10.1002/jts.21719 ward? J Behav Ther Exp Psychiatry 49(Pt A):1–4. doi:10.1016/j. Tarlow EM, Haaga DAF (1996) Negative self-concept: specificity to jbtep.2015.05.006 depressive symptoms and relation to positive and negative affec- Krauel K, Duzel E, Hinrichs H, Rellum T, Santel S, Baving L (2009) tivity. J Res Pers 30:120–127 Emotional memory in ADHD patients with and without comor- Uekermann J, Kraemer M, Abdel-Hamid M, Schimmelmann BG, bid ODD/CD. J Neural Transm (Vienna) 116(1):117–120. Hebebrand J, Daum I, Kis B (2010) Social cognition in attention- doi:10.1007/s00702-008-0154-0 deficit hyperactivity disorder (ADHD). Neurosci Biobehav Rev Lubke GH, Hudziak JJ, Derks EM, van Bijsterveldt TC, Boomsma DI 34(5):734–743. doi:10.1016/j.neubiorev.2009.10.009 (2009) Maternal ratings of attention problems in ADHD: evidence van der Meer D, Hartman CA, Richards J, Bralten JB, Franke B, for the existence of a continuum. J Am Acad Child Adolesc Psy- Oosterlaan J, Hoekstra PJ (2014) The serotonin transporter gene chiatry 48(11):1085–1093. doi:10.1097/CHI.0b013e3181ba3dbb polymorphism 5-HTTLPR moderates the effects of stress on Mathews A, MacLeod C (2002) Induced processing biases have attention-deficit/hyperactivity disorder. J Child Psychol Psychia- causal effects on anxiety. Cognit Emot 16(3):331–354. try 55(12):1363–1371. doi:10.1111/jcpp.12240 doi:10.1080/02699930143000518 van Oostrom I, Franke B, Rijpkema M, Gerritsen L, Arias-Vásquez A, Mathews A, MacLeod C (2005) Cognitive vulnerability to emotional Fernández G, Tendolkar I (2012) Interaction between BDNF Val- disorders. Annu Rev Clin Psychol 1(1):167–195. doi:10.1146/ 66Met and childhood stressful life events is associated to affective annurev.clinpsy.1.102803.143916 memory bias in men but not women. Biol Psychol 89(1):214–219. McGaugh JL (2004) The amygdala modulates the consolidation of doi:10.1016/j.biopsycho.2011.10.012 memories of emotionally arousing experiences. Annu Rev Neuro- Vogel S, Gerritsen L, van Oostrom I, Arias-Vásquez A, Rijpkema M, sci 27:1–28. doi:10.1146/annurev.neuro.27.070203.144157 Joëls M, Fernández G (2014) Linking genetic variants of the min- McLeer SV, Callaghan M, Henry D, Wallen J (1994) Psychiatric disor- eralocorticoid receptor and negative memory bias: interaction ders in sexually abused children. J Am Acad Child Adolesc Psy- with prior life adversity. Psychoneuroendocrinology 40:181–190. chiatry 33(3):313–319. doi:10.1097/00004583-199403000-00003 doi:10.1016/j.psyneuen.2013.11.010 Nolen-Hoeksema S, Wisco BE, Lyubomirsky S (2008) Rethink- Vrijsen JN, Becker ES, Arias-Vásquez A, van Dijk MK, Speckens A, ing rumination. Perspect Psychol Sci 3(5):400–424. Oostrom I (2014a) What is the contribution of different cognitive doi:10.1111/j.1745-6924.2008.00088.x biases and stressful childhood events to the presence and number Onnink AM, Zwiers MP, Hoogman M, Mostert JC, Kan CC, Buitelaar of previous depressive episodes? Psychiatry Res 217(3):134–142. J, Franke B (2014) Brain alterations in adult ADHD: effects of doi:10.1016/j.psychres.2014.02.033 gender, treatment and comorbid depression. Eur Neuropsychop- Vrijsen JN, van Oostrom I, Arias-Vásquez A, Franke B, Becker ES, harmacol 24(3):397–409. doi:10.1016/j.euroneuro.2013.11.011 Speckens A (2014b) Association between genes, stressful child- Perry BD (2008) Child maltreatment: a neurodevelopmental perspec- hood events and processing bias in depression vulnerable individ- tive on the role of trauma and neglect in psychopathology. In: uals. Genes Brain Behav 13(5):508–516. doi:10.1111/gbb.12129 Beauchaine T, Henshaw SP (eds) Child and adolescent psycho- Vrijsen JN, Tendolkar I, Arias-Vásquez A, Franke B, Schene AH, pathology. Wiley, Hoboken, pp 93–129 Fernández G, van Oostrom I (2015a) Interaction of the 5-HTTLPR Polanczyk G, Rohde LA (2007) Epidemiology of attention-deficit/ and childhood trauma influences memory bias in healthy individu- hyperactivity disorder across the lifespan. Curr Opin Psychiatry als. J Affect Disord 186:83–89. doi:10.1016/j.jad.2015.06.008 20(4):386–392. doi:10.1097/YCO.0b013e3281568d7a Vrijsen JN, Vogel S, Arias-Vásquez A, Franke B, Fernández G, Becker Simon V, Czobor P, Bálint S, Mészáros A, Bitter I (2009) Prevalence ES, van Oostrom I (2015b) Depressed patients in remission show and correlates of adult attention-deficit hyperactivity disorder: an interaction between variance in the mineralocorticoid receptor meta-analysis. Br J Psychiatry 194(3):204–211. doi:10.1192/bjp. NR3C2 gene and childhood trauma on negative memory bias. Psy- bp.107.048827 chiatr Genet 25(3):99–105. doi:10.1097/YPG.0000000000000081 Sobanski E, Brüggemann D, Alm B, Kern S, Deschner M, Schubert Vrijsen JN, Hertel PT, Becker ES (2016) Practicing emotionally biased T, Rietschel M (2007) Psychiatric comorbidity and functional retrieval affects mood and establishes biased recall a week later. impairment in a clinically referred sample of adults with attention- Cognit Therapy Res 40(6):764–773 deficit/hyperactivity disorder (ADHD). Eur Arch Psychiatry Clin Vrijsen JN, van Amen CT, Koekkoek B, van Oostrom I, Schene AH, Neurosci 257(7):371–377. doi:10.1007/s00406-007-0712-8 Tendolkar I (2017) Childhood trauma and negative memory bias Stein MB, Koverola C, Hanna C, Torchia MG, McClarty B (1997) as shared risk factors for psychopathology and comorbidity in a Hippocampal volume in women victimized by childhood sexual naturalistic psychiatric patient sample. Brain Behav 7(6):e00693. abuse. Psychol Med 27(4):951–959 doi:10.1002/brb3.693 Steinberg JA, Gibb BE, Alloy LB, Abramson LY (2003) Childhood Watson D, Clark LA, Tellegen A (1988) Development and validation of emotional maltreatment, cognitive vulnerability to depression, brief measures of positive and negative affect: the PANAS scales. and self-referent information processing in adulthood: reciprocal J Pers Soc Psychol 54(6):1063–1070 relations. J Cognit Psychotherapy 17:347–358 Stevens SE, Sonuga-Barke EJ, Kreppner JM, Beckett C, Castle J, Colvert E, Rutter M (2008) Inattention/overactivity following 1 3
ADHD Attention Deficit and Hyperactivity Disorders – Springer Journals
Published: Oct 28, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud