Transfer of training in alphabet arithmetic

Area of Science:

• Cognitive Psychology
• Cognitive Neuroscience
• Mathematics Education

Background:

• Skilled adults are hypothesized to solve simple addition problems using a fast counting procedure.
• Practice typically leads to transfer of learning, but studies found no generalization for simple addition, challenging the fast counting theory.

Purpose of the Study:

• To investigate generalization in an alphabet arithmetic task to determine if counting-based procedures produce generalization.
• To challenge or support fast counting theories of skilled adults' simple addition.

Main Methods:

• Two experiments were conducted with 48 participants examining generalization in an alphabet arithmetic task.
• Alphabet arithmetic problems (e.g., B + 5 = C D E F G) were used to test counting generalization.
• Reanalysis of previously published addition generalization experiments (combined n = 172) was performed.

Main Results:

• Both experiments demonstrated robust generalization in alphabet arithmetic, evidenced by faster response times for practiced sequences.
• Generalization occurred when test problems shared overlapping letter augends and answer sequences with practiced problems.
• Experiment 2 showed generalization even with unpracticed letters if the answer sequence was practiced.
• Reanalysis of addition experiments found no evidence of facilitation for problems with matching augends and counting sequences.

Conclusions:

• Counting-based procedures in alphabet arithmetic exhibit clear generalization.
• The absence of generalization of practice effects in genuine addition challenges fast counting theories.
• Findings suggest that the mechanisms underlying simple addition may differ from those in alphabet arithmetic.