Ellis Weismer & Hesketh (1993)

Comparison Research

 

SOURCE:  Ellis Weismer, S., & Hesketh, L. J. (1993). The influence of prosodic and gestural cues on novel word acquisition by children with specific language impairment. Journal of Speech and Hearing Research, 36, 1013-1025.

 

REVIEWER(S): pmh

DATE:  11.15.13

 

ASSIGNED GRADE FOR OVERALL QUALITY:  B- (B+ is the highest possible grade given the design.

 

TAKE AWAY:  This is not an intervention study. Rather it is an investigation of the effectiveness of prosodic and gestural cues on the comprehension and production of words. It provides good evidence that kindergarteners with SLI (at least 1 standard deviation below the mean) and TD children can improve comprehension and production of nonsense nouns when they are presented at slower rates.  They can also improve comprehension of nonsense locative prepositions if they are presented verbally and with gestures.

 

1.  What type of evidence was identified?

                                                                                                           

a.  What was the type of evidence?  Prospective, Nonrandomized Group Design with Controls 

b.  What was the focus of the research? Clinically Related (i.e., not an intervention study but dealt with intervention issues.)      

c.  What was the level of support associated with the type of evidence?  Level = B+

                                                                                                           

2.  Group membership determination:

                                                                                                           

a.  If there were groups, were participants randomly assigned to groups?  No. The groups were typically developing (TD) children and children with specific language impairment (SLI).  All Ps were exposed to all conditions (repeated measures.)

 b.  If there were groups and participants were not randomly assigned to groups, were members of groups carefully matched? N/A

                                                                    

3.  Were the experimental conditions concealed?               

a.  from participants?  No

b.  from administrators of experimental conditions?  No

c.  from analyzers/judges?  No

                                                                    

 

4.  Were the groups adequately described? 

a.         How many participants were involved in the study?

•  total # of participant:  16

•  # of groups: 2

•  # of participants in each group:  8

•  List names of groups:  typically developing/normal language (NL); specific language impairment (SLI)

 

b.  The following variables were described:                               

•  age:  mean SLI = 71.6 months      ; mean NL = 70.6

•  gender:  both groups = 6m, 2f

•  cognitive skills:  both groups = normal nonverbal cognitive skills; however NL was significantly higher than SLI

•  expressive language:  NL (mean percentile on TOLD-2 = 70) was significantly higher than SLI (mean percentile on TOLD-2 = 22.6)

•  receptive language:

–  NL (mean percentile on PPVT-R = 70) was significantly higher than SLI (mean percentile on PPVT-R = 16.1)

     –  NL (mean percentile on TOLD-2 = 64.9) was significantly higher than SLI (mean percentile on TOLD-2 = 23.4)

•  MLU:  NL (mean MLU = 5.7) was significantly higher than SLI (mean MLU = 3.7)

•  Physical, motor, emotional skill:  WNL for both groups

•  Hearing:  WNL for both groups

•  Vision:  WNL for both groups

 

c.   Were the groups similar before intervention began?  No. NL and SLI different on several linguistically related measure, which is expected. They also differed in nonverbal cognitive skills. This was corrected for in the statistical analyses.

                                                         

d.  Were the communication problems adequately described?  Yes

•  disorder type:  (Specific language impairment (at least 1 standard deviation below the mean; language skills were not commensurate with cognitive skills)

•  functional level

 

5.  What were the different conditions for this research?

a.  Subject (Classification) Groups:  SLI and NL

                                                               

b.  Experimental Conditions:        

•  Rate (fast- 5.9 syllable per second; normal- 4.4 syllables per second; slow – 2.8 syllables per second)

•  Stress (neutral; emphatic)

•  Visual (verbal with gestures; verbal without gestures)

 

6.   Were the groups (i.e., 5a) controlled acceptably? Yes

 

7.  Were dependent measures appropriate and meaningful?  Yes

a.  The dependent measures are

Dependent Measure #1:  Number of correct productions of novel words

Dependent Measure #2.  Number of correct responses on a comprehension task using novel words.

b.  The subjective dependent measures are

  Dependent Measure #1:  Number of correct productions of novel words

Dependent Measure #2.  Number of correct responses on a comprehension task using novel words.

c.  None of the the dependent/ outcome measures are objective

 

8.  Were reliability measures provided?

                                                                                                             

a.  Interobserver for analyzers?  Yes.

