Electroconvulsive therapy regulates emotional memory bias of depressed patients

Author’s Accepted Manuscript Electroconvulsive therapy regulates emotional memory bias of depressed patients Tongjian Bai, Wen Xie, Qiang Wei, Yang Chen, Jingjing Mu, Yanghua Tian, Kai Wang www.elsevier.com/locate/psychres

PII: DOI: Reference:

S0165-1781(17)30210-X http://dx.doi.org/10.1016/j.psychres.2017.07.069 PSY10703

To appear in: Psychiatry Research Received date: 5 February 2017 Revised date: 20 June 2017 Accepted date: 29 July 2017 Cite this article as: Tongjian Bai, Wen Xie, Qiang Wei, Yang Chen, Jingjing Mu, Yanghua Tian and Kai Wang, Electroconvulsive therapy regulates emotional memory bias of depressed patients, Psychiatry Research, http://dx.doi.org/10.1016/j.psychres.2017.07.069 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Electroconvulsive therapy regulates emotional memory bias of depressed patients Tongjian Bai a, b1, Wen Xie c1, Qiang Wei a, b, Yang Chen c, Jingjing Mu c, Yanghua Tian a, b, d*

a

, Kai Wang a, b, e*

Department of Neurology, the First Affiliated Hospital of Anhui Medical University, Hefei,

Anhui Province, China; b

Collaborative Innovation Center of Neuropsychiatric Disorders and Mental Health, Anhui

Province, China; c

Anhui Mental Health Center, Hefei, Anhui Province, China;

d

National Clinical Research Center for Mental Disorders, China;

e

Department of Medical Psychology, the First Affiliated Hospital of Anhui Medical

University, Hefei, Anhui Province, China; * Corresponding author at: Department of Neurology, First Hospital of Anhui Medical University, 218 Jixi Road, Hefei, Anhui Province, PR China. Tel.: +86 551 62923704; fax: +86 551 62923704. Email address: [email protected] (Y. Tian), or [email protected] (K. Wang). Abstract: Emotional memory bias is considered to be an important base of the etiology of depression and can be reversed by antidepressants via enhancing the memory for positive stimuli. Another antidepressant treatment, electroconvulsive therapy (ECT), has rapid antidepressant effect and frequently causes short-term memory impairment. However, it is unclear about the short-term effect of ECT on memory bias. In this study, the incidental memory task with

1

T. Bai and W. Xie equally contributed to this work;

emotional pictures were applied to evaluate the emotional memory of twenty depressed patients at pre- and post-ECT (three days after ECT) compared to twenty healthy controls. The depressive symptoms were evaluated using the Hamilton rating scale of depression (HRSD). Before ECT, patients showed decreased recognition memory for positive pictures compared to controls and remembered negative pictures more easily than positive pictures in the recognition task. In patients, the main effect of session (pre-ECT and post-ECT) was significant for both recognition and recall memory with reduced memory performance. The interaction between valence (positive, neutral and negative) and session was significant for recognition memory, indicating that negative memory was impaired more severely than positive memory. Our study indicates that ECT relieves depressive symptoms and regulates emotional memory through more severe impairment on memory for negative stimuli.

Keywords: Depression; Memory biases; Electroconvulsive therapy 1. Introduction Depression is a common, chronic psychiatric disorder, which is a serious burden to patients, their family and society. However, the mechanism of depression is still unclear, although a lot of hypotheses have been proposed for many years, such as the monoamine hypothesis of depression (Asberg et al., 1976) and cognitive theories of depression (Beck et al., 1979). Cognitive theories of depression hold that cognitive biases in information processing play a crucial role in the etiology and maintenance of depressive disorders. The cognitive biases consider that depressed patients partially remember, perceive and attend to affectively negative materials or selectively attenuate the information processing for positive

materials. In addition, cognitive biases in depressed patients have been associated with increased risk of relapse (Bouhuys et al., 1999). Many antidepressive treatments are thought to exert their effects by reversing these biases (Roiser et al., 2012). The memory bias is an important element of the cognitive model of depression (Beck, 2008) and associated with the occurrence and maintenance of depression (Hasler et al., 2004). To be more specific, the emotional memory bias manifests in two types: bias toward negative materials and bias away from positive materials (Roiser et al., 2012). Many studies suggest that depressed patients preferentially remember negative materials (Direnfeld and Roberts, 2006; Gilboa-Schechtman et al., 2002; Hamilton and Gotlib, 2008). Some other studies reveal that depressed patients present the absence of normal bias toward remembering positive materials which are showed in healthy participants (Ellwart et al., 2003; Harmer et al., 2009). Intriguingly, antidepressant drugs regulate the emotional memory in both depressive and healthy participants by increasing the memory for positive stimuli rather than reducing the memory for negative stimuli in behavioral measures (Harmer et al., 2011; Harmer et al., 2009; Harmer et al., 2004). Besides, similar results were also found in neuroimaging studies (Miskowiak et al., 2007). Tendolkar and colleagues concluded that the memory bias reversal was associated with the enhanced amygdala processing for positive stimuli during memory retrieval rather than reduced processing for negative stimuli (Tendolkar et al., 2011). Indeed, the emotional memory bias has been correlated with the abnormal activity of the amygdala (Disner et al., 2011; Hamilton and Gotlib, 2008; Young et al., 2016), as well as the abnormity of the hippocampus, a region crucial to episodic memory (Ai et al., 2015; Goosens, 2011; Toki et al., 2014). Taken together, depressed patients exhibit memory bias toward negative

