COM3404- Short Answers #3
HappyRainFrog1
NVC_Chapter11_v9updated.pptxNonverbal Communication
Chapter 11
The Effects of Vocal Cues That Accompany Speech
Definitions
Prosody (term preferred in linguistics)
Paralanguage (term preferred in psychology)
Both terms describe variations in the voice that accompany speech and clarify its meaning
Paralanguage is more inclusive because it includes vocal behavior that is not attached to speech
Definitions (cont’d)
Clarifying and disambiguating verbal message
Pausing between clauses
Stopping between sentences
Distinguishing questions from declarative sentences
Emphasis to clarify meaning:
The dog that I want to buy
The dog that I want to buy
The dog that I want to buy
Definitions (cont’d)
Conveying additional meaning and information:
About the message (joking)
About the speaker (female)
About the speaker’s state (tense)
About the relationship (formal)
Definitions (cont’d)
Paralanguage may stand alone
May contradict words
May convey a parallel message, simultaneous with the words
Paralanguage can also be called voice tone, voice quality, vocal cues
How are Cue Channels Weighed in the Total Message?
What “counts” more when listeners form impressions?
Mehrabian’s early studies (e.g., speakers say positive or negative words with positive or negative tone or facial expression):
Face > voice tone > words
These were artificial experiments
Can often be true, but not always
How are Cue Channels Weighed in the Total Message (cont’d)
Words count the most when the words are the ‘bottom line’
Teacher says “you failed the test”
Doesn’t matter whether tone is positive or negative
How are Cue Channels Weighed in the Total Message (cont’d)
Words also count the most when they are the easiest and most reliable to judge
Words likely to be less ambiguous than nonverbal cues
Example: When guessing someone’s thoughts and feelings, words have a lot of the information you need to be accurate so you attend to them first
How are Cue Channels Weighed in the Total Message (cont’d)
Nonverbal cues may matter more if encoder is trying to lie or hide something or if there is a big discrepancy between verbal and nonverbal
Nonverbal cues can convey ‘leakage’: information the encoder did not want to convey
Generally, verbal and nonverbal work together more than one ‘counting’ more than the other
The Ingredients and Methods of Studying Paralanguage
Acoustic properties
Speech rate
Fundamental frequency (pitch)
Intensity (loudness)
Rhythm, contour, variation (in the above)
Non-words – mmm, uhh
Non-sounds – pauses between words and phrases; speech latency (switching pause)
Other phenomena – accents, nonfluencies (speech errors), laughing, sighing, interrupting
Content-Masked Speech
How to separate verbal from vocal information?
Standard content (alphabet, standard passage) – only works for certain situations
Foreign language
Random splicing
Electronic filtering (band-pass filters)
Choosing a Level of Analysis
Acoustic measurements
Example: fundamental frequency
Mid-level descriptions (ratings)
Examples: nasality, breathiness, expressiveness
Global impressions
Examples: angry, bored, sexy
Vocal Cues and Speaker Recognition
How do we recognize speakers
Listening
Visual comparison of spectrograms (plot of vocal energy in different frequency bands)
Recognition by computers
Listening is very accurate
But depends on various factors
Familiarity, delay, duration, disguised voices, distinctiveness, dialect
Vocal Cues and Speaker Recognition (cont’d)
A more objective method of speaker identification involves the spectrogram (AKA voiceprint)
A visual picture of a persons speech, similar to a fingerprint
Unlike fingerprints, can show large variability from one measurement to another, and can appear very similar
Not yet reliable enough to be used in trails
Vocal Cues and Personality
Extraversion
More fluency, faster rate, louder speech, more dynamic contrast, higher pitch, more variable pitch
Masculinity
Less expressive, lower pitches, slower, louder voices, poor enunciation
Vocal Cues and Personality (cont’d)
Type A (‘coronary-prone personality’)
Fast speech, uneven speech rate, short latencies, interruptions, loud/explosive voice, hard or staccato voice
