关键词: 失象症, 心理表象, 认知策略, 深度学习

Abstract: Aphantasia , a unique psychological condition in which certain individuals, in the absence of external sensory input, cannot involuntarily generate or recreate symbolic images and sensory experiences in their minds. Due to the non-universality of human experiences, there is a general belief that the ability to internally simulate visual sensory experiences, known as visual imagery, is shared by everyone. However, in reality, there are substantial differences in human visual imagery, a fact that has only gradually gained attention in recent years. Aphantasia, as an extreme case of individual differences in mental imagery, challenges traditional theories in cognitive psychology, highlighting that not everyone possesses vivid visual images.
The study reviews the psychological phenomenon of aphantasia, including its characteristics, assessment methods, and cognitive strategies. It discusses the limitations of subjective self-report tools used to assess mental imagery, such as the Vividness of Visual Imagery Questionnaire (VVIQ), and examines objective experimental techniques like binocular rivalry and advanced brain imaging to confirm the validity of aphantasia. By combining subjective self-report assessments with objective experimental paradigms, the study delves deeply into the neural and cognitive mechanisms underlying aphantasia. The findings reveal that, despite their inability to generate visual images, individuals with aphantasia typically employ compensatory strategies, such as non-visual strategies like verbal descriptions, to overcome their deficits in visual imagery. These strategies are not only evident in the realms of imagination and memory but also manifest in areas such as spatial abilities, metacognition, emotional experiences, dreaming, creativity, and synesthesia. This discovery challenges the traditional notion that imagination and memory primarily rely on visual processing, suggesting instead that these cognitive functions can be supported by other sensory and cognitive mechanisms. The diversity in cognitive strategies enriches our understanding of human thought and offers new perspectives on how cognitive tasks can be effectively performed in the absence of visual input.
Furthermore, the study discusses the intersection of deep learning models and cognitive neuroscience, expanding its innovative approach by exploring the role of deep learning models in understanding aphantasia. The development of deep learning models not only advances the convergence of cognitive science and artificial intelligence but also provides new avenues for uncovering the neural computational mechanisms underlying aphantasia. By utilizing deep neural networks to simulate the hierarchical structure and distributed processing of the ventral visual pathway, the study presents a novel approach to modeling the cognitive patterns associated with aphantasia. This interdisciplinary method not only deepens our understanding of aphantasia but also contributes to discussions on human cognitive diversity within the fields of artificial intelligence and cognitive science. The application of deep learning models represents a significant advancement in aphantasia research, offering a new method for investigating how the brain compensates for the lack of mental imagery. These models provide a framework for simulating the cognitive processes of individuals with aphantasia, allowing researchers to explore how non-visual strategies are implemented at the neural level. This approach opens new avenues for studying the neural computations that support alternative cognitive strategies, potentially aiding in the development of more adaptable and versatile artificial intelligence systems that can mimic these strategies.
The implications of this research extend beyond the study of aphantasia, offering valuable insights into the broader field of cognitive diversity. By understanding how individuals with aphantasia complete tasks without visual imagery, the study highlights the importance of considering cognitive diversity in both cognitive science and artificial intelligence. The findings suggest that human cognition is not monolithic but rather encompasses a wide range of strategies and processes, each suited to different individuals' cognitive architectures. This perspective has significant implications for the development of more inclusive and adaptable AI systems that can accommodate diverse cognitive profiles.
The study proposes several directions for future research. A key area is the exploration of multisensory modalities in aphantasia, which could provide further insights into how individuals with this condition integrate information across different sensory domains. Additionally, the study suggests that future research should focus on developing new deep learning models that can more accurately simulate the cognitive patterns associated with aphantasia. These models could be used to investigate the neural mechanisms underlying this condition, offering a more comprehensive understanding of how the brain processes information in the absence of visual imagery. Moreover, the research emphasizes the potential of using these models to explore the diversity of cognitive processing in the human brain. By simulating different cognitive strategies, researchers can gain a deeper understanding of the neural computations that support various forms of cognition. This approach could also contribute to the development of more human-like intelligent systems, which are capable of adapting to different cognitive styles and strategies.

Key words: aphantasia, mental imagery, cognition strategy, deep learning