An electroconvulsive therapy procedure impairs reconsolidation of episodic memories in humans

Journal name:
Nature Neuroscience
Volume:
17,
Pages:
204–206
Year published:
DOI:
doi:10.1038/nn.3609
Received
Accepted
Published online

Despite accumulating evidence for a reconsolidation process in animals, support in humans, especially for episodic memory, is limited. Using a within-subjects manipulation, we found that a single application of electroconvulsive therapy following memory reactivation in patients with unipolar depression disrupted reactivated, but not non-reactivated, memories for an emotional episode in a time-dependent manner. Our results provide evidence for reconsolidation of emotional episodic memories in humans.

At a glance

Figures

  1. ECT disrupts reconsolidation.
    Figure 1: ECT disrupts reconsolidation.

    Memory scores on the multiple choice test are expressed as percentage correct (y axis). One day after reactivation and ECT, memory for the reactivated story (solid bars) was not different from chance (group A) and was impaired compared with the non-reactivated story (open bars). Notably, consistent with reconsolidation being a time-dependent process, testing memory immediately after ECT revealed no difference in recognition scores for the reactivated story compared with the non-reactivated story (group B). Furthermore, in the absence of ECT (group C) memory for the reactivated story was enhanced relative to the non-reactivated story. The dotted line indicates chance level (25%) and error bars represent s.e.m. Scores (%) for each story per group: group A (n = 13 subjects) reactivated mean = 25.35, s.e.m. = 1.57; non-reactivated mean = 33.83, s.e.m. = 3.36; group B (n = 13 subjects) reactivated mean = 33.34, s.e.m. = 2.61; non-reactivated mean = 33.31, s.e.m. = 3.11; group C (n = 13 subjects) reactivated mean = 48.90, s.e.m. = 4.64; non-reactivated mean = 39.55, s.e.m. = 3.50. ** significant group × reactivation effect (P < 0.001), *significant effect (P = 0.012, P = 0.001), n.s. = non-significant (P = 0.994); circles represent individual data points.

  2. Study design.
    Supplementary Fig. 1: Study design.

    Patients were assigned to one of three groups (A, B, C). During a first study session all groups were shown two emotional slide-show stories. During a second session memory for one of the two stories was reactivated. Immediately after memory reactivation patient in groups A and B received ECT. In patients of group B memory was tested immediately upon recovery from ECT (Test, blue). In patients of groups A and C memory was tested one day after reactivation (Test, red and orange respectively).

  3. Stimulus material
    Supplementary Fig. 2: Stimulus material

    Patients were presented with two slide shows that form arousing episodic stories of negative valence. Top: the original “Cahill Story”, bottom: the newly developed story. Both stories consist of 11 slides and each slide is accompanied by an auditory narrative.

  4. Memory reactivation scores
    Supplementary Fig. 3: Memory reactivation scores

    All groups showed evidence of memory reactivation, i.e., memory performance at reactivation was above chance level as indexed by the memory reactivation score (one-sample t-test across all groups, (t(36) = 7.53, P < 0.001), and groups did not differ in memory reactivation scores (y-axis; Kruskal-Wallis for group (A,B,C), H(2) = 1.77, P = 0.412; group A (red) mean: 3.08, s.e.m.: 0.35; group B (blue): mean: 2.62, s.e.m.: 0.96; group C (orange) mean: 3.23; s.e.m.: 1.24). Therefore, the observed between-group differences in reactivated memories are not due to differences in strength of memory reactivation, and adequate memory reactivation principally allows the initiation of a reconsolidation process. Dashed line indicates chance level (25%), error bars depict s.e.m.

  5. DSST results
    Supplementary Fig. 4: DSST results

    DSST scores in minutes (y-axis). A group (A, B, C) x time point (study, test) repeated measures ANOVA on DSST scores revealed no main effect of time point (F1, 33 = 1.72, P = 0.199), group (F2, 33 = 0.69, P = 0.509) or group x time point interaction (F2, 33 = 1.64, P = 0.209). Thus, General cognitive functioning does not differ between groups. Hence, group differences in memory performance are unlikely to be due to group differences in general cognitive functioning. Error bars depict s.e.m.

