Interface

Philadelphia Naming Test (PNT)

A 175-item picture-naming test that characterizes naming ability and error patterns.

By Claire White

What does the PNT measure?

The PNT presents 175 line drawings of everyday objects and animals, one at a time, and records the person’s spoken name for each. It measures one thing in depth rather than many things briefly. That thing is lexical access, the step between knowing what you are looking at and retrieving the word for it. The value of the test is less in the count of correct answers than in what the wrong answers look like.

What does a PNT score mean?

Naming accuracy is the number of items named correctly out of 175. The PNT publishes no severity bands, so the count is read as a position on a continuum and tracked against the same person’s earlier scores. The error profile carries the rest of the information. Every response is assigned to one of six types.

The six response types in the Philadelphia Naming Test scoring scheme
Response typeWhat it is
CorrectThe target word, produced within the deadline
SemanticA real word related in meaning to the target
FormalA real word similar in sound to the target
MixedA real word related to the target in both meaning and sound
UnrelatedA real word related to the target in neither meaning nor sound
NonwordA response that is not a real word

The shape of that distribution is what treatment planning and aphasia research use. It is reference information for the care team, not a diagnosis.

How the PNT is administered

Administered by
A clinician or researcher, who advances the pictures and records each response
Session structure
10 practice trials, then 175 items with a 30 second deadline each. A break is allowed after item 88, and the test may be split across two sessions.
Equipment
A computer running the slide presentation, an audio recorder, and the score sheet
Cost
Free to take here.
Published by
Moss Rehabilitation Research Institute
Short forms
Two 30-item forms, and a 30-item computer-adaptive version that runs in under 10 minutes

Which conditions is the PNT used for?

Related instruments

PNT questions, answered

How is the PNT scored?

Every response is assigned to one of six types: correct, semantic, formal, mixed, unrelated, or nonword. The count of correct responses out of 175 gives naming accuracy. The distribution across the five error types gives the error profile, which is the part clinicians and researchers use to reason about where lexical access is breaking down.

How long does the PNT take?

The instructions set a 30 second deadline for each of the 175 items rather than a total time, because how long it runs depends almost entirely on how long the person needs. A break is allowed after item 88, and the test can be split across two sessions. If it is split, that gets noted, because it changes how perseverations are scored.

Is the PNT free?

Yes. It is free to take here. The full test, the administration instructions, the scoring protocol, and a glossary are published by the Moss Rehabilitation Research Institute.

Is there a short version of the PNT?

Three. Two matched 30-item short forms, PNT30-A and PNT30-B, and a 30-item computer-adaptive version that runs in under 10 minutes. All three estimate overall naming accuracy well. They are less suited to building an error profile, where the full 175 items carry information the short forms do not.

What is a normal PNT score?

People without aphasia name nearly all 175 items correctly, which is why the test is scored as a count rather than sorted into severity bands. There are no published severity bands for the PNT. The score is read as a position on a continuum and, more usefully, tracked against the same person's earlier score.

Does the PNT diagnose aphasia?

No. It measures one part of language in depth and produces reference information. Diagnosis stays with your clinician.

Evidence, psychometrics and provenance

Developed in the Language and Aphasia Lab at the Moss Rehabilitation Research Institute. The scoring scheme and its rationale were published by Roach and colleagues in 1996.

Psychometrics

No single test-retest coefficient is published for the full 175-item form. What the literature reports instead is the stability of the accuracy score and of the error profile across repeat administrations.

Test-retest reliability
PNT accuracy was extremely reliable over one week in 25 people with aphasia (Walker and Schwartz, 2012)
Agreement across administrations
Krippendorff's alpha 0.98 for correct responses and 0.79 for formal errors, across two full administrations (Swiderski et al., 2023)
Short forms
Two matched 30-item forms, PNT30-A and PNT30-B, correlate closely with the full test (Walker and Schwartz, 2012)

References

  1. 1.Roach A, Schwartz MF, Martin N, Grewal RS, Brecher A. The Philadelphia Naming Test: scoring and rationale. Clin Aphasiol. 1996;24:121-133.
  2. 2.Walker GM, Schwartz MF. Short-form Philadelphia naming test: rationale and empirical evaluation. Am J Speech Lang Pathol. 2012;21(2):S140-S153.
  3. 3.Swiderski AM, Hula WD, Fergadiotis G. Accuracy of naming error profiles elicited from adaptive short forms of the Philadelphia Naming Test. J Speech Lang Hear Res. 2023;66(4):1351-1364.
  4. 4.Philadelphia Naming Test: administration instructions and materials. Moss Rehabilitation Research Institute, Thomas Jefferson University. Source of the session structure and response deadline.

This assessment uses a validated instrument and is reference information, not a diagnosis.