materials or away from positive materials. However, the antidepressant drugs tend to reverse memory bias through increasing memory for positive materials. Although drugs and psychological therapies were widely applied in the treatment of depression, there are approximately 30% of patients resistant to drugs and psychological therapies (Stimpson et al., 2002). Patients with severe suicidal tendencies need a rapid response, which cannot be acquired by drugs (Katz et al., 2004). For these patients, electroconvulsive therapy (ECT) is a satisfactory treatment with rapid and efficient antidepressive effect (Husain et al., 2004; Kellner et al., 2012). Unfortunately, several although not all studies suggested that ECT frequently causes cognitive side-effects, mainly the memory impairment (Payne and Prudic, 2009; Squire, 1986). The cognitive impairment restrict the widespread use of ECT, although it may be mainly limited into a short period after ECT (Semkovska and McLoughlin, 2010). However, there is no consistent conclusion about the exact mechanism of its therapeutic and adverse effects. Recently, it is revealed that ECT can impair reconsolidation of memories for emotionally aversive stories in depressed patients (Kroes et al., 2014). Imaging studies emphasize the neurotrophic effects of ECT on the limbic structures mainly including the hippocampus and the amygdala (Joshi et al., 2016; Tendolkar et al., 2013), which are involved in the pathophysiology and memory bias of depression (MacQueen and Frodl, 2011; Pulcu and Elliott, 2015). In a word, ECT can rapidly relieve the abnormal emotion of depressed patients and frequently have short-term memory impairment. However, few studies have examined the effect of ECT on the memory bias in depression. Therefore, the present study mainly aimed to determine the influence of ECT on the emotional memory within a short post-treatment

period in the patients who suffered from depression using incidental memory tests with affective pictures. Based on the previous studies, we hypothesized that ECT would impair discriminatively memory for the emotionally different materials with more memory deficits for negative materials.

2. Materials and methods 2.1. Subjects Depression group: we recruited patients from the Anhui Mental Health Center who were prescribed ECT by psychiatrists because of their resistance to drug therapies or severe suicidal tendencies. All patients met the criteria for a primary Diagnostic and Statistical Manual of Mental Disorders-IV diagnosis of depressive disorder (APA, 1994). We also excluded patients according to the exclusion criteria that include the history of ECT in the last 3 months, substance abuse, score < 24 on Mini Mental State Examination (MMSE), years of schooling < 5, life threatening somatic disease, past or current neurological illness or other mental disorders. Finally, twenty remaining patients were thus included for the study. Healthy controls: we also recruited twenty healthy participants who met the same exclusion criteria as the depressed patients except the diagnosis of depression. They were matched to depressed patients in terms of age, sex, years of schooling. All healthy participants were performed all tests and assessment mentioned above. The study was approved by the Anhui Medical University Ethics Committee, and we obtained informed consent from all patients and healthy controls. 2.2. Clinical symptom assessment

We administered the 17-item Hamilton Depression Rating Scale (HRSD) to assess the severity of depression before and after ECT course. The MMSE was also conducted before and after ECT. These scales were completed about 15 hours before the first ECT administration and on the third day after the last ECT administration. 2.3. Incidental memory tests The incidental memory test included three phases, i.e., unintentional learning, incidental recall and recognition phase. The emotional stimuli for this study were pictures selected from the Chinese Affective Picture System (Bai et al., 2005). Ninety photographs were selected, the half of which were chosen for the learning phase (15 photographs depicting emotionally neutral scenes, 15 photographs emotionally positive scenes and 15 photographs depicting emotionally negative scenes) and the other half were used as distractors for the recognition phase. During the unintentional learning phase, each picture was presented for 3s with an interval during which participants were required to rate the emotional valence (scored from 1 the most negative to 9 the most positive) with no time limit. After all pictures were presented, participants were immediately asked to complete surprise tests of free recall memory. Participants were required to tell the investigator a short description one by one, with sufficient details which made the investigator identify the specific picture. There was no time limit for recalling. A further surprise test of recognition memory was conducted about fifteen hours later. Participants were shown all 45 pictures they had viewed previously together with a new 45 pictures in random order. Mean normative control ratings for emotional valence (scored from 1 to 9 for most positive) and emotional arousal (scored from 1 minimum to 9 maximum) for set 1 were: positive photographs valence 6.98 (SD 0.5), arousal 5.23 (SD 0.82),

neutral photographs valence 5.39 (SD 0.54), arousal 3.57 (SD 0.89) and negative photographs valence 2.73 (SD 0.95), arousal 5.1 (SD 1.31). Mean normative control ratings for emotional valence and emotional arousal for set 2 were: positive photographs valence 6.70 (SD 0.77), arousal 5.48 (SD 0.71), neutral photographs valence 5.58 (SD 0.61), arousal 3.32 (SD 1.20) and negative photographs valence 2.74 (SD 0.97), arousal 4.97 (SD 1.08). Participants were required to make one of two keyboard responses depending on whether they had seen the picture or not. The first ECT was administered after memory tests. More details about ECT procedure can be found below. The same incidental memory tests were repeated on the third day after the last ECT administration. All these tests were conducted on a PC using E-Prime software. The participant’s valence rating for each picture was recorded and we calculated the mean rating for each valence category (negative, neutral and positive) in each participant. We assessed recall memory according to the oral description of participants. The scoring criteria are as follows: one point was given only when we can identify a special picture according to the enough details of participants’ description; we would not score for a description if it is too limited to identify a picture. We would give one point if the participant just offered a single depiction for a group of similar pictures. The recalled items in each category were used as analysis indicators. We evaluated the performance of recognition memory by recognition accuracy (Snodgrass and Corwin, 1988), which was equal to the difference between Hit rate and False Alarm rate. Hit rate and False Alarm rate were calculated for each participant for each valence category by dividing the number of hits and false alarms. The number of hits or false alarms was the number of old or distractor pictures identified as seen before by

participant. 2.4. ECT procedures Patients underwent modified ECT using a Thymatron System IV (Somatics, Inc, Lak e Bluff, IL) with bilateral electrode placement in the fronto-temporal area at the Anhui Mental Health Center. The number of ECT sessions was between six and twelve (three times a week). The energy of the first treatment was determined according to the age of patients. During each ECT procedure, patients were under anesthesia with propofol (1.4mg/kg) in addition to adjuvant drugs with succinylcholine (0.5mg/kg) and atropine (0.5mg). More details could be found in our previous study with the same ECT operation but different samples (Wei et al., 2014). 2.5. Data analyses The effect of depression and ECT on the clinical symptom was tested by independent-samples t-test and paired two-sample t-test, respectively. We used two-way analysis of variance (ANOVA) concerning group (patients and controls) and valence of stimuli (positive, neutral and negative) to compare the differences of rating, recalled items and recognition accuracy between depressed and healthy participants. We also performed a two-factor repeated-measure ANOVA, valence of stimuli repeated over session (pre-ECT and post-ECT) to evaluate the effect of ECT on the memory performance. If significant main effects or interactions were obtained, an independent two sample t-test (patients and controls), paired two-sample t-test (pre-ECT and post-ECT) and one-way ANOVA (positive, neutral and negative) were conducted. If the interaction of session by valence of stimuli was significant, the difference between pre-ECT and post-ECT was calculated (pre-ECT －

post-ECT) and was used to conduct a further one-way ANOVA (positive, neutral and negative). Pearson correlation analyses were performed between the score of HRSD and the performance of recall or recognition memory in patients at pre-ECT and post-ECT, as well as the difference of HRSD and memory performance. The significance level was defined as two-tailed p < 0.05.

