Language decline in subjective cognitive impairment: Early detection of dementia

Subjective cognitive decline (SCD) is a form of self-reported cognitive
impairment and one of the early predictors of Alzheimer*s disease, preceding
objective mild cognitive impairment (MCI) (Jessen et al., 2020). In recent
years, SCD has been put into focus due to its importance for early detection of
Alzheimer*s disease (AD), as well as the current neuropsychological tools* lack
of sensitivity to SCD. While it is often possible to detect early AD using
biomarkers, lumbar puncture, PET, and MRI (Jack et al., 2010), these tests are
considered to be highly invasive, extremely expensive, and not widely available
(Johnson et al., 2012). This is why it is important to find and design more
affordable, accessible non-invasive methods for early detection of dementia.
Subjective complaints do not always correlate with objective cognitive
performance (Jungwirth et al., 2004). People with SCD seem to perform similarly
on the n-back task as those without subjective complaints, but they do show
distinct hemodynamic response (Zhang et al., 2021). Neuroimaging studies have
shown some promise in documenting potential markers of SCD (Parker et al.,
2022). For instance, Kawagoe et al (2019) reported a positive correlation
between complaint severity and parieto-occipital connectivity. However, more
research is needed to confirm this evidence. Additionally, Marandi and Gazerani
(2019) argued that eye tracking is a promising tool to provide objective
biomarkers of aging and neurodegeneration. In particular, pupil dilation has
been shown to be related to prefrontal activation during a working memory task
(Yeung et al., 2021).
Language deficits have been reported for AD and aMCI. They include impaired
lexical access, impaired semantic fluency (Belleville et al., 2017) and
sentence level complexity (López-Higes et al., 2014; Sung et al., 2020),
reduced discourse cohesion, *empty speech*, and paraphasias. However, when it
comes to SCD, studies focusing only on language impairments are not numerous
and they focus mostly on verbal fluency. Authors have used lexical access tasks
to compare participants with SCD to healthy aging elderly and participants with
MCI and AD (Benito-Leon et al., 2010; Ostberg et al., 2005).

Our project aims to investigate cognitive and linguistic deficits in people
with SCD, focusing on working memory impairments, deficits of lexical access,
and deficits brought on by syntactic complexity. We will integrate clinical
linguistic findings with brain activation measured by fNIRS and EEG, and pupil
dilation measured by eye-tracking. This will be a longitudinal study with a
1-year follow-up. In this way, we plan to create a neurolinguistic profile of
SCD using neuropsychological and neuroimaging data.

In this study, we hypothesize that:
1. Older adults with SCD will perform better on n-back task as compared to
those with MCI, but similarly to healthy controls (HC).
2. Older adults with SCD will show greater connectivity of the frontoparietal
network (FPN) connectivity than HC and people with MCI, indicating a
compensatory cognitive effort (Si et al., 2020).
3. Similarly, older adults with SCD will also show greater pupil dilatation
changes than HC and people with MCI.
4. Action naming will cause bigger lexical access difficulties in individuals
with SCD and MCI than HC, due to greater reliance on executive functions and
attentional control (Macoir et al., 2019). Prediction: HC < SCD < MCI.
5. Linguistically more challenging structures (passives and relative clauses)
will be more sensitive to cognitive decline. We predict that the SCD group will
demonstrate more difficulties with complex sentence processing than HC, but
less difficulties than the MCI group.

This will be a longitudinal observational study of brain activation while doing
cognitive and linguistic tasks using fNIRS and EEG measurements, as well as
pupillometry. Cognitive screening and language assessments will be separated
and administered in two days to the three groups participating in the study.
Groups needed for this study are: healthy elderly (healthy controls), elderly
with subjective cognitive decline (SCD), and elderly with mild cognitive
impairment (MCI). The same procedure will be used in the follow-up, a year
later.
On day 1, the participants will be given all the information on the study and
cognitive screening will be administered. This usually takes around 60 minutes.
The cognitive screening will include:
o Mini-Mental State Examination (MMSE)
o 15-Word Test (15WT)
o Trail-Making Test (TMT)
o Stroop Color-Word Test (SCWT)
o WAIS Digit Span
o WAIS Symbol Digit Modalities Test (SDMT)
As well as:
o Geriatric Depression Scale-30 (GDS-30)
o Apathy Evaluation Scale (AES)

On day 2 cognitive and linguistic tasks will be administered. Since a more
sensitive way to measure language processing is required, functional
near-infrared spectroscopy (fNIRS) and electroencephalography (EEG), which
allow for real-time measurement of neural activity, and eye-tracking
(pupillometry) will be used to support the study and capture subtle changes.
Before the four tasks are administered, the spontaneous speech test and the
resting state EEG will be done. Resting state EEG includes three rounds of *1-
minute eyes closed/1-minute eyes open* protocol, lasting 6 minutes in total.
Tasks
In the current study, one cognitive and four linguistic tasks will be
administered:
(I) Spontaneous speech
(II) N-back task
(III) Object and action naming,
(IV) Elicited imitation of complex sentences,
(V) Comprehension of reversible and irreversible active and passive sentences.

The study will consist of two sessions: the first for explaining the study to
the participant and neuropsychological testing - screening whether all
inclusion criteria are met (and no exclusion criteria). This will usually take
no longer than 60min. The second session concerns the spontaneous speech test
(5 minutes), resting state and cognitive and linguistic testing using fNIRS,
EEG, and eye-tracking measurements, with a total duration of 90 min. After a
year, Session 1 and Session 2 will be repeated (YFUs1 and YFUs2).

The experiment will not involve more than minimal risks for the participants.
NIRS, EEG, and pupillometry are standard non-invasive brain activation
measuring techniques with no known negative effects on health. The study is not
intended to benefit the participants directly. Participants will receive
compensation/a present for their contribution.