A conceptual model deriving transfer from all-or-none learned common and falsely-common task elements and the product-moment correlation¶

Theory · All theories
Evidence · 4 claims (4 for) · 2 studies (2 causal), q1–q2 · 0 of 2 report an effect size · 4 claims rest on one study

Description¶

The model treats tasks as sets of mutually independent hypothetical elements whose extension corresponds to average mastery time. Transfer T is derived as proportional to the number of commonly identified elements minus falsely identified ones, and the product-moment correlation between tasks is derived under an all-or-none learning assumption, yielding T proportional to the unsquared correlation. The article argues this "does not require detailed task analyses, let alone a S-R decomposition" and makes psychometric research available to learning theory and instruction.

Design Implications¶

Context¶

Requirements¶

Both learning tasks must already be partially but not completely mastered (0 < p < 1, 0 < t < 1) for a meaningful correlation coefficient.
The single c-elements and f-elements must be learned at the same rate as all other elements, e.g., by equal-interval arrangement or equal sampling probability.

Constraints¶

Correlation coefficients are highly sensitive to subject sampling.
The model is not yet concerned with individual differences in transfer; it uses individual differences in performance to predict average transfer.

Target Learners¶

school-age learners in psychometric task settings (fifth graders in the validation study)

Target Learning Objectives¶

prediction of transfer of learning between cognitive tasks

Claims¶

Where identical elements exist in two situations, training in one produces positive transfer to the other [+W] — attached 2026-10-24 from Poffenberger et al. (1915), which proposed "Specific stimulus-response bonds and identical elements as the mechanism of transfer of training"; tests this page's relationship.
Where a test requires breaking previously formed bonds and forming new ones, training produces interference rather than transfer [+W] — attached 2026-10-24 from Poffenberger et al. (1915), which proposed "Specific stimulus-response bonds and identical elements as the mechanism of transfer of training"; tests this page's relationship.
Where there are no identical bonds between stimulus and response in two processes, training in one produces neither transfer nor interference in the other [+W] — attached 2026-10-24 from Poffenberger et al. (1915), which proposed "Specific stimulus-response bonds and identical elements as the mechanism of transfer of training"; tests this page's relationship.
Practice effects in visual reproduction are specific: they transfer to new conditions only in proportion to identical ability-conditioning factors. [+W]

Examples¶

Systematic Instructional Design

Key Sources¶

Flammer, A. (1973). A Conceptual Model Relating Transfer of Learning and Correlation: Theoretical Paper No. 43. Wisconsin Research and Development Center for Cognitive Learning. https://eric.ed.gov/?id=ED085637

Page metadata

Field	Value
`type`	`theory`
`title`	`A conceptual model deriving transfer from all-or-none learned common and falsely-common task elements and the product-moment correlation`
`description`	`The model treats tasks as sets of mutually independent hypothetical elements whose extension corresponds to average mastery time.`
`status`	`review`
`canonical`	`True`
`generated`	by `claude/unspecified` on `2026-09-25`
`trust`	unverified — no one has confirmed this page's content
`sources`	1 — `flammer-1973`

Frontmatter is the page's machine-readable record — see the Schema & Guide.