Memory

• Coding refers to the type of information being stored (visual, auditory or semantic).
• Research to support our understanding of the coding of short term memory.

  Baddeley found that participants with accoustically similar lists had worse immediate recall because the short term memory is accoustically encoded so the information was confused. However, participants with semantically similar lists did worse after 30 minutes because the long term memory is semantically encoded so the information confuses.
• Capacity refers to the amount of information held.
• Research to support our understanding of the capacity of short term memory.

  Miller asked participants to recall a list of numbers and increased the amount of numbers each time. He suggested that the capacity of short term memory is 7 +/- 2, so we can't hold more than 9 items in our short term memory.

  HOWEVER, Cohen found we can only remember 4 pieces of information, and Jacobs found a digit span of 9.3/7.3.
• Duration refers to the length of time information stays in a store.
• Research to support our understanding of the duration of short term memory.

  Peterson and Peterson gave participants a trigram (random letters). They were asked to recall trigram after 3,6,9,12,15,18 seconds. During countings, they were given a random number and asked to count back in 3s or 4s. After a longer period of time, ppts couldn't remember the trigram. HOWEVER, the stimuli were meaningless. Lacks external validity.
• Research to support our understanding of the duration of long term memory.

  Bahrick asked 392 participantsaged 17-74 to name 50 people from their yearbook photos and free recall people from graduating class. Photo recognition was good (90% of last 15 year grads, 70% of last 48 year grads). Photo recognition was worse (60% and 30%). Supports lifetime duration of long term memory.
• Coding, capacity and duration of sensory register.

  Coding: sense specific.

  Capacity: all sensory input.

  Duration: 1/4 to 1/2 a second.
• Coding, capacity and duration of short term memory.

  Coding: mainly auditory.

  Capacity: 7 +/- 2 items.

  Duration: 18-30 seconds.
• Coding, capacity and duration of long term memory.

  Coding: mainly semantic.

  Capacity: unlimited.

  Duration: unlimited.
• What does the multi-store model of memory suggest?

  Suggests memory is a flow of information through systems. Sensory information enters sensory register, moves to short term memory when the individual pays attention. If rehearsed sufficiently well, it passes on to long term memory. If not, memories are lost either because new info pushes them out (displacement) or they fade away (decay).
• Case study for multi-store model of memory.

  Clive Wearing - impaired STM (duration of 7 seconds) and unable to rehearse new information. Couldn't transfer new long term memories. Difficulty retrieving long term memories. Suggests STM and LTM are separate stores as they are impaired differently.

  KF - motor accident. LTM intact, STM impaired (2-3 items).

  HM - epilepsy. STM intact (30s duration), LTM damaged (couldn't move anything to LTM).
• Declarative long term memories are things you know that you know and can explain. Non-declarative memories are things you don't really know how you do it, and you can't explain how to do it.
• Three types of long term memories.

  Episodic, semantic, procedural.
• Episodic long term memories.

  Episodes of your life. Strength of the memory is dictated by your emotional response during encoding. They have a time stamp.
• Semantic long term memories.

  The things you know - facts, concepts, understandings. These can become semantic as the experience becomes more about what you have learned from the experience.
• Procedural long term memories.

  Memories of how to do things that we may struggle to explain. These are non-declarative and don't require conscious thought.
• Research to support the idea that long term memory is not a separate store.

  Clive Wearing - poor episodic memory (what his children are doing), reasonable semantic memory (meaning of words), excellent procedural memory (playing the piano). HOWEVER, issue with applying findings from a man with brain damage to the general population.

  Tulving - PET scans. When asked episodic memory, left pre-frontal cortex became active. Semantic memories, right pre-frontal cortex.
• Research to support real world applications of our understanding of long term memory.

  Belleville (2006) had older people with cognitive impairments do one of two conditions: training group to rehearse/recall episodic memories, and a control group. Those who trained had improved episodic recall, which is good because it is the most likely to deteriorate. Can be used to put remeniscence/memory boxes in care homes.
• The working memory model aims to explain short term memory as a system with multiple components.
• What are the elements of the working memory model.

