No Effects of Non-invasive Brain Stimulation on Multiple Sessions of Object-Location-Memory Training in Healthy Older Adults

Abstract

Object-location memory (OLM) is known to decline with normal aging, a process accelerated in pathological conditions like mild cognitive impairment (MCI). In order to maintain cognitive health and to delay the transition from healthy to pathological conditions, novel strategies are being explored. Tentative evidence suggests that combining cognitive training and anodal transcranial direct current stimulation (atDCS), both reported to induce small and often inconsistent behavioral improvements, could generate larger or more consistent improvements or both, compared to each intervention alone. Here, we explored the combined efficacy of these techniques on OLM. In a subject-blind sham-controlled cross-over design 32 healthy older adults underwent a 3-day visuospatial training paired with either anodal (20 min) or sham (30 s) atDCS (1 mA, temporoparietal). Subjects were asked to learn the correct object-location pairings on a street map, shown over five learning blocks on each training day. Acquisition performance was assessed by accuracy on a given learning block in terms of percentage of correct responses. Training success (performance on last training day) and delayed memory after 1-month were analyzed by mixed model analysis and were controlled for gender, age, education, sequence of stimulation and baseline performance. Exploratory analysis of atDCS effects on within-session (online) and between-session (offline) memory performance were conducted. Moreover, transfer effects on similar trained (visuospatial) and less similar (visuo-constructive, verbal) untrained memory tasks were explored, both immediately after training, and on follow-up. We found that atDCS paired with OLM-training did not enhance success in training or performance in 1-month delayed memory or transfer tasks. In sum, this study did not support the notion that the combined atDCS-training approach improves immediate or delayed OLM in older adults. However, specifics of the experimental design, and a non-optimal timing of atDCS between sessions might have masked beneficial effects and should be more systematically addressed in future studies.

Keywords: aging; associative learning; cognitive training; consolidation; episodic memory; transcranial direct current stimulation; transfer effects; visuospatial memory.

Figure 1

Study overview. (A) Schematic of the associative object-location learning paradigm (LOCATO). During acquisition, each trial comprised a picture of a schematized street map with one building. Buildings (objects) occurred on one correct and 10 incorrect positions (locations). Subjects had to learn correct object-location-pairings over the course of multiple learning blocks and indicate (button press) in each trial, whether a building was in a “correct” location (Yes or NO). Memory for object-location-associations were assessed by two different cued recall tests [item recognition (IR), 3-alternative-forced choice test (3-AFC)]. During IR correct object-location-associations were intermixed with new (incorrect) object-location associations and subjects indicated by button press if presented position is “correct.” In 3-AFC tests subjects specified by button press (“1,” “2,” or “3”) the correct position of the building shown above the schematic street map. (B) Timeline Training. Each of the two study blocks (cross-over design) comprised 6 sessions (2–7; first session (not shown) included baseline testing) with 3 months in-between. Training (session 3–5) consisted of three consecutive days, each comprised five learning blocks and subsequent cued recall test (IR, 3-AFC). Memory was also tested follow-up on session 6 (1-day) and 7 (1-month) post-training. Overnight retention (offline effects) was assessed by applying memory tests (IR, 3-AFC) before next training (session 4 and 5). In addition, at the beginning of each training day subjects self-rated their affective state (“Befindlichkeitsskalierung anhand von Kategorien und Eigenschaftswörtern”; BSKE, Janke et al., 2002), and provided information about their sleep (number of slept hours, sleep quality) of previous night. In pre- and post-training sessions (2,6,7) we asked for positive and negative affective state (PANAS, Watson et al., 1988) and applied additional memory tests to assess transfer effects in trained (LOCATO-15; shorter and less complex version of training paradigm, Kulzow et al., 2014), and untrained (Rey–Osterrieth Complex Figure (ROCF) Test, Knight and Kaplan, ; Rey Auditory Verbal Learning Test (AVLT), Helmstaedter et al., 2001) memory tasks. (C) Stimulation protocol. Anode (7 × 5 cm²) was attached to T6 (according to EEG 10–20 System) and return electrode (cathode: 10 × 10 cm²) contralateral above the eyebrow (supraorbital). Connector of the anode was positioned at the posterior edge distant from the return electrode. Larger size of the cathode renders the stimulation density functionally ineffective. Anodal transcranial direct current stimulation (atDCS) of 1 mA was administered during beginning of OLM-training (Session 3–5) for 20 min (“atDCS”) or 30 s (“sham”) and current was ramped up and down within 10 s. Abbreviations: IR, item recognition; 3-AFC, 3 alternative forced choice task; PANAS, Positive and Negative Affect Schedule; LOCATO-15, short version of object-location-memory task; ROCF, Rey–Osterrieth Complex Figure Test; AVLT, German version of the Rey Auditory Verbal Learning Test, d-day; mo, month; FU, Follow-up.

Figure 2

Performance during and after object-location-memory training. Response accuracy (% correct) during each learning (L) block and overnight cued recall performance (% correct) in 3-alternative forced choice (R_3AFC) and item recognition (R_IR) task assessed before next training on day 2 and 3 as well as cued recall at 1-day and 1-month follow-up is depicted. Dark filled circles (black, dark gray) represent performance of atDCS applied during training (“atDCS”), light filled circles represent performance of sham applied during training (“sham”). Behavioral online effects related to within-session performance and offline effects to between-session performance. Data are given as means and standard deviations. D, day; mo, month.