IS GOOD SPEECH PERCEPTION SUFFICIENT FOR LEARNING NEW WORDS?

Andrea Pittman, PhD CCC-A Associate Professor Arizona State University

A. Pittman Copyright © 2017 All rights reserved. STATEMENT OF SUPPORT

This work was supported by grants from:

Hearing Industry Research Consortium

Arizona Community Foundation Research Assistants (The Pitt Crew) Current: Past: Elizabeth Stewart Ian Odgear Ashley Wright Amanda Willman Jacelyn Olson Molly Hiipakka Lauren Meadows Allison Latto American Speech- Amy Stahl Madalyn Rash Language-Hearing Elizabeth Rainy Ashley Pederson Foundation Brittany Schuett

A. Pittman Copyright © 2017 All rights reserved. Isn’t good speech perception enough?

A. Pittman Copyright © 2017 All rights reserved. CHILDREN VS.

Adults with hearing loss expect amplification to help them restore communication.

Children receive amplification to help them learn to communicate.

A. Pittman Copyright © 2017 All rights reserved. LEARNING NEW INFORMATION

Children School Adults Names (people) Medical terms (condition/medications) Places/locations (addresses) Technical terminology (ipad, iphone, chip) On-the-job training Second-language learners

A. Pittman Copyright © 2017 All rights reserved. DO WE CONSIDER NEW INFORMATION IN OUR EVALUATIONS?

Hearing Handicap Inventory for Adults Scores for social and emotional impact Hearing Handicap Inventory for the Elderly Scores for situational and emotional impact Hearing Handicap Inventory Screening Questionnaire for Adults Score indicating probability of hearing impairment #3 Do you have difficulty hearing/understanding co-workers, clients, customers?

A. Pittman Copyright © 2017 All rights reserved. HOW CAN WE QUANTIFY LEARNING NEW INFORMATION?

Oxford dictionary of American English: 1000+ new entries each year 1. new words 2. new definitions to existing words

A. Pittman Copyright © 2017 All rights reserved. HOW MANY WORDS ARE THERE?

Merriam-Webster Dictionary Database started in the 1880s 15.7 million words

To be included in the database, the new word must be used in a substantial number of citations that come from a wide range of publications over a considerable period of time.

http://www.merriam-webster.com/help/faq-words-into-dictionary A. Pittman Copyright © 2017 All rights reserved. HOW MANY OF THESE WORDS DO WE KNOW?

D’Anna, Zechmeister, & Hall (1991) The average undergraduate student knows between 15,000 and 200,000 words. Their vocabulary test excluded: abbreviations hyphenated words affixes contractions interjections foreign words scientific words technological words slang

D’Anna, C.A., E. B. Zechmeister, and J.W. Hall (1991) Toward a meaningful definition of vocabulary size. Journal of Reading Behavior 23.1: 109–22. A. Pittman Copyright © 2017 All rights reserved. NEW WORDS I LEARNED THIS WEEK

Twentysomething: Someone caught in waithood, waiting for adulthood to happen

Waithood: A debilitating state of helplessness and dependency

Twixter: Someone caught between and adulthood

NEET: Not in , , or training

A. Pittman Copyright © 2017 All rights reserved. WHAT DOES THIS MEAN?

Children have a lot of word-learning to do. 50,000 word vocabulary learned over 18 years (3 to 22 years) = 7 new words everyday

Adults need to update their vocabularies as well. 1,000 new words per year = 3 new words every day

A. Pittman Copyright © 2017 All rights reserved. DO ADULTS UPDATE THEIR VOCABULARIES?

Older adults outperformed younger adults on standardized vocabulary tests (Verhaeghen, 2003).

Flynn Effect (Flynn, 1987): Scores increase with age due to a cohort effect that favors the earlier born.

-50 20 20 70 1960 1970 1980 1990 2000 2010 2020 Vocab Test (1960) Verhaeghen (2003) Aging and Vocabulary Scores: A Meta-Analysis. Psychology and Aging, 18(2), 332–339. Flynn (1987) Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101, 171–191. A. Pittman Copyright © 2017 All rights reserved. UPDATES TO VOCABULARY

Breadth

Henriksen’s (1999) lexical knowledge model:

Depth 1. Size - how many words are known 2. Depth - how well the words are known 3. Mastery - comprehension and production of the words

Ashley Wagner, Editor Oxford Dictionary of American English

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

Word Type # Example

Longstanding word with 15 Aghast: Filled with horror or shock established definition

Longstanding word with Ship: The desire of a fan for two fictional 15 new definition characters to be in a romantic relationship

Senioritis: Affliction of students in their final New word 15 year of high school or college, characterized by a decline in motivation or performance. Nonsense word 5 Desill

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

1. Voluntourism a) To freely offer services in support of a cause. b) The commercial organization of vacations and visits to places of interest. c) To enter into the military service voluntarily. d) A form of tourism in which travelers voluntarily participate in humanitarian work. e) I don’t know.

