Embodiment - language and reading Flashcards
how are word meanings represented?
represented as features, usually described in terms of other words
- if you wanted to know the meaning of the a word you would have to look it up in a mental dictionary. The meaning of the word woyld be described to you in term of other words
What is meant by the idea that linguistic meaning is grounded in bodily activity?
e.g. “you take your dog for a walk”, the comprehender might form a visual representation or mental image of the dog, or a somatosensory representation of the dog tugging on a lead, or a motor representation of taking a dog for a walk, you learn to associate these words with particular bodily experience
what perceptual features are represented?
- visual orientation (Stanfield & Zwaan, 2001)
- Object shape (Zwaan, Stanfield & Yaxley, 2002)
- Visibility (Yaxley & Zwaan, 2007)
How do people mentally represent? (Stanfield & Swaan, 2001)
if we consider the sentence “The pencil is in the cup” we assume the pencil is vertice. This is because ppts should mentally represent the orientation of the pencil based on experiences.
- in this study the picture matched the orientation implied by sentence 50% of the time
- Results showed that ppts were faster to respond and say that the object had been mentioned in the sentence when it matched the orientation
Object shape (Zwaan, Stanfield, & Yaxley, 2002)
- study used a sentence-picture verification task
- ppts would read a sentence such as ‘Mary saw the egg in an eggbox’
- ”. If they have created a mental simulation of an egg in an eggbox while reading this sentence, then their mental image should be of an egg still in its shell
- The image shown either matched their mental image or did not e.g. a cracked egg instead
- results showed that ppts were faster to indicate that the object had been mentioned in the sentence matched than when the condition in which it did not. Zwaan et al. took this as evidence that participants had mentally represented the shape of the object
Visibility (Yaxley & Zwaan, 2007)
- In this study, participants performed a sentence–picture verification task in which they read sentences about a person viewing an object (e.g., a moose) in different conditions of visibility (e.g., clean vs. fogged goggles).
- Participants then verified whether a subsequently pictured object was mentioned in the previous sentence.
- The time that participants took to do this was shorter when the resolution of the pictured object and the resolution implied by the sentence matched than when they did not.
- These results suggest that the degree of visibility implied in linguistic context can influence object recognition.
- These data suggest that readers mentally simulate the visibility of objects during language comprehension.
- The simulation of linguistic descriptions is not limited to the activation of intrinsic object properties (e.g., object shape), but also invokes the perceptibility of objects that is implied by the environmental context.
Eye-tracking (Wassenburg & Zwaan, 2010)
A study conducted by Wassenburg and Zwaan (2010) demonstrated that prior exposure to pictures affects eye movements during later reading, even though picture exposure was incidental to the reading task.
- Phase 1: Word-picture verification task = e.g. ps shown a picture and had to say if the word correctly described it
- Phase 2: Filler task (mental rotation task – 15-20 mins)
- Phase 3: eye-tracking experiment. ps eyes were monitored when they read sentences which mentioned the objects that were used in Phase 1
- Results: suggests that previous exposure to a picture of an object in a particular orientation affected reading times for phrases that described that particular object.
ERPs (Coppens, Gootjes, & Zwaan, 2012)
Coppens et al. then conducted a subsequent study, in which they examined object shape rather than object orientation, and used ERPs rather than eye-tracking
- Expt 1: Word-picture verification task
- Expt 2: Emotional word stroop task (15 mins)
- Expt 3: Sentences were presented word-by-word while EEG signals were being recorded.
- Results: There was a spike in the electrical activity produced by the brain when the shape of the object that is implied by the sentence does not match the shape of the object that was previously seen. This again provides evidence for an interaction between visual processing and language processing
fMRI study (Hauk, Johnsrude, & Pullvermüller, 2004)
Hauk et al. investigated whether action words would also activate motor areas of the brain.
