Skip to content

Latest commit

 

History

History
71 lines (35 loc) · 3.69 KB

README.md

File metadata and controls

71 lines (35 loc) · 3.69 KB

Dependent Variables (Learning Outcomes)

Learning Rate:

Defined by the slope of a regression line fit to lesson data obtained from each training session.

Programming Accuracy:

Determined based on code produced by learners after training, with the goal of creating a Rock-Paper-Scissors game. Code was assessed by averaging three raters' scores (accuracies in proportions) based on a rubric developed by a team of expert Python programmers. Numbers represent the resulting averaged accuracy in proportions, 0 to 1.

Declarative Knowledge:

Defined by a total accuracy in proportions, 0 to 1, on a 50-item multiple choice test, composed of 25 questions assessing the general purpose of functions (or semantic knowledge) and 25 questions assessing syntactic knowledge.

Predictors

Numeracy

Assessed using a Rasch-Based Nuemracy Scale which was created by evaluating 18 numeracy questions across multiple measures and determining the 8 most predictive items. Numbers represent accuracy in proportions, 0 to 1, with 1 representing perfect performance.

Fluid Intelligence:

Assessed using a shortened 18-item version of the Raven's Advanced Progressive Matrices, which was developed by splitting the original 36 questions into two, difficulty-matched subtests. Numbers represent total correct, 0 to 18, with 18 representing perfect performance.

Working Memory Span:

Assessed by averaging across the Computerized Reading Span Task, the Operation Span Task, and the Symmetry Span Task. Numbers represent a single composite score that was computed by z-transforming each individual span score and taking an average of the three scores.

Language Aptitude:

Assessed using the Modern Language Aptitude Test, a standardized measure of second-language aptitude normed for native English speaking adults. Numbers represent an individual's percentile score based on normed data, 0 to 100, with 100 representing the 100th percentile.

Vocabulary:

Assessed using the Nelson-Denny Vocabulary test. Numbers represent accuracy in percentages, 0 to 100, with 100 representing perfect performance.

Working Memory Updating:

Assessed using the 3-Back version of the N-Back Task. In this 80-item version, participants are presented with a stream of letters and are tasked with determining if the presented letter is the same or different from the letter presented three items ago. Numbers represent accuracy in proportions, 0 to 1, with 1 representing perfect performance.

Inhibitory Control:

Assessed using the Simon Task, a non-verbal measure designed to assess susceptibility to stimulus-response interference. The version used herein consisted of 75% congruent and 25% incongruent trials.

Right Fronto-Temporal Low Gamma Power:

An individual's rsEEG power in the low-gamma frequency range (30-40 Hz) recorded over the right fronto-temporal network.

Right Posterior Low Gamma Power:

An individual's rsEEG power in the low-gamma frequency range (30-40 Hz) recorded over the right posterior network.

Right Fronto-Temporal Beta Power:

An individual's rsEEG power in the beta frequency range (13-29.5 Hz) recorded over the right fronto-temporal net.

Left Posterior Beta Coherence:

An individual's rsEEG coherence in the beta frequency range (13-29.5 Hz) within the left posterior network.

Left Posterior Theta Coherence:

An individual's rsEEG coherence in the theta frequency range (4-7 Hz) within the left posterior network.

Other

EEG Usability:

Represents the usability of an individual's EEG data, as represented by the numbers of alpha peaks detected by the O2 channel. Subjects with EEG Usability of less than 1 were excluded from analyses.