3. Results 3.1 | Demographic and clinical characteristics The demographic and clinical characteristics for the two groups (patients and controls) are shown in Table 1. There were no differences between depressed patients and healthy controls in the terms of age, years of education, or gender. Pre-ECT patients scored higher than healthy controls on HRSD. After ECT patients had a significantly lower mean score of HRSD compared to pre-ECT (Fig. 1). 3.2. Valence ratings For valence ratings, the interaction between group (patients and controls) and valence of picture (negative, neutral and positive) (F (2, 114) = 4.94, p < 0.01), the main effect of valence of picture (F (1, 114) = 176.02, p < 0.001) and group (F (1, 114) = 5.26, p < 0.05) were obtained (Table 2; Fig. 2, panel A). Planed t-test implied that pre-ECT patients rated positive pictures less delightedly than control participants (t (38) = 4.26, p < 0.001), but they did not differ in their ratings of negative or neutral pictures. In depressed patients, one-way ANOVA showed that before ECT patients rated positive pictures as the most and negative pictures as the least, and a similar result was replicated in healthy controls (p < 0.001 for all).

There were no main effects for session (pre- and post-ECT) or the interaction between session and valence of pictures, which means that ECT had no effect on patients’ valence ratings (Table 3; Fig. 3, panel A). 3.3. Recall Memory 3.3.1. The effect of depression There was no significant difference between depressed patients and healthy participants on the performance of recall memory as revealed by a main effect of group (F (1, 114) = 2.44, p > 0.05) (Table 2; Fig. 2, panel B). The group by valence interaction was not significant (F (2, 114) = 0.44, p > 0.05), either. The main effect of valence was significant (F (1, 114) = 11.23, p < 0.001). Further one-way ANOVA in depression and control groups revealed a significant difference for patients (F (2, 57) = 5.12, p < 0.01) and controls (F (2, 57) = 6.95, p < 0.01). Post hoc tests showed that both depressed and healthy participants recalled positive and negative pictures better than neutral pictures (p < 0.05 for all), but there was no significant difference between the performance of recall memory for positive and negative pictures in both group. 3.3.2. The effect of ECT There was a main effect of session (F (1, 57) = 25.45, p < 0.001), but no interaction between session and valence of pictures (F (2, 57) = 0.90, p > 0.05) was found. Post hoc t-test implied that patients had a poorer recall memory for all pictures (p < 0.001 for all) (Table 3; Fig. 3, panel B). 3.4. Recognition Memory 3.4.1. The effect of depression

There was a significant effect of interaction between group and valence for recognition memory (F (2, 114) = 4.34, p < 0.05), which demonstrated that patients with depression had a selective deficit in the recognition memory of affective pictures compared to healthy participants (Table 2; Fig. 2, panel C). Post hoc t-test revealed that depressed patients had a poorer performance of recognition memory for positive pictures (t = 2.61, p<0.05) than healthy controls. No significant difference was found for the recognition of negative pictures (t = 0.57, p > 0.05) and neutral pictures (t = 1.28, p > 0.05) between depressed patients and healthy controls. The main effect of group (F (1, 114) = 0.37, p > 0.05) or valence (F (2, 114) = 2.24, p > 0.05) were not significant. 3.4.2. The effect of ECT There was a significant difference of recognition memory in depressed patients between pre- ECT and post-ECT (F (1, 57) = 53.35, p < 0.001). Post hoc t-test indicates that ECT impaired the recognition memory of depressed patients for all three valence categories (p < 0.05 for all). More importantly, we found that ECT had a selective impairment on recognition memory for different valences as demonstrated by a significant group by stimuli interaction (F (1, 57) = 4.86, p < 0.05) (Table 3; Fig. 3, panel C). Further one-way ANOVA of the difference between pre- and post-ECT suggested that ECT had a more severe impairment on the memory for negative stimuli than positive stimuli (post hoc test with Bonferroni, p < 0.01). 3.5. Correlation analyses There were no correlations between HRSD score and any of memory performance before and after ECT. There were no significant correlations between the change of depressive

symptoms and the change of memory performance. There were also no significant relationships between the emotional memory for three valenced categories at pre-ECT and clinical factors (duration of illness, duration of depressive episode and the baseline Hamilton score). We further explored whether patients’ memory performance related to their age, educational years or MMSE score, but no significant correlations was found.

4. Discussion In this study, we examined the memory bias for emotional information in depressed patients. We found that depressed patients exhibited worse memory performance for positive pictures during both recall and recognition tasks compared to healthy controls. They also showed worse memory performance for positive pictures than their memory for negative pictures during the recognition task. More significantly, our research revealed that ECT regulated the emotional memory in depressed patients with more severe impairment on the memory for negative stimuli. The memory bias in depression has been demonstrated in previous researches. Considerable studies have found enhanced memory for negative materials in depressed patients (Bradley et al., 1995; Dunbar and Lishman, 1984; Hamilton and Gotlib, 2008; Rinck and Becker, 2005). In this study, however, we reported reduced positive memory in depressed patients compared to healthy controls. Indeed, the deficits of positive memory in depression are obvious. Some studies, which used affective words for the recall task (Harmer et al., 2009; Sloan et al., 2001) and facial expressions for the recognition task (Gilboa-Schechtman et al., 2002), also reported decreased positive memory in depression. More studies showed that

