Chunk formation in immediate memory and how it relates to data compression.
Level 3 - non-randomized controlled study
Laboratory experimental cognitive study (by design analogy, not clinical CEBM).
PubMed 27367593 · doi:10.1016/j.cognition.2016.05.024
What was done
Untrained participants completed an immediate recall task presenting sequences of stimuli with varying degrees of association. Information compressibility was used to measure single-trial chunkability and evaluate whether immediate memory rapidly recodes redundant list structures into compact representations.
What was found
Lists lacking structure produced an immediate memory span of about three items. Memory span increased linearly as structure was added. The quantity of information retained was maximal for the most compressible sequences, particularly when presentation order facilitated compression. The abstract reports no numerical effect sizes, dispersion measures, or statistical test metrics.
Why it matters
This study suggests that chunk formation can occur dynamically in immediate memory via rapid compression of novel structural regularities, rather than depending strictly on pre-existing long-term memory traces.
Limits
The abstract omits sample size, participant demographics, stimulus modality/materials, and specific statistical values or effect sizes. Testing was limited to immediate recall.
Cited by
- supports Structuring information logically rather than presenting it as an unstructured list improves cognitive processing and memory retention.