A single g factor is not necessary to simulate positive correlations between cognitive tests.

Area of Science:

• Psychometrics
• Cognitive Psychology
• Factor Analysis

Background:

• The interpretation of positive correlations among diverse cognitive tests is debated.
• One perspective supports a single general ability factor ('g' factor).
• An alternative view posits multiple dissociable underlying traits.

Purpose of the Study:

• To evaluate whether multiple independent abilities can produce a positive manifold.
• To test if consistent factor influence across tests is key to positive correlations.
• To determine if positive correlations differentiate between single and multiple factor models.

Main Methods:

• Simulated test performance data.
• Generated correlation matrices from models with one or eight independent factors.
• Manipulated the consistency of factor influence across tests.

Main Results:

• Both single and multiple independent factor models produced positive correlations.
• The degree of consistency in factor operation across tests influenced the strength of positive correlations.
• Increased consistency led to a greater tendency for positive correlations in both models.

Conclusions:

• The presence of a positive manifold in cognitive test correlations does not exclusively support a single general factor model.
• Multiple independent factors, when operating consistently, can also explain observed positive correlations.
• The findings suggest a need to reconsider the interpretation of positive manifold in abilities testing.