Back to all research

The perceptual and phenomenal capacity of mental imagery

DOI: 10.1016/j.cognition.2017.02.004
Keogh, R., & Pearson, J. (2017). The perceptual and phenomenal capacity of mental imagery. Cognition, 162, 124–132. doi:10.1016/j.cognition.2017.02.004

Abstract

Despite the brain’s immense processing power, it has finite resources. Where do these resource limits come from? Little research has examined possible low-level sensory contributions to these limitations. Mental imagery is a fundamental part of human cognition that bridges cognition with sensory representations. Hence, imagery serves as a good candidate sensory process for probing how low capacity limitations might extend down the processing hierarchy. Here we introduce a novel technique to measure the sensory capacity of mental imagery, while removing the need for memory and any direct subjective reports. Contrary to our dynamic phenomenological experience, we demonstrate that visual imagery is severely limited by the perceptual and phenomenal bottleneck of visual representation. These capacity limits appear to be independent of generation time, depend on visual feature heterogeneity, are attenuated by concurrent retinal stimulation and are endowed with good metacognition. Additionally, the precision of visual representation declines rapidly with the number of stimuli, which is governed by a simple power law. We anticipate that this assay will be important for mapping the limits of human information processing.

Authors

  • Rebecca Keogh14
  • Joel Pearson28

Understanding the Limits of Mental Imagery

Overview/Introduction

Our brains are incredibly powerful, yet they have limits. This study explores where these limits come from, focusing on mental imagery—a key part of how we think and perceive the world. Mental imagery connects our thoughts with sensory experiences, making it a perfect area to investigate these limitations. The research introduces a new method to measure how much information our mental imagery can handle, without relying on memory or subjective reports.

Methodology

  • Participants: 72 individuals aged 18 to 35 participated in various experiments.
  • Technique: Participants were asked to imagine 1 to 7 colored patterns (Gabor patches) for 6 seconds. Placeholders indicated where and what to imagine, minimizing memory use.
  • Testing: After imagining, participants viewed a brief visual display to test how well they imagined the patterns. They reported which pattern they saw as dominant.
  • Analysis: Data were analyzed using statistical methods to understand the capacity and precision of mental imagery.

Key Findings

  • Capacity Limits: Mental imagery is limited by the brain's ability to represent visual information. These limits are not affected by how long the imagery is generated.
  • Feature Dependence: The capacity depends on the variety of visual features. More diverse features lead to reduced capacity.
  • Precision Decline: As the number of imagined items increases, the precision of each image decreases, following a simple mathematical rule.
  • Metacognition: Participants were aware of the vividness of their mental images, and this awareness matched the measured capacity limits.
  • External Influence: Background visual information can reduce the strength of mental imagery.

Implications

  • Cognitive Science: These findings challenge the idea that thought is independent of sensory systems, supporting the notion that our thoughts are grounded in sensory experiences.
  • Practical Applications: Understanding these limits can aid in designing better educational tools and cognitive therapies by aligning with how our brains naturally process information.

Limitations

  • Sample Size and Diversity: The study involved a relatively small and specific demographic, which may limit the generalizability of the findings.
  • Experimental Conditions: The controlled lab setting may not fully represent real-world scenarios where mental imagery is used.
In conclusion, this research highlights the inherent limits of mental imagery, suggesting that these constraints are tied to the brain's sensory processing capabilities. This insight opens new avenues for exploring how we think and perceive, with potential applications in various fields from education to artificial intelligence.