Categorization training increases the perceptual separability of novel dimensions.

Area of Science:

• Cognitive Psychology
• Perception Science
• Human Cognition

Background:

• Perceptual separability is crucial for understanding how humans process information and is fundamental to concepts like feature independence and configural processing.
• Despite its importance, the origins of dimensional separability remain largely unknown, with limited research on whether it can be modified through experience.
• Existing research indicates that categorization training can facilitate the learning of new perceptual dimensions.

Purpose of the Study:

• To investigate whether perceptual separability of dimensions increases with categorization training.
• To examine the impact of categorization training on Garner interference and marginal invariance, key indicators of perceptual separability.
• To test the hypothesis that dimensional separability is not a fixed characteristic but can be learned.

Main Methods:

• Utilized General Recognition Theory (GRT) to assess perceptual separability.
• Experiment 1 involved pre-training participants in a categorization task and measuring Garner interference and marginal invariance.
• Experiment 2 employed a model-based analysis within GRT to evaluate separability changes post-categorization training.

Main Results:

• Participants undergoing categorization pre-training exhibited reduced Garner interference effects.
• Pre-trained participants showed fewer violations of marginal invariance, indicating increased perceptual separability.
• Model-based GRT analysis confirmed that categorization training significantly enhanced the perceptual separability of relevant dimensions.

Conclusions:

• Categorization training demonstrably increases the perceptual separability of dimensions.
• Findings challenge the prevailing assumption that feature separability and independence are innate and unchangeable.
• This research suggests that perceptual dimensions can become more separable through learned experience and task-specific training.