Hoque (2008)

ANALYSIS GUIDELINES

Comparison Research

 

NOTE: The focus of the investigation is not on intervention. Accordingly, no summary of intervention is included in the review.

 

KEY:

 

ASD = Autism Spectrum Disorder

DS = Down syndrome

EI = early intervention

NA = Not Applicable

P = participant or patient

pmh = Patricia Hargrove, blog developer

SLP = speech-language pathologist

 

 

SOURCE: Hoque, M. E. (2008). Analysis of speech properties of neurotypicals and individuals diagnosed with autism and Down syndrome. Proceedings of the 10th International ACM SIGACCESS Conference on Computers and Accessibility (Assets ’08). ACM, New York, NY, USA, 311-312. ARTICLE: DOI=http://dx.doi.org/10.1145/1414471.1414554

 

REVIEWER(S): pmh

 

DATE: August 28, 2016

 

ASSIGNED GRADE FOR OVERALL QUALITY: Not graded; this investigation was concerned with the nature of prosodic impairments in individuals diagnosed with Autism Spectrum Disorder (ASD) or Down syndrome (DS.)

 

TAKE AWAY: This brief report in conference proceedings of an investigation involving a small number of participants provides preliminary support for differential use of prosody among NT Ps, ASD Ps, and DS Ps. NT Ps exhibited longer speaking turns than ASD and DS peers and they were most likely to produce pause features that corresponded to typical pause strategies. DS Ps produced speech with higher energy values than DS or NT peers. In addition, ASD (but not DS Ps) and NT Ps produced the same number of rising and falling edges in conversation. However, both the ASD and DS Ps differed from the NT peers in their execution of rising and falling edges with NT Ps producing higher maximum. Interestingly, the DS Ps productions of edges also were higher than the ASD Ps.

 

 

  1. What type of evidence was identified?

                                                                                                           

  • What was the type of design? Comparison Research

 

  • 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? Not Applicable (NA)

 

 

  1. Were experimental conditions concealed?

                                                                                                           

  • from participants (Ps)? NA

                                                                    

  • from administrators of experimental conditions? NA

                                                                    

  • from analyzers/judges? NA

                                                                    

 

  1. Were the groups adequately described? No

 

–   How many participants were involved in the study?

  • total # of Ps: 6
  • # of groups: 3
  • List names of groups and the number of Ps in each group:

     – Autism Spectrum Disorder (ASD) = 3

– Down syndrome (DS) = 1

– Neurotypical (NT) = 2

 

  • Did all groups maintain membership throughout the investigation? Yes, there was only one session.

                                                                                

– CONTROLLED CHARACTERISTICS—none described                 

 

– DESCRIBED CHARACTERISTICS

  • education of Ps: All ASD and DS Ps attended a nonprofit school that provided early intervention (EI) services to children with disabilities
  • severity of ASD: mild or moderate

 

– Were the groups similar? Unclear

                                                         

– Were the communication problems adequately described?

 

 

  1. What were the different conditions for this research?

                                                                                                             

– Subject (Classification) Groups? Yes

  • ASD
  • DS
  • NT

                                                               

– Experimental Conditions? No

 

–  Criterion/Descriptive Conditions? No

 

 

  1. Were the groups controlled acceptably?

 

 

  1. Were dependent measures appropriate and meaningful? Yes

                                                                                                             

– DEPENDENT MEASURES:

 

NOTE: The investigator noted that more than 50 features associated with segmental and suprasegmental aspects of speech were analyzed but only the following prosody related features are listed here

 

PITCH FEATURES

  • OUTCOME #1: Minimum pitch
  • OUTCOME #2: Maximum pitch
  • OUTCOME #3: Mean pitch
  • OUTCOME #4: Pitch standard deviation
  • OUTCOME #5: Absolute value of pitch
  • OUTCOME #6: Quantiles (pitch)
  • OUTCOME #7: Ratio between voiced and unvoiced frames

 

INTENSITY FEATURES

  • OUTCOME #8: Minimum intensity
  • OUTCOME #9: Maximum intensity
  • OUTCOME #10: Mean intensity
  • OUTCOME #11: Intensity standard deviation
  • OUTCOME #12: Quantiles (intensity)

 

RATE FEATURE

  • OUTCOME #13: Speaking rate

 

PAUSE FEATURES

  • OUTCOME #14: Number of pauses in an utterance
  • OUTCOME #15: Percent of Unvoiced Frames
  • OUTCOME #16: Number of voice breaks
  • OUTCOME #17: Maximum duration of pauses
  • OUTCOME #18: Average duration of pauses
  • OUTCOME #19: Total duration of pauses

 

DURATION FEATURE

  • OUTCOME #20: Average duration per turn

 

None of the dependent measures were subjective.

