Accuracy-based measures provide a better measure of sequence learning than reaction time-based measures

Area of Science:

• Cognitive science
• Neuroscience
• Psychology

Background:

• The Serial Reaction Time Task (SRTT) is a common tool for assessing motor sequence learning.
• Traditional SRTT performance metrics, like reaction time (RT) difference scores, may be limited by floor effects and not fully capture learning.
• Alternative measures, such as ratio RT and predictive accuracy, may offer more sensitive assessments of sequence learning.

Purpose of the Study:

• To introduce and validate the Predictive Sequence Learning Task (PSLT) as an alternative to the SRTT.
• To compare the validity of different performance measures (RT difference, ratio RT, predictive accuracy) from both SRTT and PSLT in assessing sequence learning.
• To investigate the relationship between sequence learning measures and fluid abilities, as well as age-related performance declines.

Main Methods:

• A within-subjects design comparing performance on SRTT and PSLT in 99 adult participants.
• Assessment of sequence learning using reaction time (RT) difference scores, ratio RT measures, and predictive accuracy.
• Measurement of participants' fluid abilities.

Main Results:

• RT difference scores from both SRTT and PSLT showed no significant relationship with fluid abilities.
• Ratio RT measures from both tasks and PSLT accuracy measures were positively correlated with fluid abilities.
• PSLT accuracy measures revealed an age-related decline in sequence learning performance.

Conclusions:

• Traditional SRTT RT difference scores may be insensitive to meaningful variations in sequence learning due to floor effects.
• PSLT, particularly its accuracy measures, offers a more sensitive and valid assessment of sequence learning.
• PSLT accuracy measures show potential for evaluating cognitive abilities and age-related changes in learning.