# Baseline task performance predicts mini-mental state examination improvement after individuals with dementia practice an N-back task with bilateral transcranial direct current stimulation

**Authors:** Carlos Roncero, Alex Popov, Bruna Seixas-Lima, Durjoy Lahiri, Fatemeh Abnavi, Howard Chertkow

PMC · DOI: 10.1371/journal.pone.0332372 · PLOS One · 2025-10-29

## TL;DR

This study finds that people with dementia who perform worse at baseline show the most improvement in cognitive tasks after tDCS treatment.

## Contribution

The study demonstrates that baseline performance predicts tDCS effectiveness in dementia patients, particularly with bilateral stimulation.

## Key findings

- Participants with slower baseline response times improved more on N-Back and MMSE after tDCS.
- Bilateral tDCS was more effective for MMSE improvement compared to unilateral stimulation.
- Baseline performance is a key factor in determining tDCS effectiveness for cognitive improvement.

## Abstract

This research explores the relationship between baseline performance in dementia patients and its impact on the effectiveness of transcranial direct current stimulation (tDCS) for improving executive function. Past research suggests that those participants who are more impaired at baseline will be those who then demonstrate the greatest improvements with tDCS. In addition, we examined if bilateral or unilateral stimulation would be more effective for improving the general level of cognition of these participants.

The study involved 24 participants, with 18 diagnosed with Alzheimer’s disease (AD) and 6 with Frontotemporal Dementia (FTD). These participants engaged in three rounds of N-back task training sessions, each separated by two months. The sessions included three types of stimulation: 4mA unilateral stimulation targeting the left Dorsolateral Prefrontal Cortex (DLPFC), 4mA bilateral stimulation targeting both the left and right DPFC, and sham stimulation. Additionally, a control group of elderly individuals without dementia was used to establish a benchmark for the normal average response time (RT) of the administered N-Back task. Evaluations, comprising of the N-Back task and the Mini-Mental State Examination (MMSE), were conducted before stimulation, at the final session, and two weeks post-stimulation in all rounds.

Participants were retrospectively allocated into those whose baseline N-Back response times (RTs) were found on average to be within the normal range versus those who performed more slowly than the normal range. For the group that performed within the normal range, no significant improvement was observed on the N-Back task after a series of sessions involving active tDCS. In contrast, the participants with slower than normal RTs at baseline demonstrated improvement both on the N-Back task (faster times) and the MMSE (higher scores) after receiving active tDCS, although accompanying improvement on the MMSE task was restricted to the condition where participants received bilateral tDCS.

This study replicates a pervasive finding in the tDCS literature: participants who perform worse at baseline often demonstrate the greatest improvements with tDCS. Furthermore, these results have clinical implications, in that tDCS may be most beneficial in individuals when clear deficits or impairments are present. Attempts to treat less severe participants might fail to show improvement with tDCS. Ultimately, baseline condition should be routinely considered when evaluating tDCS results.

## Linked entities

- **Diseases:** dementia (MONDO:0001627), Alzheimer’s disease (MONDO:0004975), Frontotemporal Dementia (MONDO:0010857)

## Full-text entities

- **Diseases:** FTD (MESH:D057180), AD (MESH:D000544), dementia (MESH:D003704)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12571271/full.md

## Figures

5 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12571271/full.md

## References

82 references — full list in the complete paper: https://tomesphere.com/paper/PMC12571271/full.md

---
Source: https://tomesphere.com/paper/PMC12571271