Limits...
Differential activation of human core, non-core and auditory-related cortex during speech categorization tasks as revealed by intracranial recordings.

Steinschneider M, Nourski KV, Rhone AE, Kawasaki H, Oya H, Howard MA - Front Neurosci (2014)

Bottom Line: While non-primary auditory cortex located on the posterolateral superior temporal gyrus (PLST) is clearly involved in acoustic-to-phonetic pre-lexical representations, it is unclear what role this region plays in auditory object formation.We found that activity in posteromedial HG and early activity on PLST was robust to sound stimuli regardless of their context, and minimally modulated by tasks.Activity in prefrontal cortex appears directly involved in word object selection.

View Article: PubMed Central - PubMed

Affiliation: Departments of Neurology and Neuroscience, Albert Einstein College of Medicine Bronx, NY, USA.

ABSTRACT
Speech perception requires that sounds be transformed into speech-related objects with lexical and semantic meaning. It is unclear at what level in the auditory pathways this transformation emerges. Primary auditory cortex has been implicated in both representation of acoustic sound attributes and sound objects. While non-primary auditory cortex located on the posterolateral superior temporal gyrus (PLST) is clearly involved in acoustic-to-phonetic pre-lexical representations, it is unclear what role this region plays in auditory object formation. Additional data support the importance of prefrontal cortex in the formation of auditory objects, while other data would implicate this region in auditory object selection. To help clarify the respective roles of auditory and auditory-related cortex in the formation and selection of auditory objects, we examined high gamma activity simultaneously recorded directly from Heschl's gyrus (HG), PLST and prefrontal cortex, while subjects performed auditory semantic detection tasks. Subjects were patients undergoing evaluation for treatment of medically intractable epilepsy. We found that activity in posteromedial HG and early activity on PLST was robust to sound stimuli regardless of their context, and minimally modulated by tasks. Later activity on PLST could be strongly modulated by semantic context, but not by behavioral performance. Activity within prefrontal cortex also was related to semantic context, and did co-vary with behavior. We propose that activity in posteromedial HG and early activity on PLST primarily reflect the representation of spectrotemporal sound attributes. Later activity on PLST represents a pre-lexical processing stage and is an intermediate step in the formation of word objects. Activity in prefrontal cortex appears directly involved in word object selection. The roles of other auditory and auditory-related cortical areas in the formation of word objects remain to be explored.

No MeSH data available.


Related in: MedlinePlus

Task effects on responses to speech stimuli in HG. Responses to three types of stimuli (female voices, animals, numbers; left, middle and right column, respectively) are shown for three representative recording sites in HG (rows). See Figure 1A for location of the recording sites. Colors (blue, green, and red) represent different task conditions. Lines and shaded areas represent mean high gamma ERBP and its standard error, respectively. Purple bars denote time windows where responses to the target stimuli were significantly larger than those to the same stimuli in the tones task (q < 0.01). Horizontal box plots denote the timing of behavioral responses to the target stimuli (medians, 10th, 25th, 75th, and 90th percentiles).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4128221&req=5

Figure 2: Task effects on responses to speech stimuli in HG. Responses to three types of stimuli (female voices, animals, numbers; left, middle and right column, respectively) are shown for three representative recording sites in HG (rows). See Figure 1A for location of the recording sites. Colors (blue, green, and red) represent different task conditions. Lines and shaded areas represent mean high gamma ERBP and its standard error, respectively. Purple bars denote time windows where responses to the target stimuli were significantly larger than those to the same stimuli in the tones task (q < 0.01). Horizontal box plots denote the timing of behavioral responses to the target stimuli (medians, 10th, 25th, 75th, and 90th percentiles).

Mentions: Whereas activity along most of HG was strongly modulated by the acoustic attributes of the sounds, responses in the high gamma range were only weakly affected by the target detection tasks (Figure 2). The left column in Figure 2 compares neural activity to the same set of stimuli (female voices) in three blocks: when they were targets, when they were non-targets in the tone detection block, or when they were non-targets in a semantic task (numbers). A low-amplitude increase in high gamma was seen beginning within 600–650 ms after stimulus onset when female voices were the targets [site (a)], overlapping in time with the subject's behavioral response. A similar effect was seen for responses to the animals and numbers when they were the targets. However, the onset of the task-related high gamma modulation in these semantic categorization conditions was even slower than that occurring during voice identification task (q < 0.01 at 750–800 ms after stimulus onset; middle and right columns of Figure 2).


