Session 11

• kinesthetic feedback

  sense of movement and position in joints & tendons
• internal: proprioceptive feedback


  - information regarding length and tension of muscle and velocity of muscle stretch

  - monitored by muscle spindles
• muscle spindles


  - sensors of muscle movement; delivery info quickly

  - gamma loop (gives feedback) efferent fibers gobackto the muscles

  - take sensation up to brain where gamma loop signals back down if necessary

  - aids in initial acceleration of movement, overshoot, and tension regulation

  - afferent: from center to periphery
• levels of feedback awareness


  - both subconscious (still get info since it's a motor pattern that's learned but not aware of it) and conscious sensations (gets to cortex if sensation is strong enough)
• over-learned pattern


  - speech is an over-learned pattern

  - if a child keeps doing an error incorrectly, it's harder to fix; body will always go to it
• neural innervation of oral sensations


  - trigeminal (v), facial (vii), glossopharyngeal (ix), accessory (ix), hypoglossal (xii)
• internal feedback (IF)


  - delivery of info from brain about motor commands prior to motor response itself

  - faster feedback; when a motor skill is repeated, it becomes faster as energy needed decreases and muscles can anticipate
• purpose of speech models


  - simplify complex phenomena: taking complex information and data and putting into something simpler

  - try to account for data (more it can account for, the stronger the model is)

  - predict experimental results
• closed loop models (associated chain model)


  - where feedback from periphery was essential in order to move through linguistic string; need the feedback before sending down next signal

  - no data to support this model

  - not enough time for feedback to go from the periphery and back again before next segment
• open loop models (cone)


  - syllable or linguistic unit is programmed and sent down as unit (peripheral feedback not needed)

  - accounts for coarticulation
• mixed models (combo)


  - phrase is organized/preprogrammed in the brain, (including suprasegmentals)

  - send down info but also at the periphery can make fine tune adjustments

  - all speech production muscles primed for speech

  - signal sent down to hit certain target area, and gamma feedback loop provides refinement
• intersubject variability


  - across speakers

  - similar to interstate = across states
• intrasubject variability


  - within each speaker
• classes of speech distinctions


  - vowels (easiest to identify)

  - diphthongs (2 vowels)

  - semivowels

  - nasal consonant

  - stop consonants

  - fricative consonants

  - affricate consonants

  - suprasegmental/prosodic features
• vowels


  - longer duration, intensity, & clearer formants

  - voiced, high-intensity

  - identification relies on 1st and 2nd formants

  - formant transitions to and from neighboring sounds
• diphthongs


  - easy to identify: voiced, high-intensity sounds

  - 2 vowels; quick shift from one to the next

  - identification based on rapidly changing formant freq. (rapid vocal tract change)
• semivowels


  - rapidly changing formant frequencies (faster than diphthongs) w vs. j

  - direction of changes in F2 frequency

  - changes in F2 are context dependent (key difference b/w /w/ and /j/)
• /r/ vs. /l/


  - frequency characteristics of F2, F3

  - context dependent changes in F3: falling and/or rising for /r/; level for /l/

  - main difference is with 3rd formant
• nasal consonants (manner)


  - low intensity formants caused by anti-resonances

  - nasal murmur (low F1)

  - nasalization of surrounding vowel

  - F1 very low because of longer tube
• to or from /m/


  - lowest frequency and shortest duraction
• to or from /n/


  - mid frequency and average duration
• to and from /ɜ/


  - highest frequency and longest duration
• where is there more information in a linguistic string?


  - always at the beginning

  - more important as it tells you the most information
• fricatives


  - presence of an aperiodic source of extended duration caused by forcing airflow through constriction

  - higher frequency noise more forward in vocal tract

  - voiced fricatives have 2 sound sources
• stops


  - presence of silent/gap

  - transient burst: /p/ - low freq.; /t/ - high freq.

  - shortest to longest VOT: < 25 msec (b,d,g); > 25 msec (p,t,k)

  - place of articulation moves posterior VOT >

  - context dependent patterns of F2 transitions in neighboring vowels

  - presence of aspiration
• affricates


  - presence of a silent closure interval, transient release burst; rapid rise/fall time
• voice onset time (VOT)


  length of time between the release of a stop consonant and the onset of vocal cord vibration for the following vowel
• which feedback mechanism is used the least for speech

  sent
• what is an essential part of duration


  juncture
• what is the importance of suprasegmentals

  allows for more context and meaning
• what is the lumbar effect


  when you increase your vocal intensity when there is more noise
• what is the purpose of the lumbar effect


  increased vocal intensity so the listener can understand
• what will most feedback form tactile be from?

  lips and alveolor ridge
• what are examples of tactile used in speech


  - cheeks

  - lips

  - tongue

  - alveolar ridge

  - palates
• what does TMJ stand for

  jaw
• what is the main joint for kinesthetic feedback

  tmj (temporomandibular joints)
• what kind of feedback is kinesthetic feedback

  internal feedback
• what is internal feedback important for


  learning a new language
• proprioceptive feedback


  information regarding length and tension of muscles, and velocity of muscle stretch
• proprioceptive word correlation


  muscle