High Reward Makes Items Easier to Remember, but Harder to Bind to a New Temporal Context
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
Learning through reward is central to adaptive behavior. Indeed, items are remembered better if they are experienced while participants expect a reward, and people can deliberately prioritize memory for high- over low-valued items. Do memory advantages for high-valued items only emerge after deliberate prioritization in encoding? Or, do reward-based memory enhancements also apply to unrewarded memory tests and to implicit memory? First, we tested for a high-value memory advantage in unrewarded implicit- and explicit-tests (Experiment 1). Participants first learned high or low-reward values of 36 words, followed by unrewarded lexical decision and free-recall tests. High-value words were judged faster in lexical decision, and more often recalled in free recall. These two memory advantages for high-value words were negatively correlated suggesting at least two mechanisms by which reward value can influence later item-memorability. The ease with which the values were originally acquired explained the negative correlation: people who learned values earlier showed reward effects in implicit memory whereas people who learned values later showed reward effects in explicit memory. We then asked whether a high-value advantage would persist if trained items were linked to a new context (Experiments 2a and 2b). Following the same value training as in Experiment 1, participants learned lists composed of previously trained words mixed with new words, each followed by free recall. Thus, participants had to retrieve words only from the most recent list, irrespective of their values. High- and low-value words were recalled equally, but low-value words were recalled earlier than high-value words and high-value words were more often intruded (proactive interference). Thus, the high-value advantage holds for implicit- and explicit-memory, but comes with a side effect: High-value items are more difficult to relearn in a new context. Similar to emotional arousal, reward value can both enhance and impair memory.
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 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 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