depressed patients exhibited the absence of better memory for positive materials compared to negative materials, which could frequently be seen in healthy individuals (Denny and Hunt, 1992; Gotlib et al., 2004). More importantly, a meta-analysis study suggested that the absent positive memory bias was more reliable than the enhanced negative memory in depression (Burt et al., 1995). Similarly, the present study showed reduced memory for positive information compared to negative information in depressed patients, but not in healthy controls. Our results showed that ECT had a significant impairment on three valenced stimuli-negative, positive and neutral stimuli. This finding is in line with some reports that showed the memory impairment in depressed patients after ECT (Schat et al., 2007; Squire, 1986). In fact, the short-term memory deficits after ECT have been implicated in several previous studies using standardized neuropsychological tests, such as Rey Auditory Verbal Learning and Cambridge Neuropsychological Test Automated Battery (Bodnar et al., 2016; Falconer et al., 2010). Though, of course, intact memory was also found in another study using Cambridge Neuropsychological Test Automated Battery (Maric et al., 2016). One potential factor contributing to such difference is the older participant in the former study (mean age 52 versus 47), which was associated significant cognitive deficits after ECT (Sackeim et al., 2007). However, younger participant in our study also reported memory impairment for affective stimulus. One possible reason is that the emotional memory depend on more brain regions, including amygdala, frontal and temporal lobes, as well as their interactions (LaBar and Cabeza, 2006) and the function of these regions are susceptible to ECT (Abbott et al., 2014; Bai et al., 2017; Redlich et al., 2017).

It is important to note, however, that our result showed a modulatory role of ECT in memory bias by means of decreasing negative memory in depression. Indeed, ECT-mediated impairment of negative memories has been demonstrated in depression (Kroes et al., 2014) and post-traumatic stress disorder (Gahr et al., 2014). This effect may be used to treat post-traumatic stress disorder in the future (Andrade et al., 2016). Differently, some previous studies showed enhanced positive memory after the application of antidepressant drugs in depressive (Harmer et al., 2009) and healthy individuals (Harmer et al., 2004). Tendolkar and colleagues concluded that antidepressant drugs reverse the memory bias with the method of enhanced processing for positive stimuli during memory retrieval rather than reduced processing for negative stimuli (Tendolkar et al., 2011). The difference of emotional memory performance between ECT and antidepressant drug implied the disparate therapeutic mechanism of these treatments. However, antidepressant medication was used in depressed patients during ECT treatment, which would produce a mixed effect for our results. Concomitant pharmacotherapy has been demonstrated to improve the antidepressive effect and reduced the cognitive adverse effects of ECT (Sackeim et al., 2009). On the other hand, the dose and the type of antidepressants were little changed throughout the ECT course in present study, which may lessen the mixed effect. Of course, it is necessary to included patients with no antidepressant medication in future research. The previous studies have found that the amygdala and the hippocampus are associated with the regulation of emotional memory in animals and humans (Hamann et al., 1999; Kemppainen et al., 2002; Smith et al., 2006; van Tol et al., 2012). For depressed patients, a large body of researches have shown that the function of the amygdala and the hippocampus

are closely related to the memory bias in depression (Disner et al., 2011). Hamilton and Gotlib suggested that the negative memory bias in depressed patients was associated with the excessive activation of the right amygdala and the functional connectivity between the amygdala and the hippocampus during the encoding phase for negative stimuli (Hamilton and Gotlib, 2008). Moreover, the aberrant responses of the amygdala to emotional stimuli in depressed patients can be altered by antidepressants (Anand et al., 2007; Fu et al., 2008; Sheline et al., 2001). There is also plenty of evidence that ECT has a significant effect on the function and the structure of the hippocampus and the amygdale (Abbott et al., 2014; Joshi et al., 2016; Tendolkar et al., 2013), although few researches revealed the effect of ECT on the neural response for emotional stimuli directly. Further studies are needed to explore the neural substrates associated with the effect of ECT on the emotional memory in depressed patients. The bilateral ECT with relatively more frequency (3 times per week) was used in our study, this relatively strong dosage may be associated the memory impairment. Indeed, the number of ECT treatments is associated with greater cognitive impairment (Sackeim et al., 2007). Less frequent ECT application (2 times per week) has been demonstrated less cognitive impairment, particularly memory impairment, than ECT with the frequency of 3 times per week (Shapira et al., 2000). However, there was also evidence that show no difference between ECT with the frequency of 3 times per week and 2 times per week (Maric et al., 2016). In addition, the placement of electrode placements is another key factor which impacts the cognitive outcome of ECT. Specially, the bitemporal ECT often causes more pronounced and persistent cognitive impairments than bifrontal and right unilateral ECT

(Bailine et al., 2000; Sackeim et al., 1993). There are also evidence showed no substantially different between the profile of bitemporal and bifrontal ECT (Kellner et al., 2010), used in present study. Hence, future studies are needed to evaluate the effect of different electrode placement in ECT on emotional memory. It is noteworthy that propofol with a significant impact on memory consolidation (Veselis et al., 2009; Veselis et al., 2008) may also produce similar effects. For example, some studies found that the participants in the propofol group showed inferior recognition memory performance for negative stimuli than neutral stimuli (Pryor et al., 2015). However, they infused propofol at subclinical doses to the participants conducting the encoding task simultaneously to examine which process of memory formation was damaged by propofol. Taking a different approach, we assessed the memory of patients after the last administration of propofol more than seventy two hours. Many studies revealed that the pretreatment of propofol for patients treated by ECT or animals treated by electroconvulsive stimulation (ECS) had a protective effect on the impairment of memory (Imashuku et al., 2014; Luo et al., 2014; Luo et al., 2011). For example, some studies suggested that rats managed by ECS pretreated with normal saline showed worse learning-memory performance than those pretreated with propofol (Luo et al., 2011). Besides, animal studies have shown that the administration of propofol at the anesthetic dose can enhance emotional memory consolidation (Hauer et al., 2011). Indeed, further investigations are needed to explain the effect of repetitious administration of propofol on the emotional memory. There are also several additional limitations. First, we got a small sample size in the ECT group, which reduced the reliability of results. Second, considering the scarcity of

patients who met our standards, the participants enrolled into the ECT group consisted of patients with major depression and treatment-resistant depression. Further studies are needed to measure the effect of ECT on the emotional memory in specific types of depression. Last but not the least, we did not observe the effect of ECT on emotional memory after a long-term follow-up. The results from the current study indicate that ECT regulates the emotional memory of depressed patients with more severe impairment on the memory for negative stimuli than positive stimuli. Investigations of behavioral changes of ECT on the emotional memory in depression may help to elucidate the possible therapeutic mechanism of ECT in depression. More work is necessary to clarify the neural substrates of partial impairment on the emotional memory in depressed patients.