  The central executive, the visuospatial sketchpad, the phonological loop, the phonological store and the episodic buffer.
• What is the central executive?

  Allocates information to subsystems. It also deals with mental functions like arithmetic and problem solving.
• What is the visuospatial sketchpad?

  The inner eye. It stores and processes information in a visual/spatial form and can be used for navigation.
• What is the phonological loop?

  Deals with spoken and written material. It is subdivided into the phonological store and the articularly control process.
• What is the phonological store?

  The inner ear. Processes speech perception and stores spoken words we hear for 1-2 seconds.
• What is the articulatory control process?

  The inner voice. Processes speech production and rehearses and stores verbal information from the phonological store.
• What is the episodic buffer?

  Acts as a backup store that communicates with both long term memory and components of the working memory model.
• Research to support the working memory model.

  KF - (Shallice & Warrington) motor injury. LTM functioned normally, but STM was impaired. Could only hold 1-2 items from a list, and he forgot letters and digits quicker when he received them auditorally than visually. Normal STM span for meaningful sounds (doorbell, phone ringing). Uneven pattern of deficits suggests evidence for separate STM stores, therefore supporting the working memory model.
• Limitations of the working memory model.

  • Most experiments are lab studies which lack ecological validity.
• Interference theory suggests we can't accurately recall information because it becomes disrupted during encoding. This happens most when information is similar.
• Proactive interference is when old information makes it difficult to recall new information.
• Retroactive interference is when new information makes it difficult to recall old information.
• Research to support retroactive interference.

  McGeogh & McDonald (1931) participants given a list of words and assigned to one of six conditions (synonyms of original list, antonyms, unrelated adjectives, nonsense syllables, 3 digit numbers and control group). Ppts in synonym condition did worse when given a second list because it confused with original list. Lab experiment so high levels of control, but low ecological validity.
• Research to support interference as an explanation for forgetting.

  Baddeley & Hitch (1977) had rugby players name teams they had played in the last number of weeks. Those who had played more games in the last 4 weeks could remember fewer teams they played than someone who played fewer games.

  • Real life situation - higher validity.
• Retrieval failure is forgetting information due to an absence of cues. Cues can either be context dependent (the external environment) or state dependent (your internal state).
• Research to support context dependent retrieval failure.

  Godden & Baddeley had 16 divers learn information either on land or underwater. They then had to recall the information either on land or underwater. Those who learned and recalled in the same context had better recall.

  • Meaningless stimuli - lacks ecological validity.
• Research to support state dependent retrieval failure.

  Overton (1972) had participants learn information either drunk or sober and then recall the information either drunk or sober. Recall was better when participants recalled the information in the same state they had learnt it in.

  • Demand characterstics if ppts knew they would have to recall information later, potentially made more effort to learn it.
  • Potentially unethical.
• Limitations of retrieval failure as an explanation for forgetting.

  • Incomplete explanation - it doesn't consider all types of memory, for example, procedural memories. We don't know if these can be forgotton or not.
• Research to support the idea of leading questions having an impact on memory.

  Loftus & Palmer had 45 American students (not representative) take part in a lab experiment. They watched 7 films of traffic accidents in a random order (no order effects). Each participant did one of 5 conditions. They were asked 'how fast were the cars going when they ******** each other?' When the verb 'contacted' was used, there was a mean estimate speed of 31.8. When the verb was 'smashed' the mean was 40.8. Suggests leading questions have an impact on memory.
• Research to support the idea of post even discussion having an impact on memory.

  Gabbert et al (2003) had 60 students and 60 older residents participate (representative). 40 were tested individually and watched a video of a girl stealing money from a wallet. The remaining 80 were tested in pairs, each person in the pair watched a different video (one where she stole, one where she didn't). 71% recalled things they hadn't seen. 60% thought she was guilty even if they watched the video where she didn't steal.

  • May have thought they were supposed to be witnessing a crime.