The investigators will conduct a randomized, double-blinded, sham-controlled trial of approximately 60 patients with minor stroke and post-stroke mild cognitive impairment (psMCI). Participants will be individually randomized on enrollment using a random number generator to treatment with anodal tDCS + computerized cognitive treatment (CCT) versus sham + CCT (approximately 30 patients in each arm). Clinical evaluation including assessment of cognition will be performed pre- and post-intervention by individuals on the study team blinded to the participant's intervention. Participants will also undergo functional neuroimaging with magnetoencephalography (MEG) pre- and post-intervention (1, 3, and 6 months post-stroke to evaluate for initial and longer-term effects of treatment on cerebral activation patterns and functional connectivity). Neuroimaging and clinical outcomes will be assessed to determine the effect of tDCS versus sham + CCT on psMCI.

Vascular cognitive impairment, ranging from vascular mild cognitive impairment to vascular dementia, is a leading cause of progressive cognitive dysfunction second only to Alzheimer Disease. While the accumulation of ischemic infarcts or a large cortical stroke can result in permanent cognitive dysfunction, a single small stroke can also result in disabling impairment. The investigators have shown that small lesions, regardless of the location, result in acute cognitive decline. Similar to those with progressive vascular cognitive impairment, post-stroke MCI (psMCI) patients display slowed reaction times and dysfunction across multiple cognitive domains. Many significantly recover over the first 6 months. However, the heterogenous recovery and uncertainty regarding prognosis can lead to major life changes, such as early retirement or selling of homes, that substantially impact quality of life.

Magnetoencephalography (MEG) recordings in those with psMCI show temporal dispersion consistent with generalized disruption of cognitive networks during resting state, irrespective of lesion location. Evaluation of frequency spectra show bilaterally decreased beta power in the frontoparietal lobes correlating with impaired reaction times. Functional connectivity analyses at 6 months demonstrate increased inter-hemispheric connections that may explain or reflect a patient's improvement. It is currently unknown whether specific cognitive networks are involved, though based on the pattern of clinical deficits it would be reasonable to hypothesize that the frontoparietal network, responsible for executive function and higher level cognitive tasks, is disrupted to a greater extent than the limbic, responsible for emotion and memory. Neural modulation with transcranial direct current stimulation (tDCS) increases the likelihood of neural firing, strengthening connectivity by promoting long-term potentiation and facilitating task performance. As anodal tDCS only induces the firing of neurons near threshold, it follows that it is most effective and longest acting when paired with a task that engages focal activation. This is best seen in patients with aphasia undergoing speech-language therapy.

In this study, the investigators will determine the utility of A-tDCS in conjunction with computerized cognitive therapy (CCT) to treat psMCI. The investigators will recruit approximately 60 patients with subacute minor stroke and randomize the patients to A-tDCS administered over the ipsilesional frontoparietal cortex versus sham plus 15 sessions of cognitive therapy using a widely used online platform focused most on executive functioning and processing speed based on the investigators' preliminary work targeting deficits most specific to those with psMCI. The investigators hypothesize that tDCS will augment both generalized connectivity as well as the connectivity of specific networks targeted during training (most notably the frontoparietal) and that patients will show increased clinical improvement acutely after therapy that will last for months after treatment.

Along with clinical evaluation, the investigators will use MEG to evaluate cerebral activation patterns and connectivity pre- (1 month post-infarct) and post- (3 and 6 months post-infarct) intervention. The investigators will collect longitudinal MEG and clinical data through a multicenter collaboration of experts in the fields of stroke, dementia, and functional neuroimaging.

Inclusion Criteria:

1. Adults (≥18 years) presenting with neurological symptoms due to acute ischemic stroke (symptom onset within the week prior to admission).
2. Evidence on brain MRI of acute ischemic stroke (imaging negative strokes and TIAs will be excluded).
3. Native English speaker (by self-report) prior to stroke.
4. NIHSS <8 at initial follow-up visit (approximately 30 days post-stroke).
5. mRS 0-2 at initial follow-up visit.

Exclusion Criteria:

1. Primary intracerebral hemorrhage- as evidenced by blood on head CT or MRI.
2. Presence of proximal large vessel occlusion.
3. Cortical exam findings including aphasia or neglect.
4. Prior report or history of dementia or undertreated psychiatric illness.
5. Uncorrected hearing or visual loss.
6. Inability to attend treatment or follow-up sessions.
7. Inability to travel to College Park (UMD) for MEG recording sessions.
8. Presence of any of the following that would lead to significant artifact on MEG: cardiac pacemaker, intracranial clips, metal implants or external clips within 10mm of the head, metal implants in the eyes (unlikely given that all patients will have an MRI and criteria are similar).
9. Claustrophobia, obesity, and/or any other reason leading to difficulty staying in the MEG machine for up to 1 hour.

Results in aggregate will be posted; however, de-identified individual participant data will be made available to other investigators and potential collaborators upon reasonable request of the PI.

One year following completion of data collection and publication of results.