Horley et al. (2010)

January 9, 2017

 

 

ANALYSIS 

Comparison Research

NOTE: The focus of the investigation is on the nature of prosodic disorders. Accordingly, no summary of intervention is included in the review.

 KEY: 

DAT = Dementia of the Alzheimer’s Type

eta = partial eta squared

F0 = fundamental frequency

MLU = mean length of utterance

NA = Not Applicable

P = participant or patient

pmh = Patricia Hargrove, blog developer

SD = standard deviation

SLP = speech-language pathologist

 

 

SOURCE: Horley, K., Reid, A., & Burnham, D. (2010). Emotional prosody perception in Dementia of the Alzheimer’s Type. Journal of Speech, Language, and Hearing Research, 53, 1132-1146.

 

 

REVIEWER(S): pmh

 

DATE: January 8, 2017

 

ASSIGNED GRADE FOR OVERALL QUALITY: The overall quality of the evidence for intervention is not graded because the investigation is concerned with impairment.

 

TAKE AWAY: Speakers with moderate Dementia of the Alzheimer’s Type (DAT) when compared to typically aging (TA) participants (Ps) produced similar average fundamental frequencies (F0) for affective prosody tasks. TA Ps, however, produced significantly more pitch variation (i.e., F0 standard deviations) than did their DAT peers. For rate, the DAT and TA Ps did not differ during a modeling task but Ps with DAT produced significantly slower sentences during a reading task. The ordering of difficultly of producing the targeted emotions was similar for the DAT and TA groups. With respect to perception, Ps with DAT consistently underperformed compared to TA peers.

  

  1. What type of evidence was identified?

                                                                                                           

– What was the type of design? Comparison Research; Prospective, Nonrandomized Group Design with

 

– What was the focus of the research? Clinically Related

                                                                                                           

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

 

                                                                                                           

  1. Group membership determination:

                                                                                                           

  • If there were groups, were participants randomly assigned to groups? No. There could not be random assignment to groups because the groups consisted of TA Ps and Ps who had been diagnosed with DAT.

 

  • If there were groups and Ps were not randomly assigned to groups, were members of groups carefully matched? Yes

                                                                    

 

  1. Were experimental conditions concealed?

                                                                                                           

  • from participants? No

                                                                    

  • from administrators of experimental conditions? No

                                                                    

  • from analyzers/judges? No

                                                                    

 

  1. Were the groups adequately described? Yes

 

– How many participants were involved in the study?

  • total # of Ps: 40
  • # of groups:  2
  • List names of groups and the number of Ps in each group:

     – Dementia of the Alzheimer’s Type (DAT) = 20

– Typically Aging (TA) = 20

 

– Did all groups maintain membership throughout the investigation? Yes

                                                                                

–  CONTROLLED CHARACTERISTICS

                                                                                                                       

  • age:

     – DAT = 71- 92 years, mean = 80 years

     – TA = 70- 84 years, mean = 78 years

  • gender: both groups 10m, 10f
  • expressive language:

– DAT – those with diagnosis of aphasia excluded

     – TA – no noted communication disorders

  • mood disorder: excluded from both groups
  • Socio-economic status: “generally matched” (p. 1135)
  • educational level of participants: “generally matched” (p. 1135)
  • neurological status:

– TA – no known neurological impairment

  • age at referral

 

– DESCRIBED CHARACTERISTICS

  • cognitive skills:

– DAT – diagnosed as DAT

  • language skills:

– DAT – diagnosed as DAT

  • memory:

– DAT – diagnosed as DAT

  • diagnosis:

DAT = Alzheimer’s with Late Onset; majority attended services for people with moderate to severe dementia

  • reading level of participants (Ps): all Ps read at least at the first and second grade level

 

– Were the groups similar? Yes ______  

                                                         

– Were the communication problems adequately described? No

  • disorder type: DAT
  • functional level: moderate DAT

 

 

  1. What were the different conditions for this research?

                                                                                                             

  • Subject (Classification) Groups? Yes

Cognitive Status:

– DAT       

     – TA

                                                               

  • Experimental Conditions? Yes

     – Tasks:

  • Production

               ∞ Modeling

               ∞ Reading

  • Comprehension/Perception

              ∞ Affect Identification (prosody and semantic content)

             ∞ Neutral Content Affect Identification

Emotions:

             ∞ Anger

             ∞ Sadness

             ∞ Surprise

             ∞ Happiness

 

  • Criterion/Descriptive Conditions?

 

 

  1. Were the groups controlled acceptably? Yes

 

 

  1. Were dependent measures appropriate and meaningful? Yes

                                                                                                             

– Dependent measures:

 

  • PRODUCTION TASKS (Modeling and Reading Aloud)
  • OUTCOME #1: Average pitch (i.e., mean fundamental frequency of F0 )of each utterance
  • OUTCOME #2: Pitch modulation (i.e., F0 standard deviation, SD) of each utterance
  • OUTCOME #3: Speaking rate (i.e., syllables per second) of each utterance

 

  • PERCEPTION TASKS (Affect Identification using Prosody and Linguistic Content and Neutral Content Affect Identification)
  • OUTCOME #4: Percentage of correct identification of target emotion

 

OUTCOME #4 (Percentage of correct identification of target emotion) was  subjective.

 

– The outcomes that were objective are

  • OUTCOME #1 [Average pitch (i.e., mean fundamental frequency of F0 )of each utterance]
  • OUTCOME #2 [Pitch modulation (i.e., F0 standard deviation, SD) of each utterance]
  • OUTCOME #3 [Speaking rate (i.e., syllables per second) of each utterance]

 

 

  1. Were reliability measures provided?

                                                                                                            

  • Interobserver for analyzers?

