Abstract:Mounting evidence has elucidated the sequential information processing trajectory, progressing from low-level perceptual to higher-level conceptual features during visual object recognition and reversing during memory retrieval. However, the extent to which the processing stream can be modulated by selective attention, as dictated by task requirements, remains unclear. The present study combined convolutional neural network (AlexNet), drift-diffusion model and multivariate decoding analysis to address this question. By using single-trial-based multivariate decoding analysis, we tracked the onset and peak time of feature representations across various visual perceptual and mnemonic categorization tasks, encompassing color, animacy, and size feature categorization. The findings of this study revealed several key outcomes: (1) A significant increase in reaction time was observed for the visual perception of color, animacy, and size features. (2) The findings from AlexNet and the drift-diffusion model revealed that, despite a parallel processing nature in these three features, those with larger decision boundaries necessitate subjects to spend more reaction time in making decisions about them. (3) In perceptual tasks, selective attention induced a reversal in the onset time of perceptual and conceptual features in the occipital and parietal lobe, resulting in the earlier detection of conceptual features compared to perceptual features. Conversely, in mnemonic tasks, selective attention led to the precedence of the peak time of perceptual features over conceptual features in the frontal lobe, signifying the reactivation of perceptual features with the highest fidelity prior to conceptual features. (4) Phase coupling analysis indicated a reversed information flow in the beta band during the task preparation stage in visual perception and memory retrieval tasks. Additionally, interactions between the occipital and temporal lobes in the theta band during the task preparation stage were found to promote an earlier onset time of target features in the visual perception stage. Collectively, these results underscore the parallel processing nature of features and demonstrate that the speed of feature information accumulation can be modulated by task requirements, with a more pronounced effect observed during perceptual compared to mnemonic stages. ### Competing Interest Statement The authors have declared no competing interest.

Visual attention modulates the transition from fine-grained, serial processing to coarser-grained, more parallel processing: A computational modeling study

A Hierarchical Model of Visual Processing Simulates Neural Mechanisms Underlying Reflexive Attention.

The influence of visual attention on letter recognition and reading acquisition in Arabic

Cortical and Subcortical Mechanisms of Orthographic Word-form Learning

Selective attention modulates feature accumulation speed during perception and memory retrieval

Visual word learning in skilled readers of English

Severe processing capacity limits for sub-lexical features of letter strings

Monolingual and bilingual infants' attention to talking faces: evidence from eye-tracking and Bayesian modeling

A general-purpose mechanism of visual feature association in visual word identification and beyond

Early Development of Visual Attention: Change, Stability, and Longitudinal Associations

Auditory attention influences trajectories of symbol–speech sound learning in children with and without dyslexia

A neuroanatomically grounded Hebbian-learning model of attention-language interactions in the human brain

Dynamic Modulation of Representational Trajectories Through Selective Attention

Visual attention span as a predictor of reading fluency and reading comprehension in Arabic

Spatial attention in written word perception

Understanding visual attention with RAGNAROC: A reflexive attention gradient through neural AttRactOr competition.

Ongoing Cognitive Processing Influences Precise Eye-Movement Targets in Reading

Slow perceptual processing at the core of developmental dyslexia: a parameter-based assessment of visual attention

Reading instruction causes changes in category-selective visual cortex

Attentional Modulation of Visual Spatial Integration: Psychophysical Evidence Supported by Population Coding Modeling

Developmental refinements to neural attentional state during semantic memory retrieval through adolescence