 

– All of the dependent measures were objective.

 

 

  1. Were reliability measures provided?

                                                                                                            

  • Interobserver for analyzers?  No

 

  • Intraobserver for analyzers? No

 

  • Treatment or administration fidelity for investigators? No

 

 

  1. Description of design:

 

  • Pairs of Ps engaged in a conversation while seated across from one another a table. Each conversational pair comprised one NT P and either one P with ASD or one P with DS. Each member of the pair was recorded separately.

 

  • The NT P selected a topic and each P’s speech was recorded onto its own channel.

 

  • More than 50 segmental and suprasegmental features were calculated from each of the recorded samples.

 

  • The investigator analyzed the data using feature mining algorithms. These algorithms were used to identify similarities and differences among the three P groups (ASD, DS, NT.)

 

  1. What were the results of the statistical testing?

 

– OUTCOMES (For the most part, the results are reported in feature grouping. The outcomes associated with each feature group are provided to assist in comprehensibility.)

 

PITCH FEATURES—Ps with ASD exhibited lower maximum falling and rising edges than the P with DS or the NT Ps.

  • OUTCOME #1: Minimum pitch
  • OUTCOME #2: Maximum pitch
  • OUTCOME #3: Mean pitch
  • OUTCOME #4: Pitch standard deviation
  • OUTCOME #5: Absolute value of pitch
  • OUTCOME #6: Quantiles (pitch)
  • OUTCOME #7: Ratio between voiced and unvoiced frames

 

INTENSITY FEATURES— Ps with DS exhibited higher intensity features than Ps with ASD or NT Ps.

  • OUTCOME #8: Minimum intensity
  • OUTCOME #9: Maximum intensity
  • OUTCOME #10: Mean intensity
  • OUTCOME #11: Intensity standard deviation
  • OUTCOME #12: Quantiles (intensity)

 

RATE FEATURE

  • OUTCOME #13: Speaking rate

 

PAUSE FEATURES—NT Ps used pauses more appropriately than Ps with either ASD or DS but ASD Ps produced a similar number of pauses as NT peers.

  • OUTCOME #14: Number of pauses in an utterance
  • OUTCOME #15: Percent of Unvoiced Frames
  • OUTCOME #16: Number of voice breaks
  • OUTCOME #17: Maximum duration of pauses
  • OUTCOME #18: Average duration of pauses
  • OUTCOME #19: Total duration of pauses

 

DURATION FEATURE

  • OUTCOME #20: Average duration per turn—The utterance durations of Ps with NT were longer than the durations of Ps with either ASD or DS.

 

PROSODIC FEATURES THAT WERE SIMILAR ACROSS THE GROUPS:

  • Speaking rate
  • Number of pauses per turn
  • Maximum duration of pauses

 

PROSODIC FEATURES THAT WERE DISSIMILAR ACROSS THE GROUPS:

  • Minimum pitch
  • Mean pitch
  • Maximum pitch
  • Mean intensity
  • Maximum intensity
  • Minimum intensity
  • Energy

 

(add additional outcomes as appropriate)

 

– What was the statistical test used to determine significance? Feature Mining Algorithm—Waikato Environment for Knowledge Analysis (WEKA)

 

– Were effect sizes provided? No

 

– 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:
  • On average in conversations, NT Ps exhibited longer speaking turns than ASD and DS peers. The investigator noted that is supports the view that NT speakers tend to dominate conversations with ASD and DS partners.

 

  • NT speakers were most likely to produce pause features that corresponded to typical pause strategies.

 

  • On average in conversations, DS Ps produced speech with higher energy values than DS or NT peers. The investigators posited that this supports characterizations of DS speakers as being easily excited.

 

  • On average in conversations, ASD (but not DS Ps) and NT Ps produced the same number of rising and falling edges in conversation.

 

  • However, both the ASD and DS Ps differed from the NT peers in their execution of rising and falling edges with NT Ps producing higher maximum edges. Interestingly, the DS Ps also were higher than the ASD Ps. This corresponds with previous characterizations of ASD Ps.

 

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: