PSYC 3380_Chapter11 (DONE).pdf

PSYC 3380 – Cognitive Neuroscience Speaking Lecturer: Ezgi Palaz, MSc Speech and Language We transfer ideas from one individual to other with vibration of molecules in the air. In addition to auditory processing we are able to produce, perceive and comprehend speech → langu...

PSYC 3380 – Cognitive Neuroscience Speaking Lecturer: Ezgi Palaz, MSc Speech and Language We transfer ideas from one individual to other with vibration of molecules in the air. In addition to auditory processing we are able to produce, perceive and comprehend speech → language Language is a social engagement. We deduce what others know or not, what they believe etc. Do animals have language? Washoe learned sign language, Kanzi used lexigrams Golden seabrights have different call signals for different conditions Alex the parrot: https://www.youtube.com/watch?v=qGT5JEeOryQ Speech Production vs Comprehension Spoken Word Recognition We match the acoustic form to a stored set of spoken words in vocabulary The storage is called phonological lexicon Matching process is called lexical access What are the access units? ◦ Phonemes? Syllables? Stress patterns? Timing? Access units are debated but the consensus is that speech recognition involves competition between similar sounding words Cohort Model In lexical access, many spoken words are initially considered as candidates, but words get eliminated as more evidence accumulates. For example: The sound “e” activates all words beginning with this sound (cohort of words). As more information is revealed (e.g. “ele”)> fewer possibilities are activated (e.g., “elephant,” “electricity”). Uniqueness point is reached (“eleph”) in which the evidence is consistent with only a single word. Cohort Model Uniqueness point: The point at which the acoustic input unambiguously corresponds to only one known word. Time taken to recognize a word depends on how early or late the uniqueness point occurs Cohort Model Linguistic factors that influence recognition: Frequency of a word (how often it is used in daily life) Imageability of a word (how concretely a word can be imagine: table vs. truth) Words in Context Cohort model explains how a single word is recognized. Words are normally spoken in context of a discourse EEG studies show that in a sentence, out of context words induce a N400 response. «I take my coffee with milk and dog» In contrast, gramatical errors induce P600 response. «I takes my coffee with milks and sugars» Aphasia Disorders of language due to brain damage on the left hemisphere Causes problems in speech perception, speech production, writing Language Specialization Broca’s patient Leborgne had inability to speak intelligently (the repetition of the utterance “tan, tan, tan...” Broca claimed dedicated language center in the brain: Broca’s area Located in the frontal lobe, damage causes problems with fluent speech, speech production: Broca’s aphasia https://www.youtube.com/watch?v=JWC-cVQmEmY Language Specialization Wernicke claimed an area in temporal lobe is also dedicated to language: Wernicke’s area Difficulties comprehending language. No problem with fluent speaking but the speech is not coherent (devoid of content, word salad) https://www.youtube.com/watch?v=3oef68YabD0 Broca’s vs Wernicke’s Sentence Comprehension Words have meaning (semantics) and syntactic roles (grammatical classes such as nouns and verbs). Syntax: The order and structure of the words within a sentence. Enables the listener to figure out who is doing what to whom. Broca’s aphasia is related to agrammatism: “loss of grammar.” Broca’s aphasia patients have trouble when meaning of a sentence depends on syntax (sentences A & B) ◦ A: John pushed James ◦ B: James pushed John ◦ C: John was pushed by James Broca’s Area Two functional sub-divisions of Broca’s area: The posterior division (BA44): related to syntactic complexity The anterior division (BA45): related to WM, meaning Syntax and semantics are separable (N400 vs P600) but not compeletely independent, and Broca’s area can be viewed as an integration site. Retrieval of Spoken Words When producing speech, three types of information need to be retrieved: Lexicalization: the selection of a word based on the meaning that one wishes to convey (related to pragmatics; based on the listener’s knowledge (e.g., “it” vs. “animal”) Grammatical properties (i.e., noun, verb, adjective) Form of the Word (syllables, phonems) Speech Errors Freudian slips: The substitution of one word for another that reflects the speaker’s hidden intentions Malapropisms: A speech error that consists of a word with a similar phonological form to the intended word. → Karagöz Spoonerisms: A speech error in which initial consonants are swapped between words. Tip-of-the-tongue phenomenon: Inability to retrieve the correct word even if a person knows, conceptually, the word that they wish to say Speech Errors Anomia: Word-finding difficulties due to brain damage. Constant state of tip-of-the-tongue phenomenon Proper name anomia: Severe difficulties in retrieving proper names → knows the boyband, can’t name Articulation Final stage of speech Associated with basal ganglia and insula Damage to insula results in apraxia for speech (difficulties in shaping the vocal tract). People with apraxia for speech know what it is that they want to say and have normal muscle tone of the speech articulators, but have problems with the production of consonants, vowels, and prosody. Articulation Final stage of speech Associated with basal ganglia and insula Damage to basal ganglia results in dysarthria (impaired muscular contractions).