Quotation withheld

Quotation withheld George A. Miller

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Quotation withheld 1956 Psychological Review 10.1037/h0043158 miller1956-magical-number miller1956-magical-number.txt, the project's staged copy ccab0731ccb64c6fe88ddab7062994ec06e99126 open-access psychology

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them The recurring limit of about seven in absolute judgment, attention and immediate memory, and whether those limits are one thing or several. Quotation withheld because this record is not cleared for redistribution. existence Quotation withheld because this record is not cleared for redistribution. span of absolute judgment and span of immediate memory Quotation withheld because this record is not cleared for redistribution. review review Quotation withheld because this record is not cleared for redistribution. mechanisms the human observer behaves like a limited-capacity communication channel: absolute judgment saturates at a channel capacity of about 2.5 bits (~7 categories) Quotation withheld because this record is not cleared for redistribution. Treating the observer as a communication channel makes the limit measurable: input information is raised until transmitted information stops rising, and that asymptote is the capacity Quotation withheld because this record is not cleared for redistribution. Immediate memory holds a fixed number of chunks rather than a fixed number of bits, which is why it behaves unlike absolute judgment Quotation withheld because this record is not cleared for redistribution. Because the limit is in chunks, packing more information into each chunk raises how much can be held without raising the count Quotation withheld because this record is not cleared for redistribution. Recoding is the operation that builds bigger chunks: group the input, name the group, remember the name Quotation withheld because this record is not cleared for redistribution. Adding dimensions and accepting cruder judgments on each raises the span of absolute judgment far beyond seven Quotation withheld because this record is not cleared for redistribution. the recurrence of 'seven' across channel capacity and memory span may be merely coincidental what this limits: the recurrence of seven across channel capacity and memory span: THERE SEEMS TO BE A REASONABLE SUSPICION THAT IT IS NOTHING MORE THAN A COINCIDENCE Quotation withheld because this record is not cleared for redistribution. The suggested span of perceptual dimensionality (about ten) is stated to have no supporting objective evidence. what this limits: the span of perceptual dimensionality: THERE IS NO OBJECTIVE EVIDENCE TO SUPPORT THIS SUSPICION Quotation withheld because this record is not cleared for redistribution. Some reported channel-capacity values are probably underestimates because data were pooled across subjects before computing transmitted information. what this limits: the channel-capacity values: THESE VALUES ARE APT TO BE SLIGHTLY TOO LOW BECAUSE THE DATA FROM ALL SUBJECTS WERE POOLED BEFORE THE TRANSMITTED INFORMATION WAS COMPUTED Quotation withheld because this record is not cleared for redistribution. The central theoretical unit, the chunk, is not precisely defined. what this limits: every claim counted in chunks: WE ARE NOT VERY DEFINITE ABOUT WHAT CONSTITUTES A CHUNK OF INFORMATION Quotation withheld because this record is not cleared for redistribution. The subitizing comparison rests on a guess, since the number-estimation data were never analysed in informational terms. what this limits: the subitizing comparison: THE DATA ON NUMBER ESTIMATES HAVE NOT BEEN ANALYZED IN INFORMATIONAL TERMS, and the four-bit figure is offered as a guess Quotation withheld because this record is not cleared for redistribution. The paper's own closing position is that the recurring seven is probably coincidence, not one underlying limit what this limits: the seven itself: I SUSPECT THAT IT IS ONLY A PERNICIOUS, PYTHAGOREAN COINCIDENCE Quotation withheld because this record is not cleared for redistribution. The unit that the memory limit is counted in is admitted to be undefined what this limits: the same scope as limitations[3] - the unit the memory limit is counted in is undefined Quotation withheld because this record is not cleared for redistribution. The chunk independence holds only over the range tested so far, which the author states rather than assumes what this limits: the chunk-independence claim: AT LEAST OVER THE RANGE THAT HAS BEEN EXAMINED TO DATE Quotation withheld because this record is not cleared for redistribution. The span of seven is claimed for unidimensional judgments specifically, not judgments in general what this limits: the span of seven specifically: I MAINTAIN THAT FOR UNIDIMENSIONAL JUDGMENTS THIS SPAN IS USUALLY SOMEWHERE IN THE NEIGHBORHOOD OF SEVEN - not judgments in general Quotation withheld because this record is not cleared for redistribution. The information-theoretic machinery is deliberately paraphrased away, so the argument as presented cannot be checked against the formalism it rests on what this limits: the argument as presented: IN THE INTERESTS OF COMMUNICATION I HAVE SUPPRESSED THE TECHNICAL DETAILS OF INFORMATION MEASUREMENT Quotation withheld because this record is not cleared for redistribution. analyzing the limits of human perception and memory using information theory (channel capacity, bits) Quotation withheld