Non-Invasive Brain Stimulation for Medication-Resistant Auditory Hallucinations in Schizophrenia Patients

25% of schizophrenia (SZ) patients report chronic auditory hallucinations (AH) despite being
on medication. Previous research proposed that hyperactivity of the left temporo-parietal
cortex (11) and hypoactivity in left dorsolateral prefrontal cortex (DLPFC) (12) are
associated with the onset of AH. While temporal cortex hyperactivity has been extensively
manipulated with neuromodulation interventions trying to reduce AH, results have not been
consistent (1,13). Recent transcranial direct current stimulation (tDCS) work that targeted
both regions above by applying excitatory stimulation over the left DLPFC and inhibitory
stimulation on the left temporo-parietal cortex reported a 30% reduction in AH in patients
with medication refractory AH that lasted up to 3 months after treatment (14). The
investigators are interested to further investigate whether the use of paired
neuromodulation (P-NM) with tDCS in co-targeting top-down executive control with bottom-up
sensory processes would result in a larger reduction in AH. To date, there are no studies
that have used P-NM to examine the magnitude of plasticity of executive control on sensory
processing in SZ patients. This type of P-NM would be relevant since AH are thought to be
the result of a lack of top-down control over perceptual distortions of internally generated
speech. In the forced-attention dichotic listening task (DL), two different syllables are
simultaneously presented to the left and the right ear. Subjects are required to pay
attention to and report either the left or right ear stimulus. Due to the preponderance of
the contralateral pathways, the right ear syllable is projected to the left temporal lobe
for processing, while the left ear stimulus is projected to the right temporal lobe and has
to be transferred across the corpus callosum to be processed in the left hemisphere. Healthy
controls show a right ear advantage (REA) while SZ patients do not (15). DL task performance
requires top-down control over auditory cortex and has been associated with increased DLPFC
activation in controls (16). Worse performance in this task has been associated with greater
severity of AH (17).

SPECIFIC AIMS: The overall aim is to examine the efficacy of tDCS intervention combined with
sensory gating training (paired neuro-modulation; P-NM) to reduce hallucinations (as
measured by Positive and Negative Syndrome Scale and the Auditory Hallucination Rating
Scale). The investigators will collect pilot data to explore the magnitude of DLPFC
plasticity during executive control on auditory processing using P-NM and its effect on
hallucination-related behavior. The investigators will combine tDCS to enhance left DLPFC
activity with auditory stimuli presented to the right ear. The investigators will use the DL
task to assess plasticity of top-down control over auditory sensory perception before and
after intervention. The investigators will correlate DL task performance to the magnitude of
left DLPFC plasticity and with AH severity.

The investigators hypothesize that (a) DLPFC enhancement will facilitate REA in SZ patients
by supporting inhibition of attention to irrelevant syllable and facilitating attention to
the right ear syllable; (b) continued pairing of left DLPFC activity and auditory processing
will reduce auditory hallucinations (as measured by Positive and Negative Syndrome Scale and
the Auditory Hallucination Rating Scale).

Methods: The investigators will recruit 50 SZ patients with persistent severe AH despite
medication as measured by the Positive and Negative Syndrome Scale and the Auditory
Hallucination Rating Scale (AHRS). To examine the effects of P-NM over time, patients will
receive transcranial direct current stimulation (tDCS) during five days, twice a day while
they practice the dichotic listening task (DL; Hugdahl 2013). During the DL task, patients
will be presented with consonant-vowel syllables via headphones. The syllables will consist
of paired presentations of the six stop-consonants /b/, /d/, /g/, /p/, /t/, and /k/ together
with the vowel /a/ to form dichotic consonant-vowel syllable pairs of the type /ba - ga/,/ta
- ka/ etc. The syllables were paired with each other for all possible combinations, thus
yielding 36 dichotic pairs, including the homonymic pairs. Patients will be given two
different instructions. In one instruction condition they will be told to focus attention to
and report from the right ear, and if think they hear something in the left ear, this should
be ignored ("forced-right" condition). In the other condition, patients will be asked to
focus attention to and report from the left ear, and if they think that they hear something
in the right ear, this should be ignored ("forced-left" condition).

Participants will receive either active tDCS treatment or sham stimulation while performing
the DL task. Half of them (n=25) will receive either (a) electrical stimulation (2mA) for 20
minutes to the left DLPFC to enhance top-down control and improve the tuning or gating of
the extraneous information or (b) while the other half will receive sham stimulation to
control for placebo effects. Active tDCS and sham tDCS will be performed with the same tDCS
equipment. The difference is that while active tDCS will be configured to reach constant 2mA
stimulation for 20 minutes, sham tDCS will be a very brief 2mA stimulation for 30 seconds.
During sham tDCS the subject believes he/she is being stimulated normally, but there should
not be any real effects. This same procedure will be repeated during five days, twice a day.
At the end of day 5 all patients will complete the Positive and Negative Syndrome Scale and
the AHRS to evaluate change in AH. Patients will be asked to complete these scales 3, 6, and
9 months after tDCS intervention to examine long-term effects.

Inclusion Criteria:

- Current DSM-V-defined diagnosis of SZ or Schizoaffective Disorder as assessed by the
Structured Clinical Interview for Axis I DSM-V Disorders (SCID-I/P)

- Ages 18-45

- Competent and willing to sign consent form

Exclusion Criteria:

- Any serious neurological or endocrine disorder or any medical condition or treatment
with neurological sequelae (i.e. stroke, tumor, loss of consciousness of more than 30
min, HIV)

- Diagnostic and Statistics Manual (DSM-V) criteria for mental retardation or axis I
psychiatric disorder, subjects may have a lifetime but not current diagnosis of
depression

- Primary current substance use disorder diagnosis on any substance except for caffeine
or nicotine - nicotine use will be recorded but will not be an exclusion criterion

- Medical condition which requires treatment with a medication with psychotropic
effects

- Significant risk of suicidal or homicidal behavior

- Documented loss of consciousness (LOC) for longer than 30 minutes or LOC with
neurological sequelae

- History of electro-convulsive therapy

- Pregnancy