  6. Story phase results
    Supplementary Fig. 5: Story phase results

    Memory scores (y-axis) per phase for group A. Emotional narratives accompanied both slide-stories learnt by patients. Each story can be separated into three phases (x-axis) of which the middle is considered most emotional and results in enhanced memory when compared to the same images accompanied by a neutral narrative. Testing for a reactivation (reactivated story (solid bars), non-reactivated story (open-bars)) x phase (1,2,3) effect within group A revealed a main effect of reactivation (F1, 12 = 8.75 P = 0.012), but no main effect of phase (F2, 24 = 0.48, P = 0.624), or reactivation x phase interaction (F2, 24 = 2.40, P = 0.112). Thus we observe no interaction between relative emotionality of the studies material and the disturbance of reactivated memory. Error bars depict s.e.m.

  7. Memory performance is associated with illness and ECT parameters
    Supplementary Fig. 6: Memory performance is associated with illness and ECT parameters

    We assessed whether screening scores or elements of ECT treatment were related to memory performance. Top: Cumulative illness rating scale scores (CIRS, y-axis) correlated with memory performance (x-axis) over all groups (Pearson r = –0.32, N = 38, P = 0.047), thus the lower the comorbid physical problems or illnesses the better memory performance. Middle: Limiting the analyses to group A and group B, we tested whether ECT treatment parameters were related to memory performance. An independent t-test showed that memory impairment was greater for bifrontotemporal electrode placement compared to right unilateral stimulation (t(24) = 2.28, P = 0.032, right unilateral mean: 36.27, s.e.m: 2.45; bifrontotemporal mean: 28.91, s.e.m.: 1.96). Bottom: Given that bilateral stimulation leads to more memory impairment, we specifically tested for a modulation of reconsolidation by electrode placement in group A. No effect was observed (P > 0.05), but this null finding may reflect the size of our sample (unilateral N=4, bilateral N=9). The reconsolidation impairment observed in group A was still evident when controlling for electrode placement. The relation between electrode placement and memory performance suggests that the observed memory impairments are a result of the electrical stimulation and/or the convulsion itself and not other elements of the ECT treatment such as the anaesthesia. Error bars depict s.e.m.