because this record is not cleared for redistribution. Shannon's information measure, the channel and its capacity, carried into psychology - and the paper is explicit that the experiments would not exist without it Quotation withheld because this record is not cleared for redistribution. A translation table offered in the other direction, so psychologists could read the information measure as the variance they already knew Quotation withheld because this record is not cleared for redistribution. span of immediate memory span of immediate memory Quotation withheld because this record is not cleared for redistribution. recoding recoding Quotation withheld because this record is not cleared for redistribution. seven - the span of absolute judgment: about seven categories distinguishable along one dimension, a limit in BITS Quotation withheld because this record is not cleared for redistribution. seven - the span of immediate memory: about seven items retained, a limit in CHUNKS and almost independent of the bits in each Quotation withheld because this record is not cleared for redistribution. seven - the span of attention: about six objects taken in at a glance, a third limit with a similar number and a different mechanism Quotation withheld because this record is not cleared for redistribution. amount of information - as the paper translates it for the reader: variance, in the ordinary statistical sense Quotation withheld because this record is not cleared for redistribution. amount of TRANSMITTED information - covariance or correlation between input and output, a different quantity that shares most of its name Quotation withheld because this record is not cleared for redistribution. claim_type: existence a second reading reaching the same value. Recorded: existence — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. phenomenon: span of absolute judgment and span of immediate memory a second reading reaching the same value. Recorded: span of absolute judgment and span of immediate memory — note: a second reading reaching the same value from a different sentence. Not applied - a person decides. Quotation withheld because this record is not cleared for redistribution. builds-on → information theory (channel capacity) Quotation withheld because this record is not cleared for redistribution. replicates → Hayes (10), memory-span experiment with five kinds of test material Quotation withheld because this record is not cleared for redistribution. replicates → Hake and Garner (8), absolute judgment of visual position Quotation withheld because this record is not cleared for redistribution. builds-on → Külpe (14), abstraction experiment Quotation withheld because this record is not cleared for redistribution. extends → information theory as imported into psychology Quotation withheld because this record is not cleared for redistribution. critiques → the assumption that immediate memory is limited by the same quantity as absolute judgment differs in: THE AXIS IS INFORMATION PER ITEM. 'Quotation withheld' The critique works by deriving a prediction that varies with information per item - precisely the quantity the criticised assumption holds constant. Quotation withheld because this record is not cleared for redistribution. span of attention —distinct-from→ span of absolute judgment span of attention span of absolute judgment strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. chunk —distinct-from→ bit chunk bit strength of the relation: 1 Quotation withheld because this record is not cleared for redistribution. span of immediate memory —operationalised-as→ chunk span of immediate memory chunk strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span relates memory span to a fixed number of chunks; operationalised-as is a loose reading. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · recoding —causes→ chunk recoding chunk EVIDENCE CORRECTED: RE-GROUNDED. The old span describes WHAT recoding IS - 'There are many ways…' - and names only one endpoint. It carries NO EFFECT CUE AT ALL, so a 'causes' edge standing on it would fail the gate's own causes test today; it predates that rule. The replacement names both endpoints and states the effect: '…recoding INCREASES the bits per chunk'. Both spans are Miller's and both are accurate; the second is evidence for THIS TYPE and the first is not. Selected from an internal tool after reading six offers and DECLINING FIVE - cook2013's candidate carried a glued sentence-case heading, schmidhuber2015's was a contents listing, and grilli2016's and another row of this record's existing spans were STRONGER for their types than the offers (grilli2016's sentence on modular block structure; another row of this record's on a span of a fixed number of chunks). A both-endpoint span is not an improvement if it is weaker evidence for the type. — strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but endpoint by anaphora: 'this recoding' = recoding the input into a code with fewer chunks (previous sentence). Quotation withheld because this record is not cleared for redistribution. noise —contrasted-with→ transmitted information noise transmitted information strength of the relation: 0.9 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span separates output variance due to input from noise; contrasted-with is a loose reading. Quotation withheld because this record is not cleared for redistribution. recoding —causes→ span of immediate memory recoding span of