Are recurrent associative memories good models of decision making? Modelling discrimination decisions from different perspectives
Bibliographic record
Abstract
Discrimination decisions are at the forefront of human cognition. For this reason, many different types of models aim to predict how they are made. In this research, we compared the discrimination capabilities of a Recurrent Associative Memory (RAM) with the predictions of an accumulator model to show that, although the discrimination processes of both model classes differs, both make similar predictions regarding trends in the results. We did this by measuring the performances of a RAM within the context of a discrimination task using different stimuli (i. e., letters and randomly generated stimuli) and fitting the obtained results with an accumulator model possessing a coactive architecture. The experimental conditions varied with regard to the correlation between the tested stimuli, the amount of redundancy of the stimuli used in a trial, and the number of total stimuli presented to the network in the learning phase. Results showed that high inter-stimulus correlation led to slower recall speed, and that low redundancy also resulted in slower recall speed. Results also indicated that an increased number of exemplars contained in the network's memory increased recall speed for the letter stimuli but randomly generated stimuli received no apparent benefits. Ultimately, exploring neural networks and accumulator models jointly provides a broader and deeper understanding of the cognitive processes behind discrimination decisions.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".