Neural region features of rapid serial visual presentation(RSVP)for target detection
10.7644/j.issn.1674-9960.2024.10.004
- VernacularTitle:快速序列视觉呈现目标检测的脑区特征解析
- Author:
Qian ZHOU
1
;
Baozeng WANG
;
Zijian YUAN
;
Yang YANG
;
Siwei LI
;
Jin ZHOU
;
Changyong WANG
Author Information
1. 军事科学院军事医学研究院,北京 100850
- Keywords:
rapid serial visual presentation;
concealment conditions;
target detection;
event-related potential;
brain computer interface
- From:
Military Medical Sciences
2024;48(10):744-752
- CountryChina
- Language:Chinese
-
Abstract:
Objective To study the differences in features of event-related potentials(ERPs)and target detection accuracy between five brain regions(frontal,temporal,central,parietal,and occipital)in target detection tasks based on rapid serial visual presentation(RSVP)brain computer interface(BCI)under six target concealment conditions.Methods Twelve participants were selected for the study,whose scalp electroencephalogram(EEG)signals were collected under the six concealment conditions using a NeuroScan SynAmps2 EEG acquisition system.The ERP waveforms,P300 amplitudes and latencies,among other things,were compared across the five brain regions.The hierarchical discriminant component analysis(HDCA)algorithm was used to classify the EEG signals while the differences in classification accuracy were probed across the five brain regions.Results(1)Under the six concealment conditions,target images elicited distinct ERP waveforms in all the five brain regions;(2)For P300 amplitudes,the temporal region exhibited the smallest values;(3)Regarding P300 latencies,the parietal and central regions showed longer durations than other brain regions(except for small camouflage and small occlusion conditions);(4)In terms of classification accuracy,the parietal and central regions outperformed other brain regions(except for the large camouflage condition).Conclusion The selection of parietal and central channels can offer a new perspective for enhancing the performance in concealed target detection based on RSVP-BCI,and is expected to spark new ideas for the design of miniaturized,simple and wearable BCI devices.