Conflicts of interest none.

Acknowledgments This study was supported by funding from National Nature Science Foundation of China (No. 81471117 and No. 81671354), the National Basic Research Program of China (No.2015CB856400) and National Key Technology Research and Development Program of the Ministry of Science and Technology of China (No. 2015BAI13B01).

References

Abbott, C.C., Jones, T., Lemke, N.T., Gallegos, P., McClintock, S.M., Mayer, A.R., Bustillo, J., Calhoun, V.D., 2014. Hippocampal structural and functional changes associated with electroconvulsive therapy response. Transl Psychiatry 4, e483. Ai, H., Opmeer, E.M., Veltman, D.J., van der Wee, N.J., van Buchem, M.A., Aleman, A., van Tol, M.J., 2015. Brain Activation During Emotional Memory Processing Associated with Subsequent Course of Depression. Neuropsychopharmacology 40 (10), 2454-2463. Anand, A., Li, Y., Wang, Y., Gardner, K., Lowe, M.J., 2007. Reciprocal effects of antidepressant treatment on activity and connectivity of the mood regulating circuit: an FMRI study. J Neuropsychiatry Clin Neurosci 19 (3), 274-282. Andrade, C., McCall, W.V., Youssef, N.A., 2016. Electroconvulsive therapy for post-traumatic stress disorder: efficacy, mechanisms and a hypothesis for new directions. Expert Review of Neurotherapeutics 16 (7), 749-753. APA, 1994. Diagnostic and Statistical Manual of Mental Disorder, 4th edn. American Psychiatric Press, Washington, DC. Asberg, M., Thoren, P., Traskman, L., Bertilsson, L., Ringberger, V., 1976. "Serotonin depression"--a biochemical subgroup within the affective disorders? Science 191 (4226), 478-480. Bai, L., Ma, H., Huang, Y.X., Luo, Y.J., 2005. The development of native Chinese affective picture system-a pretest in 46 college students. Chin. Mental Health J. 19, 719-722. Bai, S., Galvez, V., Dokos, S., Martin, D., Bikson, M., Loo, C., 2017. Computational models of Bitemporal, Bifrontal and Right Unilateral ECT predict differential stimulation of brain regions associated with efficacy and cognitive side effects. Eur Psychiatry 41, 21-29. Bailine, S.H., Rifkin, A., Kayne, E., Selzer, J.A., Vital-Herne, J., Blieka, M., Pollack, S., 2000. Comparison of bifrontal and bitemporal ECT for major depression. Am J Psychiatry 157 (1), 121-123. Beck, A.T., 2008. The evolution of the cognitive model of depression and its neurobiological correlates. Am J Psychiatry 165 (8), 969-977. Beck, A.T., Rush, A.J., Shaw, B.F., Emery, G., 1979. Cognitive Therapy of Depression. Guilford Press, New York. Bodnar, A., Krzywotulski, M., Lewandowska, A., Chlopocka-Wozniak, M., Bartkowska-Sniatkowska, A., Michalak, M., Rybakowski, J.K., 2016. Electroconvulsive therapy and cognitive functions in treatment-resistant depression. World J Biol Psychiatry 17 (2), 159-164. Bouhuys, A.L., Geerts, E., Gordijn, M.C., 1999. Depressed patients' perceptions of facial emotions in depressed and remitted states are associated with relapse: a longitudinal study. J Nerv Ment Dis 187 (10), 595-602. Bradley, B.P., Mogg, K., Millar, N., White, J., 1995. Selective processing of negative information: effects of clinical anxiety, concurrent depression, and awareness. J Abnorm Psychol 104 (3), 532-536. Burt, D.B., Zembar, M.J., Niederehe, G., 1995. Depression and memory impairment: a meta-analysis of the association, its pattern, and specificity. Psychol Bull 117 (2), 285-305.