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

2. Affirmative a) To have an effect on; to make a difference to b) Agreeing with a statement or a request c) Making an assertion d) To move someone emotionally e) I don’t know

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

OVERALL PERFORMANCE 1.0 p>.05

Age Groups n 0.8 20-39 yrs 68 0.6 40-59 yrs 63 0.4 0.46 0.47 0.47

60-74 yrs 13 Proportion Correct 0.2 Total 144

0.0 30 50 70 Generation

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

ESTABLISHED NEW NONSENSE WORDS WORDS 1 p>.05 p=.011 p=.001 0.88 0.87 0.8 0.83

0.6 0.65 0.59 0.52 0.46 0.4 0.42

0.37 Proportion Proportion Correct 0.2

0 30 50 70 30 50 70 30 50 70

Pittman and Stahl (in process) Generation A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

Conclusions: 1. Vocabulary knowledge is generational 2. Younger adults know fewer (or less about) longstanding words than older adults 3. Older adults know fewer new words than younger adults

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. UPDATES TO ADULT VOCABULARY

Implications: 1. To communicate effectively with younger generations, older adults will need to learn new words. 2. Adults who are at risk for learning new words (e.g., hearing loss) may have the greatest difficulty with every-day communication.

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. WHAT ARE THE EFFECTS OF AGE AND HEARING LOSS ON WORD LEARNING?

PITTMAN, STEWART, ODGEAR & WILLMAN (IN REVIEW)

A. Pittman Copyright © 2017 All rights reserved. NEW WORD DETECTION AND LEARNING MODEL

Known Words

no yes

Recognize? Next Word

Integration Configuration Acoustic Pattern Analysis

Stimulus Input

A. Pittman Copyright © 2017 All rights reserved. HYPOTHESIS

If detection and learning of new words is retained in the adult years, then… 1) adults should perform as well as children 2) the effects of hearing loss should have the same effect in both adults and children

A. Pittman Copyright © 2017 All rights reserved. METHODS

Adults: 15 NH (50-67 years) 17 HL (52-78 years)

Children 20 NH (8-12 years) 20 HL (8-12 years)

A. Pittman Copyright © 2017 All rights reserved. OPTIMIZED AMPLIFICATION

Tested with two forms of high-frequency amplification:

Thresholds <80 dB HL at 8 kHz 1) Narrowband (standard) 2) Wideband

A. Pittman Copyright © 2017 All rights reserved. OPTIMIZED AMPLIFICATION

Tested with two forms of high-frequency amplification:

Thresholds >80 dB HL at 8 kHz 1) NLFC off (standard) 2) NLFC on

A. Pittman Copyright © 2017 All rights reserved. WORD LEARNING MODEL

Known Words

no yes

Recognize? Next Word

Integration Configuration Acoustic Pattern Analysis

Stimulus Input

A. Pittman Copyright © 2017 All rights reserved. WORD RECOGNITION BEST PERFORMANCE

100 HIC HIA

80

60

40

20

Frequency (% correct) Amplification Frequency

- High 0 0 20 40 60 80 100 Standard Amplification (% correct) Non-linear frequency compression Bandwidth A. Pittman Copyright © 2017 All rights reserved. WORD RECOGNITION

100 Children Adults 80

60

40

Performance (% correct) correct) (% Performance 20

0 Unaided Aided Normal Hearing Loss Hearing A. Pittman Copyright © 2017 All rights reserved. NEW WORD DETECTION AND LEARNING MODEL

Known Words

no yes

Recognize? Next Word

Integration Configuration Acoustic Pattern Analysis

Stimulus Input

A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION 1.0

Clocks tick on time.

Birds rike long worms.

Dats catch slow bice.

0 1 22 3

A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION BEST PERFORMANCE

100 HIC HIA

80

60

40

Freq Amplification (% correct) (% Amplification Freq -

20 High

0 0 20 40 60 80 100 Standard Amplification (% correct) Non-linear frequency compression Bandwidth A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION 1.0

100 Children Adults 80

60

40

Performance (% correct) correct) (% Performance 20

0 Unaided Aided Normal Hearing Loss Hearing A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION

Conclusions 1. Adults detect new words as well as, or better than, children 2. Like children, hearing loss reduces their ability to detect unknown words 3. Like children, amplification improves their non- word detection

A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION 2.0

Clocks tick on time.

Birds rike long worms.

Dats catch slow bice.

1 2 3 4

Next Pittman, Daliri, et al (in process) A. Pittman Copyright © 2017 All rights reserved. SIGNAL DETECTION THEORY

STIMULUS

Nonsense Real FALSE HIT ALARM

Nonsense (MISPERCEPTION)

CORRECT RESPONSE MISS Real (REPAIR) REJECTION

Pittman, Daliri, et al (in process) A. Pittman Copyright © 2017 All rights reserved. EFFECTS OF HEARING LOSS

Adults Children 5 5

4 4

3 3

2 2

Prime Prime

- - D 1 D 1

0 0

-1 -1 NHA NHC Unaided Unaided -2 -2 0 20 40 60 80 0 20 40 60 80 Binarual Free-Field PTA (.5,1,2,4 kHz) Binaural Free-Field PTA (.5,1,2,4 kHz)

