Cognition test battery: Adjusting for practice and stimulus set effects for varying administration intervals in high performing individuals

ABSTRACT

Introduction

Practice effects associated with the repeated administration of cognitive tests often confound true therapeutic or experimental effects. Alternate test forms help reduce practice effects, but generating stimulus sets with identical properties can be difficult. The main objective of this study was to disentangle practice and stimulus set effects for Cognition, a battery of 10 brief cognitive tests specifically designed for high-performing populations with 15 unique versions for repeated testing. A secondary objective was to investigate the effects of test-retest interval on practice effects.

Methods

The 15 versions of Cognition were administered in three groups of 15–16 subjects (total N = 46, mean±SD age 32.5 ± 7.2 years, range 25–54 years, 23 male) in a randomized but balanced fashion with administration intervals of ≥10 days, ≤5 days, or 4 times per day. Mixed effect models were used to investigate linear and logarithmic trends across repeated administrations in key speed and accuracy outcomes, whether these trends differed significantly between administration interval groups, and whether stimulus sets differed significantly in difficulty.

Results

Protracted, non-linear practice effects well beyond the second administration were observed for most of the 10 Cognition tests both in accuracy and speed, but test-retest administration interval significantly affected practice effects only for 3 out of the 10 tests and only in the speed domain. Stimulus set effects were observed for the 6 Cognition tests that use unique sets of stimuli. Factors were established that allow for correcting for both practice and stimulus set effects.

Conclusions

Practice effects are pronounced and probably under-appreciated in cognitive testing. The correction factors established in this study are a unique feature of the Cognition battery that can help avoid masking practice effects, address noise generated by differences in stimulus set difficulty, and facilitate interpretation of results from studies with repeated assessments.

KEYWORDS:

• Cognition
• practice
• learning
• test difficulty
• testing interval

View correction statement:

Correction

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here.

Notes

1. Different “versions” of a test refer to different stimulus sets. Another term that is frequently used for “version” in cognitive testing is “form”.

2. When we call tests “unique” here and throughout the manuscript, we refer to the test stimuli and their order. The 15 versions of each test still have the same psychometric properties, i.e., they are based on the same rules, they come with the same instructions, and they target the same construct, but each with novel and different content.

Additional information

Funding

This work was funded by the National Space Biomedical Research Institute (NSBRI) through grant NBPF00012 (NASA NCC 9-58) and by the National Aeronautics and Space Administration (NASA) through grant NNX14AH98G