4 - Cognition & Emotion Flashcards
What are the dimensions of the two-dimensional state-space / emotional grid?
Valence: positive/pleasant & negative/unpleasant.
Arousal: calm & aroused.
What did Bradley et al. (1992) show about arousal-valence stimuli and memory encoding?
- It is valence or arousal? R/ship with memory.
- Showed subjects a large assortment of emotional and neutral pictures.
- rated the pictures along the dimensions of valence and arousal.
- An unexpected free-recall test was administered both immediately afterwards and at one year following the rating sessions.
- Immediate & delayed recall was better for pics rated as high on arousal, regardless of valence.
- Recognition memory performance was also faster for arousal stimuli, regardless of valence.
Takeaway: arousal functions as a kind of elaborative encoding - events assoc with high arousal are likely to be important for survival.
What did the emotional Stroop task show about a potential attentional bias of people with anxiety?
The key finding is that anxious (trait) individuals show more interference than nonanxious individuals when naming the colour of the ink that threat-related words (e.g., danger, kill) are printed in relative to neutral words.
How did McLeod et al.’s (1986) dot-probe task explore attentional biases in anxious and depressed individuals?
Design:
- Pts asked to locate dot quickly after presented with a fixation cross > 2 words presented simultaneously > dot (either upper or lower section of the screen)
- Congruent trial: dot and negative word same spot.
- Incongruent trial: dot and neutral word same spot.
- possible to determine whether visual attention had shifted toward or away from the emotionally threatening stimulus based on RT.
Takeaway:
- participants who score highly on measures of state or trait anxiety and clinically anxious patients show an attentional bias towards threat-related material.
- Faster when threat & dot are congruent (not shifting attention)
- Slow when threat & dot are incongruent (shifting attention)
What is the general structure of a semantic network model of semantic knowledge?
- Knowledge represented in a hierarchical semantic network of interconnected elements (nodes).
- Distance b/w the nodes represents similarity b/w the items.
- Definition of a concept is in terms of its connections with other concepts.
- ‘superordinate node’ (living thing) > ‘subordinate’ node (plant) > etc.
Quillian (1968) refers to the cognitive economy of his semantic network model of knowledge - where category membership at each level of the hierarchy entails a number of properties that are shared by the members of the more specific categories. Why did is this model “economic”? give an example starting from the ‘living thing’ superordinate category.
Storing and retrieving information based on a hierarchical structure is more efficient if you do not need to store every detail for everyone single object. Rather, you can store shared features as one node.
Example: Animals > Birds > Canarys.
The properties shared by all animals could be stored only once, at the concept of animal, e.g. moves & breathes. The properties shared by all birds stored only once, e.g. wings, beaks. And then properties unique to canaries stored once also.
Then, by consulting stored taxonomic relations, a subject could determine that more particular types of animals (e.g., canaries) should inherit the properties of animals.
What is the mechanism of spreading activation in the semantic network of knowledge (Collins & Quillian)? What implications does this idea have on new knowledge?
- Spreading activation of one category representation spreads to taxonomically superordinate concepts & vice versa - i.e. the object is categorised as ‘canary’, activation spreads to bird & animal, along with properties stored at each concept.
- new facts could be stored with the appropriate category node, and would then automatically be inherited by all subordinate concepts.
- Similarly, a new sub-category, such as a new species of fish, could be added as a subordinate of the more general category fish, and existing knowledge about the properties of fish would automatically generalise to it
How does semantic priming effect provide evidence for spreading activation? - Nurse-Doctor Priming. (Meyer & Schvaneveldt, 1971)
- Lexical decision task - respond ‘yes’ if both strings are words or otherwise ‘no’.
- Pairs were two related words, to unrelated, or word & non-word.
- The critical comparison was between trials in which pairs of words were either related or non-related in meaning (but matched in length and frequency).
- RESULT: RTs to respond “word” were faster when the two words were related in meaning than when they were not related in meaning.
- Takeaway: semantic priming effect was argued to reflect the automatic process of spreading activation between related concepts in a semantic memory network.
Describe Bower’s Semantic Network Theory of Emotion & Cognition in reference to emotional nodes, inhibitory connections & spreading activation.
- emotions are represented as nodes in a network.
- Emotion nodes (ENs) connected to related concepts, words, events, autonomic responses & other emotions, e.g. Happy — Sad.
- Inhibitory connections between opposing emotions, e.g. Happy activated would inhibit sad.
- Spreading activation from ENs to associated semantic nodes, e.g. Happy –> smile.
What evidence did Bower provide of the Mood Congruent Memory (MCM) effect? (Happy-Sad Inducement & story).
Mood Congruent Memory (MCM):
- Induce mood via hypnosis; happy or sad groups.
- read two stories: happy & sad content
- congruent: happy-inducement & happy story
- incongruent: happy-inducement & sad story
- Tested recall for the conent of two stories.
- The recall is best when mood & content are congrunent (match) i.e. MCM effect
What evidence did Bower provide of the Mood Dependent Memory (MDM) effect? (Happy-Sad Inducement & Learn Words).
Mood Dependent Memory (MDM):
- Similar by different from MCM paradigm:
- Induced mood
- Learn neutral words (e.g. basket) - not interested congruency.
- The induced mood at retrieval that is either the same or different to the encoding mood.
- The recall is better if the mood at retrieval matches mood at encoding
- mood acts as a cue for retrieval; mood dependent memory effect -
- STATE DEPENDENT MEMORY: emotional states serves as retrieval cues to aid memory
How doe semantic network models of emotion & memory explain negative cognitions that occur in Clinical Depression?
- Depressed mood lowers the threshold of associated concepts & events, making them more available for retrieval - affective priming
- As well as priming the negatively related stimuli.
- Positively valenced concepts and memories are inhibited - feeding a cycle of depressive thoughts.
Teachman et al. Reading:
- What are the 4 different features of automaticity?
Question from Meredith - “What evidence is there - based on Stroop & dot-probe tasks - that led the researchers to the conclusion that ADs profile of automaticity is defined by uncontrollable, unconscious, and unintentional elements, whilst MDDs profile defined by uncontrollable automaticity?
- Unconcious; lacks awareness of stimuli.
- efficient; processing stimuli requires minimal attentional resources.
- unintentional; no goal needed.
- uncontrollable; difficult to avoid or stop.
What is the difference between Mood Congruent Memory and Mood Dependent Memory effects?
MCM refers to recall being most effective when mood & content are congruent (match).
Whereas, MDM refers to recall being more effective if the mood at retrieval matches mood at encoding. i.e. STATE DEPENDENT MEMORY: emotional states serves as retrieval cues to aid memory