Pittman, Daliri, et al (in process) A. Pittman Copyright © 2017 All rights reserved. EFFECTS OF AMPLIFICATION

Adults Children 5 5

4 4

3 3

2 2

Prime Prime

- - D 1 D 1

0 0

-1 -1 NHA NHC Aided Aided -2 -2 0 20 40 60 80 0 20 40 60 80 Binarual Free-Field PTA (.5,1,2,4 kHz) Binaural Free-Field PTA (.5,1,2,4 kHz)

Pittman, Daliri, et al (in process) A. Pittman Copyright © 2017 All rights reserved. BONE-CONDUCTION COUPLING

2.0

1.5 (d (d prime) 1.0

Performance Performance 0.5

0.0 Softband Direct

Pittman (in process) A. Pittman Copyright © 2017 All rights reserved. COCHLEAR IMPLANT +HEARING AID

2.0

1.5

prime)

- (d 1.0

Performance Performance 0.5

0.0 CI only Bimodal

Pittman & Luo (in process) A. Pittman Copyright © 2017 All rights reserved. NON-WORD DETECTION 3.0

# of nonsense Example words phrase 0 He was short. He was dort. 1 Ve was short. He tuz short. Ve tuz short. 2 He tuz dort. Ve was dort. 3 Ve tuz dort.

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. EFFECTS OF NOISE

4.0

3.5 51 NH adults (19-50 years) 3.0 2.5

Steady-state noise 2.0 dprime 1.5

1.0

0.5

0.0 0 3 6 9 12 15 Q SNR

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. EFFECTS OF NOISE

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3

PROPORTION CORRECT PROPORTION 0.2 NU6 REAL 0.1 NONSENSE 0.0 0 3 6 9 12 15 Q SNR

Pittman and Stahl (in process) A. Pittman Copyright © 2017 All rights reserved. NEW WORD DETECTION AND LEARNING MODEL

Known Words

no yes

Recognize? Next Word

Integration Configuration Acoustic Pattern Analysis

Stimulus Input

A. Pittman Copyright © 2017 All rights reserved. RAPID WORD LEARNING

A. Pittman Copyright © 2017 All rights reserved. RAPID WORD LEARNING

100 −푛/푐 90 푃푐 = 1 − 0.84푒 80 70 60 (% (% Correct) Learning Speed: 50 3 = 1 trial (perfect learning) 40 2 = 10 trials 30 1 = 100 trials PERFORMANCE PERFORMANCE 20 0 = 1000 trials (no learning) 10

0

5

35 85 15 25 45 55 65 75 95

135 105 115 125 145 TRIALS

A. Pittman Copyright © 2017 All rights reserved. WORD LEARNING BEST PERFORMANCE

3 HIC HIA

2

1

Freq Amplification (learningspeed) Amplification Freq

- High 0 0 1 2 3 Standard Amplification (learning speed)

Non-linear frequency compression Wide-band amplification A. Pittman Copyright © 2017 All rights reserved. RAPID WORD LEARNING

3.0 Children Adults 2.5

2.0

1.5

1.0

Performance (learning speed) (learning Performance 0.5

0.0 Unaided Aided Normal Hearing Loss Hearing

A. Pittman Copyright © 2017 All rights reserved. IS RAPID WORD LEARNING REAL WORD LEARNING?

96 NH Adults (19-43 years)

Training Retention Post Test

Wright, Pittman, Wright (in process) A. Pittman Copyright © 2017 All rights reserved. RAPID WORD LEARNING IS REAL WORD LEARNING

100 90 80 70 60 50 40 30

Performance (% correct) Performance 20 Training 10 Retention 0 0 50 100 150 200 250 300 Trials

Wright, Pittman, Wright (in process) A. Pittman Copyright © 2017 All rights reserved. WHAT HAVE WE LEARNED?

1. New word learning is not a perishable skill. Adults are able to learn new words as rapidly as children. 2. Hearing loss impairs/slows word learning equally and significantly in both children and adults. 3. Normal or near-normal performance can be achieved with optimal amplification.

A. Pittman Copyright © 2017 All rights reserved. IS GOOD SPEECH PERCEPTION SUFFICIENT FOR LEARNING NEW WORDS?

3.0 HIC HIA 2.5 NHC NHA 2.0

1.5

LearningSpeed 1.0

0.5

0.0 0 20 40 60 80 100 Word Recognition

A. Pittman Copyright © 2017 All rights reserved. smegs pum fleep gorn soovie nunny grazda morses ranfed smiving hwiking bretty ekleep wattem gorning kello doppers stoobie taybe wrugger rin pomen tuncle mooking zigger winber dwendy sama pasker franding sishing wattem tanti unel manka yosie batrol kissle rathit drave cloums the end loctor higger cougin beels trouzle mayfe kerkle hatty riger thocket prozen tieces transa finter mivshen kather peagie henkred vummy rumpdin pimi ifstid ranfed tryving brenny ricken potton ofli ropim voing bipping samder mither sasepal tifer smiving chamer beefar etote chorket lelling veking geaunen piving poctor kinner kidnesh sefond samaA. Pittmandazi Copyright © 2017ketting All rights reserved. rouple