Denny, E.B., Hunt, R.R., 1992. Affective valence and memory in depression: dissociation of recall and fragment completion. J Abnorm Psychol 101 (3), 575-580. Direnfeld, D.M., Roberts, J.E., 2006. Mood congruent memory in dysphoria: the roles of state affect and cognitive style. Behav Res Ther 44 (9), 1275-1285. Disner, S.G., Beevers, C.G., Haigh, E.A., Beck, A.T., 2011. Neural mechanisms of the cognitive model of depression. Nat Rev Neurosci 12 (8), 467-477. Dunbar, G.C., Lishman, W.A., 1984. Depression, recognition-memory and hedonic tone a signal detection analysis. Br J Psychiatry 144, 376-382. Ellwart, T., Rinck, M., Becker, E.S., 2003. Selective memory and memory deficits in depressed inpatients. Depress Anxiety 17 (4), 197-206. Falconer, D.W., Cleland, J., Fielding, S., Reid, I.C., 2010. Using the Cambridge Neuropsychological Test Automated Battery (CANTAB) to assess the cognitive impact of electroconvulsive therapy on visual and visuospatial memory. Psychol Med 40 (6), 1017-1025. Fu, C.H., Williams, S.C., Cleare, A.J., Scott, J., Mitterschiffthaler, M.T., Walsh, N.D., Donaldson, C., Suckling, J., Andrew, C., Steiner, H., Murray, R.M., 2008. Neural responses to sad facial expressions in major depression following cognitive behavioral therapy. Biol Psychiatry 64 (6), 505-512. Gahr, M., Schönfeldt-Lecuona, C., Spitzer, M., Graf, H., 2014. Electroconvulsive therapy and posttraumatic stress disorder: first experience with conversation-based reactivation of traumatic memory contents and subsequent ECT-mediated impairment of reconsolidation. J Neuropsychiatry Clin Neurosci 26 (3), e38-39. Gilboa-Schechtman, E., Erhard-Weiss, D., Jeczemien, P., 2002. Interpersonal deficits meet cognitive biases: memory for facial expressions in depressed and anxious men and women. Psychiatry Res 113 (3), 279-293. Goosens, K.A., 2011. Hippocampal regulation of aversive memories. Curr Opin Neurobiol 21 (3), 460-466. Gotlib, I.H., Kasch, K.L., Traill, S., Joormann, J., Arnow, B.A., Johnson, S.L., 2004. Coherence and specificity of information-processing biases in depression and social phobia. J Abnorm Psychol 113 (3), 386-398. Hamann, S.B., Ely, T.D., Grafton, S.T., Kilts, C.D., 1999. Amygdala activity related to enhanced memory for pleasant and aversive stimuli. Nat Neurosci 2 (3), 289-293. Hamilton, J.P., Gotlib, I.H., 2008. Neural substrates of increased memory sensitivity for negative stimuli in major depression. Biol Psychiatry 63 (12), 1155-1162. Harmer, C.J., de Bodinat, C., Dawson, G.R., Dourish, C.T., Waldenmaier, L., Adams, S., Cowen, P.J., Goodwin, G.M., 2011. Agomelatine facilitates positive versus negative affective processing in healthy volunteer models. J Psychopharmacol 25 (9), 1159-1167. Harmer, C.J., O'Sullivan, U., Favaron, E., Massey-Chase, R., Ayres, R., Reinecke, A., Goodwin, G.M., Cowen, P.J., 2009. Effect of acute antidepressant administration on negative affective bias in depressed patients. Am J Psychiatry 166, 1178-1184. Harmer, C.J., Shelley, N.C., Cowen, P.J., Goodwin, G.M., 2004. Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition. Am J Psychiatry 161 (7),

1256-1263. Hasler, G., Drevets, W.C., Manji, H.K., Charney, D.S., 2004. Discovering endophenotypes for major depression. Neuropsychopharmacology 29 (10), 1765-1781. Hauer, D., Ratano, P., Morena, M., Scaccianoce, S., Briegel, I., Palmery, M., Cuomo, V., Roozendaal, B., Schelling, G., Campolongo, P., 2011. Propofol enhances memory formation via an interaction with the endocannabinoid system. Anesthesiology 114 (6), 1380-1388. Husain, M.M., Rush, A.J., Fink, M., Knapp, R., Petrides, G., Rummans, T., Biggs, M.M., O'Connor, K., Rasmussen, K., Litle, M., Zhao, W., Bernstein, H.J., Smith, G., Mueller, M., McClintock, S.M., Bailine, S.H., Kellner, C.H., 2004. Speed of response and remission in major depressive disorder with acute electroconvulsive therapy (ECT): a Consortium for Research in ECT (CORE) report. J Clin Psychiatry 65 (4), 485-491. Imashuku, Y., Kanemoto, K., Senda, M., Matsubara, M., 2014. Relationship between blood levels of propofol and recovery of memory in electroconvulsive therapy. Psychiatry Clin Neurosci 68 (4), 270-274. Joshi, S.H., Espinoza, R.T., Pirnia, T., Shi, J., Wang, Y., Ayers, B., Leaver, A., Woods, R.P., Narr, K.L., 2016. Structural Plasticity of the Hippocampus and Amygdala Induced by Electroconvulsive Therapy in Major Depression. Biol Psychiatry 79 (4), 282-292. Katz, M.M., Tekell, J.L., Bowden, C.L., Brannan, S., Houston, J.P., Berman, N., Frazer, A., 2004. Onset and early behavioral effects of pharmacologically different antidepressants and placebo in depression. Neuropsychopharmacology 29 (3), 566-579. Kellner, C.H., Greenberg, R.M., Murrough, J.W., Bryson, E.O., Briggs, M.C., Pasculli, R.M., 2012. ECT in treatment-resistant depression. Am J Psychiatry 169 (12), 1238-1244. Kellner, C.H., Knapp, R., Husain, M.M., Rasmussen, K., Sampson, S., Cullum, M., McClintock, S.M., Tobias, K.G., Martino, C., Mueller, M., Bailine, S.H., Fink, M., Petrides, G., 2010. Bifrontal, bitemporal and right unilateral electrode placement in ECT: randomised trial. Br J Psychiatry 196 (3), 226-234. Kemppainen, S., Jolkkonen, E., Pitkanen, A., 2002. Projections from the posterior cortical nucleus of the amygdala to the hippocampal formation and parahippocampal region in rat. Hippocampus 12 (6), 735-755. Kroes, M.C., Tendolkar, I., van Wingen, G.A., van Waarde, J.A., Strange, B.A., Fernandez, G., 2014. An electroconvulsive therapy procedure impairs reconsolidation of episodic memories in humans. Nat Neurosci 17 (2), 204-206. LaBar, K.S., Cabeza, R., 2006. Cognitive neuroscience of emotional memory. Nat Rev Neurosci 7 (1), 54-64. Luo, J., Min, S., Wei, K., Cao, J., Wang, B., Li, P., Dong, J., Liu, Y., 2014. Propofol prevents electroconvulsive-shock-induced memory impairment through regulation of hippocampal synaptic plasticity in a rat model of depression. Neuropsychiatr Dis Treat 10, 1847-1859. Luo, J., Min, S., Wei, K., Li, P., Dong, J., Liu, Y.F., 2011. Propofol protects against impairment of learning-memory and imbalance of hippocampal Glu/GABA induced by electroconvulsive shock in depressed rats. J Anesth 25 (5), 657-665. MacQueen, G., Frodl, T., 2011. The hippocampus in major depression: evidence for the