immediate memory status: CONJECTURED at this point in the paper - the authors say they SUSPECT it; the demonstration follows · direction on the measure: increases · strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says a recoding procedure COULD be used to increase the span; hedged. Quotation withheld because this record is not cleared for redistribution. span of absolute judgment —distinct-from→ span of immediate memory span of absolute judgment span of immediate memory strength of the relation: 1 · note: RE-GROUNDS another row of this record: the old span named ONE endpoint. Quotation withheld because this record is not cleared for redistribution. recoding —causes→ chunk recoding chunk the relation itself is separately evidenced · strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span says recoding increases the bits per chunk; 'causes chunk' drops the quantity. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · about 7 — the limit on the span of absolute judgment and immediate memory EVIDENCE CORRECTED: RE-GROUNDED. The evidence was THE PAPER'S OWN TITLE - 'Quotation withheld' - which names the quantity and asserts nothing. A title is furniture in the same sense as a reference line: it says what the paper is about. The replacement is Miller STATING the quantity in his own voice, defining the span of absolute judgment and putting it 'in the neighborhood of seven'. The old span survived every check because an internal tool carried a literal 0x08 where a word boundary was meant, so the <h1> was never stripped from the body tags are checked against., measured by the project, verified here against the record's own title field. Quotation withheld because this record is not cleared for redistribution. about seven — the span of absolute judgment, in categories, for unidimensional stimuli Quotation withheld because this record is not cleared for redistribution. about seven items — the span of immediate memory, in chunks - a different quantity that happens to have the same number Quotation withheld because this record is not cleared for redistribution. about six objects — the span of attention, at a glance - the third of the three limits Quotation withheld because this record is not cleared for redistribution. at least 150 — the span of absolute judgment once many dimensions are used with crude judgments on each Quotation withheld because this record is not cleared for redistribution. one bit — the unit, defined as a choice between two equally likely alternatives Quotation withheld because this record is not cleared for redistribution. The author asks whether the discontinuity at seven reflects the same basic process that limits unidimensional judgments to about seven categories. Quotation withheld because this record is not cleared for redistribution. EVIDENCE CORRECTED (RE-GROUNDED) · The author distinguishes bits of information from chunks of information; the number of bits is constant for absolute judgment. EVIDENCE CORRECTED: RE-GROUNDED. The published string was in NEITHER copy of the paper. Ours began 'between bits…'; the paper says 'I HAVE FALLEN…'. The span also ran into the page break 'informa- [p. 93] tion', so it now stops at the clause before it. Quotation withheld because this record is not cleared for redistribution. the span of absolute judgment is about seven categories for unidimensional stimuli Quotation withheld because this record is not cleared for redistribution. adding dimensions with crude judgments on each raises that span to at least 150 Quotation withheld because this record is not cleared for redistribution. the span of immediate memory is almost independent of the number of bits per chunk Quotation withheld because this record is not cleared for redistribution. recoding into larger chunks raises how much information the span can carry without raising the count Quotation withheld because this record is not cleared for redistribution. the author declines to unify the recurring sevens and calls the coincidence Pythagorean Quotation withheld because this record is not cleared for redistribution. amount of information in the stimuli amount of information in the stimuli Quotation withheld because this record is not cleared for redistribution. amount of transmitted information amount of transmitted information Quotation withheld because this record is not cleared for redistribution. chunk chunk Quotation withheld because this record is not cleared for redistribution. recoding recoding Quotation withheld because this record is not cleared for redistribution. bit bit Quotation withheld because this record is not cleared for redistribution. channel capacity channel capacity Quotation withheld because this record is not cleared for redistribution. span of absolute judgment span of absolute judgment Quotation withheld because this record is not cleared for redistribution. recoding recoding Quotation withheld because this record is not cleared for redistribution. amount of information amount of information Quotation withheld because this record is not cleared for redistribution. mean mean Quotation withheld because this record is not cleared for redistribution. standard-deviation standard-deviation Quotation withheld because this record is not cleared for redistribution. statistical-significance statistical-significance Quotation withheld because this record is not cleared for redistribution. variance variance Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.1037/h0043158.