References

  1. Misanin, J.R., Miller, R.R. & Lewis, D.J. Science 160, 554555 (1968).
  2. Przybyslawski, J. & Sara, S.J. Behav. Brain Res. 84, 241246 (1997).
  3. Nader, K., Schafe, G.E. & LeDoux, J.E. Nature 406, 722726 (2000).
  4. Nadel, L. & Land, C. Nat. Rev. Neurosci. 1, 209212 (2000).
  5. Dudai, Y. & Eisenberg, M. Neuron 44, 93100 (2004).
  6. Kroes, M.C.W., Strange, B.A. & Dolan, R.J. J. Neurosci. 30, 39593963 (2010).
  7. Kroes, M.C.W. & Fernández, G. Neurosci. Biobehav. Rev. 36, 16461666 (2012).
  8. Schiller, D. & Phelps, E.A. Front. Behav. Neurosci. 5, 24 (2011).
  9. Strange, B.A., Kroes, M.C., Fan, J. & Dolan, R.J. Front. Behav. Neurosci 4, 175 (2010).
  10. Squire, L.R. Am. J. Psychiatry 134, 9971001 (1977).
  11. Karpicke, J.D. & Roediger, H.L. III. Science 319, 966968 (2008).
  12. Hupbach, A., Hardt, O., Gomez, R. & Nadel, L. Learn. Mem. 15, 574579 (2008).
  13. Inda, M.C., Muravieva, E.V. & Alberini, C.M. J. Neurosci. 31, 16351643 (2011).
  14. Raio, C.M., Carmel, D., Carrasco, M. & Phelps, E.A. Curr. Biol. 22, R477R479 (2012).
  15. Nader, K. & Hardt, O. Nat. Rev. Neurosci. 10, 224234 (2009).
  16. Sara, S.J. Front. Behav. Neurosci. 4, 185 (2010).
  17. Squire, L.R., Slater, P.C. & Chace, P.M. Behav. Biol. 18, 335343 (1976).
  18. Rubin, R., Fried, R. & Franks, C. in Advances in Behavior Therapy, 1968 (eds. R. Rubin & C. Franks) 3744 (Academic Press, New York, 1969).
  19. Gamache, K., Pitman, R.K. & Nader, K. Neuropsychopharmacology 37, 27892796 (2012).
  20. Milton, A.L. & Everitt, B.J. Eur. J. Neurosci. 31, 23082319 (2010).
  21. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (Dsm-Iv-Tr) (American Psychiatric Association, Washington, DC, 2000).
  22. Hupbach, A., Gomez, R., Hardt, O. & Nadel, L. Learn. Mem. 14, 4753 (2007).
  23. Lisanby, S.H., Maddox, J.H., Prudic, J., Devanand, D.P. & Sackeim, H.A. Arch. Gen. Psychiatry 57, 581590 (2000).
  24. Squire, S.R. & Slater, P.C. Am. J. Psychiatry 135, 13161320 (1978).
  25. Linn, B.S., Linn, M.W. & Gurel, L. J. Am. Geriatr. Soc. 16, 622626 (1968).
  26. Montgomery, S.A. & Asberg, M. Br. J. Psychiatry 134, 382389 (1979).
  27. Cahill, L., Prins, B., Weber, M. & McGaugh, J.L. Nature 371, 702704 (1994).
  28. van Stegeren, A.H., Everaerd, W., Cahill, L., McGaugh, J.L. & Gooren, L.J.G. Psychopharmacology (Berl.) 138, 305310 (1998).
  29. Wechsler, D. Wechsler Adult Intelligence Scale-Iii (Psychological Corporation, San Antonio, 1997).
  30. Cahill, L. & McGaugh, J.L. Conscious. Cogn. 4, 410421 (1995).

Download references

Author information

Affiliations

  1. Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, The Netherlands.

    • Marijn C W Kroes,
    • Indira Tendolkar,
    • Guido A van Wingen &
    • Guillén Fernández
  2. Department of Psychiatry, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.

    • Indira Tendolkar
  3. Department of Psychiatry and Psychotherapy, University Hospital Essen, Essen, Germany.

    • Indira Tendolkar
  4. Department of Psychiatry, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands.

    • Guido A van Wingen
  5. Department of Psychiatry, Rijnstate Hospital, Arnhem, The Netherlands.

    • Jeroen A van Waarde
  6. Laboratory for Clinical Neuroscience, Centre for Biomedical Technology, UPM, Madrid, Spain.

    • Bryan A Strange
  7. Department of Cognitive Neuroscience, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.

    • Guillén Fernández

Contributions

All of the authors contributed to the design of the study. M.C.W.K. acquired and analyzed the data. M.C.W.K., B.A.S. and G.F. wrote the manuscript. I.T., G.A.v.W. and J.A.v.W. reviewed and approved the manuscript.

Competing financial interests

I.T. has received speaker fees from SERVIER Nederland Pharma B.V.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Study design. (33 KB)

    Patients were assigned to one of three groups (A, B, C). During a first study session all groups were shown two emotional slide-show stories. During a second session memory for one of the two stories was reactivated. Immediately after memory reactivation patient in groups A and B received ECT. In patients of group B memory was tested immediately upon recovery from ECT (Test, blue). In patients of groups A and C memory was tested one day after reactivation (Test, red and orange respectively).

