Some computational analyses of the PBK test: Effects of frequency and lexical density on spoken word recognition

Ted A. Meyer, David B. Pisoni

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Objective: The Phonetically Balanced Kindergarten (PBK) Test (Haskins, Reference Note 2) has been used for almost 50 yr to assess spoken word recognition performance in children with hearing impairments. The test originally consisted of four lists of 50 words, but only three of the lists (lists 1, 3, and 4) were considered 'equivalent' enough to be used clinically with children. Our goal was to determine if the lexical properties of the different PBK lists could explain any differences between the three 'equivalent' lists and the fourth PBK list (List 2) that has not been used in clinical testing. Design: Word frequency and lexical neighborhood frequency and density measures were obtained from a computerized database for all of the words on the four lists from the PBK Test as well as the words from a single PB-50 (Egan, 1948) word list. Results: The words in the 'easy' PBK list (List 2) were of higher frequency than the words in the three 'equivalent' lists. Moreover, the lexical neighborhoods of the words on the 'easy' list contained fewer phonetically similar words than the neighborhoods of the words on the other three 'equivalent' lists. Conclusions: It is important for researchers to consider word frequency and lexical neighborhood frequency and density when constructing word lists for testing speech perception. The results of this computational analysis of the PBK Test provide additional support for the proposal that spoken words are recognized 'relationally' in the context of other phonetically similar words in the lexicon. Implications of using open-set word recognition tests with children with hearing impairments are discussed with regard to the specific vocabulary and information processing demands of the PBK Test.

Original languageEnglish (US)
Pages (from-to)363-371
Number of pages9
JournalEar and hearing
Volume20
Issue number4
DOIs
StatePublished - Aug 1 1999

Fingerprint

Hearing Loss
Speech Perception
Vocabulary
Automatic Data Processing
Research Personnel
Databases
Recognition (Psychology)

ASJC Scopus subject areas

  • Otorhinolaryngology
  • Speech and Hearing

Cite this

Some computational analyses of the PBK test : Effects of frequency and lexical density on spoken word recognition. / Meyer, Ted A.; Pisoni, David B.

In: Ear and hearing, Vol. 20, No. 4, 01.08.1999, p. 363-371.

Research output: Contribution to journalArticle

@article{d4a625a662454da19816355cbdc47859,
title = "Some computational analyses of the PBK test: Effects of frequency and lexical density on spoken word recognition",
abstract = "Objective: The Phonetically Balanced Kindergarten (PBK) Test (Haskins, Reference Note 2) has been used for almost 50 yr to assess spoken word recognition performance in children with hearing impairments. The test originally consisted of four lists of 50 words, but only three of the lists (lists 1, 3, and 4) were considered 'equivalent' enough to be used clinically with children. Our goal was to determine if the lexical properties of the different PBK lists could explain any differences between the three 'equivalent' lists and the fourth PBK list (List 2) that has not been used in clinical testing. Design: Word frequency and lexical neighborhood frequency and density measures were obtained from a computerized database for all of the words on the four lists from the PBK Test as well as the words from a single PB-50 (Egan, 1948) word list. Results: The words in the 'easy' PBK list (List 2) were of higher frequency than the words in the three 'equivalent' lists. Moreover, the lexical neighborhoods of the words on the 'easy' list contained fewer phonetically similar words than the neighborhoods of the words on the other three 'equivalent' lists. Conclusions: It is important for researchers to consider word frequency and lexical neighborhood frequency and density when constructing word lists for testing speech perception. The results of this computational analysis of the PBK Test provide additional support for the proposal that spoken words are recognized 'relationally' in the context of other phonetically similar words in the lexicon. Implications of using open-set word recognition tests with children with hearing impairments are discussed with regard to the specific vocabulary and information processing demands of the PBK Test.",
author = "Meyer, {Ted A.} and Pisoni, {David B.}",
year = "1999",
month = "8",
day = "1",
doi = "10.1097/00003446-199908000-00008",
language = "English (US)",
volume = "20",
pages = "363--371",
journal = "Ear and Hearing",
issn = "0196-0202",
publisher = "Lippincott Williams and Wilkins",
number = "4",

}

TY - JOUR

T1 - Some computational analyses of the PBK test

T2 - Effects of frequency and lexical density on spoken word recognition

AU - Meyer, Ted A.

AU - Pisoni, David B.

PY - 1999/8/1

Y1 - 1999/8/1

N2 - Objective: The Phonetically Balanced Kindergarten (PBK) Test (Haskins, Reference Note 2) has been used for almost 50 yr to assess spoken word recognition performance in children with hearing impairments. The test originally consisted of four lists of 50 words, but only three of the lists (lists 1, 3, and 4) were considered 'equivalent' enough to be used clinically with children. Our goal was to determine if the lexical properties of the different PBK lists could explain any differences between the three 'equivalent' lists and the fourth PBK list (List 2) that has not been used in clinical testing. Design: Word frequency and lexical neighborhood frequency and density measures were obtained from a computerized database for all of the words on the four lists from the PBK Test as well as the words from a single PB-50 (Egan, 1948) word list. Results: The words in the 'easy' PBK list (List 2) were of higher frequency than the words in the three 'equivalent' lists. Moreover, the lexical neighborhoods of the words on the 'easy' list contained fewer phonetically similar words than the neighborhoods of the words on the other three 'equivalent' lists. Conclusions: It is important for researchers to consider word frequency and lexical neighborhood frequency and density when constructing word lists for testing speech perception. The results of this computational analysis of the PBK Test provide additional support for the proposal that spoken words are recognized 'relationally' in the context of other phonetically similar words in the lexicon. Implications of using open-set word recognition tests with children with hearing impairments are discussed with regard to the specific vocabulary and information processing demands of the PBK Test.

AB - Objective: The Phonetically Balanced Kindergarten (PBK) Test (Haskins, Reference Note 2) has been used for almost 50 yr to assess spoken word recognition performance in children with hearing impairments. The test originally consisted of four lists of 50 words, but only three of the lists (lists 1, 3, and 4) were considered 'equivalent' enough to be used clinically with children. Our goal was to determine if the lexical properties of the different PBK lists could explain any differences between the three 'equivalent' lists and the fourth PBK list (List 2) that has not been used in clinical testing. Design: Word frequency and lexical neighborhood frequency and density measures were obtained from a computerized database for all of the words on the four lists from the PBK Test as well as the words from a single PB-50 (Egan, 1948) word list. Results: The words in the 'easy' PBK list (List 2) were of higher frequency than the words in the three 'equivalent' lists. Moreover, the lexical neighborhoods of the words on the 'easy' list contained fewer phonetically similar words than the neighborhoods of the words on the other three 'equivalent' lists. Conclusions: It is important for researchers to consider word frequency and lexical neighborhood frequency and density when constructing word lists for testing speech perception. The results of this computational analysis of the PBK Test provide additional support for the proposal that spoken words are recognized 'relationally' in the context of other phonetically similar words in the lexicon. Implications of using open-set word recognition tests with children with hearing impairments are discussed with regard to the specific vocabulary and information processing demands of the PBK Test.

UR - http://www.scopus.com/inward/record.url?scp=0032782082&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032782082&partnerID=8YFLogxK

U2 - 10.1097/00003446-199908000-00008

DO - 10.1097/00003446-199908000-00008

M3 - Article

C2 - 10466571

AN - SCOPUS:0032782082

VL - 20

SP - 363

EP - 371

JO - Ear and Hearing

JF - Ear and Hearing

SN - 0196-0202

IS - 4

ER -