Localisation and Auditory Scene Analysis Flashcards
Localising sound - binaural cues
Binaural - requires comparison of singles in left and right ears and are vital for signalling location of a sound in azimuth (left and right plane)
- interaural time differences (ITD)
- interaural level differences (ILD)
Localising sounds monaural cues
Work with one ear and localise elevation (up and down plane) and distance of sound
- filter properties of pinnae
- intensity and reverberation
Binaural - interaural time differences
Relative time at which sound arrives at two ears depends on its location in azimuth
A sound source thats straight ahead = distance to each ear is the same therefore no difference in time ( immediately behind, in front, above)
Sound source to the right = difference distances therefore sound reaches left ear later than right
What does range of ITD’s depend on?
Speed of sound (constant through air)
Distance between two ears (larger heads = larger ITD’s)
Maximum in humans = 0.6ms
Requires precise signalling of timing - phaselocking therefore most useful for low frequencies and abrupt onset sounds
Binaural - interaural level differences
Relative sound pressure level reaching two ears also depends on location of source in azimuth
- reduction in sound levels for the far ear, due to acoustic shadow created by head
- therefore occurs for high frequency sounds (6000Hz) but not for low (200Hz)
What is the physiology of binaural processing?
Processing of ITD’s and ILD’s start within the brainstem in superior olivary complex (superior olive)
Binaural localisation cues processed by different neurones located in superior olive
- lateral superior olive (LSO) sensitive to ILD’s
- medial superior olive (MSO) neurones sensitive to ITD’s
Strengths and Weaknesses of binaural cues
+ ITD’s (low frequencies) and ILD’s (high frequencies) provide complementary information about azimuth location
- provide ambiguous information about elevation and distance of sound source
Monaural localisation cue - elevation
Elevation:
filter properties of pinnae - sound reflects off parts of external ear, the relative intensity of different frequency sound wave changes - changes sound source elevation (and azimuth)
- individuals different ear shapes and patterns filter frequencies in different ways
- experiments with artificially altering ear shape with plastic moulds impair the ability to localise sound elevation but brain adapts after period of time
Monaural cue - distance
Relative intensity
- sound intensity decreases with distance, closer objects will tend to have greater amplitude than further
Reverberation
- the way in which sound reflects off objects provides cue to distance
- multiple reflectors combine to produce a persistence of sound - reverberation
The distance of source alters relative intensity and timing of direct and reverberant sounds
Reverberation potential problems - precedence effect
Not sure where the actual location of the sound source
Precedence effect - when similar sounds arriving in quick succession from different locations = localised according to direction of first sound - provided the delay is short, only single sound is perceived (less 10-20ms)
Scene analysis in natural environment
The natural environment has multiple sound sources
Auditory system undergoes scene analysis to make sense of sounds:
- segregates components of sound that comes from different sound sources
- group components of sound that come from same sound source
What are the strategies for auditory grouping?
- similar to visual e.g., common fate, proximity
- spectrum grouping = combining different frequency sound components that occur at same time
- sequential grouping = combining sequences of sounds over time
Spectrul grouping
grouped single sound as probably the same sound event
Harmonicity = if component is mistuned to other components it will be heard as separate sound
Common frequency change (common fate) - frequency components that change together tend to be grouped together
Sequential grouping
Auditory stream segregation
- Changing characteristics of sounds can bias us to either group the sequence or segregate into multiple streams
- Similarity of pitch - often similar when from same source. Increasing stream difference promotes stream segregation
- Temporal proximity - sound s that occur in rapid progression tend to be produced by same source
- similarity of timbre grouping
- continuity - sounds that stay constant or change smoothly are often produced same source - perceived as continuous even when interrupted by noise
- phoenic restoration - interrupting speech with silence makes it difficult to understand - gaps containing noise makes it easier to understand