- participants completed a ‘localizer’ task, in which they performed actions with their feet, fingers, or tongue, while they were in the MRI scanner, so that the researchers could localize brain areas that were activated when participants performed these actions.
- Participants then performed a passive reading task where they were presented with words referring to face, arm, or leg actions, such as ‘lick’, ‘pick’, or ‘kick’.
- Results showed overlap in the brain regions that were activated by actual finger movements and those that were activated for arm-related words
- overlap between brain regions that were activated by foot movements and those that were activated by leg-related words
- These results demonstrate that the reading of words referring to actions performed with different body parts activates the motor and premotor cortex regions associated with actually performing those actions.
Action words in context (Raposo et al. 2009
Raposo et al. also conducted an fMRI study but extended the work of Hauk et al. to look at action words in sentence contexts, and in contexts in which the word does not actually describe a physical action, as in idioms.
- Then during the experiment, participants were presented with sentences that described an actual action being performed with the hands or feet, such as “The fruit cake was the last one, so Claire grabbed it”, or “The muddy children trampled over Sarah’s clean floor”.
- They were also presented with sentences that contained the same action words, but because of the idiomatic nature of the context, no actual action was performed. Examples would be “The job offer was a great chance to Claire grabbed it
- In addition, they were presented with action words in isolation
- If action words automatically activate their motor properties irrespective of their context, as may be assumed on a “fire together wire together” view, then they expected to observe activation in the motor/premotor cortex for all three conditions – that is, action verbs in isolation, and both literal and idiomatic sentences
- . If, however, the meaning of words is modulated by the sentential context, they expected the neural processing of verbs to vary depending on the context.
- Raposo et el. found significant activation in motor regions when action verbs were presented in isolation, and, to a lesser extent, in literal sentence contexts
Action-sentence compatibility effect Glenberg & Kaschak (2002)
- examined sentences which implied directional hand movements, such as “You handed Courtney the notebook”, or “Courtney handed the notebook to you”.
- They proposed that such action sentences are understood by the reader mentally simulating the actions of giving or receiving a notebook
- participants were facing three buttons, which were arranged in line away from them and in front of them. A participant’s finger was placed on the centre button at the start of each trial. When a string of words was presented, participants simply had to judge whether the string made a grammatical sentence or not (50/50)
- Results showed that participants were faster to move their hands in the direction that was implied in the sentence.
Attempts to replicate action-sentence compatibility effects across 18 labs (Morey et al. (2022))
- Used adapted version of Glenberg and Kaschak’s (2002) task.
- Effects failed to replicate across all 18 attempts.
Action-sentence compatibility effect (Zwaan, Taylor, & de Boer,2010)
- Zwaan et al. examined sentences which were presented embedded within a continuous narrative about a bank robbery, with the aim of making more natural-sounding stimuli.
- participants made the story appear phrase-by-phrase by turning a dial.
- Half of the participants advanced the story by turning the dial clockwise, and half by turning it anticlockwise.
- In half of the critical sentences, there was a description of an action being carried out that required a manual rotation action (e.g., starting a car, locking a door, turning up the volume on a police radio), and half depicted someone intending to act in such ways, such as “He wanted to start the car.”
- Results showed action-sentence compatibility effects for sentences that described actual actions
Body specificity hypothesis (Casasanto, 2009)
According to theories of embodied cognition (e.g. Barsalou, 1999), thoughts comprise mental simulations of bodily experiences. If this is true:
- People with different kinds of bodies must represent language differently.
- People with different bodily characteristics, who interact with their physical environments in systematically different ways, should form correspondingly different mental representations.
Handedness effects
Brain activity when participants imagine performing actions with their hands:
- an fMRI study conducted by Willems et al. showed that when participants imagined performing actions that are usually performed with the dominant hand (such as writing), for right-handed people, you get more activity in the left hemisphere of the brain, whereas for left-handed people you get more activity in the right hemisphere
- Similar activity observed when reading action verbs (Willems, Hagoort, & Casasanto, 2010).
