STM + LTM

• STM capacity

  7 +/- 2 items - new information comes in and pushes out old (displacement) - Miller - digit span technique - reading a series of digit sets that progressively get longer - asked to immediately repeat back in right order - found could recall on average 5-9 items - concluded STM has limited capacity and new information displaces old - can be increased by chunking - items grouped together in larger chunks to remember
• STM Duration

  information can only be held in STM for limited amount of time - 18-30 seconds - can be extended through rehearsal (repeating information over and over - maintenance rehearsal) - Peterson and Peterson - 24 students listen to consonant trigram (3 letters) - immediately after heard heard 3 digit number and asked to count back in 3s - asked to recall trigram after 3,6,9,12,15 or 18 seconds - found highest level of recall was after 3 seconds (90%) - decreased as duration increased (2% recall at 18 seconds)
• STM Coding

  mainly acoustic (sound) - try to hold information in STM by saying them over and over out loud - Baddeley - found participants presented with acoustically similar list and asked for immediate recall made more errors than acoustically dissimilar list - concluded because confusion on the way words sounded - suggests STM codes information acoustically
• LTM capacity

  unlimited - forgetting due to interference or retrieval failure - anokhin - number of possible neural connections in brain is '1 followed by 10.5 million kilometres of noughts' - concluded no human brain yet exists who can use all potential of brain - LTM capacity is limitless
• LTM Duration

  unlimited - don't need to rehearse to retain - Barrack - 392 participants aged 17-74 name old classmates from high School (free recall) - given 50 photos from yearbook and asked to recognise (recognition) - found 15 years after graduation free recall 60% accurate - 48 years after dropped to 30% - recognition 15 years after 90% - 48 years 70% - suggests LTM unlimited duration if LTM can't be recalled may just need assistance to be retrieved
• LTM coding

  mainly semantic (meaning) - need to understand something to retain in LTM - Baddeley - participants shown semantically similar word list - asked for recall 20 minutes later - made more errors than when shown semantically dissimilar word list - concluded confusion of meaning of words
• MEMORY RESEARCH EVALUATION - high control over variables

  research conducted in controlled setting of laboratory allowed extraneous variables to be controlled - e.g. noise, temperature or time of day - high internal validity
• MEMORY RESEARCH EVALUATION - easy to replicate

  controlled setting of laboratory one study can be used for further replications - e.g. same word list can be used in several trials of same study - ensures findings are reliable
• MEMORY RESEARCH EVALUATION - lacks mundane realism

  materials used meaningless and don't reflect kinds of information we use STM and LTM for remembering in real world - e.g. remembering lists or numbers doesn't reflect real life memory activities - not representative of real life STM and LTM use
• MEMORY RESEARCH EVALUATION - low ecological validity

  research takes place in artificial environment of a lab so memory ability may be different from in real world - e.g. may remember more at home or at work - can't be generalised to real life settings