  2. Supplementary Figure 2: Stimulus material (569 KB)

    Patients were presented with two slide shows that form arousing episodic stories of negative valence. Top: the original “Cahill Story”, bottom: the newly developed story. Both stories consist of 11 slides and each slide is accompanied by an auditory narrative.

  3. Supplementary Figure 3: Memory reactivation scores (25 KB)

    All groups showed evidence of memory reactivation, i.e., memory performance at reactivation was above chance level as indexed by the memory reactivation score (one-sample t-test across all groups, (t(36) = 7.53, P < 0.001), and groups did not differ in memory reactivation scores (y-axis; Kruskal-Wallis for group (A,B,C), H(2) = 1.77, P = 0.412; group A (red) mean: 3.08, s.e.m.: 0.35; group B (blue): mean: 2.62, s.e.m.: 0.96; group C (orange) mean: 3.23; s.e.m.: 1.24). Therefore, the observed between-group differences in reactivated memories are not due to differences in strength of memory reactivation, and adequate memory reactivation principally allows the initiation of a reconsolidation process. Dashed line indicates chance level (25%), error bars depict s.e.m.

  4. Supplementary Figure 4: DSST results (30 KB)

    DSST scores in minutes (y-axis). A group (A, B, C) x time point (study, test) repeated measures ANOVA on DSST scores revealed no main effect of time point (F1, 33 = 1.72, P = 0.199), group (F2, 33 = 0.69, P = 0.509) or group x time point interaction (F2, 33 = 1.64, P = 0.209). Thus, General cognitive functioning does not differ between groups. Hence, group differences in memory performance are unlikely to be due to group differences in general cognitive functioning. Error bars depict s.e.m.

  5. Supplementary Figure 5: Story phase results (33 KB)

    Memory scores (y-axis) per phase for group A. Emotional narratives accompanied both slide-stories learnt by patients. Each story can be separated into three phases (x-axis) of which the middle is considered most emotional and results in enhanced memory when compared to the same images accompanied by a neutral narrative. Testing for a reactivation (reactivated story (solid bars), non-reactivated story (open-bars)) x phase (1,2,3) effect within group A revealed a main effect of reactivation (F1, 12 = 8.75 P = 0.012), but no main effect of phase (F2, 24 = 0.48, P = 0.624), or reactivation x phase interaction (F2, 24 = 2.40, P = 0.112). Thus we observe no interaction between relative emotionality of the studies material and the disturbance of reactivated memory. Error bars depict s.e.m.

  6. Supplementary Figure 6: Memory performance is associated with illness and ECT parameters (67 KB)

    We assessed whether screening scores or elements of ECT treatment were related to memory performance. Top: Cumulative illness rating scale scores (CIRS, y-axis) correlated with memory performance (x-axis) over all groups (Pearson r = –0.32, N = 38, P = 0.047), thus the lower the comorbid physical problems or illnesses the better memory performance. Middle: Limiting the analyses to group A and group B, we tested whether ECT treatment parameters were related to memory performance. An independent t-test showed that memory impairment was greater for bifrontotemporal electrode placement compared to right unilateral stimulation (t(24) = 2.28, P = 0.032, right unilateral mean: 36.27, s.e.m: 2.45; bifrontotemporal mean: 28.91, s.e.m.: 1.96). Bottom: Given that bilateral stimulation leads to more memory impairment, we specifically tested for a modulation of reconsolidation by electrode placement in group A. No effect was observed (P > 0.05), but this null finding may reflect the size of our sample (unilateral N=4, bilateral N=9). The reconsolidation impairment observed in group A was still evident when controlling for electrode placement. The relation between electrode placement and memory performance suggests that the observed memory impairments are a result of the electrical stimulation and/or the convulsion itself and not other elements of the ECT treatment such as the anaesthesia. Error bars depict s.e.m.

PDF files

  1. Supplementary Text and Figures (2,034 KB)

    Supplementary Figures 1–6, Supplementary Tables 1 and 2

Additional data