Stimulation of the left dorsolateral prefrontal cortex with slow rTMS enhances verbal memory formation

Abstract

Encoding of episodic memories relies on stimulus-specific information processing and involves the left prefrontal cortex. We here present an incidental finding from a simultaneous EEG-TMS experiment as well as a replication of this unexpected effect. Our results reveal that stimulating the left dorsolateral prefrontal cortex (DLPFC) with slow repetitive transcranial magnetic stimulation (rTMS) leads to enhanced word memory performance. A total of 40 healthy human participants engaged in a list learning paradigm. Half of the participants (N = 20) received 1 Hz rTMS to the left DLPFC, while the other half (N = 20) received 1 Hz rTMS to the vertex and served as a control group. Participants receiving left DLPFC stimulation demonstrated enhanced memory performance compared to the control group. This effect was replicated in a within-subjects experiment where 24 participants received 1 Hz rTMS to the left DLPFC and vertex. In this second experiment, DLPFC stimulation also induced better memory performance compared to vertex stimulation. In addition to these behavioural effects, we found that 1 Hz rTMS to DLPFC induced stronger beta power modulation in posterior areas, a state that is known to be beneficial for memory encoding. Further analysis indicated that beta modulations did not have an oscillatory origin. Instead, the observed beta modulations were a result of a spectral tilt, suggesting inhibition of these parietal regions. These results show that applying 1 Hz rTMS to DLPFC, an area involved in episodic memory formation, improves memory performance via modulating neural activity in parietal regions.

Fig 1. Experimental design.

Arrows on brain model indicate stimulation site (DLPFC = purple, vertex = orange). Participants were asked to study 2 lists of 10 words over 12 runs. During encoding of List 2, 45 pulses of 1 Hz rTMS were applied to the left DLPFC (MNI coordinates: −45, 6, 39) or vertex. Memory performance was assessed as percentage of correctly recalled words per list. The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex; rTMS, repetitive transcranial magnetic stimulation; TMS, transcranial magnetic stimulation.

Fig 2. Behavioural memory performance in experiment 1.

(A) Serial position curve for List 1 words. Error bars depict standard errors of the mean. (B) Raincloud plots of average memory performance for List 1 words across all blocks with paired boxplots [16]. Coloured area within the box plots indicate the standard error, while the circles depict individual data points. (C) Serial position curve for List 2 words. Error bars depict standard errors of the mean. (D) Memory performance for List 2 words. (E) Difference in average memory performance between the DLPFC and vertex condition for each list (List 2 = Stimulation). The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex; n.s., not significant; TMS, transcranial magnetic stimulation.

Fig 3. EEG results (only later remembered trials analysed).

(A) Time frequency plot for the difference between DLPFC and vertex during List 2 encoding averaged over electrode cluster demonstrating a significant negative difference (i.e., less power for DLPFC compared to vertex) between the DLPFC and vertex group in the beta frequency range poststimulus. Dashed line indicates word onset. (B) Topographies depicting beta power (13 to 30 Hz) difference between DLPFC and vertex stimulation in time windows of interest (pre: −0.5 s to −0.05 s; post = 0 to 1 s). White circles depict significant negative electrode cluster poststimulus. Black circles show electrodes within the negative cluster showing a positive difference prestimulus. (C) Time course of beta power (13 to 30 Hz) averaged over the negative electrode cluster shown in B. Shaded area represents standard error of the mean. Black dashed line indicates word onset. Grey dashed lines depict time bins. (D) Beta power difference (List 2 − List 1) over significant negative electrode cluster split by rTMS. Error bars show standard error of the mean. Data were split into 6 nonoverlapping time bins: [−1 s to −0.55 s]; [−0.5 s to −0.05 s]; [0 s to 0.45 s]; [0.5 s to 0.95 s]; [1 s to 1.45 s]; [1.5 s to 1.95 s]. The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex; EEG, electroencephalography; rTMS, repetitive transcranial magnetic stimulation.

Fig 4. FOOOF analysis shows that beta effects are nonoscillatory in nature.

(A) Schematic representation of the different components in a given power spectrum. The black line represents a typical power spectrum that is to be separated. The blue line is the corresponding log function following removal of the periodic peaks, thereby representing aperiodic properties of the signal. (B) Power spectra separated by each condition. Shaded area indicates standard error. (C, D) Line plots of the mean aperiodic component before and after item presentation for the DLPFC and vertex condition, respectively. The right axis relates to the plotted post–pre difference (dotted line). The x-axis has been extended for illustrative purposes to highlight the differences in slopes between the difference conditions. The actual fit was performed on data in the 1–40 Hz range. The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex.

Fig 5. Behavioural memory performance in experiment 2.

(A) Serial position curve for List 1 (N = 23). (B) Raincloud plots of memory performance for List 1 words (difference between DLPFC and vertex stimulation). Coloured area within the box plots indicate the standard error, while the circles depict individual data points. (C) Serial position curve for List 2. (D) Raincloud plots of memory performance for List 2 words (difference between DLPFC and vertex stimulation). The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex; TMS, transcranial magnetic stimulation.

Fig 6. Meta-analysis of the behavioural results of the first and second experiment.

Forest plots of the meta-analytically combined DLPFC effect for List 1 (A) and List 2 (B) words. Error bars represent 95% confidence intervals; effect sizes were calculated as Hedge’s g. The data and scripts used to generate this figure can be found at https://osf.io/dyxjv/. DLPFC, dorsolateral prefrontal cortex; FE, fixed effects.