An fMRI study on neural basis of occluded image interpretation under different mental workloads
Poster No:
947
Submission Type:
Abstract Submission
Authors:
Bao Li1, Li Tong1, Chi Zhang1, Panpan Chen1, Hui Gao1, Long Cao1
Institutions:
1PLA Strategic Support Force Information Engineering University, Zhengzhou, Henan Province
First Author:
Co-Author(s):
Panpan Chen
PLA Strategic Support Force Information Engineering University
Zhengzhou, Henan Province
PLA Strategic Support Force Information Engineering University
Zhengzhou, Henan Province
Introduction:
When images become occluded, the visual system receives reduced information input, which increases the difficulty of recognizing objects (Rajaei et al., 2019). Nevertheless, experts in visual imaging are able to accurately interpret images with limited visual information, and this skill plays an important role in daily life (Bain, Wareing and Henderson, 2017). This study aims to investigate the neural mechanisms underlying the interpretation of occluded image under different mental workloads using functional magnetic resonance imaging (fMRI) techniques.
Methods:
A total of 64 participants (32 females) were enlisted for this study, and their behavioral and fMRI data were collected during the image interpretation task. To explore the cognitive ability for interpreting images under varying mental workloads, we designed three levels of image occlusion tasks (10%, 70%, and 90% occlusion) to elicit different levels of workload (low, mid and high workload) (refer to Figure 1A) (Li et al., 2023). Each participant completed 2 runs, and each run consisting of 30 blocks, and each task in 10 randomly selected blocks (Figure 1B). The experiment entailed a simple binary decision from the participants, who were tasked with determining whether the viewed image depicted aircraft A or B (Figure 1C).
We conducted a first-level Generalized Linear Model (GLM) analysis to assess variations in brain activation among all participants under different levels of mental workload (Monti, 2011). Utilizing the behavior data, we categorized all participants into two groups based on their performance: a high-ability group and a low-ability group. Subsequently, a second-level GLM analysis was performed to explore the differences in brain activity between the two groups during task execution. A significance threshold for the statistical results was set at p<0.05, corrected by FDR (Benjamini and Hochberg, 1995).
We conducted a first-level Generalized Linear Model (GLM) analysis to assess variations in brain activation among all participants under different levels of mental workload (Monti, 2011). Utilizing the behavior data, we categorized all participants into two groups based on their performance: a high-ability group and a low-ability group. Subsequently, a second-level GLM analysis was performed to explore the differences in brain activity between the two groups during task execution. A significance threshold for the statistical results was set at p<0.05, corrected by FDR (Benjamini and Hochberg, 1995).
·Figure 1. The fMRI experimental paradigm. (A) Occlusion Image Set. (B) The whole process of MRI scans. (C) Task operations in each trail.
Results:
The first-level GLM analysis indicated that tasks with higher mental workloads were associated with increased activation in the dorsal anterior cingulate cortex (dACC) , inferior occipital gyrus (IOG), middle occipital gyrus (MOG) and occipital fusiform gyrus (OFG) (Figures 2A-B). Notably, dACC activation continued to strengthen as the workload level escalated from mid to high (Figure 2C).
The second-level GLM analysis revealed that the high-ability group exhibited higher activation in the dACC, supplementary motor area (SMA), middle frontal gyrus (MFG), superior occipital gyrus (SOG), MOG, inferior parietal lobe (IPL) and insula when performing the image recognition task (Figure 2D).
The second-level GLM analysis revealed that the high-ability group exhibited higher activation in the dACC, supplementary motor area (SMA), middle frontal gyrus (MFG), superior occipital gyrus (SOG), MOG, inferior parietal lobe (IPL) and insula when performing the image recognition task (Figure 2D).
·Figure 2. The GLM analysis results. (A-C) Brain activation maps in comparison between tasks with different occlusion levels. (D) Brain activation maps in comparison between the the two groups.
Conclusions:
Our study uncovered the critical involvement of the dorsal anterior cingulate cortex (dACC) and occipital gyrus in performing image interpretation tasks under high mental workload. Comparisons across different ability groups further demonstrated that increased activation in the dACC, occipital gyrus, and insula was associated with superior image interpretation abilities. In conclusion, these findings enhance our understanding of the neural mechanisms involved in the interpretation of occluded images under different mental workloads.
Higher Cognitive Functions:
Higher Cognitive Functions Other 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Keywords:
Cognition
FUNCTIONAL MRI
Vision
Other - image interpretation, changing workloads, ability prediction.
1|2Indicates the priority used for review
Provide references using author date format
Benjamini, Y. and Hochberg, Y. (1995) ‘Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing’, Journal of the Royal Statistical Society: Series B (Methodological), 57(1), pp. 289–300. Available at: https://doi.org/10.1111/j.2517-6161.1995.tb02031.x.
Li, B. et al. (2023) ‘Brain Functional Representation of Highly Occluded Object Recognition’, Brain Sciences, 13(10), p. 1387. Available at: https://doi.org/10.3390/brainsci13101387.
Monti, M.M. (2011) ‘Statistical Analysis of fMRI Time-Series: A Critical Review of the GLM Approach’, Frontiers in Human Neuroscience, 5, p. 28. Available at: https://doi.org/10.3389/fnhum.2011.00028.
Rajaei, K. et al. (2019) ‘Beyond core object recognition: Recurrent processes account for object recognition under occlusion’, PLOS Computational Biology. Edited by L. Isik, 15(5), p. e1007001. Available at: https://doi.org/10.1371/journal.pcbi.1007001.