Transfer of training from one working memory task to another: behavioural and neural evidence
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
N-back working memory (WM) tasks necessitate the maintenance and updating of dynamic rehearsal sets during performance. The delayed matching-to-sample (dMTS) task is another WM task, which in turn involves the encoding, maintenance, and retrieval of stimulus representations in sequential order. Because both n-back and dMTS engage WM function, we hypothesized that compared to a control task not taxing WM, training on the n-back task would be associated with better performance on dMTS by virtue of training a shared mental capacity. We tested this hypothesis by randomly assigning subjects (N = 43) to train on either the n-back (including 2-back and 3-back levels) or an active control task. Following training, dMTS was administered in the fMRI scanner. The n-back group performed marginally better than the active control group on dMTS. In addition, although the n-back group improved more on the less difficult 2-back level than the more difficult 3-back level across training sessions, it was improvement on the 3-back level that accounted for 21% of the variance in dMTS performance. For the control group, improvement in training across sessions was unrelated to dMTS performance. At the neural level, greater activation in the left inferior frontal gyrus, right posterior parietal cortex, and the cerebellum distinguished the n-back group from the control group in the maintenance phase of dMTS. Degree of improvement on the 3-back level across training sessions was correlated with activation in right lateral prefrontal and motor cortices in the maintenance phase of dMTS. Our results suggest that although n-back training is more likely to improve performance in easier blocks, it is improvement in more difficult blocks that is predictive of performance on a target task drawing on WM. In addition, the extent to which training on a task can transfer to another task is likely due to the engagement of shared cognitive capacities and underlying neural substrates-in this case WM.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it