convergence of the bench and bedside in psychiatric research? Molecular Psychiatry 16, 252-264. Maric, N.P., Stojanovic, Z., Andric, S., Soldatovic, I., Dolic, M., Spiric, Z., 2016. The acute and medium-term effects of treatment with electroconvulsive therapy on memory in patients with major depressive disorder. Psychol Med 46 (4), 797-806. Miskowiak, K., Papadatou-Pastou, M., Cowen, P.J., Goodwin, G.M., Norbury, R., Harmer, C.J., 2007. Single dose antidepressant administration modulates the neural processing of self-referent personality trait words. NeuroImage 37 (3), 904-911. Payne, N.A., Prudic, J., 2009. Electroconvulsive therapy: Part I. A perspective on the evolution and current practice of ECT. J Psychiatr Pract 15 (5), 346-368. Pryor, K.O., Root, J.C., Mehta, M., Stern, E., Pan, H., Veselis, R.A., Silbersweig, D.A., 2015. Effect of propofol on the medial temporal lobe emotional memory system: a functional magnetic resonance imaging study in human subjects. Br J Anaesth 115 Suppl 1, i104-i113. Pulcu, E., Elliott, R., 2015. Neural origins of psychosocial functioning impairments in major depression. The Lancet Psychiatry 2 (9), 835-843. Redlich, R., Burger, C., Dohm, K., Grotegerd, D., Opel, N., Zaremba, D., Meinert, S., Forster, K., Repple, J., Schnelle, R., Wagenknecht, C., Zavorotnyy, M., Heindel, W., Kugel, H., Gerbaulet, M., Alferink, J., Arolt, V., Zwanzger, P., Dannlowski, U., 2017. Effects of electroconvulsive therapy on amygdala function in major depression - a longitudinal functional magnetic resonance imaging study. Psychol Med, 1-11. Rinck, M., Becker, E.S., 2005. A comparison of attentional biases and memory biases in women with social phobia and major depression. J Abnorm Psychol 114 (1), 62-74. Roiser, J.P., Elliott, R., Sahakian, B.J., 2012. Cognitive mechanisms of treatment in depression. Neuropsychopharmacology 37 (1), 117-136. Sackeim, H.A., Dillingham, E.M., Prudic, J., Cooper, T., McCall, W.V., Rosenquist, P., Isenberg, K., Garcia, K., Mulsant, B.H., Haskett, R.F., 2009. Effect of concomitant pharmacotherapy on electroconvulsive therapy outcomes: short-term efficacy and adverse effects. Arch Gen Psychiatry 66 (7), 729-737. Sackeim, H.A., Prudic, J., Devanand, D.P., Kiersky, J.E., Fitzsimons, L., Moody, B.J., McElhiney, M.C., Coleman, E.A., Settembrino, J.M., 1993. Effects of stimulus intensity and electrode placement on the efficacy and cognitive effects of electroconvulsive therapy. N Engl J Med 328 (12), 839-846. Sackeim, H.A., Prudic, J., Fuller, R., Keilp, J., Lavori, P.W., Olfson, M., 2007. The cognitive effects of electroconvulsive therapy in community settings. Neuropsychopharmacology 32 (1), 244-254. Schat, A., van den Broek, W.W., Mulder, P.G., Birkenhager, T.K., van Tuijl, R., Murre, J.M., 2007. Changes in everyday and semantic memory function after electroconvulsive therapy for unipolar depression. J ect 23 (3), 153-157. Semkovska, M., McLoughlin, D.M., 2010. Objective cognitive performance associated with electroconvulsive therapy for depression: a systematic review and meta-analysis. Biol Psychiatry 68 (6), 568-577. Shapira, B., Tubi, N., Lerer, B., 2000. Balancing speed of response to ECT in major depression and adverse cognitive effects: role of treatment schedule. J ect 16 (2),

97-109. Sheline, Y.I., Barch, D.M., Donnelly, J.M., Ollinger, J.M., Snyder, A.Z., Mintun, M.A., 2001. Increased amygdala response to masked emotional faces in depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry 50 (9), 651-658. Sloan, D.M., Strauss, M.E., Wisner, K.L., 2001. Diminished response to pleasant stimuli by depressed women. J Abnorm Psychol 110 (3), 488-493. Smith, A.P., Stephan, K.E., Rugg, M.D., Dolan, R.J., 2006. Task and content modulate amygdala-hippocampal connectivity in emotional retrieval. Neuron 49 (4), 631-638. Snodgrass, J.G., Corwin, J., 1988. Pragmatics of measuring recognition memory: applications to dementia and amnesia. J Exp Psychol Gen 117 (1), 34-50. Squire, L.R., 1986. Memory functions as affected by electroconvulsive therapy. Ann N Y Acad Sci 462, 307-314. Stimpson, N., Agrawal, N., Lewis, G., 2002. Randomised controlled trials investigating pharmacological and psychological interventions for treatment-refractory depression. Systematic review. Br J Psychiatry 181, 284-294. Tendolkar, I., van Beek, M., van Oostrom, I., Mulder, M., Janzing, J., Voshaar, R.O., van Eijndhoven, P., 2013. Electroconvulsive therapy increases hippocampal and amygdala volume in therapy refractory depression: a longitudinal pilot study. Psychiatry Res 214 (3), 197-203. Tendolkar, I., van Wingen, G., Urner, M., Verkes, R.J., Fernandez, G., 2011. Short-term duloxetine administration affects neural correlates of mood-congruent memory. Neuropsychopharmacology 36 (11), 2266-2275. Toki, S., Okamoto, Y., Onoda, K., Matsumoto, T., Yoshimura, S., Kunisato, Y., Okada, G., Shishida, K., Kobayakawa, M., Fukumoto, T., Machino, A., Inagaki, M., Yamawaki, S., 2014. Hippocampal activation during associative encoding of word pairs and its relation to symptomatic improvement in depression: a functional and volumetric MRI study. J Affect Disord 152-154, 462-467. van Tol, M.J., Demenescu, L.R., van der Wee, N.J., Kortekaas, R., Marjan, M.A.N., Boer, J.A., Renken, R.J., van Buchem, M.A., Zitman, F.G., Aleman, A., Veltman, D.J., 2012. Functional magnetic resonance imaging correlates of emotional word encoding and recognition in depression and anxiety disorders. Biol Psychiatry 71 (7), 593-602. Veselis, R.A., Pryor, K.O., Reinsel, R.A., Li, Y., Mehta, M., Johnson, R., Jr., 2009. Propofol and midazolam inhibit conscious memory processes very soon after encoding: an event-related potential study of familiarity and recollection in volunteers. Anesthesiology 110 (2), 295-312. Veselis, R.A., Pryor, K.O., Reinsel, R.A., Mehta, M., Pan, H., Johnson, R., Jr., 2008. Low-dose propofol-induced amnesia is not due to a failure of encoding: left inferior prefrontal cortex is still active. Anesthesiology 109 (2), 213-224. Wei, Q., Tian, Y., Yu, Y., Zhang, F., Hu, X., Dong, Y., Chen, Y., Hu, P., Hu, X., Wang, K., 2014. Modulation of interhemispheric functional coordination in electroconvulsive therapy for depression. Transl Psychiatry 4, e453. Young, K.D., Siegle, G.J., Bodurka, J., Drevets, W.C., 2016. Amygdala Activity During Autobiographical Memory Recall in Depressed and Vulnerable Individuals:

Association With Symptom Severity and Autobiographical Overgenerality. Am J Psychiatry 173 (1), 78-89.

Fig. 1. Depressive severity in healthy controls and pre- and post-ECT depressed patients. Bars indicate the standard error of the mean. ** p < 0.001. Abbreviation: HDRS, Hamilton Depression Rate Scale; ECT, electroconvulsive therapy. Fig. 2. The difference between healthy controls and pre-ECT depressed patients in valence rating (A), recall memory (B) and recognition memory (C). Patients and healthy controls showed the highest rating toward positive stimuli and least rating toward negative stimuli. Pre-ECT patients showed less rating toward positive stimuli than healthy controls (A). Pre-ECT patients and healthy controls showed better recall memory toward positive and negative stimuli than the neutral (B). Pre-ECT patients showed reduced recognition memory toward positive stimuli than healthy controls. Within group, pre-ECT patients showed reduced recognition memory toward positive stimuli than the negative (C). Fig. 3. The difference between pre- and post-ECT depressed patients in valence rating (A), recall memory (B) and recognition memory (C). There was no effect on patients’ valence ratings by ECT (A). ECT reduced patients recall memory for positive, neutral and negative stimuli with no discrimination (B). ECT reduced patients recall memory for positive, neutral and negative stimuli. The reduction of recognition memory for negative stimuli was more than the positive (C).

Table 1. Demographic and clinical data of depressed patients and healthy controls Subjects Number of subjects

Patients (n=20)

Controls (n=20)

20

20

p value

37.2 ± 8.92

36.5 ± 9.40

3 / 17

3 / 17

10.20 ± 3.60

9.80 ± 4.06

0.74

Pre-ECT

19.40 ± 5.72

1.30 ± 1.53

< 0.001

Post-ECT

2.85 ± 1.87

Age (mean ± SD) Gender (male/female) Education level (mean ± SD)

0.82

HRSD scores (mean ± SD)

Age of onset

< 0.001a

31.70 ± 8.61

Episodes (First/ Recurrence)

8 / 12

Treatment times (mean ± SD)

8.05 ± 1.54

Medications SSRIs (patients number)

11

SNRIs

6

SARIs

5

NaSSAs

3

Antipsychotics

8

Two-sample t-tests were used for age, education and depressive symptoms comparisons between healthy controls and depressed patients pre-ECT. a The comparison was conducted between pre- and post-ECT patients using paired t-test. HRSD = Hamilton Rating Scale for Depression; SSRIs = selective serotonin reuptake inhibitors; SNRIs = serotonin–norepinephrine reuptake inhibitors; SARIs = serotonin antagonist/ reuptake inhibitors; NaSSAs = norepinephrine and specificity serotonergic antidepressants.

Table 2. The valence rating and memory performance in depressed patients and healthy controls Patients (n=20) Controls (n=20) Statistics a Positive Neutral Negative Positive Neutral Negative F p Valence rating 6.42 5.31 2.96 7.62 5.52 2.81 (1.24) 4.94 0.009 (1.01) (0.73) (1.20) (0.75) (0.91) Recalled items 4.10 2.55 4.45 4.95 3.15 4.55 (1.70) 0.44 0.644 (2.20) (1.50) (2.21) (1.67) (1.42) Recognition 0.37 0.45 0.53 0.52 0.40 0.49 (0.20) 4.34 0.015 accuracy (0.20) (0.19) (0.17) (0.17) (0.15) Means (standard deviations) are provided. a The data were analyzed using two-way ANOVA concerning group (Patient and Control) and valence of stimuli (positive, neutral and negative), the significant interaction between group and valence of picture appear in bold (p < 0.05). Table 3. The valence rating and memory performance in pre- and post-ECT depressed patients Pre-ECT (n=20) Post-ECT (n=20) Statistics a Positive Neutral Negative Positive Neutral Negative F p Valence rating 6.42 5.31 2.96 6.74 5.53 2.89 0.64 0.533 (1.01) (0.73) (1.20) (1.10) (1.20) (1.26) Recalled items 4.10 2.55 4.45 3.10 1.60 2.80 0.90 0.413 (2.20) (1.50) (2.21) (2.73) (1.53) (1.90) Recognition 0.37 0.45 0.53 0.25 0.18 0.15 57.00 0.011 accuracy (0.20) (0.19) (0.17) (0.20) (0.22) (0.27) Means (standard deviations) are provided. a The data were analyzed using two-factor repeated-measure ANOVA, valence of stimuli repeated over session (pre-ECT and post-ECT). Significant interaction between session and valence of picture appear in bold (p < 0.05).

Highlights ● There are emotional memory bias in depression and ECT has significant effect on emotion and memory in depression; ● We evaluate the emotional memory in depression pre and post ECT; ● ECT reduces memory for all stimuli but has more severe impairment for negative stimuli; ● ECT regulates memory bias in depression through more severe impairment for negative stimuli;