Attractiveness
More resonant, less monotonous, less nasal, lower in pitch (in men)
Dominance
Loudness, interruptions, lower pitch
Vocal Cues and Group Perceptions
People tend to evaluate speech samples across the following dimensions
Sociointellectual status - high or low social status, blue or white collar, rich or poor, and literate or illiterate
Aesthetic quality - pleasing or displeasing, nice or awful, sweet or sour, beautiful or ugly
Dynamism - aggressive or unaggressive, active or passive, strong or weak, loud or soft
Vocal Cues and Sociodemographic Characteristics
Gender
Fundamental frequency is an important acoustic cue in speaker gender identification
Women speak somewhat more tentatively than men
Age
Age is judged quite accurately from the voice
Social Class/Status/Dominance
Speaker’s status can be identified quite quickly from the voice
Those high in dominance generally speak more loudly and have a less variable pitch
Other-Directed Speech
Baby talk (“motherese”) directed to babies and pets
Slow, sing-song, simple, repetitive, high
Baby talk directed to baby-faced children
Baby talk directed to the elderly or impaired (AKA elderspeak)
Vocal Cues and Emotion
In animals and birds, vocal cues convey important information
Identity, relationship, alarm, territory, physical states, emotions
In humans, many emotions can be judged from vocal cues
Vocal Cues and Emotion (cont’d)
How are emotions conveyed through the voice?
Evidence based on analysis of actual voices, synthesized tones, and music: converging results
Vocal Cues and Emotion (cont’d)
Examples
Joy/elation: higher pitch and pitch range, higher loudness, faster rate
Sadness/depression: lower pitch, lower loudness, downward contours, slower rate, longer pauses
State anxiety: higher pitch, tremor, ‘non-ah’, speech dysfluencies (sentence change, repetition, stutter, incoherent sounds, etc.)
(‘Ah’ speech dysfluencies don’t indicate anxiety)
Impact of Vocal Quality
Physician and therapist studies
Angry tone predicts less therapy success
Bored tone correlated with less clinical competence
Warm and supportive doctor tone correlated with more patient satisfaction
Dominant and low-anxious doctor tone correlated with being sued more (surgeons)
Impact of Vocal Quality (cont’d)
Workplace studies
Vocal empathy and enthusiasm correlated with better ratings of sales effectiveness by superiors
Faster speech, more pitch variability, fewer pauses, and lower pitch correlated with better performance ratings by superiors
Vocal Cues, Comprehension, and Persuasion
Good delivery matters (for retention, persuasion, credibility)
Vocal variety is good (not monotone)
Avoid excessive dysfluencies
Persuasion
More speech loudness, less halting speech, more pitch variation
Faster speech: Why? Possibilities: listener attributes competence; listener too busy to develop counterarguments; listener too distracted to develop counterarguments
Vocal Cues and Turn Taking in Conversations
Turn Yielding
Questioning tone at end of utterance
Drawl
Lower pitch
Pause
Filler-trail off (so, ah, you know, like)
Vocal Cues and Turn Taking in Conversations (cont’d)
Turn Requesting
Audible intake of breath
Interrupt
Stutter start
Fast back-channel responses (uh-huh, yeah)
Vocal Cues and Turn Taking in Conversations (cont’d)
Turn Maintaining
Increased speech rate
Decreased silent pauses
Increased volume
Increased frequency of filled pauses (‘ah’ speech dysfluencies)
Vocal Cues and Turn Taking in Conversations (cont’d)
Turn Denying
Increased back channel responses
Silence
Hesitations, Pauses, Silence and Speech
Location of pauses
Grammatical, such as between clauses (aural punctuation)
Nongrammatical (some other psychological function)
Types of pauses
Filled pause (uh, um, ah)
Unfilled (silent) pause
Before speaking (=speech latency)
During one’s speech
Mimicking and Reciprocity in the Vocal Channel
Emotional contagion via the voice
(‘catching’ the other person’s emotion)
Length of utterance influenced by other’s vocal and other behaviors
Length of utterance
Head nods
Mm-hmms
Length of pauses influenced by other’s length of pauses