Dependent Measure #1:  Number of correct productions of novel words—96%

Dependent Measure #2.  Number of correct responses on a comprehension task using novel words—100%

 

b.  Intraobserver for analyzers?  No

 

c.  Treatment fidelity for investigators?  No

 

9.  Statistical design:

 

•  repeated measures:  between subjects = Groups ; within subjects = Rate, Stress, Visual Cues

•  separate MANOVA for each condition (Rate, Stress, Visual) and task/dependent variable (comprehension, production)

 

10.  What were the results of the statistical (inferential) testing?

 

a.  The comparisons that are significant  (p ≤ 0.05) are

•  For Dependent Measure #1:  Number of correct productions of novel words

–  RATE:

1.  NL and SLI groups produced significantly more novel words at slow rate compared to fast rate.

2.  NL and SLI groups produced significantly more novel words at normal rate compared to fast rate

–  STRESS

1.  NL produced significantly more novel target words than SLI

•  Dependent Measure #2.  Number of correct responses on a comprehension task using novel words.

–  RATE

1.  NL and SLI groups comprehended more novel words at slow rate compared to fast rate.

–  VISUAL CUES

1.  NL and SLI groups comprehended more novel words when investigator presented the novel target word with gestures compared to without gestures.

b.  The statistical test used to determine significance for all the comparisons were

•  MANOVA

•  Tukey

c.  Were confidence interval (CI) provided?  No

 

11.  Brief summary of clinically relevant results: 

•  Slowed speaking rate of the investigator resulted improved comprehension and production of novel target words by children who were NL and those who were SLI.

•  Gestures on the part of the investigator resulted improved comprehension of novel target words by children who were NL and those who were SLI.

                                               

ASSIGNED GRADE FOR OVERALL QUALITY OF EXTERNAL EVIDENCE:  B-

 

SUMMARY OF PROCEDURES

 

PURPOSE:  To investigate the influence of rate, stress, and visual cues on comprehension and production of novel words

POPULATION:  kindergarteners with SLI and those who are typically developing.

 

MODALITY TARGETED:  comprehension and production

 

ELEMENTS OF PROSODY USED AS INDEPENDENT VARIABLE: TERVENTION: rate, stress

 

OTHER ASPECTS OF LANGUAGE/COMMUNICATION TARGETED (Dependent variable):  comprehension and production of novel words

DOSAGE:  individual sessions, 40 minute sessions, 3 sessions for the experimental tasks and administration of TOLD-2

 

GENERAL PROCEDURE:

1.  E showed P a toy figure named Sam and claimed that since he was from outer space, they were going to learn some of Sam’s words.

2.  All the target words were nonsense (novel) one-syllable words comprised of early developing sounds.

3.  RATE CONDITION:  E used prerecorded natural female speech producing target sentences at 3 rates:

–  fast- 5.9 syllable per second;

–  normal- 4.4 syllables per second;

–  slow – 2.8 syllables per second.

The target word was an entity.

4.  STRESS CONDITION:  E used rerecorded natural female speech producing target sentences with

–  normal stress patterns or

–  a pattern in which the target novel word was emphatically stressed.

The target word was an entity.

5.  VISUAL:  E presented the sentence stimuli live.

–  In one condition, she presented the sentence only verbally;

–  in the other condition she presented the word verbally with an accompanying gesture that signified a location.

The target word was a location.

6.  E administered the 3 conditions to each of the children multiple times using the following procedure.

•  Exposure Task:  E presented Sam and 3 target novel items/locations.

•  Production Task:  E elicited the production of each of the 3 novel target words.  If correct P received a penny and positive feedback. If wrong, E noted P was wrong but did not use the name of the target word.

•  Comprehension Task:  E added a 4th item to the array of novel items/locations. If correct P received a penny and positive feedback. If wrong, E noted P was wrong but did not use the name of the target word.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: