Template theories suggest that the lack of description of patterns is necessary. They represent unanalyzed pattern data that is contrasted with alternative patterns with the implementation of degrees of overlap as a measure of similarity. This theory has some significant drawbacks. First, such an analysis would require both compared patterns to be in the same position and orientation. Second, the two sets would have to be of the same size. Third, the template theory may expose similarities but does not identify differences. Fourth, this theory does not provide an alternative method of analysis of the two compared patterns, meaning that only one examination will be available when many exist.
The feature theory focuses on the description of the parts and characteristics of a pattern. This theory is important in the field of perceptual learning, which occurs during the discovery and collection of features that distinguish two patterns from each other. There can be difficulty in finding appropriate sets of factors, and certain criteria have been suggested to standardize the analysis of data through feature theory. First, the collected features should be critical to the pattern. Second, these aspects should not change. Third, the feature should display a unique pattern in every component of the pattern. Fourth, the set of features should be small.
An experimental procedure predicted confusion in comparing letters with four-year-old children. With prior knowledge that children can make errors in perceptual matching, the letters which could be confused were listed (Reed, 2013). However, even these young children made few errors with some letters not being confused at all. The feature theory allowed for an improved prediction due to being able to analyze both differences and similarities. Through a number of studies with participants of different ages, it was revealed that the feature theory is superior in the case of predicting confusion errors.
References
Reed, S., K. (2013). Cognition: Theories and Application. (9th Ed.). Wadsworth, Cengage Learning.