Boston Naming Test (BNT 1983)

85 drawings, then 60, then a multiple-choice backup

The test began as part of the authors' aphasia work at the Boston Veterans Administration Hospital and Boston University. It drew on earlier naming tests, especially the Oldfield Object Naming Test of 1971, which weighed word frequency, familiarity and visual complexity.⁸ The 60 drawings were chosen from the 85 to span a range of difficulty.⁵

197885 ITEMS

Experimental edition85 simple line drawings; patients were asked to give their common names.¹

198360 ITEMS

The standard formCut to 60 drawings, ordered by word frequency from bed to abacus.⁴

200160 ITEMS

BNT-2Same 60 items, plus a 15-item short form and a multiple-choice version for when naming fails.⁴

DATESThe experimental edition is dated 1976, 1978 or 1982 depending on the source, and the second edition 2000 or 2001.^2,1,8 The 60-item edition was published by Lea & Febiger in 1983; the BNT-2 is published in the US by PRO-ED.^1,6

EXPLORE · ONE PATIENT, DIFFERENT RULES

Same answers, different scores

Different scoring approaches can produce large differences in total scores.¹ Below is a fictional record of 30 items. Tap any item to change how it was answered, then switch the scoring rule and the stopping rule.

What counts as correct

When to stop

S: Named unaidedC: Named after a stimulus cueP: Named only after a phonemic cueF: Not named

Not given (stopped)

Score with these rules15/ 30

Same record, every combination

Unaided·6 = 12Unaided·8 = 14Unaided·∞ = 14Unaided·6 = 15Unaided·8 = 18Unaided·∞ = 18Also·6 = 18Also·8 = 22Also·∞ = 22

Range: 12 to 22 points from one set of answers

A fictional 30-item record, not BNT items or norms. The real test has 60 items, and its own rules on where to start aren't modelled here.

"Houston, we have a problem"

That was the title of a survey of how clinicians actually used the test. Answers varied on which version, which norms and which rules.¹

VERSIONS IN USE (SURVEY)

2001 60-item with multiple choice

20%

15-item short form

10%

Original 85-item (1978)

<3%

Our source excerpt doesn't give the share using the 1983 60-item version. Only slightly more than half followed the published stopping rule.¹

84adults aged 18 to 59 in the original adult norms, alongside children⁵

1,684educated, above-average participants pooled from 14 studies; may overestimate what people with less education should score⁴

24% · 18%of surveyed clinicians used the Mitrushina (1999) and Heaton (1991) norms respectively¹

1,044people (719 healthy, 325 with Alzheimer's) in one study: age, education and gender affected every short form⁵

Where it's used

Alzheimer's diseaseAmong the most frequently used tools in evaluating known or suspected Alzheimer's.²

Aphasia and strokeUsed to identify language disturbance after stroke or other brain injury.³

Primary progressive aphasiaDescribed in a 2023 trial protocol as one of the most sensitive tests for language impairment in semantic-variant PPA.⁷

Spanish speakersNormative data for Latin American Spanish-speaking adults were published in 2015.³

References

[1]Survey results regarding use of the Boston Naming Test: Houston, we have a problem. ResearchGate.researchgate.net

[2]The Boston Naming Test: calculation and utilization of an estimated 60-item score from 30- and 15-item administrations. ResearchGate.researchgate.net

[3]Olabarrieta-Landa, L., et al. (2015). Standard form of the Boston Naming Test: normative data for the Latin American Spanish speaking adult population. NeuroRehabilitation. IOS Press.content.iospress.com

[4]Boston Naming Test. In Encyclopedia of Clinical Neuropsychology. Springer reference entry (institutional mirror).link.springer.com (mirror)

[5]An empirically derived short form of the Boston Naming Test. Archives of Clinical Neuropsychology. ScienceDirect.sciencedirect.com

[6]Boston Naming Test scoring and documentation. BastionGPT (2026).bastiongpt.com

[7]Transcranial magnetic stimulation pilot in primary progressive aphasia: protocol NCT05860647. ClinicalTrials.gov (PDF).cdn.clinicaltrials.gov

[8]Boston Naming Test. Grokipedia (2026).grokipedia.com

HOW TO CITE THIS PAGE

What's Your IQ. (2026). Boston Naming Test (1983). In Historical IQ Tests Archive. Retrieved September 30, 2026, from /en/historical-iq-tests/boston-naming-test-1983

EDITORIAL RECORD

First published2026

Last reviewed30 September 2026

Sources8, all from visible search results

Open questionsExperimental edition date (1976/78/82); BNT-2 year (2000/01); share using the 1983 form

Canonical URLwhats-your-iq.com/en/historical-iq-tests/boston-naming-test-1983

Related historical tests

The mirror image1959

Peabody Picture Vocabulary Test

The PPVT says a word and asks for the matching picture; the BNT shows a picture and asks for the word. One tests understanding, the other retrieval.

InteractiveTake it →

Where BNT results sit1955

Wechsler Adult Intelligence Scale

Naming scores are usually read against broader ability. A full adult battery like the WAIS gives that context.

ReferenceRead more →

Another single-skill test1962

d2 Test of Attention

Like the BNT, the d2 targets one narrow ability. Both are used in clinical trials as focused outcome measures.

InteractiveTake it →

Ageing and norms1981

WAIS-R

The WAIS-R showed how much age and cohort shape scores. Naming scores also vary with age, education and gender, which is why BNT norms matter so much.

ReferenceRead more →

Scoring rules matter1960

Hamilton Depression Rating Scale

Hamilton's scale shows how the way items are summed can mislead. The BNT survey found the same problem with inconsistent scoring rules.

ReferenceRead more →

Part of the Historical IQ Tests Archive. Editorial content under CC BY-SA 4.0. The BNT is copyrighted by its publishers; no drawings are reproduced, and only the first and last item names given in a reference source are mentioned.

BibTeX RIS

From bed to abacus.

85 drawings, then 60, then a multiple-choice backup

Same answers, different scores

"Houston, we have a problem"

Where it's used

References

Related historical tests

Words are one domain. Measure all four.