Memories of Visual Events Can Be Formed Without Specific Spatial Coordinates

Area of Science:

• Cognitive Neuroscience
• Visual Perception
• Memory Research

Background:

• Feature integration theory posits spatial attention is crucial for binding object features.
• Neurons in the anterior ventral stream have retinotopic receptive fields, suggesting location is encoded.
• Previous models predicted location information is integral to visual memory representations.

Purpose of the Study:

• To investigate the extent to which spatiotopic location accompanies remembered visual event representations.
• To test the prediction that location information is inherent in visual memory.
• To explore the relationship between object identification and spatial localization in working memory.

Main Methods:

• Participants identified and located a target letter among digits in brief visual arrays.
• Analysis focused on trials with accurate target identity but inaccurate location recall.
• Experiments manipulated target features (e.g., color) to assess location binding.

Main Results:

• Over 15% of trials showed accurate identification but poor location recall for multi-feature targets.
• Location errors stemmed from feature-location misbinding and weak spatial representations.
• Single-feature targets (color) were consistently remembered with their locations, even under masking.

Conclusions:

• Consciously accessible visual memories can form without robust spatial specificity.
• Specific location may not always be encoded into or retained in working memory.
• The necessity of spatial information for visual memory depends on stimulus complexity and features.