Language Comprehension Flashcards
Prosody
the rhythm, intonation and stress patterns in speech
Invariance problem
refers to the challenge of understanding speech despite variations in how sounds are produced by different speakers, accents, speech rates, background noise, and other factors.
Co-articulation
the tendency for a speech sound to be influenced by sounds preceding or following it
foreign accent syndrome
a rare condition, resulting from brain injury, whereby an individual produces phonetic and prosodic errors in their speech such that it sounds non-native
phonotactic constraints
the language-specific sound groupings that occur in a language
slips of the ear
when we misperceive a word or phrase in speech
categorical perception
refers to the phenomenon where listeners categorize speech sounds into discrete categories, even when the sounds vary along a continuum. This means that listeners perceive differences between sounds within a category more readily than differences between categories. For example, listeners may categorize speech sounds as either /b/ or /p/, with difficulty distinguishing between sounds that fall near the boundary between these categories. Categorical perception helps listeners make sense of the continuous variation in speech sounds, aiding in efficient speech perception and language understanding.
right ear advantage
language sounds are processed more efficiently when presented to the right ear compared to the left
phoneme restoration effect
perceptual phenomenon in speech perception where listeners “hear” missing or obscured phonemes in spoken words, filling in the gaps based on contextual cues. This occurs even when a portion of a word, typically a phoneme, is replaced by a non-speech sound (such as a cough or a tone) or removed altogether.
The McGurk effect
a perceptual phenomenon where what we see influences what we hear. For example, if we see someone saying “ga” while hearing “ba,” we may perceive a blend of the two sounds, such as “da.” This effect demonstrates how visual information can override auditory input in speech perception.
The Cohort Model of speech perception by Liberman
a theory that suggests listeners use contextual information to identify spoken words. When we hear a word, our brain initially generates a set of possible word candidates based on the initial sounds or phonemes of the word. As more sounds are heard, the set of candidates decreases until only one word remains.
TRACE model by McClelland and Elman
provides a computational framework for understanding how listeners decode speech signals and extract meaning from spoken language, taking into account the dynamic interaction between bottom-up sensory input and top-down contextual information.
The model is based on the idea of distributed processing, where information is represented and processed across interconnected nodes or units. It incorporates multiple levels of processing, including feature detection, phoneme recognition, word recognition, and semantic interpretation.
lexical access
refers to the process of retrieving words from memory during language production or comprehension. It involves accessing the mental lexicon, which stores information about words, including their meanings, forms, and grammatical properties.
Two types of models commonly used to explain lexical access are:
- Serial Search
- Autonomous search model
- Parallel access model
serial search model
These models propose that lexical access occurs in a sequential or serial manner, with information about words retrieved one at a time. An example of a serial model is the “logogen” theory, which suggests that words are activated in a serial manner based on their recognition thresholds.
autonomous search model
suggests that lexical access involves an autonomous search process, where words are retrieved based on their unique characteristics without significant interference from other words. This model implies a relatively independent retrieval process for each word.
Parallel models
propose that multiple words or word candidates are activated simultaneously during lexical access, and competition among these activated words determines which word is ultimately selected. These models emphasize the simultaneous processing and interaction of multiple lexical candidates during word retrieval.
Word recognition occurs over several stages
- Orthographical access
- Phonological access
- semantic access
During lexical access, information from all three levels is typically activated simultaneously, allowing for the integration of multiple sources of information to identify and comprehend words. These levels of representation interact dynamically during language processing, contributing to the overall process of lexical access.