 

  • Intraobserver for analyzers?

 

  • Task administration fidelity for investigators? No

 

 

  1. Description of design:
  • Ps with moderate DAT were matched to TA peers.
  • The investigators administered tasks individually to all Ps. Testing times for the Ps were:

– DAT = 15 to 20 minutes

– TA = 5 to 10 minutes.

  • Although the tasks were presented in a prescribed order (Production Tasks: Modeling, Reading; Perception Tasks: Using Prosody and Linguistic Content for Affect Identification, Neutral Context Affect Identification), the order of sentences within each task as well as order of emotions were counterbalanced across Ps.
  • Each of the tasks explored happiness, anger, surprise, and sadness. The sentence stimuli are provided in the appendixes.
  • Statistical analyses explored differences between genders, Subject Groups (DAT, TA), Emotions (Happiness, Anger, Surprise, Sadness.) Because there were no Group x Gender interactions, gender data were collapsed.

 

 

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

 

  • PRODUCTION TASKS (Modeling and Reading Aloud)

 

  • OUTCOME #1: Average pitch (i.e., mean fundamental frequency of F0 )of each utterance
  • No significant difference for DAT vs TA Groups in reading or modeling production tasks.
  • Both Groups (DAT, TA) produced significantly higher mean F0 for

∞ surprise vs happiness emotions

∞ combined surprise and happiness vs combined anger and sadness emotions

∞ anger vs sadness emotion

 

  • OUTCOME #2: Pitch modulation (i.e., F0 standard deviation, SD) of each utterance
  • Significantly greater for TA compared to DAT. (The investigators noted that for reading p = 0. 502
  • Patterns of pitch modulation were similar for DAT and TA groups.

∞ During the modeling task both groups produced more pitch modulation for anger vs sadness

∞ During the reading task both groups produced more pitch modulation

– for surprise vs happiness

– for combined surprise and happiness vs anger and sadness

 

  • OUTCOME #3: Speaking rate (i.e., syllables per second) of each utterance
  • Ps with DAT spoke at a significantly higher rate during the reading task but not during the modeling task.
  • Overall, for modeling, but not for reading,

∞ happiness was slower than surprise

  • Overall, for both modeling and reading,

∞ combined anger and sadness were slower than combined happiness and surprise

∞ sadness was slower than anger

 

  • PERCEPTION TASKS (Affect Identification using Prosody and Linguistic Content and Neutral Content Affect Identification)

 

  • OUTCOME #4: Percentage of correct identification of target emotion
  • For each of the targeted emotions, Ps with DAT scored significantly more poorly on the task that provided semantic and prosodic cues than did the TA group.
  • On the task that provided only prosodic cues, Ps with DAT scored significantly more poorly on 3 of the 4 targeted emotions than did TA peers.
  • Both groups performed more poorly on the prosody only task than on the prosody plus semantic content task

– What statistical tests were used to determine significance?

  • ANOVA: 2 (Group: DAT, TA) x 4 (EMOTION: happiness, anger, surprise, sadness)
  • Chi Square

 

– Were effect sizes provided? Yes

 

  • OUTCOME #1: Average pitch (i.e., mean fundamental frequency of F0 )of each utterance
  • Both Groups (DAT, TA) produced significantly higher mean F0 for

∞ surprise vs happiness emotions (modeling eta = 0.11, medium effect; reading eta = 0.19, large effect)

∞ combined surprise and happiness vs combined anger and sadness emotions modeling (eta = 0.22, large effect; reading eta = 0.42, large effect)

∞ anger vs sadness emotion modeling (eta = 0.24, large effect; reading eta = 0.18, large effect)

 

  • OUTCOME #2: Pitch modulation (i.e., F0 standard deviation, SD) of each    utterance
  • Significantly greater for TA compared to DAT (modeling eta = 0.16, large effect; reading eta = 0.10, medium effect)
  • Patterns of pitch modulation were similar for DAT and TA groups.

∞ During the modeling task both groups produced more pitch modulation for anger vs sadness (eta = 0.28, large effect)

∞ During the reading task both groups produced more pitch modulation

– for surprise vs happiness (eta = 0.13, medium effect)

– for combined surprise and happiness vs anger and sadness (eta = 0.56, large effect)

 

  • OUTCOME #3: Speaking rate (i.e., syllables per second) of each utterance
  • Ps with DAT spoke at a significantly higher rate during the reading task (eta = 0.39, large effect).
  • Overall, for modeling, but not for reading,

∞ happiness was slower than surprise (eta = 0.39, large effect)

  • Overall, for both modeling and reading,

∞ combined anger and sadness were slower than combined happiness and surprise [eta = 0.13, medium effect (modeling); eta = 0.43, large effect (reading)]

∞ sadness was slower than anger [eta = 0.47, large effect (modeling); eta = 0.46, large effect (reading)]

 

– Were confidence interval (CI) provided? No

 

  1. Summary of correlational results:  NA

 

 

  1. Summary of descriptive results: Qualitative research NA

 

 

  1. Brief summary of clinically relevant results:
  • Speakers with moderate DAT when compared to TA peers produced similar average fundamental frequencies (F0) for affective prosody tasks.

 

  • TA Ps, however, produced significantly more pitch variation (i.e., F0 standard deviations) than did their DAT peers.

 

  • For rate, the productions of DAT and TA Ps did not differ during a modeling task but Ps with DAT produced significantly slower sentences during a reading task.

 

  • The ordering of difficultly of producing the targeted emotions was similar for the DAT and TA groups.

 

  • With respect to perception, Ps with DAT consistently underperformed compared to TA peers.

 

 

ASSIGNED GRADE FOR QUALITY OF EXTERNAL EVIDENCE:   B