“School of Cognitive%20Sciences”
Back to Papers HomeBack to Papers of School of Cognitive%20Sciences
| Paper IPM / Cognitive%20Sciences / 16484 |
|
||||||||||||
| Abstract: | |||||||||||||
|
Background: Electroencephalography (EEG) has been commonly used to measure the
brain alterations in the early stages of Alzheimerâs Disease. However, the reported
measures are limited to the univariate changes, including activation level and frequency bands. To look beyond the activation level, we used a task-based EEG and
applied multivariate pattern analysis (MVPA) to study the changes in the pattern of
information processing.
Method: We recruited 18 participants with mild cognitive impairment (MCI), and 22
age-matched healthy controls (HC). We acquired EEG data from all participants, while
they were doing the integrative cognitive assessment (ICA) test (Khaligh-Razavi et
al., 2019). ICA is a rapid visual categorization task, in which participants are presented with a sequence of natural images of animals and non-animals, and are asked
to respond as quickly and accurately, whether images contained an animal. To analyze
the EEG data, we took advantage of MVPA techniques to measure how accurately EEG
signals can discriminate between brain representations of animals and non-animals.
Result: Univariate analysis showed a significant decline in the activation level of the
right temporal lobe in MCI compared to HC. In the left parietal lobe, we found that
the pattern of EEG responses to the presented visual stimuli was significantly different
between HC and MCI, in the absence of any difference in their level of activation. Furthermore, we found that animacy information (animal vs. non-animal categorization)
emerges later in the brain of patients with MCI (t = 67 ms ± 33 SE of mean; p-value â¤
0.03) compared to HC.
Conclusion: We demonstrated that in addition to the level of activation (i.e., mean ERP
response), the pattern of EEG responses to visual stimuli also carries information about
the status of the disease. In particular, we see that in some of the brain areas where the
mean activation shows no difference between HC and MCI, patterns of EEG responses
are significantly different and can be used to discriminate MCI from HC. Furthermore,
the results also suggest a delay or impairment in the speed of processing animacy information in MCI patients.
Download TeX format |
|||||||||||||
| back to top | |||||||||||||