Differential activation of human core, non-core and auditory-related cortex during speech categorization tasks as revealed by intracranial recordings.

Steinschneider M, Nourski KV, Rhone AE, Kawasaki H, Oya H, Howard MA - Front Neurosci (2014)

Task effects on responses to speech stimuli in HG. Responses to three types of stimuli (female voices, animals, numbers; left, middle and right column, respectively) are shown for three representative recording sites in HG (rows). See Figure 1A for location of the recording sites. Colors (blue, green, and red) represent different task conditions. Lines and shaded areas represent mean high gamma ERBP and its standard error, respectively. Purple bars denote time windows where responses to the target stimuli were significantly larger than those to the same stimuli in the tones task (q < 0.01). Horizontal box plots denote the timing of behavioral responses to the target stimuli (medians, 10th, 25th, 75th, and 90th percentiles).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4128221&req=5

Figure 2: Task effects on responses to speech stimuli in HG. Responses to three types of stimuli (female voices, animals, numbers; left, middle and right column, respectively) are shown for three representative recording sites in HG (rows). See Figure 1A for location of the recording sites. Colors (blue, green, and red) represent different task conditions. Lines and shaded areas represent mean high gamma ERBP and its standard error, respectively. Purple bars denote time windows where responses to the target stimuli were significantly larger than those to the same stimuli in the tones task (q < 0.01). Horizontal box plots denote the timing of behavioral responses to the target stimuli (medians, 10th, 25th, 75th, and 90th percentiles).
Mentions: Whereas activity along most of HG was strongly modulated by the acoustic attributes of the sounds, responses in the high gamma range were only weakly affected by the target detection tasks (Figure 2). The left column in Figure 2 compares neural activity to the same set of stimuli (female voices) in three blocks: when they were targets, when they were non-targets in the tone detection block, or when they were non-targets in a semantic task (numbers). A low-amplitude increase in high gamma was seen beginning within 600–650 ms after stimulus onset when female voices were the targets [site (a)], overlapping in time with the subject's behavioral response. A similar effect was seen for responses to the animals and numbers when they were the targets. However, the onset of the task-related high gamma modulation in these semantic categorization conditions was even slower than that occurring during voice identification task (q < 0.01 at 750–800 ms after stimulus onset; middle and right columns of Figure 2).

Bottom Line: While non-primary auditory cortex located on the posterolateral superior temporal gyrus (PLST) is clearly involved in acoustic-to-phonetic pre-lexical representations, it is unclear what role this region plays in auditory object formation.We found that activity in posteromedial HG and early activity on PLST was robust to sound stimuli regardless of their context, and minimally modulated by tasks.Activity in prefrontal cortex appears directly involved in word object selection.

View Article: PubMed Central - PubMed

Affiliation: Departments of Neurology and Neuroscience, Albert Einstein College of Medicine Bronx, NY, USA.

ABSTRACT
Speech perception requires that sounds be transformed into speech-related objects with lexical and semantic meaning. It is unclear at what level in the auditory pathways this transformation emerges. Primary auditory cortex has been implicated in both representation of acoustic sound attributes and sound objects. While non-primary auditory cortex located on the posterolateral superior temporal gyrus (PLST) is clearly involved in acoustic-to-phonetic pre-lexical representations, it is unclear what role this region plays in auditory object formation. Additional data support the importance of prefrontal cortex in the formation of auditory objects, while other data would implicate this region in auditory object selection. To help clarify the respective roles of auditory and auditory-related cortex in the formation and selection of auditory objects, we examined high gamma activity simultaneously recorded directly from Heschl's gyrus (HG), PLST and prefrontal cortex, while subjects performed auditory semantic detection tasks. Subjects were patients undergoing evaluation for treatment of medically intractable epilepsy. We found that activity in posteromedial HG and early activity on PLST was robust to sound stimuli regardless of their context, and minimally modulated by tasks. Later activity on PLST could be strongly modulated by semantic context, but not by behavioral performance. Activity within prefrontal cortex also was related to semantic context, and did co-vary with behavior. We propose that activity in posteromedial HG and early activity on PLST primarily reflect the representation of spectrotemporal sound attributes. Later activity on PLST represents a pre-lexical processing stage and is an intermediate step in the formation of word objects. Activity in prefrontal cortex appears directly involved in word object selection. The roles of other auditory and auditory-related cortical areas in the formation of word objects remain to be explored.

No MeSH data available.


Related in: MedlinePlus